Precast Concrete Structures Second Edition PDF Compressed 1 299

www.EngineeringEBooksPdf.
com
Precast
Concrete
Structures
www.EngineeringEBooksPdf.com
Precast
Concrete
Structures
Boca Raton London New York
CRC Press is an imprint of the

Taylor & Francis Group, an informa business
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2017 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed on acid-free paper

Version Date: 20160405
International Standard Book Number-13: 978-1-4987-2399-2 (Hardback)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the valid-
ity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this
form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may
rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti-
lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy-
ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the
publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://
www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923,
978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging‑in‑Publication Data
Names: Elliott, Kim S.

Title: Precast concrete structures / Kim S. Elliott.
Description: Boca Raton : Taylor & Francis, CRC Press, 2017. | Includes
bibliographical references and index.
Identifiers: LCCN 2016010309 | ISBN 9781498723992
Subjects: LCSH: Precast concrete construction.
Classification: LCC TA683.7 .E43 2017 | DDC 624.1/83414--dc23
LC record available at https://lccn.loc.gov/2016010309
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface to the First Edition xiii

Preface to the Second Edition xvii
Acknowledgement xxi
About the Author xxiii
Notation xxv
1 What is precast concrete 1

1.1 Why is precast different? 1
1.2 Precast concrete structures 6
1.3 Why choose a precast structure? 12
References 19
2 Materials used in precast structures 21

2.1 Concrete 21
2.1.1 Concrete for precast production 21
2.1.2 Concrete for precast floor elements 28
2.1.3 Young’s modulus and modular ratios 31
2.1.4 Shrinkage and creep 32
2.1.5 Fire resistance and axis distance to rebars and tendons 36
2.1.5.1 Columns 37
2.1.5.2 Unbraced columns 42
2.1.5.3 Walls 42
2.1.5.4 Beams 43
2.1.5.5 Reduced additional axis distance Δa 44
2.1.5.6 Floor slabs 46
2.2 Mix design 47
2.2.1 Exposure, cover and strength 47
2.2.2 Mortars and grouts 49
2.2.3 Dry pack 50
2.2.4 Aggregates 50
2.2.5 Substitute materials 50
2.2.6 Admixtures 51
2.2.7 Water 52
vi Contents
2.3 Steel reinforcement 52

2.3.1 Rebars 52
2.3.2 Mesh 54
2.3.3 Stainless steel rebars 55
2.3.4 Prestressing tendons 55
2.4 Structural steel, welding, inserts and bolts 57
2.4.1 Structural steelwork 57
2.4.2 Welding 58
2.4.3 Cast-in fixings and lifting devices 59
2.4.4 Bolting 59
2.4.5 Cast-in sockets and anchorage 60
2.5 Non-cementitious materials 60
References 61
3 Precast frame analysis 65

3.1 Types of precast concrete structures 65
3.2 Simplified frame analysis 71
3.3 Substructuring methods 76
3.3.1 Two-dimensional plane frames 76
3.3.2 Design loads on beams and frames 79
3.3.2.1 Serviceability limit state 81
3.3.2.2 Ultimate limit state 83
3.3.2.3 Instability limit state 84
3.3.2.4 Accidental limit state 85
3.3.3 Gravity and horizontal ultimate loads on frames 85
3.3.3.1 Permanent, variable and wind actions 85
3.3.3.2 Horizontal forces due to imperfections 87
3.3.4 Beam subframe 90
3.3.5 Upper floor column subframe 92
3.3.6 Ground floor column subframe 92
3.4 Connection design 94
3.5 Stabilising methods 97
3.6 Comparison of standard designs to BS 8110 and Eurocodes 108
3.6.1 Reinforced concrete rectangular beam 108
3.6.2 Reinforced concrete rectangular column 111
3.6.3 Prestressed concrete slab 115
3A Appendix A: Summary of Eurocode EC2: Design of concrete structures
– General rules and rules for buildings, BS EN 1992, Part 1-1 120
Summary of Eurocode EC2: Design for fire BS EN 1992, Part 1-2 122
2B Appendix B: Summary of relevant items in PD 6687-1:2010 122
References 124
4 Precast concrete floors 125

4.1 Precast concrete flooring options 125
4.2 Flooring arrangements 136
Contents vii
4.2.1 Definitions of precast floor units and slab fields 136

4.2.2 Lateral load distribution in discrete precast floor units 138
4.2.3 Discritization of floor slab loads in individual units 141
4.3 Structural design of individual units 147
4.3.1 Flexural capacity 148
4.3.2 Serviceability limit state of flexure 150
4.3.3 Serviceability limit state of flexure: Calculation model 153
4.3.4 Serviceability limit state of bending 154
4.3.5 Ultimate limit state of flexure 161
4.3.6 Deflection 166
4.3.7 Ultimate shear capacity 170
4.3.7.1 Shear capacity in the uncracked region, V Rd,c 171
4.3.7.2 Shear capacity in the flexurally cracked region V Rd,cr 175
4.3.7.3 Shear capacity of hollow core units due
to combined stress in the webs 175
4.3.8 Bearing capacity 178
4.4 Design of composite floors 181
4.4.1 Precast floors with composite toppings 181
4.4.2 Flexural analysis for composite prestressed concrete elements 183
4.4.2.1 Serviceability state of stress 183
4.4.2.2 Relative shrinkage between in situ
topping to precast unit 187
4.4.2.3 Ultimate limit state of bending 188
4.4.2.4 Ultimate shear capacity of composite sections 190
4.4.3 Propping 192
4.4.3.1 Deflections in composite slabs 194
4.4.4 Interface shear stress in composite slabs 196
4.5 Calculation of crack width in prestressed concrete sections 200
4.6 Composite plank floor 202
4.6.1 Types of composite planks: reinforced and prestressed 202
4.6.2 Design for flexure during handling and in service 205
4.6.3 Shear capacity 208
4.6.4 Interface shear stress due to Stage 2 loads 213
References 214
5 Precast concrete beams 215

5.1 Types of precast beams 215
5.2 Non-composite reinforced concrete beams 222
5.2.1 Ultimate moment of resistance 223
5.2.1.1 Type I beam 223
5.2.1.2 Type II beam 225
5.2.2 Calculating and from the design moment 225
5.2.3 Ultimate shear resistance 226
5.2.4 Minimum areas and bar diameters and spacing 232
5.2.5 Section properties in cracked and uncracked sections 234
5.2.6 Boot design 236
viii Contents
5.3 Composite reinforced beams 239

5.3.1 Design in flexure of composite reinforced concrete beams 243
5.3.2 Deflections in composite r.c. beams 244
5.4 Non-composite prestressed beams 252
5.4.1 Flexural design 255
5.4.2 Ultimate flexural design 262
5.4.3 Anchorage length for ultimate limit state in beams with strands 266
5.4.4 Shear in prestressed beams 266
5.5 Composite prestressed beam design 269
5.5.1 Flexural design 270
5.6 Propping 278
5.7 Horizontal interface shear 280
5.8 Precast reinforced concrete deep beam walls 283
5.8.1 Definitions and background information 283
5.8.2 Equivalent strut-and-tie analogy 285
5.8.3 Deep beam walls with cantilevers 298
5.8.4 Bottom-loaded deep beam walls 304
5.8.5 Continuous spans in deep beam walls 308
References 310
6 Precast concrete columns 313

6.1 Precast concrete columns 313
6.1.1 Geometry, strength and general requirements 313
6.1.2 Fire requirements for reinforced concrete columns 318
6.1.2.1 Method A for first-order e fi /h ≤ 0.15 319
6.1.2.2 Method B for first-order e fi /h ≤ 0.25 319
6.1.2.3 Unbraced columns 320
6.2 Column design 321
6.2.1 Column design for factory handling 322
6.2.2 Column design for pitching and erection 324
6.2.3 Ultimate limit state design of columns 329
6.2.4 Ultimate column N–M capacity 332
6.2.5 Design rules in BS EN 1992-1-1 for columns in precast structures 336
6.2.6 Slender columns and second-order moments 337
6.2.7 Columns subjected to biaxial bending moments 340
6.2.8 Reduction factors for imposed load in several storeys 341
6.2.9 Columns in unbraced structures 346
6.2.10 Columns in partially braced structures 352
References 357
7 Shear walls 359

7.1 Horizontal and vertical stability of precast concrete structures 359
7.2 Precast concrete shear walls 359
7.3 Distribution of horizontal loading 362
7.4 Infill shear walls 374
Contents ix
7.4.1 Precast concrete infill walls 376

7.4.2 Brickwork (or blockwork) infill walls 383
7.5 Cantilever walls 388
References 394
8 Horizontal floor diaphragms 395

8.1 Introduction to floor diaphragms 395
8.2 Shear transfer mechanism 407
8.3 Edge profile and tie steel details 411
8.4 Design of floor diaphragm 413
8.4.1 Limiting stresses 413
8.4.2 Reinforcement design 416
8.5 Shear stiffness 417
8.6 Diaphragm action in composite floors with structural toppings 420
References 428
9 Joints and connections 429

9.1 Definitions 429
9.2 Basic mechanisms 431
9.3 Compression joints 433
9.3.1 Bearing in plain concrete 438
9.3.2 Concentrated loads in bearing 444
9.3.3 Reinforced and plate-reinforced concrete bearings 454
9.3.4 Bearing pads 460
9.4 Shear joints 462
9.4.1 Shear adhesion and bonding 463
9.4.2 Shear friction 463
9.4.3 Shear keys 467
9.4.4 Dowel action 469
9.4.5 Mechanical shear devices 473
9.5 Tension joints 477
9.6 Pin-jointed connections 482
9.6.1 Pinned connections between vertical and horizontal elements 484
9.6.2 Simply supported slabs on beams or walls 486
9.7 Moment-resisting connections 488
9.7.1 Design philosophy for moment-resisting connections 488
9.7.2 Structural elements in a moment-resisting connection 491
9.7.3 Floor connections at load-bearing walls 496
9.7.4 Beam-to-column face connections 498
9.7.4.1 Welded plate connector 500
9.7.4.2 Steel billet connector 502
9.7.5 Semi-rigid beam-to-column face connections 504
9.7.5.1 Theoretical derivations for the connection
flexural strength and stiffness 508
References 510
x Contents
10 Beam and column connections 513

10.1 Types of beam and column connections 513
10.2 Beam-to-column connections 519
10.2.1 Hidden connections to continuous columns – Type IA 526
10.2.2 Column insert design at Z 530
10.2.3 Additional reinforcement welded to inserts 538
10.2.4 Narrow-plate column inserts 543
10.2.5 Connections to columns using corbels 544
10.2.5.1 Mechanical connections to corbels 546
10.2.5.2 Shallow corbels 546
10.2.5.3 Deep corbels 554
10.3 Beam end shear design 556
10.3.1 Shallow recess 560
10.3.2 Deep recess 562
10.3.3 Steel shear boxes 570
10.4 Column foundation connections 575
10.4.1 Columns on base plates 577
10.4.2 Columns in pockets 584
10.4.3 Columns on grouted sleeves 593
10.4.4 Columns on steel shoes 595
References 598
11 Ties in precast concrete structures 599

11.1 Ties in precast concrete structures 599
11.2 Design for robustness and avoidance of progressive collapse 602
11.3 Categorization of buildings 607
11.4 The fully tied solution 608
11.5 Tie forces 614
11.5.1 Tie forces in codes and the National Annex 614
11.5.2 Non-linear static analysis 616
11.5.3 Horizontal floor and beam ties 618
11.5.4 Horizontal ties to columns 621
11.5.5 Vertical ties 623
References 627
12 Design exercise for 10-storey precast skeletal frame 629

12.1 Description of project 629
12.1.1 Feasibility study information 629
12.1.2 Imposed characteristic loads, densities and partial load factors 636
12.1.3 Bearings and production and construction data 636
12.1.4 Materials 636
12.2 Ultimate load combinations 637
Contents xi
12.3 Solution 637

12.3.1 Introduction 637
12.3.2 Frame stability 637
12.3.3 Sway due to imperfection at foundation 641
12.3.3.1 Frame stability in NS direction 641
12.3.3.2 Frame stability in EW direction 642
12.3.4 Sway due to imperfection at first to second floor 644
12.3.4.1 Frame stability in the NS direction 644
12.3.4.2 Frame stability in the EW direction 645
12.3.5 Sway due to imperfection above the third floor 645
12.3.5.1 Frame stability in the NS direction 645
12.3.5.2 Frame stability in the EW direction 645
12.3.6 Precast hollow core floor diaphragm 645
12.3.6.1 NS direction with diaphragm spanning EW 647
12.3.6.2 EW direction with diaphragm spanning NS 648
12.4 Floor slab design 648
12.4.1 Hollow core floors 648
12.4.2 Floor slab at cantilever 651
12.5 Edge beams 654
12.5.1 Edge beam design for column centres = 9.0 m 655
12.5.2 Edge beam design for column centres = 8.0 m 657
12.5.3 Gable edge beam design for column centres = 9.0 m 658
12.6 Internal beams 659
12.6.1 Internal beam design for column centres = 9.0 m 660
12.6.2 Composite prestressed concrete beam
for column centres = 9.0 m 661
12.6.3 Composite prestressed concrete beam for column centres = 8.0 m 663
12.6.4 Composite prestressed concrete beam
for column centres = 9.0 m 664
12.7 Column design 665
12.7.1 General information 665
12.7.2 Slenderness checks 666
12.7.3 Edge column at 9.0 m centres: fifth to
seventh and seventh to ninth floors 666
12.7.4 Edge column at 9.0 m centres: Foundation
to third floor and third to fifth floors 668
12.7.5 Design of double-sided steel billet at
edge beam–column connection 670
12.7.6 Internal column between 9.0 and 8.0 m
centres: Fourth to sixth floors 671
12.7.7 Internal column between 9.0 and 8.0 m
centres: Foundation to second floor 674
12.7.8 Design of double-sided steel billet at
internal beam–column connection 675
xii Contents
12.8 Stability ties 676

12.8.1 Horizontal floor and beam ties 676
12.8.2 Edge column ties 677
12.8.3 Vertical ties to edge and internal columns 677
12.9 Staircase 677
12.9.1 Flight 677
12.9.2 Double landing-flight 679
12.9.3 Floor landing 681
Index 683
Preface to the First Edition
In 1990, the then chairman of the British Precast Concrete Federation (BPCF), Geoff
Brigginshaw, asked me what level of teaching was carried out in British universities in pre-
cast concrete construction for multistorey buildings. The answer, of course, was very little,
and it remains that way today in spite of considerable efforts by the BPCF and sections of
the profession to broadcast the merits, and pitfalls, of precast concrete structures. Having
given lectures at about 25 UK universities in this subject, the author estimate that less than
5% of our civil/structural engineering graduates know about precast concrete and less than
this have a decent grounding in the design of precast concrete structures. Why is this so?
The precast concrete industry commands about 25% of the multistorey commercial and
domestic building market if frames, floors and cladding (facades) are all included. In higher
education (one step away from the market) precast education commands between zero and
(about) 5% of the structural engineering curriculum. This in turn represents only about 1/8
of a civil engineering course. The 5% figure claimed earlier could indeed be an overestimate.
The reasons are twofold:
1. British lecturers are holistic towards structural engineering.

2. British lecturers have no information in this subject.
This book aims to solve these suggestions simultaneously. Suggestion no. 2 is more readily
solved. This book is, unfortunately, one of the very few textbooks in this subject area aimed
at students at a level which they can assimilate in their overall structural engineering learn-
ing process. It does this by considering design at both the macro and micro levels – global
issues such as structural stability, building movement and robustness are dissected and ana-
lysed down to the level of detailed joints, localised stress concentrations and sizes of bolts
and welds.
Suggestion no. 1 is more complex. Having been acquainted with members of the FIB1
(formerly FIP2) Commission on Prefabrication, it has come to my notice the differing atti-
tudes towards the education of students in certain forms of building construction – precast
concrete being one of them (and timber another). In continental Europe, leading precast
industrialists and/or consultants hold academic posts dedicated to precast concrete con-
struction. Chairs are even sponsored in this subject. In South America, lecturers, students
and practitioners hold seminars where precast concrete is a major theme. It is not uncommon
1 FIP, Federation International de la Prefabrication; an international, but predominantly European, organisation

for the welfare and distribution of information on prefabricated concrete.
2 FIB, Federation Internationale du Beton, born from a merger of FIP with CE; an international, but predomi-
nantly European, organisation for the welfare and distribution of information on structural concrete.
xiii
xiv Preface to the First Edition
for there are as many as 10 master’s students to study this subject in a civil engineering
department. In the United States, collaborative research between consultants, precast manu-
facturers and universities is common, as the number of papers published in the PCI Journal
testifies.
The attitude in Britain is more holistic and less direct. First, basic tuition is given in solid
mechanics, structural analysis and material properties. Students are required to be capa-
ble of dealing with structural behaviour – independent of the material(s) involved. Second,
given the fundamental principles of design (and a reminder that code equations are often
simulations and their data conservative), students can assimilate any design situation, with
appropriate guidance. This may be true for structural steelwork and cast in situ concrete
structures where the designers may (if they wish) divorce themselves from the fabricator and
contractor. It is not true for precast concrete (and timber) structures where the fabricator
and site erector form part of the ‘design team’.
Precast concrete design is an iterative procedure, linking many aspects of architecture,
design, detailing, manufacture and site erection together in a 5-point lattice.
Many students will be familiar with these names, but a few will see or hear them in a
single lecture. Some of the links are quite strong. Note the central role of ‘designing’ (this
does not mean wL 2 /8, etc.) in establishing relationships with architectural requirements,
detailing components and connections and manufacturing and erecting the said compo-
nents at their connections. Could similar diagrams be drawn for structural steelwork or cast
in situ concrete structures?
Further, there are a number of secondary issues involving precast concrete construction.
Prefabrication of integrated services, automation of information, temporary stability and
safety during erection all result from the primary links.
Some of these are remote from ‘designing’. The illustration reminds us of their presence
in the total structure. The design procedure will eventually encompass all of these aspects.
This book is aimed at providing sufficient information to enable graduates to carry out
structural design operations, whilst recognising the role of the designer in precast concrete
construction. Its content is in many parts similar to but more fundamental than the author’s
book, Multistorey Precast Concrete Framed Structures (Blackwell Science 1996). The
Blackwell book assumed that the readers already have acquired a prior knowledge of the
building industry and some experiences in designing concrete structures. This book takes a
backward step to many of the design situations and does not always uphold the hypotheses
given. Reference to the Blackwell book may therefore be necessary to support some of the
design solutions.
The design examples are carried out to BS8110, and not EC2, as might be expected from a
textbook published today. The reason for this is that the clauses relevant to precast concrete
in EC2 have yet to find a permanent location. Originally Part 1.3 was dedicated to precast
concrete, but this was withdrawn and its content merged into the general code Part 1.1. For
this reason specific design data relevant to precast concrete are not available.
The author is grateful to the contributions made by the following individuals and organisa-
tions: the members of the FIB Commission on Prefabrication, in particular Arnold Van Acker
(Addtek Ltd., Belgium), Andre Cholewicki (BRI Warsaw), Bruno Della Bella (Precompressi
Centro Nord, Italy), Ruper Kromer (Betonwerk + Fertigteil-Technik (Germany), Gunner
Rise (Stranbetong, Sweden), Nordy Robbens (Echo, Belgium) and Jan Vambersky (TU Delft
& Corsmit, Netherlands); to a few organisations, namely, Trent Concrete Ltd. (United
Kingdom), Bison Ltd. (United Kingdom), SCC Ltd. (United Kingdom), Tarmac Precast
Ltd. (United Kingdom), Tarmac Topfloor Ltd. (formerly Richard Lees, United Kingdom),
Techcrete Ltd. (United Kingdom and Ireland), Composite Structures (United Kingdom),
British Precast Concrete Federation, British Cement Association and Reinforced Concrete
Preface to the First Edition xv
Council (United Kingdom), Betoni (Finland), Bevlon (Netherlands), C&CA (Australia),

Cement Manufacturers Association (southern Africa), CIDB (Singapore), Andrew Curd and
Partners (United States), Echo Prestress (South Africa), IBRACON (Brazil), Grupo Castelo
(Spain), National Precast Concrete Association (Australia), Nordimpianti Otm (Italy),
Hume Industries (Malaysia), Prestressed Concrete Institute (United States), Spaencom
D es
ign
re
te ctu
c hi
Ar
erect
Site
ion
De
tail
ing
Major links
ture
ufac
Man
Structural zones
and facades
ign
Te stab
Ar
D es
m ili
ch
po ty
ite
ra
ct
ry
ur
e
facade and
details
s
Ser vice
on
Site erecti
ng
ta ili
rie of
De
del ence
s
ive
u
Seq
re
Manufactu
Inte
gr
ser v ated
ices
xvi Preface to the First Edition
Betonfertigteile GmbH (Germany), VarioPlus (Germany), Spancrete (United Staes), AB

Stranbetong (Sweden) and Tammer Elementti Oy (Finland); to his research assistants Wahid
Omar, Ali Mahdi, Reza Adlparvar, Dennis Lam, Halil Gorgun, Kevin Paine, Aziz Arshad,
Adnan Altamimi, Basem Marmash and Marcelo Ferreira; and to his secretarial assistant
Caroline Dolby.
Preface to the Second Edition
For 14 years since the publication of the first edition of this book (April 2002), precast con-
crete elements and structures have become
(i) Taller (36-storey skeletal frame, Belgium, Figure 3) and 54-storey wall frame (the
Netherlands)
(ii) Longer (50 m long prestressed concrete beams, Figure 4)
(iii) Deeper (1000 mm deep prestressed hollow core floor units produced in Italy in 2014)
(iv) Shallower (span/depth ratio approaching 40 for prestressed composite and continuous
beams)
(v) Stronger (grade C90/105 used in columns in buildings such as in Figure 3)
Also, they have generally become more ambitious and creative in terms of integrating
structural engineering and architecture, using semi-rigid connections to enhance frame
stability in 10-storey sway frames, using high-strength hidden connections up to about
700 kN, applying automation and robotics in manufacturing, and using recycled concrete
as aggregates, PFA, GGBS, reclaimed mixing water from cement slurry and sustainable
materials. At a recent seminar chaired by myself on behalf of the fibUK Group, David Scott,
FREng, of Laing O’Rourke, made his angle on prefabrication as ‘Concrete in the Age of
Digital Enlightenment’.
These advancements have been complimented by an increase in the available literature,
not only from myself (Goodchild, Webster & Elliott 2009, Elliott & Jolly 2013) but also
from other books from Germany, Brazil, United Kingdom, and several excellent bulletins
from fib Commission 6 on Prefabrication, together with 8 European product standards
covering a wide range precast concrete elements (hollow core floor slabs, walls, stairs, etc.).
Structural design engineers have never had it so good, and yet the important knowledge
and understanding concerning the behaviour of precast concrete structures continues to
remain the same, as the author mentioned in the preface of his book Multi-Storey Precast
Concrete Framed Structures (Elliott 1996), a ‘black box’. Even our critics now acknowledge
that the realm of precast concrete and off-site manufacture of building components and
systems in steel, timber, plastics, etc., has taken major steps forwards in the past 15 years;
but there is still work to do beyond the textbook. Seminars on precast concrete are seldom
organised, and teaching in universities or technical colleges is almost non-existent (apart
from Belgium, the Netherlands, Scandinavia and odd pockets elsewhere). Whilst our sec-
ond edition of the book Multi-Storey Precast Concrete Framed Structures (Elliott & Jolly
2013) was aimed at the already accomplished structural concrete designer, this book is once
again intended for undergraduate, postgraduate and young structural engineers. Hopefully
this book will be used as an accompanying text (it will never form a core text in today’s UK
syllabus) to undergraduate courses in concrete structures. With one or two exceptions, the
xvii
xviii Preface to the Second Edition
North Galaxy Buildings, Belgium (c Ergon, Belgium).
Prestressed concrete beams 55 m long by VBI, the Netherlands.
author does not find precast concrete being taught in the United Kingdom; it’s still perceived
as a specialist building system. He is frustrated by this – he finds himself advising postgradu-
ate students to study in Delft or Eindhoven in the Netherlands or at the FEBE organised
courses in Belgium (Hasselt, Ghent, ECAM Institut Supérieur Industriel Brussels), leading to
an MSc in Precast Concrete, or even with Prof. Marcelo Ferriera in Brazil in order to gain
a sound background in this subject. My own half-day seminars to two-day workshops have
Preface to the Second Edition xix
The author’s precast concrete workshop in Seoul, South Korea, organised by Samsung Precast in 2013.
been organised worldwide (16 countries) from South Korea (Figure 5) to Malaysia, South
Africa, Brazil and Europe, but now that those events have ceased there in an even bigger
gap in the exposure to design and good practice, particularly for young/30-something
structural engineers.
This book aims to fill many of these shortcomings. It is based on the Eurocodes (EC0,
EC1, EC2 and briefly EC3) and where relevant the EC Product Standards (e.g. pulling in
of tendons and splitting stresses in hollow core slabs in EN 1168). Finding the relevant
clauses in Eurocode EC2 can be a nightmare; for example, (a) the service loading and per-
missible service stress in prestressed concrete in tension for exposure XC1 (so clearly given
in BS 8110) requires clauses 7.3.2.(4), 7.3.2(2) and Table 3.1; (b) iteration is necessary, for
example, slenderness checks on columns require the area of steel, ω in clause 5.8.3.1, to be
known before the design can commence; and (c) it requires 26 clauses and/or equations in
Chapters 3, 5 and 10 and Appendix B to determine the total losses of prestress. This book
pieces everything together, both strategically throughout the chapters and logically in the
text and via the design examples. Although the chapters follow the same sequence as in the
first edition, a new feature is introduced, that is, the complete design of a 10-storey precast
concrete framed building (Chapter 12). This focuses on stability, sway due to imperfec-
tions in the frame and floor diaphragm, load combinations and component design. It might
(if they wish) be worth to study this chapter first (without necessarily having background
knowledge on the design techniques) to appreciate the procedures that make up the entity
and to consequently work through the chapters in sequence.
Worked examples to BS8110 are not present because (i) the first edition is still available
and relevant (except γms = 1.15 and f yk = 500 [or 600] N/mm 2) and (ii) parallel EC2 versus
BS8110 examples are given in the second edition of Multi-Storey Precast Concrete Framed
Structures. Having said that, codes do not dominate precast concrete design and detailing
as much as in other building mediums – the focus is more on principles, understanding and
construction methods – and with those attributes the rest is easy.
xx Preface to the Second Edition
Goodchild, C., Webster, R. M. and Elliott, K. S. 2009. Economic Concrete Frame Elements to
Eurocode 2, The Concrete Centre, Camberley, UK, 182 p.
Elliott, K. S. and Jolly, C. K. 2013. Multi-storey precast concrete framed structures, 2nd ed., John Wiley,
London, 750 p.
Acknowledgement
As before the author is grateful to the contributions and assistance made by the following
individuals and organisations (in addition to those named in the first edition): to the members
of the fib Commission on Prefabrication, in particular Arnold Van Acker (Belgium), Andre
Cholewicki (Warsaw, Poland), Barry Crisp (Australia), Bruno Della Bella (Precompressi
Centro Nord, Italy), Iria Donaik (Cassol Pre-Fabricados, Brazil), Bjorn Engström (Chalmers
University, Sweden), Marcelo Ferreira (UFSCAR, University Sao Carlos, Brazil), George
Jones (CDC Ltd., United Kingdom and Ireland), Holgar Karutz (Concrete Plant International,
Germany), Stef Maas (FEBE, Belgium), Alessandra Ronchetti (Secretary, Italy) and Jan
Vambersky, (Netherlands); to a few organisations, namely, Bison Manufacturing Ltd. (United
Kingdom), British Precast Concrete Federation (United Kingdom), Reinforced Concrete
Council/Concrete Centre/MPA (United Kingdom) and Prestressed Concrete Institute
(United States); to some helpful people specifically Stephen McAvoy (Acheson & Glover
Ltd., Northern Ireland, United Kingdom), Simon Bullivant and Richard Morton (Bullivant
Taranto Ltd., Northern Ireland, United Kingdom), Mark Magill, Simon Kelly and Seamus
McKeague (Creagh Concrete Ltd., Northern Ireland, United Kingdom), Zuhairi Abd Hamid
and Ahmad Hazim Abdul Rahim (CREAM, Malaysia), Jon Pendleton (Coltman Precast,
United Kingdom), Kamel Bensalem (Charcon Construction Solutions), Tang Lai Peng (Eastern
Pretech, Malaysia), Daniel Petrov (Echo Prestress, South Africa), Cliff Billington (Invisible
Connections, United Kingdom), Charles Goodchild (MPA/Concrete Centre, United Kingdom),
Khoo Tian (O-Stable Panel, Malaysia), George McAllister and Mark Maloney (Preconco,
Barbados), Jureik Im and Yongnam Kim (Samsung C&T Corp., South Korea), Mounir El
Debs (University of Sao Paulo, Brazil), Haraldo (T&A Pre-Fabricada, Brazil) and Colin Jolly
(co-author, United Kingdom); to his colleagues in the Civil Engineering Department at the
University of Nottingham’s campuses in Nottingham and Malaysia, in particular Gwynne
Davies (retired) and John Owen and technicians Nigel Rook, Bal Loyla and Mike Langford;
and to his research assistants Ali Mahdi (United Kingdom), Reza Adlparvar (Tehran, Iran),
Sarakot Hasan (Iraq), Dennis Lam (University of Bradford, United Kingdom), Halil Gorgun
(University of Dicle, Turkey), Kevin Paine (University of Bath, United Kingdom), Ho Lee
(Sangju National University, South Korea), Jong-Young Song (Samsung C&T Corp., South
Korea), Kamaluddin Rashid and Aziz Arshad (JKR, Malaysia), Farah Abdul Aziz (UPM,
Malaysia) and Izni Ibrahim and Roslli Noor (UTM, Malaysia).
This book is dedicated to the late Dato Dr. Wahid Omar, my first PhD student, the former
Director General of JKR (Public Works Dept., Kuala Lumpur) and my friend in the United
Kingdom and Malaysia for 28 years.
Permission to reproduce extracts from British Standards is granted by BSI. British
Standards can be obtained in PDF or hard copy formats from the BSI online shop:
www.bsigroup.com/Shop or by contacting BSI Customer Services for hardcopies only:
Tel: +44 (0)20 8996 9001, Email: cservices@bsigroup.com.
xxi
About the Author
Dr. Kim S. Elliott is a consultant to the precast industry in the United Kingdom and Malaysia.
He was senior lecturer in the School of Civil Engineering at Nottingham University, United
Kingdom, from 1987 to 2010 and was formerly at Trent Concrete Structures Ltd., United
Kingdom. He is a member of fib Commission 6 on Prefabrication where he has made con-
tributions to six manuals and technical bulletins, and he is the author of the book Multi-
Storey Precast Concrete Framed Structures (1996, 2013) and co-authored the Concrete
Centre’s Economic Concrete Frame Manual (2009). He was chairman of the European
research project COST C1 on Semi-Rigid Connection in Precast Structures (1992–1999).
He has lectured on precast concrete structures 45 times in 16 countries worldwide (includ-
ing Malaysia, Singapore, South Korea, Brazil, South Africa, Barbados Austria, Poland,
Portugal, Spain, Scandinavia and Australia) and at 30 UK universities.
xxiii
Notation
a Axis distance to bar; distance to point load; lever arm from topping to precast
centroids; axial load factor in biaxial column design; distance to bracing ele-
ment from shear centre; length of bearing ledge in corbels; vertical deflection
due to elongation of catenary tie
aʹ Distance to point load in cantilever
ab Edge distance or half of centre-to-centre distance between bars
ac Distance to forces from centre line of column; distance to load from column
face in corbels
aC Δcrit Critical value of deflection in catenary action under accidental loading
ae Effective distance to load from main rebars in corbels and connections
aeff Required structural bearing length a in corbels
amean am Mean axis distance of a group of bars
asd asd,m Side axis distance to bars or groups of bars
av Distance to load from face of column in corbels
a1 Net (structural) bearing length
a2 a 3 Ineffective bearing lengths or permitted deviations at supports
αI αII Deformation parameters in serviceability design
b b1 b2 Breadth, in stage 1 and 2 loading
bc Total breadth of cores (in hollow core slab)
be Breadth of precast floor unit
beff Breadth of in situ topping
bf Breadth of flanges in beams or slabs
bi Interface contact length between beams and slabs or infill
bl Bearing length of localised support (e.g. pad, plate)
bmin Breadth of column and beam (in fire)
bp Breadth of bearing; breadth of localised support (e.g. pad, plate); bearing
breadth of the infill at the top of column pockets
bt Mean width of section in the tension zone
bv bw Breath of upstand in beam
bw Breath of web(s)
b1 Effective breadth of bearing
b1 b2 Depth of bays in horizontal floor diaphragm
c Cohesion factor in interface shear stress
c cov Cover to bars
cd Edge distance to rebars or gap between rebars
c1 c2 Column depths
xxv
xxvi Notation
d Effective depth to tension bars from compression face; net depth of floor slab in
horizontal floor diaphragm; root depth of shear key; distance to holding down
bolts in tension from compression face
dʹ Effective depth to compression bars; depth to the centre of the bottom bars in
deep beam wall; distance to holding down bolts from compression edge of plate
dʺ Effective depth to links (in beam boot or nibs)
dc Effective depth from top of composite section
dct Net depth to dowel bar in shear
deff Effective height of bottom flange of beam
df Effective depth from column pocket to the edge of the foundation
dh Effective depth to tension bars in nibs
dn Depth to centroid of concrete in compression
dt Distance from the top of the insert to the uppermost link in connections
dT Effective depth to tendons in tension (ultimate limit state)
d1 d2 Effective depth in stage 1 and 2 loading
d2 Axis distance to tension bars from tension face in columns; axis distance to
starter bars in base plates
d2 d 3 Ineffective bearing lengths (e.g. in nibs and corbels)
e Eccentricity to centre of pressure or load from shear centre; eccentricity due to
ultimate loads = M Ed/N Ed
eʹ Net eccentricity of upper and floor loads of deep beam wall
eb eh Eccentricity of point loads acting over breadth and depth of section
e fi Eccentricity due to fire loads = M0Ed,fi/N0Ed,fi
ei Eccentricity due to imperfection
enet Net eccentricity in column due to eccentric loads
eo Eccentricity due frame or floor effects; minimum eccentricity
etot Total eccentricity eo + ei
e y ez Eccentricity of load due to moments about y- and z-axes in columns
enib Eccentricity of load due to load acting on nib of deep beam wall
eup Eccentricity of load due to upper storey deep beam wall
ew e due to wind pressure alone
e2 e2,mean Eccentricity due to second-order deflections in vertical elements; mean value
for multiple columns
f Horizontal shear force per unit length
f b fm Strength of brick element and mortar
fb General term for final stress in bottom fibre of prestressed section
f b1 f t1 f b2 f t2 Bottom and top fibre stress due to stage 1 and 2 bending moments
f t2ʹ Top of topping stress due to stage 2 bending moment
f bd f b,req Design and required concrete bond strength
f bed Design strength of bedding material
fc Mean cylinder strength of concrete required for lifting
fcd fcdi Design strength of concrete; in situ/mortar bedding
fcd,pl Design strength of plain concrete
fci Design stress in infill beneath base plate
fck Characteristic cylinder strength of concrete
fcki fck of in situ or infill concrete
fck(ti) fck(ti) Value of fck at time t and at installation ti
fcm fcm,cube Mean value of fck and fck,cube
fcm(t) Value of fcm after time t
fctd fctdi Design strength of concrete and in situ concrete in tension = fctm/γm
Notation xxvii
fctd(t) Value of fctd at transfer or at time t

fct,eff Value of fctm in crack width calculations
fctk,0.05 The (lowest) 5% fractile value of fctm
fctm fctmi Mean strength of concrete and in situ in tension
fcu fck,cube Characteristic cube strength of concrete
f Ed Ultimate design bearing stress (or σEd)
f Edw Ultimate design stress in welds
fk Characteristic strength of brickwork in compression
fneoprene Design bearing strength of neoprene
f p f p,max Final stress in tendons (ultimate limit state); maximum ultimate stress in
tendons
f LOP f pʹ Stress in tendons at εLOP; at intercept in σ-ε diagram
f pbt Bond stress at transfer
f pd Design strength of prestressing tendon = f pk/γm
f pk Characteristic strength of prestressing tendon
f p0,1k 0.1% proof strength of prestressing tendon
f Rd f Rdu Ultimate bearing stress; for partially loaded area
fs As-measured yield strength of rebar; service stress in rebars
fsc Design stress in column bars, yield or elastic
fu Ultimate tensile strength of hot rolled structural steel
fub Ultimate strength of bolts
fvk Ultimate shear strength of brickwork
fy Yield strength of hot rolled structural steel
f yd Design strength of steel rebar = f yk/γm
f yk Characteristic yield strength of steel rebar
f ykh Value of f yk of tie bars welded to bearing plates
f yk,b Buckling strength of steel rebar
f ywk Value of f yk of stirrups
fu Ultimate tensile strength of steel sections
fub fuw Ultimate tensile strength of bolts and dowels and welds
fy Yield strength of steel sections (see pyd for design values of sections, bolts, etc.)
f ybk f yw Yield strength of bolts and dowels and welds
g Gap (between ends of elements)
gk Pressure of permanent (dead) load per unit area
h Storey height; height of deep beam wall; depth of section; depth of slab
ha Equivalent height of deep beam wall
hagg Maximum nominal size of coarse aggregate
heff Member thickness or depth (in fire)
hft hfb Depth of top and bottom flanges
hi Pressure due to imperfection force; effective height of unbraced column
hl Overall depth of lattices in half slab
ho Notional depth or thickness of element
hs Slab depth in shrinkage effects calculation; depth of upstand in beams
hy hc Depth of face and total depth of corbel
hʹ Height of infill wall (lw in BS EN 1992-1)
h1 Minimum length of shear key
h1 hs Slab thickness in fire calculation
i Radius of gyration of section
k1 k 2 Column end ‘flexibility’ factors (ratio of column/beam stiffness); factors in
moment redistribution ratio
xxviii Notation
k Depth factor for shear strength; moment distribution factor; strength factors in
brick infill wall design
kc k Factors for minimum area of reinforcement
kn Size coefficient for shrinkage
kp(θcr) Stress ratio in prestressing tendons in fire
kT Limit for prestress ratio after initial losses
l Façade length of building; span; effective span for l/d ratio; clear height of verti-
cal element, length of bracing element; depth of pocket in column foundation
lb Bearing length (parallel with the span of the element) (termed a in the code)
lbd lb,rqd Design anchorage (bond) length; structural anchorage length required
le Characteristic length (of beam) of uniform stress
leff Effective span
ln Clear distance between the faces of the supports; actual length of beam
lo Effective length of lattice bars in half slab; clear span of deep beam wall
lp Length of the overlap in lapped bars or hooked bars; length of beam of con-
stant curvature
lpt Transmission length; lap length for strand in accidental loading
lpt2 Design value of lpt at ultimate limit state
lr Distance between the column or wall ties
ls Elastic elongation length of tie bar; straight part of a bar in anchorage
ltd Length of spot welded transverse bar
lw Shear span; bearing width (perpendicular with the span of the element) (called
b1 in the code); length of weld (not leg length)
lx Distance to shear plane from end of prestressed element; floor to ceiling height
lʹ Cantilever span
l0 Effective length or height of vertical element
l0 , fi l0 in fire situation
l0t Distance l0 between torsion restraints
m Bending moment; modular ratio (long term); number of the elements contribut-
ing to imperfections; distance from holding down bolts to starter bars in base
plates
mh Horizontal bending moment in floor plate
m(t) Value of modular ratio m at time t
m1 m2 Long-term modular ratio for beam and in situ infill in two-stage design
n Normalised column axial load factor; number of storeys above a loaded ele-
ment; number of bars in cantilever wall
nbal Balanced stress ratio in column design
no Number of storeys including basements
nu Ultimate axial stress ratio in column design
p Line pressure in billet connections
pbq Shear strength of bolts and dowels
pd Accidental load per area or unit length
pybd pywd Design strengths of bolts and dowels and welds
pyk(20°C) Strength of prestressing tendons at ambient temperature
q Mean wind pressure; line load due to wind plus imperfection
qEd Variable (live) load per unit length
qk Pressure of variable (live) load per unit area
q1 q3 Perimeter and internal floor tie forces
r rb Radius of curvature in second-order and deflection calculations
ri ro Internal and external (outside) bend radius of bars
Notation xxix
rm Column end moment ratio = M01/M02

rsup γsup Partial safety factor for variations in prestressing
s Cement factor; distance between bars or links; distance of load F from the
nearest lateral support in hollow core slabs; spacing of floor ties
sr,max Crack spacing
t Time; thickness; depth of topping; thickness of wall (hw in BS EN 1992-1-1);
total breadth of the in situ concrete side walls of column pockets
te Equivalent depth or thickness in fire of hollow core slabs
to Time when load is first applied
ts Time to removal from mould (or detensioning for prestress)
tw Throat thickness of weld
t 0T Equivalent age at transfer for creep
u Perimeter of element; edge distance to partial area or point load
v Shear force per unit length; shear deflections
v v1 Strength reduction factor for concrete cracked in shear
vʹ Concrete strength reduction factor
vEd vEd1 Shear stress, due to stage 1 loads
vEdi Design interface shear stress in horizontal floor diaphragm
vhx Shear stress in joints in horizontal floor diaphragm
v min Minimum ultimate shear resistance
vRdi Interface shear resistance in horizontal floor diaphragm
vRv Interface shear resistance of a smooth surface
w Width of the compressive strut
wb wt w1 Width of compressive struts in deep beam wall
wEd Ultimate uniformly distributed load per unit length
wEd,max wEd,min Maximum and minimum values of wEd for patch loading
wh w u Self-weight precast plank, upper precast plank and topping, in half-slab
design per unit length
wk Pressure of wind load per unit area; crack width
w max Limiting crack width
ws ws1 ws2 Service loads, stage 1 and imposed stage 2
w0 Self-weight of element per unit length
wʹ Diagonal length of infill wall
x Distance between props; distance from face of column to centre of load
x X x1 x2 Depth to neutral axis, in stage 1 and 2 loading
xu xc Depth to centroidal axis, flexurally uncracked and cracked sections
xpc1 Depth to centroidal axis in partially cracked beam section
Distance to the centroid of the areas
xʹ yʹ Distances to the centroid of stiffness
y Distances between bracing elements parallel to loading
yb y t Distances to centroid of section from bottom to top
yb,c Distance to centroid of composite section from bottom
yb,co yt,co Compound values of yb yt
ys Height to centroid of all tendons
yT Height to centroid of tendons in tension zone
z z1 z 2 Lever arm, in stage 1 and 2 loading
zcp Eccentricity of prestressing force
zh Lever arm between lattices in half slab
zo Lever arm in corbels
Ac Cross-sectional area
xxx Notation
Ac,co Compound value of Ac

Acc Cross-sectional area of precast and in situ topping
Acc,co Compound value of Acc
Ac,eff Concrete area in tension zone
Aci Cross-sectional area of in situ concrete
Ac(y) Concrete section area above height y from bottom of section
Ac1 Ac0 Partially loaded resistance and loaded areas
Ad Design value of accidental load; area of diagonal bars in half joints
Agt Percentage total elongation in rebars at maximum force
Aj Area of interface joint
Ap Ap1 Ap2 Area of prestressing tendons, in stage 1 and 2 loading
ApT Area of prestressing tendons in tension (ultimate limit state)
A s A s1 A s2 Area of rebars in tension, in stage 1 and 2 loading
A sʹ Area of rebars in compression
A sc Compressive strut reinforcement in deep corbels
A sd Area of dowels to resist horizontal force
A s,db,min Minimum area of reinforcement in deep beam wall
A s,dist Area of transverse (distribution) steel
A sh Area of rebars to resist horizontal force
A shd A shd1 A shd2 Area of rebars or strands in horizontal floor diaphragm, due to bending
moments and shear forces
A svh Additional area to A shd in horizontal floor diaphragm
A s,main A s,link Area of main bars and links in corbels
A si Area of interface shear dowels, stirrups; area of row i bars
A sj Area of coupling bars in horizontal floor diaphragm
A s,l Area of lattice bars in half slab
A s,min Minimum area of reinforcement in tension or compression required
A s,max Maximum area of reinforcement allowed
A st Area of transverse bars (lacer bars)
A sv Area of vertical ties
A sw Area of shear stirrups
Av A l Area of reinforcing loops and lacer bars between infill wall and column
B Depth of slab in horizontal floor diaphragm
C Column curvature factor; compressive force
C min Minimum cover for durability
C nom Nominal cover for durability
Cpt(y) Factor according to position of tendons (V Rd,c(y) calculation)
CRd Compressive strut resistance
CRd,c Shear stress constant
D Depth of precast slabs in horizontal floor diaphragm
Ecm Concrete Young’s modulus (secant value)
Ecm(t) Ecm at time t
Ecmi Ecm of in situ or infill concrete
Ecm,long Ec,ef Long-term value of Ecm
Ec,ef1 Ec,ef2 Values of Ec,ef for stage 1 and 2 design
E Ed,fi Fire action (load)
ERd,t,fi Fire resistance
Ei Young’s modulus of infill wall
Ep Young’s modulus for prestressing strand or wire
Epʹ Young’s modulus after εLOP
Notation xxxi
Es Young’s modulus for steel or rebars

F Force; horizontal shear force in columns
Fbst Ultimate bursting force (EC2 uses T)
Fbt Ultimate force in single bar at bends
Fbtd End bearing capacity of a spot welded transverse bar of length ltd
FEd Ultimate end reaction; ultimate forces above and below inserts; ultimate
horizontal force in column pockets
FEdh Ultimate horizontal force in bottom reinforcement of connections
F Linear load in hollow core floor slab distribution; force in holding down bolts
Fc Ultimate force in concrete in compression
FcR Ultimate compressive capacity of strut in corbels
Fc,t Horizontal interface shear force due to stage 2 loads
Fd Ultimate force in diagonal bars in half joints
Fq Ultimate shear resistance of dowels or bars
FR Three-line edge support for hollow core floor slab load F
FRd,t Ultimate tensile capacity of top bars in corbels
Fs Ultimate force in bars or tendons; tension force in column bars
Fsc Ultimate compressive force in column bars
Fs,db Ultimate force due to A s,db in deep beam wall
Fshʹ Restoring force due to restraint of free shrinkage
Ft Ftʹ Basic and modified stability (catenary) tie force
Ft,beam Catenary tie force collected at or in beams; notional force in gable edge
beams
Ft,col Stability tie force in columns
Fwd Ultimate tensile capacity of the spot weld; tie force in corbels
Fy Ultimate yield force in column bars
F1 F 2 F3 Ultimate reactions in column pockets
G′ Effective shear modulus of horizontal floor diaphragm
Gk Characteristic permanent (dead) load
H Accidental load; horizontal tie force; total height of building; height of col-
umn; reaction forces in bracing elements
Hcrit Critical height for unbraced skeletal frames
Hi Horizontal force due to imperfection; sum of horizontal loads due to wind
and imperfection
H Rd Horizontal tie steel capacity
H Rv Horizontal resistance of infill wall
I Ic Ix-x Second moment of area of section, concrete and about axis x–x
Ic,co Compound values of Ic
Iu Icr Ief Uncracked, cracked and effective values of I
K Bending moment factor M Ed /fck bd2
K1 K1w Value of K for stage 1 and for breadth of webs in stage 1
Kʹ Limiting value of K
K Curvature factor; span/depth factor
Kr K φ Column curvature factors for axial load and creep
K1 K 2 Kef Uncracked, cracked and effective flexural rigidity
Ks Shear stiffness of longitudinal joint in horizontal floor diaphragm; nor-
malised stiffness of connection to pin-ended beam
L Effective span (for deflection calculations); overall length of element; over-
hang of base plate from edge of column
Lb Depth of bearing plate (deep beam wall); bearing length
xxxii Notation
Le Effective diagonal length of infill wall

Ls Characteristic length of the tie bar in horizontal floor diaphragm
L sb Nodal distance from beam end connections
Lt Width of bearing plate (deep beam wall)
Lʹ Length of infill wall (b in BS EN 1992-1)
L2 L3 L4 Length of pressure zones in insert/billet connections
M Bending moment
Mcol Net ultimate moment at column node due to patch loading
Mcr Mcr,c Cracking moment of resistance; for composite section
ME Design bending moment in semi-rigid connection
M ER Required moment capacity in semi-rigid connection
M Ed Ultimate design moment
M Ed1 M Ed2 Value of M Ed for stage 1 and 2 loading (composite design)
M Ed,h Value of M Ed due to self-weight of precast plank
MwEd Value of M Ed due to wind loading
M Ed,fi M0Ed,fi Design moment in fire situation
M Ed,h M Ed in horizontal floor diaphragm
Mh Bending moment in horizontal floor diaphragm
Mi Overturning ultimate moment due to imperfections
Mk Applied bending moment at serviceability
M max M min Maximum and minimum column moments at a node
Mnet Net column moment Mmax - Mmin
Mprop Service moment at position of props
M RC Moment of resistance in semi-rigid connection
M Rd Ultimate moment of resistance
M Rd,c M Rd for composite section
M Rd,fi Ultimate moment of resistance in fire situation
M Rd2 M Rd3 M Rd for stages 2 and 3
M Rdy M Rdz Values of M Rd in y and z planes in biaxial column design
M Rd,r Increased value of M Rd with additional bars in connections
Ms Service moment
M s1 M s2 M s3 Value of M s for stages 1, 2 and 3 loading (composite design)
M sw M self Service moment due to self-weight
M sʹ M s due to frequent load combination
M sR M sR2 Serviceability moment of resistance, in stage 2
M s,QP Service moment due to quasi-permanent combination of loads
Mu Ultimate failure bending moment in experimental test
Mwk Overturning moment due to wind load
M zz Maximum ultimate moment in insert connector design
M0Ed First-order value of M Ed due to frame action
M01 M02 First-order column end moments due to frame action
M2 Second-order bending moment
N Compressive force; number of tendons; number of props
Nb N a Ultimate design axial force below and above a certain floor level
N Ed Ultimate design compressive or axial force
N Ed,fi N0Ed,fi Design axial load in fire situation
N Rd Ultimate compressive or axial resistance
NT Number of tendons in tension (ultimate limit state)
P Prestressing forces; characteristic point load in accidental loading
Notation xxxiii
PEd Single ultimate point load in deep beam wall; precompression from external
sources
Ppi Initial prestressing force
Pr Prestressing force at release
Ppm0 Prestressing force after initial loses
Ppmi Prestressing force at installation
Ppo Final prestressing force after all loses (at transfer)
PG Point dead load
Pt (lx) Prestressing force at distance lx from end of element
Qk Characteristic variable (live) load
R Reaction (force); diagonal force in bracing elements; prop reaction
R Ed R1 Support reaction at columns in deep beam wall
R Ed Ultimate diagonal force in infill wall
Re Yield strength of reinforcing bars
Rm Tensile strength of reinforcing bars
Rtr Ratio of prestressing force after initial losses (at transfer)
Rwk Ratio of prestressing force after final losses (in service)
Rv Compressive diagonal resistance of infill wall
Rvc Rvs Compressive diagonal and shear resistance of brick infill wall
S Shape factor for bearing pads
SE Rotational stiffness in semi-rigid connections
Sq Confined stress enhancement factor
S Sx-x First moment of area of section, about centroidal axis
Sy First moment of area above y from bottom of section
T Mean curing temperature; vertical tie force; bursting force
T Ed Axial tensile force in steel plate; tension capacity of steel column shoe; cat-
enary force under accidental loading
Ti Tp Internal and perimeter tie forces
Th Tb Tq Total chord tie force, due to bending moments and shear force in horizontal
floor diaphragm
T Rd Vertical tie steel capacity; catenary tie capacity
V Shear force
Vbeam Ultimate shear capacity of beam connector in infill wall design
Vdt Value of V Rd,d without partial safety factors (from tests)
V Ed Ultimate design shear force
V Ed,d V Ed in dowel bar
V Ed,h Ultimate design V Ed in horizontal floor diaphragm
V Ed,t V Ed,b Ultimate top and bottom loaded shear forces in deep beam wall
Vh Vhx Vhy Horizontal shear force in floor diaphragm, in the direction perpendicular and
parallel to horizontal load
V Rd Ultimate shear resistance
V Rd,c V Rd due to flexurally uncracked concrete section
V Rd,cr V Rd due to flexurally cracked section
V Rd,c(y) V Rd,c at distance lx from the end of the unit and at height y
V Rd,cc V Rd,crc Values of V Rd,c and V Rd,cr for composite section
V Rd,d V Rd of dowel bar
V Rd,r Increased value of V Rd with additional bars in connections
V Rd,s V Rd due to steel stirrups
V Rd,s,l V Rd,s due to lattices in half slab
xxxiv Notation
V Rds,b V Rd due to bottom steel stirrups in deep beam wall

W Elastic section modulus at tension face at serviceability; width of the hollow
core unit in horizontal floor diaphragm
W Ed Ultimate load from floor loads acting on deep beam wall
Wext Wint External work and internal deformation energy in catenary tie
Wk Characteristic wind load
Wpl Plastic modulus for steel sections
X Depth to neutral axis; depth of stress block ratio in base plate design
Distance to shear centre of bracing system
Yc Height from bottom to position to calculate V Rd,c(y)
Z b Zt Section modulus at bottom and top fibre
Zb2 Zb3 Section modulus at bottom fibre for composite stages 2 and 3 (ditto Zt)
Zz Section modulus at centroid of tendons
Zb,co Zt,co Compound values of Zb Zt
Compound values of Zt at top of the topping
α Modular ratio (short term) Es/Ecm , Ecmi /Ecm; rebar buckling parameter; angle of
compressive strut; contact length between the wall and the column
αk Lateral load distribution factors in hollow core slabs
αβ Inclination of diagonals in lattices along and perpendicular to span
αA Floor area imposed live load reduction factor
αcc Concrete strength factor
αcw Axial stress parameter for shear stress
αds1 αds2 Coefficients of cement for shrinkage strains
αh α m Factors for column height and number m contributing to imperfection
αe Modular ratio (long term) E s /Ec,ef
αl Ratio lx /lpt2
αn Imposed live load reduction factor
α2 Transmission length factor for tendons; modular ratio (long term) in stage 2
design
α1 α2 Anchorage length parameters
α1 α2 α3 Coefficients of concrete for creep
β Column or wall effective length or height factor; factor for short- or long-term
loading
βθ Inclination of compressive strut and soffit angle in corbels
βγ Inclinations of the holding bars to anchor plates
β(fcm) Strength factor for creep
βas Coefficient for autogenous shrinkage strain
βds (t,ts) Age factor for shrinkage
βRH Relative humidity factor for shrinkage
β(ti) Age at installation loading factor for creep
β (t 0) Age at release loading factor for creep
χ Stress reduction factor
δ Deflections
δs Longitudinal slip between two adjacent units in horizontal floor diaphragm
δt δt,max Transverse crack width in horizontal floor diaphragm, maximum value
δti Initial crack width between floor units in horizontal floor diaphragm
δu δy Deformation, deflection at ultimate and yielding
ε Strain
εcu εcu3 Ultimate crushing strain in concrete
εca Autogenous shrinkage strain
Notation xxxv
εcd Drying shrinkage strain

εcd,o Basic drying shrinkage strain
εcs Shrinkage strain
εLOP Strain in tendons at limit of proportionality (elastic limit)
εp Final strain in tendons
εpo Prestrain in tendons after losses
Relative shrinkage between precast and in situ concretes
εsm εcm Concrete strains in crack width calculation
εud Limiting strain for rebars
εuk Elongation of rebars at the breaking load
εs εs1 εs2 Elastic strains in reinforcing bars; in layers 1 and 2
ϕ Tendon diameter (strand, wire); mean tendon diameter; rotations between ele-
ments in connections, inclination of surface in sawtooth model for shear
ϕs,l Lattice bar diameter in half slab
γ Fixity factor in semi-rigid connections; partial safety factor (general)
γG Partial safety factors for dead load
γG,sup γG,inf Superior and inferior values of γG
γQ Partial safety factors for live load
γm Material partial safety factor (general)
γc Material partial safety factor for concrete
γmb γmv Partial safety factor for brickwork in compression and shear
γM Partial safety factor for steel sections
γMo Partial safety factor for lattice bars in compression
γM2 Partial safety factor for welds
γP Partial safety factor for prestressing action
γp,fav Favourable value of γP for effect of prestressing
γs Material partial safety factor for steel bars
γSH Partial safety factor for concrete shrinkage
γW Partial safety factors for wind load
η Factor (for stress block); initial factor for prestressing
n fi Axial load ratio if fire design
η1 η2 Casting condition and bar diameter parameters for bond
φ Tendon diameter (wire, strand); creep coefficient (general)
φef Effective creep factor (column design)
φRH Creep coefficient for relative humidity
φ(t,ti) Creep coefficient
λ Slenderness ratio; relative stiffness ratio between infill wall and frame
λcrit Critical slenderness limit in bars in compression
λlim Limiting value of λ
λfi Slenderness in fire situation = l0,fi /i
λy λ z Value of λ in y and z planes in biaxial column design
λʹ Non-dimensional slenderness = λ/λcrit
μ Coefficient of friction; ratio of initial prestress
μfi Axial load ratio in fire situation = N Ed,fi /N Rd
μʹ Effective coefficient of shear friction and wedging combined
ν Poisson’s ratio
θ Temperature; angle of rotation; beam end rotation; inclination of strut in shear
design; angle of slope of infill wall; angle of taper in column pocket
θcr Critical temperature for rebars and prestressing tendons
θi Inclination of vertical element or frame due to imperfection
xxxvi Notation
ρ Reinforcement ratio A s /Ac

ρp,eff Value of ρ in crack width calculation = Ap /Ac,eff
ρ1 Reinforcement ratio for bars extending beyond shear plane
ρo Concrete strength factor
ρ1000 Tendon relation loss at 1000 hours
σ Stress; prestress
σb σt Final prestress in concrete at bottom and top fibres
σb(t) σt(t) Prestress in concrete at transfer at bottom and top fibres
σc Concrete stress at centre of tendons after relaxation loss
σcm Ultimate bearing stress of concrete at anchorage points
σcp Prestress at centroidal axis after all losses (include γp,fav)
σcp(y) Value of σcp at height y from bottom of section
σc,p Limiting compressive stress in prestressed concrete
σct,p Limiting tensile stress in prestressed concrete for XC1 exposure = fctm
σc0 Limiting unconfined compressive stress in concrete
σEd Ultimate bearing stress
σn Normal stress
σpi Initial prestress in tendons
σpd Ultimate stress in tendons (used in bond design)
σpm0 Prestress in tendons after initial losses (at transfer)
σpo Final prestress in tendons after all losses (in service) (Note EC2 uses σp∞)
σr Prestress at release
σRd,max Limiting compressive strength or strut strength of concrete
σs Stress in rebars; stress variation in tendons after decompression in crack width
calculations
Free shrinkage stress due to
σtd Concrete bearing stress at spot welded transverse bar
τ Shear stress; elastic shear stress distribution (shape) function
τcp(y) Concrete shear stress at height y from bottom of section
ω Reinforcement ratio
ξ Loss ratio after initial losses
ψi ψ28 Creep coefficient for deflections at t or 28 days
ψ0 Characteristic combination load factor
ψ1 Frequent combination load factor
ψ2 Quasi-permanent combination load factor
ψ∞ Long-term creep coefficient
ζ Characteristic combination dead load factor; bursting force coefficient; ratio of
solid material to the whole of voided sections; tensioning stiffening distribution
coefficient
X Concrete ageing coefficient
Δ Total tolerances to allow for manufacturing and erection errors
Δa Reduced additional axis distance to tendons in fire
Δcrit aC Critical value of deflection in catenary action under accidental loading
Δσpr Prestress loss due to initial relaxation of tendons
Δσel Prestress loss due to elastic shortening
Δσp,c Initial prestress losses Δσpr + Δσel
Δσpci Prestress loss due to creep before installation
Δσpc Prestress loss due to creep after installation
Δσp,s Prestress loss due to shrinkage
Notation xxxvii
Δσpr Stress due to final relaxation of tendons (note same notation as for initial relax
loss)
Δσp,r Prestress loss due to final relaxation of tendons
Δt Deformation of bearing pad
Δti Curing time
ΔC dev Dimensional deviation for cover to bars
Δa2 Δa 3 Ineffective bearing lengths or permitted deviations at supports
Φ Reinforcing bar diameter; rebar buckling parameter; slenderness load reduc-
tion factor
Ψ Exponential creep growth factor
Abbreviations and Other Nomenclature

Exp. Expression or equation from code of practice
Eq. Equation in this book
CEM Types of cements
DBW Deep beam wall
fi Fire situation (subscript)
i In situ infill concrete, mortar or grout (subscript)
hcu Hollow core unit
M-θ Moment versus beam end rotation
MRC Moment resisting connection
PSF Partial safety factor
RHs Relative humidity (%)
Chapter 1
What makes precast concrete different from other forms of concrete construction? Whether
concrete is precast, that is statically reinforced or pretensioned (prestressed), is not always
apparent. It is only when we consider the role concrete will play in developing structural
characteristics that its precast nature becomes significant. The most obvious definition for
precast concrete is that it is concrete which has been prepared for casting, cast and cured in
a location which is not its final destination. The distance travelled from the casting site may
only be a few metres, where on-site precasting methods are used to avoid expensive haulage
(or VAT in some countries), or may be thousands of kilometres, in the case of high-value-
added products where manufacturing and haulage costs are low. The grit basted architec-
tural precast concrete in Figure 1.1 was manufactured 600 km from the site, whereas the
precast concrete columns, beams and walls shown in Figure 1.2a and b travelled less than
60 m; wall panels have been stack-cast in layers between sheets of polythene adjacent to the
final building.
What really distinguishes precast concrete from cast in situ is its stress and strain
response to external (load-induced) and internal (autogenous volumetric changes) effects.
These are collectively known as ‘actions’ in the Eurocodes, and those mainly applicable
to precast concrete structures are the ‘keynote’ code EC0 (BS EN 1990 2002), the loading
or ‘actions’ code EC1 (BS EN 1991-1-1 2002) and the ‘concrete design’ code EC2 (BS EN
1992-1-1 2004).
A precast concrete element is, by definition, of a finite size and must therefore be joined
to other elements to form a complete structure. A simple bearing ledge or corbel will suffice,
as shown in Figure 1.3. But when thermal shrinkage or load-induced strains cause volu-
metric changes (and shortening or lengthening), the two precast elements try to move apart
(Figure 1.4a). Interface friction at the mating surface prevents movement, but in doing so
creates a force F = μR which is capable of splitting both elements unless the section was
suitably reinforced (Figure 1.4b). Figure 1.5a shows an example of where frictional forces
due to relative, unreinforced movement between precast slabs and beams caused spalling in
the beam. In other cases, spurious positive bending moments due to the restraint of relative
movement or end rotation have caused cracking in the soffit of slabs, or at a beam-to-column
corbel connection, as shown in Figure 1.5b.
Flexural rotations of the suspended element (the beam) reduce the mating length lb (bearing
length), creating a stress concentration until local crushing at the top of the pillar (the column)
occurs, unless a bearing pad is used to prevent stress concentration (Figure 1.4c). If the bear-
ing is narrow, dispersal of stress from the interior to the exterior of the pillar causes lateral
tensile strain, leading to bursting of the concrete at some distance below the bearing unless the
section is suitably reinforced (Figure 1.4d).
2 Precast Concrete Structures
Architectural-structural precast concrete frame at Scottish Office, Leith, United Kingdom.

(Courtesy of Trent Concrete Ltd., Leith, UK.)
(a) Site cast using stack casting between sheets of polythene. ( )
What is precast concrete 3
( ) (b) Completed structure of mould cast columns and beams and stack cast walls.
Simple bearing nib and corbel.
(a) Unrestrained movement between two precast concrete elements. (b) Restrained movement
but without tensile cracking prevention. (c) Reduced bearing length and stress concentrations
due to flexural rotation. (d) Lateral splitting due to narrow bearings (left) (A) and preventative
(confinement) rebars (right) (B). ( )
in situ
( ) (e) Loss of bearing due to accidental actions (left) (A) and preventative dowel bar (right) (B).
(f) Loss of bearing due to settlement (left) (A) and preventative rebar loops (right) (B).
If the column is disturbed by an accidental or a structural force H such that H > μR,
the displacement u is not elastically recoverable and may lead to instability or even loss
of bearing altogether unless the bearing possesses shear capacity (Figure 1.4e). Should
the column’s foundation fail, loss of bearing will occur unless the bearing has tensile
capacity (Figure 1.4f).
These are some factors that distinguish precast concrete from other forms of construction.
(a)
(b)
(a) Spalling due to relative movement between slabs and beams. (b) Cracking due to spurious
restraint at a beam-to-column corbel connection.
A precast concrete structure is an assemblage of precast elements which, when suitably

connected together, form a three-dimensional framework capable of resisting gravita-
tion and wind (or even earthquake*) loads. The framework is ideally suited to buildings
such as offices, retail units, car parks, schools, stadia, and other such buildings requiring
minimal internal obstruction and multifunctional leasable space. The quantity of concrete
* This book does not address seismic actions.
in a precast framework is less than 4% of the gross volume of the building, and two-thirds
of this is in the floors. In the case of the shopping centre and car park shown in Figure 1.6,
the precast concrete elements supporting vertical actions (i.e. gravity loads) are columns,
beams, floor slabs, staircases, and stair-cores. The framework is ‘braced’ against horizontal
actions (i.e. lateral loads and wind pressure) using very deep columns (gable end to the left
of the photo) and diagonal bracing (front elevation). The framework shown in Figure 1.7
was built using similar elements, but because the resistance against horizontal actions is
provided by the same columns that support vertical actions, the framework and hence the
columns are classed as ‘unbraced’. The precast framework Figure 1.8 is likewise a column,
Precast concrete ‘skeletal’ braced frame, The ‘Green Apple’ retail centre and car park, Helsinki.
Precast concrete ‘skeletal’ sway frame, Europark, Rome.
Precast concrete ‘skeletal’ known as ‘semi-rigid’ frame, Recife University, Brazil.
Precast ‘skeletal’ structure with integrated architectural columns and spandrel beams, Reading
Business Park. (Courtesy of Trent Concrete Ltd., UK.)
beam and slab structure, but here the beam-to-column connections are designed as moment
resisting, and therefore together with the strength and stiffness of the beams and columns,
the resistance against horizontal actions is provided frame action, in a similar manner as
for cast in situ concrete frames. The distinguishing feature of the precast framework is that
the beam-to-column connections are rarely fully rigid, known as ‘semi-rigid’, and therefore
the columns must also resist horizontal actions as in the case of the unbraced frame in
Figure 1.7. The frameworks shown in Figures 1.9 and 1.10 were built using similar elements,
but thanks to some creative surface finishes and more expensive mouldings, this build-
ing appears to have a completely different function, both architecturally and structurally.
Precast ‘skeletal’ and ‘wall-frames’, Barwa City, Qatar.
These different types of precast frameworks, called ‘skeletal’ structures, will be discussed
and designed in this book. Although columns, beam and slabs are visible in Figure 1.10, this
is predominantly a ‘wall-frame’, distinguished from ‘skeletal’ by the use of wall panels with
the triple function of (1) the architectural façade, and resisting (2) vertical and (3) horizontal
actions. Wall-frames are not discussed or designed in this book.
Referring to Figure 1.11a, gravitation forces gk (pressure of dead loads [kN/m 2]) and qk
(pressure of live or imposed loads [kN/m 2]) are carried by floors in bending and shear,
and sometimes torsion. Floor loads are then transmitted to the beams (as shear v [kN/m]);
from the beam (as shear V [kN]) to the column; and from the column to the foundation (as
compression N [kN]). Bending moments m [kNm] may arise in the column due to beam-
to-column interaction, but these will be ignored for now because, when compared with the
moments M arising from horizontal forces Wk, they are rather small.
Referring to Figure 1.11b, the forces Wk (pressure due to horizontal wind loads [kN/m2])
act over the full façade of the building. To check for frames built out of plumb, an imperfec-
tion force Hi [kN] is added to Wk [kN] for each floor in turn. Hi is defined and used in frame
stability design in Section 6.2.6 and Figure 6.23. The imperfection line load hi = Hi/l [kN/m].
Together they are absorbed by the structure at each floor level. If the storey height is h [m]
and the length of the façade is l [m], the total force absorbed at each floor level is Wk + Hi =
wkhl + hil [kN] (except at the roof and ground where the captured height is h/2. Wk + Hi
is transmitted through the floors, now acting as plates (as horizontal shear f [kN]) to the
beams; from the beams (as horizontal shear F [kN]) to the columns; and from the columns
(as bending M [kNm] and shear Σf [kN]) to the foundations. As second-order sway deflec-
tions e2 [m or mm] increase, the moment at the foundation = M + m + Ne2 [kNm]. Because e2
is proportional to h3, e2 increases rapidly with height and quickly becomes dominant.
The sway profile of the column also depends on the degree of fixity (resistance against bend-
ing) between the column and the foundation, and between the column and beams. At best,
the foundation and beams are rigidly connected to the column and the column’s sway profile
is as shown in Figure 1.12a. At worst the foundation and beams are pinned, and the structure
Main beam
or or in
Flo Flo m
M
m m m
Min Max
v v v am
v v be
ble
Ga
m m m
m
Max Min
v v v v v
Max = γf gk + γf qk
m Min = gk
m m m
Min Max
v v v v v
N N N
m m m m
(a)
F
am
Line pressure be
wk h + hi Floor plate ble
Ga
F F
e2
Floor plate
f f N3
Wk
+ Hi F Main beam F F F F
N2
h
Wk f f
+ Hi F F F
h
H
N1
Wk f f
+ Hi F F F
h
Sway column
M + Ne2 M + Ne2 M + Ne2 illustrated
(b)
Load transfer in a skeletal unbraced (sway) frame due to (a) gravity loads and (b) horizontal loads.
is a mechanism (Figure 1.12b). In practice, the foundation is very rigid (not infinitely so) and
the beams are nearly pinned (not zero stiffness) so that the column’s profile is as shown in
Figure 1.12c. Acting as a free-standing cantilever, the capacity of the column is exhausted by
the combined actions of N, M, m and Ne 2 when the total height of the structure is H = Hcrit.
When H < Hcrit, the skeletal structure is called a ‘sway frame’ or an ‘unbraced structure’.
When H > Hcrit, walls are introduced to the skeletal structure, which then becomes a ‘no
sway frame’ or ‘braced structure’ (Figure 1.13). The walls (or other types of bracing) which
are aligned lengthways with the direction of Wk + Hi replace the function of the columns in
resisting overturning moments. Because of their massive strength and stiffness, walls will not
Pinned joint
Large stiffness beam
Rigid joint Mechanism

Column in
double
curvature
Rigid foundation
(a) (b)
Horizontal restraint
Euler strut
(c) (d)
Column sway profiles for different end conditions. (a) Rigid ended - unbraced, (b) Pinned ended -
unbraced, (c) Rigid foundation pinned beam - unbraced, (d) Pinned ended - braced.
Wall
F
vn F
f f
f
Wk
F
R Deflection Deflection
–N f
f
Wk
F wk Mn Vn
R f
I –N N
f
Wk
F
R f f
–N N
Wall F
ΣF F
Wall
(b) Plan
(a) Elevation
Load transfer in a braced frame using infill shear walls (a) together with a precast floor dia-
phragm (b). Note that in this figure and include the force and line pressure and ,
respectively due to imperfection.
be required to replace every column, in fact 1 in 20 is more likely, and as they are positioned
at strategic positions, mostly at the ends of buildings and around staircases or lift shafts.
Because they are few in number in skeletal frames, walls have little effect on the transfer of gk
and qk, and so Figure 1.11a applies. However, their effect on horizontal forces is considerable
as shown in Figure 1.13. Wk + Hi is transmitted through the floor, now acting as a horizon-
tal plate (as horizontal shear f [kN] and horizontal bending mh [kNm]) to the beams at the
position of the walls; from the beam (as shear F [kN]) to the walls; through the walls (as diago-
nal force R [kN]) to the foundation (as shear ΣF [kN] and axial reactions N [kN]).
Though shear deflections v occur, the walls are made sufficiently stiff to ensure that the
limit of v/h <1/500 will eliminate e 2 from further consideration. The sway profile of the col-
umns is no longer dependent on the fixity of the foundation or beams if the connections are
pinned, the sway profile is as shown in Figure 1.12d (Euler strut). Second-order deflections
e 2 only occur if the length of each column, actually its slenderness ratio, between storey
heights is greater than the critical values imposed by codes.
Possible variations in the number and position of walls are endless, but the principles
given here do not change. ‘Partially braced’ structures are an amalgam of unbraced, above
or below a certain level (see Figure 3.15), and braced structures. ‘Unidirectionally braced’
structures are braced in one direction and unbraced in the other.
Having resolved the moments, shears, axial forces (and possibly torsion) in the precast
elements, and taken note of the effects of shrinkage, bursting, accidental damage, loss at
bearing, etc. discussed in Section 1.1, element design follows the principles of traditional
reinforced or prestressed concrete design. However, the manufacture transportation, hoist-
ing and temporary stability of the elements and of the structure during construction must
be considered. Precast concrete frame designers cannot diverse themselves from the manu-
facture and site erection procedures. Feedback from the factory (as casting site) to the design
office and from the construction site to the factory and office is essential.
The very existence of a precast concrete industry and the numerous successful building
projects achieved using precast concrete, for the whole or just a part of the structure, is
proof that the technique is practical and economical. In global terms, the market share of
precast ‘grey’ frames (structural concrete with no architectural qualities) is probably around
5% of the multistorey business. However, for precast structures with an integrated facade
or other decorative features (Figure 1.9), the global market share is closer to 15%, being as
high as 70% in the colder climates and/or where site labour is expensive. Figure 1.14 shows
Precast concrete wall-frames for residential dwellings near to Helsinki.
Precast concrete wall panels for low- and medium-rise housing, O-Stable Panel system, Malaysia.
a building in a ‘prefabricated concrete town’, just outside Helsinki, Finland. Figure 1.15
shows a high-spec prefabricated house in Malaysia, where rain-resistant wall panels, cast
with window and door openings, are keyed into the ornamental corner columns.
Many people believe that in certain countries, especially where the ratio of labour-to-materials
or plant is low, say one man-hour pay is less than 1/50 tonne of cut-and-bent rebar or one man-
hour pay is less than 1/500 daily hire of a large mobile crane, precast cannot compete with cast
in situ concrete. Where local labour policies demand high levels of unskilled site labour, heavy
concrete prefabricates create the potential for new safety hazards due to transportation, han-
dling and temporary stability. Similarly, in countries with a strong steel industry and widespread
education in steelwork design, the popular opinion is that precast cannot compete with structural
steelwork frames. In many circumstances, however, precast concrete is the only economical and
practical solution, and if the designer is unwilling to consider precast concrete as a total solution,
the result of this is the so-called ‘mixed’ solution. Figures 1.16 through 1.19 show cases of using
precast prestressed concrete for long-spanning slabs or beams, whilst vertical supports are in
structural steelwork, tubular steelwork, cast in situ concrete and structural timber, respectively.
Providing that the stability of the structure is not impaired, the substitution of a steel
girder, truss or portal frame for a precast element is taken care of in the connection detail.
In the structure shown in Figure 1.16, shallow steel beams were used as direct substitution
for precast beams, which would otherwise have had a downstand below the floor. Savings
made from reduced headroom, and hence building and façade height, is the obvious advan-
tage of this method as the shallow steel beams cannot possibly be as economical as deeper
steel beams or prestressed concrete beams.
According to the fib bulletin (fib 2002), mixed construction is now being used in more
than 75% of new multistorey buildings in the western world, where the increased use of
precast concrete over the past 10–15 years is due to the move towards greater offsite prefab-
rication of structural elements. Some of the limitations found in precast concrete inevitably
lead to it being used with other materials in a cost-effective manner, for example to provide
structural continuity using small quantities of cast in situ reinforced concrete, or to form
long-span steel or timber roofs. Structurally, combinations may work together or indepen-
dently but together they can provide many advantages over the use of a single material.
Mixed precast and concrete with structural steelwork.
Mixed precast concrete with tubular steel bracing, Bouwhuis, the Netherlands, 2008.
The ‘key’ to success using any form of mixed or precast concrete solution is to be able to
offer the client, architect and consulting engineer a solution that is
1. Buildable
a. The construction sequence is sensible enabling other trades to dovetail into
programmes.
b. The construction is safe and temporary stability is guaranteed.
c. Economic cranes are used.
d. Labour is skilled.
Mixed precast and lightweight concrete topping on profiled metal decking, Bracken
House. (Courtesy of Trent Concrete Ltd., Leith, UK.)
Mixed construction of precast concrete floors in a timber beam–column frame. (Courtesy of

Spiroll Ltd., Derby, UK.)
2. Cost-effective
a. Building components cost-per-structural capacity is comparable to that of other
materials, for example cost/bending moment or cost/axial load capacity.
b. The overall building costs, inclusive of transport, fixing finishing, maintenance
and repair, are competitive.
c. Production is quality-assured – the lowest standard deviation on the population of
all elements, materials and methods.
d. The solution uses factory-engineered concrete.
e. Destructive load testing or nondestructive testing assurances are given.
3. Fast to erect
a. Although the precast manufacturing period may be several weeks, once started,
construction proceeds rapidly.
b. Following trades (bricklayers, electricians, joiners) move in quickly.
c. Handover or possession can be phased.
d. Clear heights and floor zones are satisfied.
e. Service routes are not interrupted.
f. Beam and column sizes are satisfied, and the bracing positions.
g. Decorative concrete, both internally and externally, may be exploited.
Prefabrication factories are not cheap to run. In some instances, it is certainly economical
to establish a prefabrication facility on site, or in an adjacent field, and work from there.
Whichever method is chosen, it is necessary to be able to do things in the factory that are
impossible to achieve on site. Figure 1.20 shows an example of predesigned, prefabricated and
bulk-purchased steel beam-to-column connectors, having a shear capacity of about 360 kN
on each face within a beam depth of around 400 mm. The end connectors alone probably
cost about 75 percent as much as the rest of the element, but in being able to transmit this
shear force in such a shallow depth the advantages to the structure as a whole are obvious.
Figure 1.21 is from the Netherlands; 55 m long prestressed precast concrete beams weighing
up to 140 tonnes are transportation by barge, taking advantage of their prefabrication and
infrastructure skills. Figures 1.22 and 1.23 show the dual advantage of prefabrication: com-
posite beam and insulated wall panel with an imitation Portland stone finish. The columns,
containing beam-to-column connectors similar to those shown in Figure 1.20, were finished
to the same standard and erected up to five storeys in a single piece.
A question is often asked at seminars and workshops – “please explain what makes
precast concrete different from traditional cast in situ concrete for use in buildings, and
what are the conditions best suited to the use of precast concrete?” The answer to the first
part may be best demonstrated by the precast columns, beams and roof units shown in
Figure 1.24. This photograph was taken 12 years after construction, although it does benefit
from internal exposure in a highly populated rail station. The answer to the second part is
Fabrication of a three-way steel-to-steel beam–column connector with 360 kN capacity on each face.
Taking advantage of factory production: 55 m long prestressed precast concrete beams at VBI,
the Netherlands.
Prefabricated beam and insulated wall element with a cast concrete–Portland stone finish.
Completion of building using elements shown in Figure 1.22 at Portland Building, University of
Nottingham, United Kingdom.
Precast column, beam and slab elements with architectural finishes at Paddington Station, London.
more complicated – it is a function not only of the form and design of the structure itself,
but of the balance between national (or even regional) rates/costs/availability of labour,
materials, factory automation and site plant. It is rarely a design decision. This subject is
addressed in by Elliott and Jolly (2013).
BS EN 1990. 2002. Eurocode: Basis of structural design, BSI, London, UK.

BS EN 1991-1-1. 2002. Eurocode 1: Actions on structures – Part 1-1: General actions – Densities,
self-weight, imposed loads for buildings, BSI, London, UK.
BS EN 1992-1-1. 2004. Eurocode 2: Design of concrete structures – Part 1-1: General rules and rules
for buildings, BSI, London, UK., February 2014.
Elliott, K. S. and Jolly, C. K. 2013. Multi-Storey Precast Concrete Framed Structures, 2nd ed., John
Wiley, London, UK., 750pp.
fib. 2002. Precast concrete in mixed construction, State-of-art report, Bulletin 43, Fédération
Internationale du Béton, Lausanne, Switzerland, 68pp.
Chapter 2
‘Precast concrete is of the highest possible quality, both in terms of strength and durabil-
ity’. The essence of this statement is captured in Figure 2.1, where concrete is accurately
delivered to every part of the mould, ensuring zero segregation, minimising voids and hon-
eycombing, using minimal vibration. However, contrast this view with the precast column
shown in Figure 2.2 and it is easy to understand the problems of workmanship, rather
than material defects, that certain parts of the industry endure. Using materials that have
passed strict quality control procedures, rapid hardening cement type CEM I class 42.5R or
52.5R according to BS EN 197 Part 1 (BS EN 197-1 2011) is mixed with excellent-quality
aggregates of known source and purity, in computer-controlled batching and mixing plant,
to produce concrete of specified workability and strength. Even the introduction of small
quantities of uncontaminated recycled concrete, usually from the factory’s own waste pro-
duction, superplasticisers and pozzolanic materials (such as pulverised fuel ash), has not
reduced this standard. To quote standard deviations (SDs) of less than 2 N/mm 2 on concrete
of the 28-day compressive cube strength of fck = 45–70 N/mm 2 would not be an exaggera-
tion. Table 2.1 lists typical mean concrete strengths fcm (usually obtained from 2 or 3 cubes)
for precast production at 20 h (detensioning prestress or stripping moulds), 3 days and
28 days, and the characteristic values for cement cubes fcu (BS EN 197-1 2011). The table
shows that even allowing for fcu = fcm – 4 N/mm 2 , concrete at 20 h just about exceeds the
requirements for cement strength at 2 days.
Concrete is specified according to BS EN 206 (BS EN 206 2013). The design strength is
denoted in the Eurocode EC2, BS EN 1992, Part 1-1, clause 3.1.2(1) (BS EN 1992-1-1 2004)
by strength classes (known as ‘grade’), which relate to the characteristic (5%) compressive
cylinder strength fck, or the cube strength fck,cube (fcu in BS 8110) in accordance with EN
206, for example grade C32/40 is the 28-day characteristic strength fck = 32 N/mm 2 and
fck,cube = 40 N/mm 2 . See Section 2.1.3 for Young’s modulus Ecm.
The strength class used in design is the cylinder strength fck up to a maximum of 90 N/mm 2
(105 cube strength) with a minimum 20 N/mm 2 for structural grade. The true stress vs
strain (σc vs εc) relationship for concrete of normal and higher mean strength fcm is shown in
Figure 2.3a, based on BS EN 1992-1-1, Fig. 3.2. Note the reduced ductility after the peak
stress of higher strength, defined as > C50/60, where the design concrete crushing strain
reduces from εcu3 = 0.0035 to 0.0026. The effect of this means that the calculated value
of x/d for a flexurally balanced section, using rebar with f yd = 500/1.15 = 434.8 N/mm 2 ,
is generally x/d = 0.0035/(0.0035 + 434.8/200,000) = 0.617, reducing to x/d = 0.545 for
fck = 90 N/mm 2 .
21
Precise delivery of concrete into the mould. (Courtesy of Tammer Elementti Oy, Finland.)
Compaction problems in precast concrete columns.
The constitutive stress–strain relationship is idealised as a parabola–rectangle in

Figure 2.3b, where the design strength fcd = fck /γm , where γm is the material partial safety
factor of 1.5 at the ultimate limit state and 1.2 for accidental situations based on the UK
national annex (NA) to BS EN 1992-1-1, Table 2.1N (NA to BS EN 1992-1-1, 2004).
Figure 2.3c shows the idealised rectangular stress block used in design, where the distance
x to the centroid of the areas beneath the curves in Figure 2.3b and c due to the design
Materials used in precast structures 23
Strength requirements for cement based on BS EN 197-1:2011 and typical concrete strengths
achieved by precast producers
42.5N 15–20 ≥ 40–55 – 50–70

42.5R 25–35 ≥ 50–70 60–90
52.5N 20–30 ≥ 50–70 ≥ 60–90
52.5R 30–35 ≥ 50–80 70–100
Key: BS EN 197 characteristic value for cement (italics); producer’s mean cube strength.
σc
High strength fcm
fcm
fck
x = to centroid of area
fcd η fcd
0.4 fcm
x x
Ecm
εc
(a) εc1 εcu1 (b) εc2 εcu2 (c) εcu3
Constitutive stress vs strain relationships for concrete in compression. (a) Actual relationships
for normal and high strengths, (b) parabola–rectangle idealization, and (c) idealized rectangular
stress block used in design.
strength ηfcd is established by national codes to be as close as possible. BS EN 1992-1-1,

Exp. 3.19 through 3.22 gives:
h= £
(2.1)
h= - - £
For example for fck = 90 N/mm 2 fcd loses 20%.

and in Figure 2.3c if the x-axis represents the depth to the neutral axis x, then the design
rectangular stress block has a depth λx, where
l= £
(2.2)
l= - - < £
That is 0.8 reduces to 0.7 at 90 N/mm 2 .
For example the ultimate force in a concrete beam of breadth b is Fc = design stress ×
area = ηfck b λx = 0.8fckbx for fck = 50 N/mm 2 reducing to 0.56fckbx for fck = 90 N/mm 2 .
Coupled with the reduced depth at the balanced section, the total force at the balanced
section is Fc = 24.68 bd and 27.47 bd for fck = 50 and 90 N/mm 2 , respectively, that is only
11% greater for an 80% increase in fck!
The strength of concrete at demoulding, or at detensioning in the case of prestressed
concrete, is specified by the producer according to manufacturing and handling techniques,
using mean compressive cube strengths fcm,cube. This takes place mostly at 16–24 h, where
there are no empirical expressions to predict this strength, which is highly dependent on
casting and curing conditions. The workability of the concrete is still specified according to
slump value, for example 25–100 mm in factory production, except in the case of concrete
provided for slip form or extrusion, which has no workability requirements (zero slump).
The strength of concrete fck(t) between 3 and 28 days can be determined either by testing
cubes/cylinders or estimated using semi-empirical equation for an average temperature of
20°C and standard curing conditions as follows:
-Ö
= (2.3)
where
fcm(t) = fck(t) + 8 N/mm 2 (2.4)
and fcm = fck + 8 N/mm 2 (representing a margin of 1.64 SDs of strength for 5% character-
istic values). Precast production quality measures achieve early strength SD ≈ 2–3 N/mm 2 ,
making the one mentioned earlier conservative.
s = 0.20, 0.25 and 0.38 for cement Class R, N and S.
For example if fck = 40 N/mm 2 at 28 days using CEM 52.5N (Class R), at say t = 3 days,
fcm(t) = 48 e0.2(1 – √28/3) = 48 × 0.663 = 31.8 N/mm 2 and fck(t) = 23.8 N/mm 2 . If the margin at
3 days is taken as 3 N/mm 2 , then for the same example fck(t) = 28.8 N/mm 2 , giving a better
estimate of more than two-thirds at 3 days.
When precasting in a modern factory, it is vital to achieve the dual requirements of
good workability and early strength. Concrete is often transported automatically from an
automatic batching plant to the mould leaving no opportunity for workability tests. It is
unlikely that such tests are required thanks to the experience of batching staff. Producers
do not vary the mix specification unnecessarily – for example Grade C40/50 concrete
might be used for a staircase even though the design calls for a lower strength of say
fck = 25 N/mm 2 .
A huge impact on precast production methods has been the use of self-compacting con-
crete (SCC), which as the name suggests will flow and compact without external vibration.
SCC admixtures conform to BS EN 206 Part 9 (BS EN 206-9 2010). After the problems of
low early strength gain were resolved around 2002, this has been widespread in all types of
frames, flooring, stairs, and terraces. Figure 2.4 shows SCC flowing up to 1.5 m in an out-
door steel mould on quite a cold day in the United Kingdom (around 7oC), achieving an 18-h
detensioning cube strength of fck,cube = 25–30 N/mm 2 . Figure 2.5 shows the finished SCC
product for seating terraces, undeniably of the highest-quality concrete. The workability of
SCC is measured using a flow table in which the diameter of flow, as shown in Figure 2.6,
is usually 600–700 mm, neglecting the small ring of cement slurry at the edges. The precast
industry has exploited the advancements made in concrete technology more than in any
other part of the construction industry.
Self-compacting concrete, flowing and compacting without segregation, used to manufacture

prestressed floor slabs.
Self-compacting concrete used to manufacture precast terrace units for stadia, by Bison
Manufacturing, United Kingdom.
For the production of standard elements such as columns and beams, concrete is cast into
clean steel or timber (sometimes concrete) moulds, accurate to ±3 mm, or less, in cross sec-
tion. Figure 2.7 shows an adjustable steel mould of around 100 m length for casting beams
(in this case, bridge beams), or for casting double-tee slabs of fixed width (usually 2.4 m)
and variable length and depth as shown in Figure 2.8. The use of clamped vibrators tuned
Measurement of the workability of self-compacting concrete by measuring the diameter of flow

on a flow table.
Production of precast prestressed concrete structural elements in steel moulds.
to the correct oscillations for the size and weight of the filled mould ensures correct com-
paction to a density of around 2400 kg/m3 (excluding reinforcement). The resulting surface
finish results in minimum porosity for maximum durability. Concrete strengths are made to
match the optimum performance of each element, such that flexural members are produced
in grade C32/40 concrete, whilst compression members are typically C40/50 for walls and
C50/60 to C90/105 for columns.
Stripping of steel moulds and lifting of prestressed element 20 h after casting.
Production of precast concrete facade elements in timber moulds.
Non-standard elements, such as the cladding panel in preparation in Figure 2.9, are cast
in timber or fibre glass moulds (epoxy-based materials may be remoulded after use). The
grade of concrete is usually C32/40. Although the surface finish from a timber mould is not
equal to that from a steel mould, it is nonetheless equal to ‘special’ as follows.
Surface finishes are specified according to BS EN 13670 ‘Execution of Concrete Structures’,
Annex F8.8 (BS EN 13670 2009). However, the background document to the NA to BS EN
1992-1-1, PD 6687-1 (PD 6687-1 2010) states that the provisions of the National Structural
Concrete Specification Edition 4, known as NSCS-4 (NSCS 2010) are considered equiva-
lent to those in BS EN 13670 for all types. The website (The Concrete Centre 2016) states
‘developed by CONSTRUCT and a technical committee that included contractors, consul-
tants and industry partners. It aims to be the standard specification for concrete frames in
the United Kingdom. The specification is designed to achieve increased efficiency through
a definitive, simple, straightforward specification which removes unnecessary conditions
and facilitates construction. It also provides guidance on tolerances and finishes’. Until
further information is available, the surface finishes will be specified in NSCS-4, clause 8.6
as formed finishes as follows:
‘Basic’ – for ground or foundations

‘Ordinary’ – to receive finishes, but not painted directly
‘Plain’ – visual concrete occasionally seen, to be painted directly
‘Special’ – architectural finishes, controlled for irregularity, arrises, colour variation,
blowholes
Precast floor slabs are manufactured using either ‘wet cast’ (semi-wet with less than 50 mm
slump) or ‘dry cast’ (semi-dry with zero slump and 5–10 s V-B time). Steel moulds are used in
the semi-wet production of reinforced hollow core units, as shown in Figure 2.10. Machine
delivered concrete makes two passes to fill the mould below and above the mandrels forming
circular hollow cores. The grade of concrete is usually C32/40.
Machine-extruded concrete is not, by definition, cast and compacted. It is instead forced
through rotating augers to pass out through steel formers, possibly with mandrels to create
voids, as shown in Figures 2.11 and 2.12. Slip-formed concrete is similarly produced through
Machine delivery of concrete to reinforced hollow core slabs at Creagh Concrete, N. Ireland,
United Kingdom.
Rotating augers create forced extrusion of concrete around the mandrels shown in Figure 2.12.
Circular mandrels create circular voids in hollow core slabs.
a machine that compacts the concrete by rows of small hammers – known as the ‘shear-
compaction’ method. To enable both the processes to happen, the concrete is ‘earth-dry’,
having a water–cement (or water-cementitious material) ratio of around 0.3. Aggregates are
carefully selected, especially the sand grading and content, and the shape and size of the
coarse aggregates, typically 14 mm down. Although not exclusive to these production meth-
ods, concrete produced in this way is usually prestressed for flexural members such as floor
slabs, such as hollow core floor units shown in Figures 2.13 and 2.14. Because prestressed
Long line slip form of prestressed hollow core floor slabs, at Bison Manufacturing, United Kingdom.
Excellent compaction of slip-formed hollow core floor units.
concrete benefits from additional strength in flexural compression, it is advantageous to use

grade C50/60 concrete; although for certain geometry and in order to balance the limiting
compressive stress (0.45 fck) and tensile stress (fctm), the common grade is C45/55. For inter-
nal exposure XC1, BS EN 1992-1-1 clause 7.3.2(4) limits the service condition to the mean
axial tensile stress (with the code reference)
Table 3.1 fctm = 0.3fck2/3 (2.5)
100
90
80
70
60
Passing (%)
50
Slipformer 1
40 Extruder: Finer grading with 55%–65% passing Extruder 2
0.5 mm sieve, and fine/total aggregate ratio
Extruder 3
30 0.35–0.40.
Extruder 4
Slipforming: 40% passing 0.5 mm sieve, and
20 fine/total aggregate ratio 0.40–0.45. Slipformer 5
Coarse aggregate 14/5 size, gravel or Extruder 6
10 limestone. Extruder 7
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Sieve (mm)
Grading curves for fine aggregate used in the extrusion and slip form of prestressed hollow core
slabs. (Data collected in the United Kingdom by the author 2004–2008.)
for concrete grade ≤ C50/60 (viz. Class 2 in BS 8110 Part 1 (BS8110 1997)). For exposure >
XC1, BS EN 1992-1-1 Table 7.1 (NAD Table A.4) requires that the service condition is ‘decom-
pression’, that is zero tension (viz. Class 1 in BS 8110-1). (See Section 4.3.1 for full analysis.)
Each producer will modify their mix according to variations in local supplies. The grad-
ing of fine aggregate is often the most critical mix parameter, especially in the dry mixes
used in slip-formed and extruded concrete. Figure 2.15 shows typical grading curves for fine
aggregate used in machine cast units, note the high percentage passing the 0.5 mm sieve size
for the extrusion machine. There is some evidence (in private work by the author) that slip-
page of strands occurs where the grading is too fine around the 1 mm sieve size.
Although the 28-day strength fck specified in the design must be achieved, the controlling
factor in mix design is the strength at demoulding, or in the case of prestressed concrete,
its strength at detensioning (transfer). Demoulding and detensioning of cast concrete takes
place at around 18 h. Detensioning of machine slip-formed or extruded concrete takes place
between 10 and 20 h. Various methods to accelerate the early strength of concrete include
rapid hardening Portland cement, chemical accelerators (calcium chloride must not be used)
and by external heating, such as steam curing and electrical heating. Even microwave oven
techniques have been tried. Care must be taken in thin-walled section, such as hollow core
floor units, to avoid thermal restraint due to large temperature differentials.
The static value for Young’s modulus is calculated according to BS EN 1992-1-1, clause
3.1.3, as:
Table 3.1 Ecm = 22 [(fck + 8)/10]0.3 (2.6)
for gravel aggregate. For limestone and sandstone aggregates, the value should be reduced
by 10% and 30%, respectively. For basalt aggregates, the value should be increased by 20%.
Concrete strengths and elastic modulii used in precast elements
Beams, shear RC 30–40 15–20 17.0–22.7 N/A 33–35 N/A

walls staircases,
wet cast floors
Columns, load- RC 40–90 20–25 22.7–51.0 N/A 35–44 N/A
bearing walls
Beams, columns PSC 40–50 25–30 22.7–28.3 3.5–4.1 35–37 31–33
Dry cast slabs PSC 45–55 25–35 25.5–19.8 3.8–4.2 36–38 31–34
a based on gravel aggregates. For limestone or sandstone, divide by 1.1 and 1.3.
Strengths and short-term elastic modulus for typical concrete used in composite construction
25 N/A 31.5 N/A

30 N/A 32.8 N/A
Precast reinforced 30 N/A 32.8 0.96–1.00
40 N/A 35.2 0.89–0.93
Prestressed 45 3.80 36.3 0.87–0.90
50 4.07 38.2 0.82–0.86
Source: From BS EN 1992-1-1, Eurocode 2: Design of concrete structures. Part 1-1: General rules and rules for buildings,
BSI, London, UK., February 2014, 2004.
based on gravel aggregates. For limestone or sandstone, divide by 1.1 and 1.3.
Table 2.2 gives the relevant information. Mix design details are given by Elliott and Jolly
(Elliott and Jolly 2013).
Where in situ concrete of strength fcki is added to precast elements to form composite
behaviour (see Sections 4.4.4 and 5.5), the relative strengths (fck) and stiffness (Ecm) of the
two materials must be considered using the modular ratio method. The most common combi-
nations of in situ-to-precast strengths are fcki/fck = 20/32 and 25/45. This leads to the modular
ratio m = Ecmi/Ecm ≈ 0.87 as shown in Table 2.3.
Shrinkage and creep are time-dependent properties of concrete. They depend on the ambi-
ent humidity (RHs %); notional depth or thickness of element = 2 × area/exposed perimeter,
ho = 2Ac/u); concrete composition (w/c, cement type, e.g. rapid hardening Class R, fck); maturity
of the concrete when the elements are stored; age (to) when load is first applied; and the
duration and magnitude of the loading.
BS EN 1992-1-1, clause 2.3.2.2 notes: Shrinkage and creep effects should generally be
taken into account for the verification of serviceability limit states. The effects should be
considered at ultimate limit states only where their effects are significant, for example in
the verification of ultimate limit states of stability where second-order effects are of impor-
tance. In other cases, these effects need not be considered for ultimate limit states, provided
that ductility and rotation capacity of the elements are sufficient. Creep effects should be
evaluated under the quasi-permanent combination of actions irrespective of the design
situation considered, that is persistent, transient or accidental. In most cases, the effects of
creep may be evaluated under permanent loads and the mean value of prestress. In other
words, creep is considered for second-order effects in column design but otherwise is used
only in the determination of long-term deformations of deflection and crack width.
In precast elements, some of the creep and shrinkage deformations take place prior to
installation, something which is not considered in the code equations, and therefore these
effects are calculated at the time of installation (typically 7–28 days) and in the long term of
500,000 h (57 years).
Shrinkage causes shortening of concrete, around 400 × 10–6 (μm) ≈ 4 mm in 10 m length,
and is considered mainly in the long term, although in composite precast design the short-term
shrinkage of the precast element is determined so that the effects of the relative shrinkage of the
in situ to the precast can be determined for shrinkage-induced curvature. First introduced in BS
EN 1992-1-1 in clause 2.3.2.2, shrinkage actions are evaluated in clause 3.1.4 and Appendix B
as follows. The shrinkage strain εcs at time t (20,833 days for long-term shrinkage) after removal
from mould (or detensioning in the case of prestressed) at time ts = typically 1 day is
Exp. 3.8 Total shrinkage strain εcs = εcd + εca (2.7)
where
Exp. 3.9 Drying shrinkage strain εcd = βds(t,ts) kn εcd,o (2.8)
Exp. 3.10 Age factor βds(t,ts) = (t – ts)/[(t – ts) + 0.04 ho1.5] (2.9)
Table 3.3
= - - £ <
- - £ < (2.10)
- - £ <
Exp. B.11 εcd,o = 0.85 [(220 + 110 αds1) exp –(αds2 fcm /fcmo)] × 10 –6 βRH (2.11)
Exp. B.12 RH factor βRH = 1.55 [1 – (RHs /100)3] (2.12)
Exp. 3.11 Autogenous shrinkage strain εca(t) = βas εca (2.13)
10.3.1.2(3) In precast elements subjected to heat curing εca(t) is negligible.
Exp. 3.13 βas(t) = 1 – e-0.2√t = 1 for the long term (where t is in days) (2.14)
-
e = - ´
Exp. 3.12 (2.15)
=
3.1.2(6) Cement classes and αds factors are

B.2 Class R = CEM 42.5 R, CEM 52.5 N and CEM 52.5 R (αds1 = 6, αds2 = 0.11)
Class N = CEM 32.5 R, CEM 42.5 N (αds1 = 4, αds2 = 0.12)
Class S = CEM 32.5 N (αds1 = 3, αds2 = 0.13)

Where shrinkage actions are required at ultimate limit state, a partial factor γSH = 1.0 is used.
Example 2.1
Calculate long-term εcs for a precast concrete element of 300 × 300 mm cross section
manufactured using grade C40/50 concrete using cement CEM I class 52.5R. The col-
umn is demoulded at 1 day and is used inside a building with RHs = 50%. Repeat using
cement class 42.5N and comment on the differences.
Solution
Equation 2.4 fcm = 40 + 8 = 48 N/mm 2
Equation 2.12 βRH = 1.55 [1 – (50/100)3] = 1.356
cl. B2 Cement class R, where αds1 = 6 and αds2 = 0.11
Equation 2.11 εcd,o = 0.85 × [(220 + 110 × 6) e−(0.11 × 4.8)] × 10–6 × 1.356 = 598 × 10–6
Equation 2.10 kn = 0.925 for ho = 2 × 90,000/1,200 = 150 mm
Equation 2.7 βds(t,ts) = (20,833 – 1)/[(20,833 – 1) + 0.04 × 1501.5] = 0.996
Equation 2.8 εcd = 0.996 × 0.925 × 598 × 10 –6 = 552 × 10 –6
Equation 2.15 εca = 2.5 × (40 – 10) × 10 –6 = 75 × 10 –6
Equation 2.7 εcs = 552 + 75 = 627 × 10 –6
This is rather a high value for precast concrete in particular and certainly much greater
than BS 8110 values closer to 400 × 10 –6. However, if the cement is changed to class N
(αds1 = 4, αds2 = 0.12) αcd,o = 0.85 × [(220 + 110 × 4) e –(0.12 × 4.8)] × 10 –6 × 1.356 = 427 × 10 –6,
a reduction of 30% will result compared with that of Class R. It is difficult to grasp how
εcd,o can be so sensitive to a change of cement from class 52.5R to 42.5N? Note that εcd,o
values are tabulated in BS EN 1992-1-1, Table 3.2 for Class N cement, giving 420 × 10 –6
for this example.
Creep, as a viscoelastic effect of the deformability of cement paste, causes changes in
the elastic modulus of concrete from the immediate value (as mentioned earlier) Ecm to
Ecm /(1 + φ(t,ti)), where φ(t,ti) = creep coefficient from time ti to t. This means that φmay
be calculated for any time interval as:
Exp. B.2 Creep coefficient φ(t,ti) = φRH . β(fcm). β(ti). βc (t,ti) (2.16)
where
Exp. B.3 φRH = {1 + [(1 – RHs /100) α1/0.1 ho1/3]}α2 (2.17)
Exp. B.8c α1 = min{1; (35/fcm)0.7}, α2 = min{1; (35/fcm)0.2} (2.18)
Exp. B.4 Strength factor β(fcm) = 16.8/√fcm (2.19)
Exp. B.5 Age at installation loading factor β(ti) = 1/(0.1 + ti 0.2) (2.20)
Exp. B.7 βc (t,ti) = {(t – ti)/[(t – ti) + βH]}0.3 (2.21)
Exp. B.8b RH factor βH (days) = 1.5 [1 + (0.012 RHs)18] ho + 250 α3 (2.22)
Exp. B.8c α3 = min{1; (35/fcm)0.5} (2.23)
The high temperature steam or otherwise curing of precast elements gives an equivalent
age of maturity age when it is demoulded at time t 0, as follows:
Exp. B.5 Age at release loading factor β (t 0) = 1/(0.1 + t 0 0.2) (2.24)
a
æ ö
Exp. B9 t =t Tç ÷ ³ (2.25)
è ø
where α = 1, 0, −1 for Class R, N, S cement, respectively.
Exp. B10 Equivalent age at transfer t 0T = Σ e−[4000/(273+T) – 13.65].Δti (2.26)
where T = mean curing temperature (typically 50°C) during curing time Δti (typically
20 h) in days.
Example 2.2
Calculate the equivalent age at release t 0 for concrete matured for 20 h under a mean
temperature of 50°C using cement class R.
Solution
Equation 2.26 t 0T = e−[4000/(273+ 50) – 13.65] (20/24) = 3.547 × 0.833 = 2.96 days
α = 1 for Class R
Equation 2.25 t 0 = 2.96 × {[9/(2 + 2.961.2)] + 1}1 = 2.96 × 2.582 = 7.65 days
Example 2.3
Calculate the long-term creep coefficient φ(t,ti) for the section in Example 2.1 and curing
maturity in Example 2.2. Repeat at t = 28, 56 and 90 days and comment on the progres-
sion of creep in the short term.
Solution
Equation 2.23 α1 = min{1; (35/48)0.7} = 0.80; α2 = 0.94; α3 = 0.85
Equation 2.17 φRH = {1 + [(1 – 50/100) 0.80/0.1 × 1501/3]} × 0.94 = 1.647
Equation 2.19 β(fcm) = 16.8/√48 = 2.425
Equation 2.24 β (t 0) = 1/(0.1 + 7.650.2) = 0.624
Equation 2.22 βH = 1.5 [1 + (0.012 × 50)18] × 150 + 250 × 0.85 = 438 days
Equation 2.21 βc (t,ti) = {(20833 – 7.65)/[(20833 – 7.65) + 438]}0.3 = 0.994
Equation 2.16 φ(t,ti) = 1.647 × 2.425 × 0.624 × 0.994 = 2.492 × 0.994 = 2.48
Equation 2.21 If t = 28 days, βc (t,ti) = {(28 – 7.65)/[(28 – 7.65) + 438]}0.3 = 0.393
Equation 2.16 φ(t,ti) = 2.492 × 0.393= 0.98
Equation 2.16 If t = 56 days, βc (t,ti) = 0.500, φ(t,ti) = 1.25
Equation 2.16 If t = 90 days, βc (t,ti) = 0.500, φ(t,ti) = 1.43
1 N R t0 = 7.65 days C40/50

2 S
3
5 C20/25
C25/30
C30/37
10
t0
C35/45
C40/50
C45/55
20 C50/60 C55/67
C60/75 C70/85
30 C80/95 C90/105
50
100
7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 100 300 500 700 900 1100 1300 1500
(∞, t0) = 2.3 h0 = 150 mm h0 = (mm)
Inside conditions – RH = 50%
Note:
1 - intersection piont between lines 4 and 5 can also
4
be above point 1
5 3 - for t0 > 100 it is sufficient accurate to assume t0
2 = 100 (and use the tangent line)
Method for determining the long-term creep coefficient. (Annotated by the author, based on BS EN
1992-1-1, Eurocode 2: Design of concrete structures. Part 1-1: General rules and rules for buildings,
BSI, London, UK., February 2014, 2004, Figure 3.1; Courtesy of British Standards Institute, London.)
The short-term creep is known to have a logarithmic relationship with time, and the
aforementioned results suggest that φ(t,ti) is a function of K loge tx as follows:
At t = 28 days, 0.2 × (Loge 28)1.3 = 0.96 (cf. 0.98)
At t = 90 days, 0.2 × (Loge 90)1.3 = 1.42 (cf. 1.43), and so on.
BS EN 1992-1-1, Fig. 3.1 presents a graphical route to φ(t,ti) as reproduced in Figure 2.16.

The additional (dashed) line gives the solution to the long-term value mentioned earlier as
φ(t,ti) ≈ 2.3, in reasonable agreement with their equations in Appendix B.
This section deals only with the relationship of fire resistance with the cover, or more precisely
the mean axis distance to the centroid of reinforcement (rebars in reinforced concrete and ten-
dons in prestressed concrete (PSC)) contributing to the bending strength of elements. It does not
address the behaviour and resistance of frameworks and connections in fire, which is beyond
the scope of the book. In the Eurocodes, fire is classed as an ‘accidental situation of fire expo-
sure.’ For concrete structures, the assessment of fire and its consequential conformity is accord-
ing to Eurocode EC2, Part 1-2 (BS EN 1992-1-2 2004) and the UK NA (NA to BS 1992-1-2
2004) ‘General Rules - Structural Fire Design.’ The fire resistance of certain precast products
is covered in product standards such as BS EN 13369 for the ‘Common Rules’ (BS EN 13369
2013), and BS EN 1168 for hollow core slabs (BS EN 1168 2005). Double-tee units are classed
as ribbed slabs and, apart from some thermal contours in the tapered ribs given in BS EN 13224
(BS EN 13224 2004), are covered in BS EN 1992-1-2 as ‘beams’ (for the ribs) and ‘slabs’ for
the top flanges. It is important to note that product standards are intended to be used for the
product itself (beams, floor units) and not for the design of the completed building or floor slab.
The fire resistance of structural elements has the following three components:
1. R: fire resistance class for the load-bearing criterion – the capacity of an element to
resist loads in the fire limit state, such as bending moments MEd,fi ≤ MRd,fi where the fire
load is taken as dead load plus the quasi-permanent fraction of the live load, typically
ψ2 = 0.3 for offices and domestic loading, and the resistance is according to the reduced
strength of steel and concrete at an elevated temperature.
2. E: fire resistance class for the integrity criterion – otherwise known as the separating
criterion based on the wall thickness or slab depth of fire-resisting compartments,
including plastering, screeds, finishes, or other non-combustible layers, for example
a 150 mm deep concrete slab with 50 mm insulating layer plus a 75 mm sand/cement
screed would have an equivalent fire thickness of 225 mm.
3. I: fire resistance class for the insulation criterion – this is also a separating criterion where
the average temperature rise over the whole of the non-exposed surface is limited to 140°C,
and the maximum temperature rise at any point of that surface does not exceed 180°C.
Fire resistance is measured by the ability of a structure or elements to satisfy the aforemen-
tioned components for 30, 60, 90, 120, 180 or 240 min in standard fire exposure. Therefore,
if an element achieves M Ed,fi = M Rd,fi after say 100 min of fire exposure, its fire class is R 90.
If the element also has sufficient thickness/depth to satisfy that requirement after say 75 min
of fire exposure, its fire class is E 60, and is able to satisfy a separating criterion for 120 min,
its fire class is I 120. The element’s combined class is then REI 60.
The design for fire resistance can be satisfied either by
(i) member analysis in BS EN 1992-1-2, Sections 2 to 4, in which the criterion for the
fire action E Ed,fi ≤ fire resistance ERd,t,fi, where E represents shear, bending, bearing,
buckling, etc.
(ii) or indirectly by the use of tables in BS EN 1992-1-2, Section 5 giving the requirements
for member thickness or depth heff, breadth of column and beam bmin, breath of webs
bw and centroidal cover distance (called ‘axis’ distance a) to steel bars, satisfying fire
classes R and E.
For (i) BS EN 1992-1-2, Section 4 member capacity is based on the reduced compressive
strength of concrete, rebars and prestressing tendons, and the spalling (loss of cross section) of
concrete at elevated temperatures. BS EN 1992-1-2 Annexes B.1 and B.2 provide a systematic
approach to spalling breadths as well as the calculation model for MRd,fi. Tabulated data for
heff, bmin, bw and a are given in BS EN 1992-1-2 Section 5 for beams and slabs (Figure 2.17),
columns and walls (Figure 2.18). Axis distance a = cover plus link diameter, neglecting the 2Ф
radius bend at the corners of the links which often causes the main bars to rise up the section.
2.1.5.1 Columns
Method A Table 5.2a. Columns subjected mainly to compression, where efi = M0Ed,fi/N0Ed,fi ≤
0.15h and in braced structures, where the effective length l0,fi ≤ 3.0 m, making this very restric-
tive using pinned beam-to-column connections where l0,fi = storey height. Note that Table 5.2a
is based on (compression strength coefficient) αcc = 1, whereas the UK value is 0.85; this makes
the data slightly conservative. Values for bmin and a are summarised in Table 2.4 for columns
exposed to fire on more than one face for values of μfi = NEd,fi/NRd, where NEd,fi is the axial load
in a fire situation = Gk + ψ2Qk loads, where ψ2 = quasi-permanent factor (defined in Table 3.2),
Compression zone
h
a a a
asd
Side
bmin e.g. mesh with transverse
(a) (b) bars in first layer
Definition of breadth min, depth and axis distance in fire situation (a) beams, (b) slabs. (Based on
BS EN 1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2: General rules – Structural
fire design, BSI, London, UK., 2004, Section 5.)
a2 a1
a
Exposedℓtoℓreℓonℓthis
Axisℓof oneℓsideℓrequiresℓa1
bending
h
a
h
bmin
(a) (b) Planℓview
Definition of breadth min, depth and axis distance in fire situation (a) columns, (b) walls.
(Based on BS EN 1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2: General rules –
Structural fire design, BSI, London, UK., 2004, Section 5.)
and NRd is the axial load capacity at room temperature. Each load case must therefore be ana-
lysed under these conditions. Note that in situations where the dead and live loads are equal
and for offices ψ2 = 0.3, then μfi ≈ 0.5. Most precast columns are designed using 30 mm cover,
8–10 mm links and 20–32 mm main bars so that typically a = 48–56 mm. Thus, bmin ≥ 200,
300 and 350 mm for R 60, R 90 and R 120, respectively. Values for the least values of bmin and
a are summarised in Table 2.4. Linear interpolation is permitted in all the fire charts.
There are no reductions in dimensions for the use of limestone aggregates (clause 5.1.(2)
only for beams and slabs).
Method B Table 5.2b. Braced columns having slenderness λfi = lo,fi/i ≤ 30, where i = radius
of gyration √I/A and efi = M0Ed,fi/N0Ed,fi ≤ 0.25h, nfi = N0Ed,fl/0.7 Nu where squash load
Nu = Ac fcd + As 0.87 fyk, and steel ratio ω = As 0.87 fyk/Ac fcd. Additional frame analysis is
required in order to determine N0Ed,fl in a fire situation using Gk + ψ2Qk loads, so that the area
Fire resistance of columns
μ μ μ
R 60 200/25 200/36 250/46
300/31 350/40
R 90 200/31 300/45 350/53
300/25 400/38 450/40
R 120 250/40 350/45 350/57
350/35 450/40 450/51
R 180 350/45 350/63 450/70
R 240 350/61 450/75 –
Source: Adapted from BS EN 1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2:
General rules – Structural fire design, BSI, London, UK., 2004, Table 5.2a for ≤ 0.15
ω = 0.3 = 0.5 = 0.7

R 60 0.5 200/25 250/35 350/40
350/25 550/25
1.0 200/25 200/40 300/50
400/25 600/30
R 90 0.5 200/45 300/45 500/50
300/25 550/25 600/40
1.0 200/40 250/40 500/50
300/25 550/25 600/45
R 120 0.5 300/45 450/50 500/60
550/25 600/25 600/50
1.0 250/50 450/45 600/60
400/25 600/30
R 180 0.5 450/50 500/60 600/75
600/25 600/50
1.0 450/50 500/60 Note 1
550/25 600/45
R 240 0.5 550/55 600/70 Note 1
600/25
1.0 500/40 600/60 Note 1
600/30
General rules – Structural fire design, BSI, London, UK., 2004, Table 5.2b for ≤ 0.25
Note 1: min > 600 mm. Particular assessment for buckling is required.
of bars As is found prior to calculating the fire resistance in an iterative manner. Values for the
least values of bmin and a are summarised in Table 2.5 for ω= 0.5 and 1.0 as the most common
value of ω is for As/Ac of 4%, then ω = 0.04 × 435/(40/1.5) = 0.65.
Method C Tables C.3, C.6 and C.9. These tables provide the data for ω = 0.1, 0.5
and 1.0 using linear interpolation. Braced columns having slenderness λfi = lo,fi /i ≤ 80
R 60 30 200/40 450/50 550/50

450/25 550/30 600/40
40 250/40 500/40 600/60
500/25 550/35
50 300/45 500/50 Note 1
550/25 550/40
R 90 30 300/50 500/55 600/80
550/25 600/40
40 350/50 550/60 Note 1
550/35 600/50
50 500/45 600/60 Note 1
550/40
R 120 30 500/50 550/50 Note 1
550/40
40 500/55 550/60 Note 1
550/45 600/55
50 500/60 600/80 Note 1
600/45
R 180 30 550/55 600/75 Note 1
40 550/60 Note 1 Note 1
50 600/70 Note 1 Note 1
General rules – Structural fire design, BSI, London, UK., 2004, Table C.6 for ≤ 0.5
Note 1: min > 600 mm. Particular assessment for buckling is required.
and efi = M0Ed,fi /N0Ed,fi ≤ 0.5h under fire conditions. If e fi > 0.5h, the section can be considered
as a beam. Unfortunately, nfi = N0Ed,fl /0.7 Nu > 0.5 requires bmin > 600 mm as well as special-
ist treatment for buckling in fire, indicated by note 1. The data are for b or h ≤ 400 mm, but
no information is given otherwise. Values for the least values of bmin and a are summarised in
Table 2.6 for ω= 0.5 and λfi ratios ≤ 50 as λfi would typically not exceed 4000/(300/√12) = 45.
Example 2.4
Determine the fire resistance for a 300 × 300 mm column reinforced using 4 no. H25 bars
which is carrying an axial load Gk = 900 kN and Qk = 600 kN and bending moment
Gk = 40 kN and Qk = 27 kNm at room temperature. The column is in a braced frame
with l 0 = 3.0 m. Use fck = 40 N/mm 2 and f yk = 500 N/mm 2 . PSF γG = 1.25 (dead load) and
γG = 1.5 (live load). ψ2 = 0.3. Cover to 8 mm dia. links = 30 mm.
Solution
a = 30 + 8 + 25/2 = 50 mm. A s = 1963 mm 2
N Ed = 1.25 × 900 + 1.5 × 600 = 2025 kN; M Ed = 1.25 × 40 + 1.5 × 27 = 90.5 kNm
N 0Ed,fi = 900 + 0.3 × 600 = 1080 kN; M 0Ed,fi = 40 + 0.3 × 27 = 48.1 kNm
e fi = M 0Ed,fi /N 0Ed,fi = 48.1/1080 = 0.044
e fi /h = 44/300 = 0.147 < 0.15 use Table 2.4 {Table 5.2a}
NRd = 90000 × 40/1.5 + 1963 × 0.87 × 500 = 3253 kN (note that fcd = fck/1.5 not
0.85 fck /1.5)
μfi = N Ed,fi /N Rd = 1080/3254 = 0.33
μ μ μ
R 60 200/25 200/36 Pass
R 90 200/31 300/45 Pass
R 120 250 /40 350/45 Pass
R 180 350/45 350/63 Fail
Answer: Fire resistance = R 120.
Note: Interpolated values in italics.
Example 2.5
Repeat Example 2.4 carrying an axial load Gk = 900 and Qk = 600 kN and bending
moment Gk = 60 and Qk = 40 kNm at room temperature. The additional moment in this
column requires 4 no. H32 bars.
Solution
a = 30 + 8 + 32/2 = 54 mm. A s = 3216 mm 2
N Ed = 2025 kN; M Ed = 1.25 × 60 + 1.5 × 40 = 135 kNm
N 0Ed,fi = 1080 kN; M 0Ed,fi = 60 + 0.3 × 40 = 72 kNm
e fi = M 0Ed,fi /N 0Ed,fi = 72/1080 = 0.067
e fi /h = 67/300 = 0.22 < 0.25 use Table 2.5 {Table 5.2b}
Nu = 90,000 × 40/1.5 + 3216 × 0.87 × 500 = 3,800 kN (note that fcd = fck /1.5 not
0.85 fck /1.5)
n fi = N 0Ed,fl /0.7 Nu = 1080/0.7 × 3800 = 0.41
ω = A s 0.87 f yk /Ac fcd = 3216 × 0.87 × 500/90,000 × 40/1.5 = 0.58
ω
R 60 0.5 200/25 250/35 2
0.58 Pass
1.0 200/25 200/40
R 90 0.5 200/45 300/45
0.58 Pass
1.0 200/40 250/40
R 120 0.5 300/45 450/50
0.58 Fail
1.0 250/50 450/45
R 180 0.5 450/50 500/60
0.58 Fail
1.0 450/50 500/60
Answer: Fire resistance = R 90.
Note: Interpolated values in italics.
2.1.5.2 Unbraced columns

Tabular data are not included for unbraced columns in BS EN 1992-1-2. This is the case for
the simplified calculation presented in Annex B.3, which is a ‘zone method’ for the analysis
of columns with significant second-order effects. The subject has not yet been officially pub-
lished in a handbook to BS EN 1992-1-2 by The Office of the Deputy Prime Minister.
In cases where the structure is unbraced, or that portion of the structure being con-
sidered cannot be considered as braced by that part of the structure remote from the
fire, then the following method is available: For initial design, the background paper to
the UK NAs to BS EN 1992-1-2 states that, at the discretion of the designer, the tabular
methods may be used for general design and critical columns checked in accordance with
either the 500°C isotherm method or the zone method. The justification given is that no
distinction was made in BS 8110 between braced and unbraced columns in fire design
and that typically column sizes in BS EN 1992-1-1 and BS EN 1992-1-2 will exceed those
determined from BS 8110. Such an approach would only be unsafe where the members
in the fire zone provide the predominant means of structural stability and in such cases
advanced calculation methods and specialist advice may be required. In conclusion, the
tabulated methods adopted for braced column may, for the present time, also be adopted
for unbraced columns.
2.1.5.3 Walls
Walls are divided into non-load-bearing (partitions) and load-bearing (structural walls or
shear and infill walls, see Section 7.4). The fire resistance of partitions is only required to
meet the thermal insulation and integrity criterion I and E. There is no structural require-
ment R or axis distance a. Minimum wall thickness tmin should not be less than that given in
Table 2.7 (based on BS EN 1992-1-2, Table 5.3) and reduced by 10% using limestone aggre-
gates. To avoid excessive thermal deformation and subsequent failure of integrity between
the wall and slab, the ratio of clear height is l/t ≤ 40.
Load-bearing solid walls are required to meet the criterion REI in terms of tmin and a
according to Table 2.7 (based on BS EN 1992-1-2, Table 5.4) for walls exposed to fire on
one or two faces for values of μfi = N Ed,fi /N Rd = 0.35 and 0.7. tmin given in Table 2.7 may also
be used for plain concrete walls (see BS EN 1992-1-1, Section 12).
Fire resistance of walls
μ μ
REI 60 80 110/10 120/10 130/10 140/10

REI 90 100 120/20 140/10 140/25 170/25
REI 120 120 150/25 160/25 160/35 220/35
REI 180 150 180/40 200/45 210/50 270/55
REI 240 175 230/55 250/55 270/60 350/60
Source: Adapted from BS EN 1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2: General rules – Structural fire
design, BSI, London, UK., 2004, Tables 5.3 and 5.4
Values of a <25 will be governed by production methods or durability.
a Number of faces exposed to fire.
2.1.5.4 Beams
The fire resistance of reinforced and prestressed beams is required to meet the structural cri-
terion R (only) in terms of the minimum breadth of beam bmin, breadth of web bw and mean
axis distance amean according to BS EN 1992-1-2, Section 5.6. For nonrectangular beams,
such as tapered or I section beams, beam breadth is defined in BS EN 1992-1-2, Fig. 5.4. To
avoid spalling of shallow bottom flanges, the effective height (Figure 2.15c) is deff ≥ d1 + 0.5
d2 ≥ bmin, where bmin is according to Table 2.8. There are some other modifications neces-
sary to a for narrow webs and for shallow bottom flanges given in BS EN 1992-1-2, Exp.
5.10. The calculation for multiple bars amean = am = ΣA si ai /ΣA si (1 to 4) and at the side asd,m =
ΣA si ai /ΣA si (5 to 8) is as illustrated in Figure 2.19. The rebar in the middle of the bottom row
may be included in asd,m , although technically its influence is less than the outer bars and it
is shared with asd,m from the other side. When reinforcement consists of rebars and tendons
with different characteristic strength, A si should be replaced by A si f yki (or A si f pki).
Fire resistance of simply supported reinforced or prestressed beams
R60 120/40 160/35 200/30 300/25 100

R90 150/55 200/45 300/40 400/35 110
R120 200/65 240/60 300/55 500/50 130
R180 240/80 300/70 400/65 600/60 150
R240 280/90 350/80 500/75 700/70 170
Source: Adapted from BS EN 1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2: General rules – Structural fire
design, BSI, London, UK., 2004, Table 5.5.
Note: For prestressed beams, increase according to 5.2(5 to 8).
may be increased, and axis distance reduced by 10% using limestone aggregates.
min
a Class WA according to NA to BS EN 1992-1-2.
Not included
As4
As3
As2 a4 Axes for a
As1 a2 a3
a1
a5
a6 Not included in
a7 mean asd
a8
Axes for asd
Calculation of mean axis distance for multiple bars. (Based on BS EN 1992-1-2, Eurocode 2:
Design of concrete structures, Part 1-2: General rules – Structural fire design, BSI, London,
UK., 2004, Section 5.)
There is a major division of data between simply supported and continuous beams.
Options for the least values of bmin in combination with amean are summarised in Table 2.8
for simply supported reinforced concrete beams (BS EN 1992-1-2, Table 5.5). The axis
distance to prestressing tendons given in Table 2.8 should be increased according to BS EN
1992-1-2, clause 5.2(5) stated as follows: For prestressing tendons, the critical temperature
for bars is assumed to be 400°C and for strands and wires to be 350°C. This assumption
corresponds approximately to Ed,fi = 0.7 Ed, f p0,1k /f pk = 0.9 and γs = 1.15 (stress level σs,
fi /f p0,1k = 0.55). If no special check according to (7) is made in prestressed tensile members,
beams and slabs, the required axis distance a should be increased by
10 mm for prestressing bars, corresponding to θcr = 400°C

15 mm for prestressing wires and strands, corresponding to θcr = 350°C
The ‘special’ check according to 5.2(7) allows the increase in a, known as Δa, to be reduced
according to the ratio of the fire load E d,fi to the ultimate load Ed and the ratio of the area of
reinforcement required/provided, for example if M Ed,fi = 80 kNm and M Ed = 125 kNm, and
if the designed area of strands Ap = 280 mm 2 and 6 no. 9.3 mm strands = 312 mm 2 are pro-
vided, then the stress used to evaluate the critical temperature of reinforcement and hence
Δa would be σp = (1770/1.15) × (80/125) × (280/312) = 884 N/mm 2 . This procedure is also
applicable to slabs or other prestressed members in bending or tension.
The axis distance to the side of beam for the corner bars (or tendon or wire) of beams with
only one layer of reinforcement is a sd = a + 10 mm. No increase is required for values of bmin
greater than those in the 3rd column of options (from the left) in Table 2.8, for example for
R 60 if bmin = 200 mm then a sd = 30 + 10 = 40 mm, but if bmin = 250 mm then asd = 30 mm.
2.1.5.5 Reduced additional axis distance Δa

BS EN 1992-1-2, Fig. 5.1, reproduced here in Figure 2.20, shows the relationship between
the critical temperature θcr for rebars and prestressing tendons. For reinforced concrete and
prestressed concrete with bonded tendons, tensile and simply supported members in bend-
ing, Clause 5.2.(6 to 8) and Equations 5.2 and 5.3 together with clause 2.4.2 allow the axis
distance a in Table 2.8 to be reduced by Δa if θcr is different at 500°C as:
Δa = 0.1 (500 – θcr) mm (2.27)
where θcr is obtained from Figure 2.20. For prestressing tendons, curve 3 may be expressed
as follows:
q = - q q =
q = - q q = (2.28)
q = - q q <
where kp(θcr) = σp,fi /pyk(20°C) (2.29)
but σp,fi = (Ed,fi /E d) (pyk(20°C)/1.15) (Ap,required /Ap,provided) (2.30)
Then from BS EN 1992-1-2, clause 2.4.2, the ratio of the fire load/ultimate load
ηfi = E d,fi /E d is the least of code Expressions 2.5a or 2.5b for the following load combinations.
1
0.8
2
ks (θcr), kp (θcr)
0.6
3 Curve 1 : Reinforcing steel

0.4
Curve 2 : Prestressing steel
(bars : EN 10138 - 4)
0.2
Curve 3 : Prestressing steel
(wires & strands EN 10138 -2

0 and -3)
0 200 400 600 800 1000 1200
θcr (°C)
Critical temperature of reinforcing and prestressing steel θ corresponding to the reduction fac-
tor (θ ) = σ / (20°C) for rebars or (θ ) = σ i/ (20°C) for prestressing tendons. (From BSEN
1992-1-2, Eurocode 2: Design of concrete structures, Part 1-2: General rules – Structural fire
design, BSI, London, UK., 2004, Figure 5.1; Courtesy of British Standards Institute.)
Exp. 2.5a. ηfi = (Gk + ψ2 Qk)/(γG Gk + γQ ψ0 Qk) (2.31)
Exp. 2.5a. ηfi = (Gk + ψ2 Qk)/(ζγG Gk + γQ Qk) (2.32)
where γG = 1.35, ζγG = 1.25 and γQ = 1.5. Refer to Table 3.2 for values of ψ0 and ψ2 .
BS EN 1992-1-2, clause 2.4.2(4) states that as only thermal deformations across the cross
section need be considered, and axial and in-plane expansions are ignored, then ηfi may be
taken as M Ed,fi /M Ed.
Example 2.6
Determine the mean axis distance a required for a 300 mm wide prestressed concrete beam
having fire resistance for 120 min. The beam is subjected to dead uniformly distributed
load (UDL) Gk = 40 kN/m and live UDL Qk = 30 kN/m. The area of tendons required
is known to be Ap = 1000 mm2. Use 12.5 mm diameter strands of 94 mm2 per strand with
pyk = 1770 N/mm2. Use office loading with ψ2 = 0.3 and ψ0 = 0.7.
Solution
Ap = 1000 mm2. No. of strands required > 1000/94 = 10.6. Use 11 no. × 94 = 1034 mm2
Ap,required /Ap,provided = 1000/1034 = 0.967
E d,fi = 40 + 0.3 × 30 = 49 kN/m
Ed = max{1.35 × 40 + 0.7 × 1.5 × 30; 1.25 × 40 + 1.5 × 30} = {85.5; 95.0} = 95 kN/m
ηfi = 49/95 = 0.516
σp,fi = 0.516 × (1770/1.15) × 0.967 = 768 N/mm 2 .
kp(θcr) = 768/1770 = 0.434 (note this can be reached by 0.516 × 0.967/1.15 = 0.434)
Then if kp(θcr) = 0.1 to 0.55, θcr = 594.4 – 444.4 kp(θcr) = 594.4 – 444.4 × 0.434 = 402°C
Δa = 0.1 (500 – 402) = 9.8 mm
From Table 2.8 for R 120, use bmin = 300 mm, a = 55 + 15 – 9.8 = 60.2 mm
The first row of strands (5 no.) could be placed at a = 40 mm, 2nd and 3rd rows (3 each)
at 80 and 120 mm giving a = (5 × 40 + 3 × 80 + 3 × 120)/11 = 72 mm > 60.2 mm.
2.1.5.6 Floor slabs

The fire resistance of reinforced solid, ribbed, waffle, flat, one-way and two-way slabs, etc.
is given in BS EN 1992-1-2, Section 5.7. Common Rules for Precast Elements BS EN 13369
clause 4.4.4.2 (BS EN 13369 2013) states that tabulated data can be found in EN 1992-1-2,
but when applicable complementary rules may be given in product standards. The fire resis-
tance of specific precast floors is covered in EC Product Standards, such as hollow core units
given in Annex. G in BS EN 1168:2005+A3:2011 (BS EN 1168 2005).
Floor slabs are required to meet the criterion REI in terms of hs given in BS EN 1992-1-2,
Fig. 5.7 and Table 5.8 ensure adequate separating function (Criterions E and I). If load-
bearing function (Criterion R) is required only the necessary slab thickness h1 assumed for
design to BS EN 1992-1-1 may be taken. Values for hs and axis distance a are according to
Table 2.9 for simply supported one-way spanning slabs and for the top flanges of double-tee
units (based on BS EN 1992-1-2, Table 5.8).
The fire resistance of prestressed concrete hollow core floor units (Figure 2.14) is a
function of depth h and mean axis distance a. Figure 2.21 shows fire-damaged soffit of
some 150 mm deep hollow core units showing exposed 5 mm diameter wire tendons.
Depth h corresponds to the minimum solid floor thickness given in Table 5.8 of BS EN
1992-1-2 and has been calculated using a conversion equation for hollow core units as
te = h√ζ, where h = actual depth and ζ = ratio of solid material (including infilled joints)
to the whole. This explains the differences of depth required for solid and hollow core
slabs in Table 2.9, for example a certain 150 mm deep hollow core unit has a fire class R
60 compared with a solid slab R 180. This is because with the voids occupying 40% of
Total thickness and axis distance for solid plank and hollow core units and composite slabs
required for fire resistance
R30 80 130 10
R60 80 130 20
R90 100 160 30
R120 120 200 40
R180 150 250 55
For prestressed slabs, increase according to BS EN 1992-1-2, clause 5.2(5 to 8).
Fire damaged soffit of prestressed hollow core floor units showing two exposed prestressing
wires. The width of the photograph scales about 1 m. The line in the middle of the photograph is
the joint between two units. Blackened remains of timber ceiling battens are seen on the right.
the area ζ = 0.6. A solid slab of √0.6 × 150 = 116 mm would be class R 90. The data in
BS EN 1168, Table G.1 are for siliceous/gravel aggregates. Granite, sandstone, greywacke
perform no worse. If calcareous or dolomitic limestone is used, the dimensions h and a are
increased by 10%. A 150 mm deep hollow core unit made of limestone aggregates with
h = 1.1 × 150 = 165 mm is class R 90.
The producer specifies the mix design according to BS EN 206 (BS EN 206 2013) and
the quantities of cement, aggregates (coarse and fine), water and admixtures & additives.
The producer will specify the early strength gain expected. The quantity of mixing water,
including the use of superplasticisers and self-compacting admixtures, is the minimum
required to ensure the correct workability and strength intended for purpose. If a new fam-
ily of concrete mixes is specified, trial mixes will be carried out to the satisfaction of the
producer to demonstrate early handling and 28-day strengths.
Unlike previous British codes, environmental exposure is directly linked to concrete
strength, and vice-versa, in addition to the usual mix design parameters of the water/
cement ratio (w/c) and cement type and content, and nominal cover (Cnom) to reinforcement.
Exposure classes for reinforcement corrosion are induced by
Carbonation (XC1 to XC4)

Chlorides (XD1 to XD3)
Chlorides from sea water (XS1 to XS3)
Freeze/thaw and chemical attack (XF1 to XF4, and XA1 to XA3)
The relationship between fck and C nom is given in:
BS 8500-1:2015 Tables A.4 (50 years’ life) and A.5 (100 years).

BS EN 1992-1-1 with the UK NA to BS EN 1992-1-1, Tables NA2 and NA3.
Note that the cover distances for exposure XS1 to XS3 have increased by around 30 mm in the
2015 revision of BS 8500-1. For the most severe XS3, the minimum concrete grade using CEM I
is C45/55 with Cnom = 85 mm, although this reduces to 50–55 mm using some blended cements.
Referring to BS EN 1992-1-1, clauses 4.4.1.1(1) and 4.4.1.3(3), nominal cover Cnom = Cmin +
ΔCdev. Where cover is regulated by quality control in the factory, ΔCdev may be taken as
5 mm. ∴Cnom = Cmin + 5 mm. Table 2.10 gives Cnom according to exposure class for three
grades of concrete from 28 to 45 N/mm2 cylinder strength, and Table 2.11 gives the minimum
water/cement ratio for the same. The minimum requirements for grade strength (N/mm2),
type CEM 1 cement content (kg/m3) and maximum water/cement ratio, for 50 years design
life, are given in Table 2.12 for exposure up to XD3.
Relationship between exposure classes, concrete strength and nominal cover for reinforced
and prestressed concrete using cement CEM I
≥
Inside enclosed buildings, low humidity. XC1 20 20 20
Permanently wet.
Long-term contact with water, not containing XC2 30 30 30
chlorides.
Inside building, moderate and high humidity. XC3/4 35 25 25
External, sheltered from rain (ground floors).
Cyclic wet and dry.
Moderate humidity. ≤10 m from carriageway. XD1 40 30 30
Slight chloride conditions.
Airborne chlorides (open car parks).
Wet, rarely dry. Immersed in water containing XD2a 45 35 35
chlorides.
Cyclic wet and dry. ≤10 m from carriageway. XD3a 45 50 45
Bridges parapet edge beams. Reinforced
pavements and car park slabs.
Airborne salt. Not direct contact to sea water. XS1a 70 55 50
Surfaces in coastal areas.
Permanently submerged. XS2a — 70 65
Marine permanently wet. Below mid-tide level.
Spray containing chlorides. Marine tidal, splash and XS3a — — 85
spray, and upper tidal zones. Exposed soffits
above sea water
Freeze/thaw. XF1-4b
Chemical attack. XA1-3b
Note a. Or concrete grade C28/35 with specified Note b. Refer to specialist
air-entrained admixture may be used with certain literature, e.g. Annex A BS8500-1
blended cements CEM II to CEM IV. Table A.4
Source: According to BS 8500-1, Concrete. Complementary British Standard to BS EN 206-1, Part 1. Method of specifying
and guidance for the specifier, BSI, London, UK., 2015.
Relationship between exposure classes, concrete strength and minimum

water/cement ratio for 50 years concrete using cement CEM I
≥
XC1 0.70 0.70 0.70
XC2 0.65 0.65 0.65
XC3/4 0.60 0.45 0.45
XD1 0.60 0.45 0.45
XD2 0.55 0.40 0.40
XD3 — 0.40 0.35
XS1 0.55 0.40 0.35
XS2 — 0.40 0.35
XS3 — — 0.35
Source: According to BS 8500-1, Concrete. Complementary British Standard to BS EN 206-1,
Part 1. Method of specifying and guidance for the specifier, BSI, London, UK., 2015.
Mix design requirements using CEM I
C20/25 240 0.70

C25/30 260 0.65
C28/35 280 0.60 300 0.60
320 0.55
C32/40 300 0.55 320 0.55
340 0.50
C35/45 360 0.45 360 0.45
C40/50 340 0.45 360 0.45
380 0.40
C45/55 380 0.35
Source: According to BS 8500-1, Concrete. Complementary British Standard to BS EN 206-1,
Part 1. Method of specifying and guidance for the specifier, BSI, London, UK., 2015.
Mortar is a sand–cement mix used for small infill and semi-dry packing of joints. Mix
design is according to BS EN 206 as quantities of cement, fine aggregates, water and admix-
tures & additives. The sand is typically graded at 0/4 mm but may include 6 mm pea gravel
for joints wider than 50 mm. BS EN 13139 (BS EN 13139 2002) lists the preferred sizes of
aggregates for mortar as (min/max) 0/2 mm, 0/4 mm, 2/4 mm and 2/8 mm. Sieve sizes and
other information is according to PD 6682-3 (PD 6682-3 2003). The strength of mortars
and grouts in compression is generally not less than the parent material at the joint, but it
may be C25/30 in other cases such as longitudinal joints between slabs.
Grout is either a cement–water mix or a mortar mix used for infilling at joints and gaps.
Grout may be used as dry pack but may also contain admixtures such as non-shrink or
expansive agents, latex bonding, or be a polyester grout or equivalent, added according to
manufacturer’s instructions.
The function of grout is either ‘structural’ or ‘non-structural’ infill. Structural grout

includes the protection of precast elements or connections in which its integrity is nec-
essary to ensure the correct structural function, including fire resistance and durability
protection. Control cubes are taken for structural grout. Non-structural grout serves no
such function.
The workability of mortar is classed as ‘very low’ with no slump. The workability of grout
is as specified on construction drawings or in the method statement and may be ‘high’ with
100–150 mm slump.
Dry pack mortar or grout may be used where it is not possible to pour grout. The grout
has a water/cement ratio of about 0.25–0.30 and is therefore mouldable in the hand. It is
placed by hand and trowel and hammered into position, which removes free air. It may be
either ‘structural’ or ‘non-structural’ as mentioned earlier, but it is not possible to make
control cubes.
EN 206 clause 5.1.3 states that normal-weight coarse aggregates from natural land
resources conform to BS EN 12620 (BS EN 12620 2002). Sea dredge aggregates are
not used. Moisture content is measured according to BS EN 1097 Part 6 (BS EN 1097-6
2000). Samples should not be taken from the base of the stock pile. BS EN 206 clause
5.5.2 states that normal-weight aggregates should have an oven-dry density between
2000 and 2600 kg/m3. BS EN 206 clause 5.2.7 states that aggregate chloride content is
≤ 0.2% by mass of cement. Chloride content is certified by the aggregate supplier, both
in bulk or in bags.
Annex E of BS EN 13139 (BS EN 13139 2002) prescribes minimum test frequencies for the
various properties and the appropriate test methods. Aggregates are checked for the reactive
silica content, which may cause alkali silica reactions with cement, sothat they remain below
a threshold value that has to be carefully experimentally assessed by the aggregate producer.
If this is not satisfied, low alkali cement with less than 0.75% alkalis may be used such that
the total alkali content of the concrete is less than 3 kg/m3.
Recycled concrete aggregates (RCAs) may be used as a substitute for natural aggregates up
to a maximum replacement of 20% by weight of coarse aggregates not passing a 4 mm sieve.
Waste hollow core floor units, Figure 2.22, produce a well-graded coarse recycled concrete
coarse aggregate (RCCA) but a very fine recycled concrete fine aggregate (RCFA), too fine to
be reintroduced in machine cast units. The RCCA is stockpiled as shown in Figure 2.23; note
that the angle of repose shows that this is as well graded and shaped as a natural aggregate.
RCFA can be used in wet-casting if not passing a 1 mm sieve. Recycled fine aggregates as dust
are not used. The water absorption of all RCA is determined according to BS EN 1097 Part 6
(BS EN 1097-6 2000) under the surface-saturated dry conditions. The chloride content of all
RCA is determined according to BS EN 1744 Part 1 (BS EN 1744-1 2009), clause 7, and is
not greater than 0.2% CaCl content by weight. Recycled crushed brick and mortar, or any
other crushed material of unknown parentage, is not used as replacement aggregate.
Waste or rejected hollow core slabs prepared for crushing towards recycled concrete aggregate.
Stockpiled recycled concrete aggregate, the angle of repose shows that the RCA is as well
graded and shaped as the natural aggregate. (Courtesy of Bison Manufacturing, Swadlincote, UK.)
Admixtures comply with BS EN 934 Parts 1 to 3 (BS EN 934-1 2008, BS EN 934-2 2009,
BS EN 934-3 2009). Dosages are not permitted to be within the range given in proprietary
data sheets provided that the performance of the concrete is verified by the manufacturer or,
ready-mix supplier. Calcium chloride is not used as an admixture.
EN 206 clause 5.1.5 states that air entrainment agents and superplasticisers, excluding
SCC, may be used to comply with BS EN 934-2, clause 4.2 as a water-reducing agent where
total air content <6%. Superplasticisers should comply with the testing standard BS EN 480
Part 1:2006+A1:2011 (BS EN 480-1 2006).
Filtration of cement slurry for recycled water, at Samsung Precast, South Korea.
Mixing water conforms to BS EN 1008 clause 3.1 (BS EN 1008 2002) and should be potable
from land sources. Sea water is not permitted. Water reclaimed from cement slurry must
be filtered, and the cement content is calculated and known from the specific gravity of the
filtered slurry. New machinery for this process is shown in Figure 2.24. Crushed ice may be
used to replace 50% of the mixing water, or according to the manufacturer’s specification.
The maximum particle size for crushed ice is not more than about 20 mm. No particles of
crushed ice are visible in the final mix.
Precast concrete elements can, if necessary, be heavily reinforced because they are cast hori-
zontally, although BS EN 1992-1-1 does not recognise this. The code permits up to 4% of
the cross section generally and 8% at laps, although this amount is rarely used in favour of
higher concrete strengths.
High tensile hot rolled ribbed bar (HT rebar) designated ‘H’ is used in 95% of cases,
even in shear links where mild steel would be suitable. The small cost difference com-
pared to the additional strength, viz. 500 vs 250 N/mm2, and the need for consistency
of habit when assembling cages, makes it more economical. Tying wire is more secure
around ribbed bar making the cage more robust. Black annealed tying wire of 17 gauge
(1.4 mm) or galvanised iron wire of similar specification is used. Wire loop ties may be
used as 16-gauge (heavy duty) or PVC coating (for use with polymeric or epoxy-coated
steel). For stainless steel rebars, tie wire is an 18 gauge stainless steel annealed wire
(1.2 mm).
High tensile steel bars conform to BS EN 10080 (BS EN 10080 2005) as B500A, B500B
and B500C. The technical specification according to the UK NA gives values of:
Re = yield strength = 500 N/mm 2 (note that strengths of 600 N/mm 2 are available in some
countries)
R m /Re = ratio of tensile strength/yield strength (also known as k = f tk /f yk) = 1.05, 1.08 and
1.15 to 1.35 for classes A, B and C in BS EN 1992-1-1:2004, Annex C.
Agt = percentage total elongation at maximum force = 2.5% (εuk = 25 × 10 –3), 5.0% and
7.5%, respectively.
The standard also gives fatigue strength, bend performance, weldability, bond strength,
shear and weld strength for use in welded fabric, tolerances and dimensions. B500 6 mm
diameter reinforcing steel does not comply with BS EN 1992-1-1 with respect to ductility.
Reinforcement shape codes are according to BS 8666 (BS 8666 2005) and details according
to ISO 3766 (BS EN ISO 3766 2003).
The actual and idealised stress vs strain relationships for reinforcing bars are given in
BS EN 1992-1-1, Figs. 3.7 and 3.8. Typically, k = f tk /f yk = 1.08. The design value for bars
f yk /γs , where γs = 1.15 for ultimate and 1.00 for accidental limit states, from NA to BS EN
1992-1-1, Table 2.1N. Then at ultimate f yd = 0.87 and f yk = 435 N/mm 2 for H500 rebar.
The bi-linear idealised and design curves are shown here in Figure 2.25a. There are two
design options known as the ‘inclined’ or ‘horizontal’ branches. The horizontal branch
(having a constant value f yd = 435 N/mm 2 for H500 rebar) is more convenient for the
design of reinforced concrete sections and has no limitation of strain in under-reinforced
sections. The drawback with the inclined branch is that even if the strain in the bars at
ultimate is say 0.01, the design stress f yd = 449 N/mm 2 , that is only 14 N/mm 2 greater than
435 N/mm 2 . Secondly, the strain is limited to εud = 0.02.
Mild steel is often used for projecting loops etc. because it is easier to hand bend on site.
Bar diameters commonly used are 8 and 10 mm for column links, 10 and 12 mm for beam
stirrups and other distribution or anti-crack bars, and 16, 20, 25, 32 and 40 mm for main
Idealized and design stress vs strain relationships. (a) Reinforcing bars (subscript ) ( )
Idealized and design stress vs strain relationships. (b) prestressing tendons (subscript ).
(Based on BS EN 1992-1-1, Eurocode 2: Design of concrete structures. Part 1-1: General
rules and rules for buildings, BSI, London, UK., February 2014, Figures 3.8 and 3.10.)
(c) Design stress vs strain diagram for standard helical strand using = 1770 N/mm2.
See Section 4.3.4 and Figure 4.23b for explanation of and ε .
flexural bars. Plain round mild steel bar of yield strength 250 N/mm 2 does not conform to
BS EN 10080 and BS EN 1992-1-1, clause 3.2.2(3).
Welded fabric, or ‘mesh’, is used in flat units such as slabs and walls. It is also used to rein-
force structural toppings to floor slabs on-site. The popular mesh size for flat panels, walls
etc. is A142 or A193 (6 or 7 mm bars at 200 mm centres in both directions). A rectangular
mesh C283 (6 mm at 100 mm centres × 5 mm at 400 mm centres) is often used in one-way
spanning units such as the flanges of double-tee slabs. For rebar and mesh, the character-
istic strength f yk = 500 N/mm 2 . The design strength f yd = f yk /γs = 500/1.15 = 438 N/mm 2 .
Young’s modulus is taken as 200 kN/mm 2 . There is no European standard for welded
mesh; it still conforms to BS 4483 (BS 4483 2005). Note that A98 and B196 mesh does not
conform to the strength of welded joints required in BS EN 1992-1-1, clause 3.2.5.
Stainless steel for reinforcement conforms to BS 6744 (BS 6744 2001). Other requirements
are according to BS EN 10088 Part 5 (BS EN 10088-5 2009). Yield strength is specified as
Re = 500 N/mm 2 and R m /Re = 1.10. For non-chloride exposure, the grade is at least 1.4301
(was BS 6744 grade 304), otherwise grades 1.4401 or 1.4436 are used (was BS 6744 grade
316). Stainless steel mesh fabric conforms to BS 4483:2005.
Two main types of steel are used for pretensioning – plain or indented (or crimped) wire, and
7-wire helical strand. Prestressing bars such as Macalloy are not common in precast factories.
Figure 2.26 shows an array of 12.5 mm diameter 7-wire strand in a prestressed beam, and
Figure 2.27 shows the same in cross section in a hollow core floor unit. Note the excellent
compaction of concrete around the strand, particularly in the crevices between the outer wires.
This is vital for good transfer of stress to the concrete. Prestressing tendons conform to BS
EN 10138 Parts 1 to 3 (BS EN 10138-1 2015, BS EN 10138-2 2015, BS EN 10138-3 2015)
(parts 1 and 3 are under ‘technical review,’ and part 2 has been ‘accepted’ in December 2015).
The standard 7-wire helical strand of 9.3, 10.9, 12.5 and 15.7 mm diameter, the standard
3-wire helical strand of 6.5 mm diameter and the super strand (Dyform) of 9.6, 12.9 and
15.2 mm diameter may be used. The 5% characteristic strength is taken as fpk = 1770 N/mm2,
except for 7 mm wire fpk = 1670 N/mm2. Super-stress strand is also available where fpk = 1770
to 1860 N/mm2, with reported strengths of up to 2100 N/mm2, and the relaxation is 2.5%.
The modulus of elasticity Ep is taken as 195 kN/mm2 for a helical strand and 205 kN/mm2
7-wire helical strand of 12.5 mm nominal diameter.
Cross section of 7-wire helical strand in hollow core units.
for an indented wire. The elongation at the breaking load (obtained from BS 5896*) is taken
as 3.5% or εuk = 35 × 10–3 > 0.0222 used in BS EN 1992-1-1 as shown in Figure 2.25b). For
hollow core floor units, the maximum diameter of the strand is 16 mm and the wire is 11 mm
according to BS EN 1168:2005+A3:2011 (BS EN 1168 2005), clause 4.1.1.1.
The choice of tendon is often a matter of the arrangement of tendons and the correct dis-
tribution of pretensioning force in a section. Large tendons should not be placed in thin wall
sections–to avoid localised splitting and bond failure, the edge cover to tendons is at least
twice the diameter. For this reason, a helical strand is preferred in larger units or where the
level of prestress is high.
A helical strand comprising 3 or 7 hot rolled wires, and individual hot rolled plain or
Belgian indented wire, is used according to BS EN 1992-1-1, clause 3.3.2.(4) as low relax-
ation Class 2. There are immediate and long-term losses of force called ‘relaxation’ due
to the stress-relieving heat treatment process. Class 2 tendons have 2.5% low relaxation,
meaning that the final stress after 1000 h relaxation is 97.5% of the original, although man-
ufacturers may quote characteristic relaxation as low as 1.6% 1000 h. Class 1 relaxation is
greater but is rarely used today. When calculating relaxation losses, BS EN 1992-1-1 splits
the losses into immediate and long-term as discussed in Section 4.3.2. Plain and indented
wire is of 5 and 7 mm diameter, and ribbed wire of 9 mm diameter may be used. Specified
strengths and elastic modulii are given in Table 2.13.
The actual and idealised stress vs strain relationships for tendons are given in BS EN
1992-1-1, Figs. 3.9 and 3.10. The design value is f pk /γs, where γs = 1.15 and 1.00 as before.
The bi-linear idealised and design curves are shown here in Figure 2.25b. Adequate ductility
in tension may be assumed if f pk /f p0,1k ≥ k = 1.1 from clause 3.3.6(7) and the UK NA, then
f p0,1k /0.9 f pk. The inclined branch is used for prestressing tendons as the additional strength
between f pd and f pk /γs is worthwhile, even if the strain is limited to 0.02. Figure 2.25c shows
the design values for strand, for example f pk = 1770 N/mm 2 . Then f pd = 0.9 f pk /1.15 = 1385
N/mm 2 and f pk /γs = 1539 N/mm 2 , an increase of stress of 154 N/mm 2 for an increase in a
* Although this standard is being superceded by BS EN 10138, the material properties of the strand supplied by
the manufacturers still conform to Tables 4 to 6 of this standard.
Types and specification of pretensioning tendons
Wire 5.0 19.6 34.7 1770 205

7.0 38.5 64.2 1670
Standard strand 9.3 52 92 1770 195
10.9 71 126 1770
12.5 93 165 1770
15.2 138 244 1770
Super strand 12.9 100 186 1860 195
15.7 150 265 1770
Drawn strand 12.7 112 209 1860 195
15.2 165 300 1820
strain εuk – εLOP = (0.0222 – 1385/195,000) = 0.0151. This gives f pmax = 1517 N/mm 2 , an
additional 132 N/mm 2 or 10% extra. The design may also be based on the actual stress vs
strain relationship, if this is known, with stress above the elastic limit reduced analogously
with BS EN 1992-1-1, Fig. 3.10.
Structural steelwork sections are used in many types of precast elements, especially at the
connections. These include rolled rectangular and square hollow sections (RHS, SHS), solid
billets, channels and angles, plates and welded tees etc. Details of how these are used in
practice are given in Sections 9.4 and 10.2. Structural sections such as universal beams and
columns (UB, UC) may be cast into precast elements to enhance strength where the rein-
forced concrete capacity is exhausted. However, this may have severe cost implications that
must be carefully examined.
Structural steelwork conforms to BS EN 10025 Part 1 (BS EN 10025-1 2004) and is
designed according to BS EN 1993 Part 1-1 (BS EN 1993-1-1 2004). The grade of hot rolled
structural steel is S275 (f y /fu = yield/ultimate = 275/430 N/mm 2), S355 (355/510) or S450
(440/550) according to BS EN 10025 Part 2 (BS EN 10025-2 2004) (these used to be known
as grades 43, 51 and 55, respectively). The grade of rectangular hollow sections is S275H
(275/430) or S355H (355/510) according to BS EN 10210-1 (EN 10210-1 1994). Its specifi-
cation includes product analysis, carbon equivalent, yield strength, tensile strength, stress v
strain data and elongation. Steel strength is according to Table 2.14.
Where steel with improved through-thickness properties is necessary according to BS EN
1993 Part 1-10 (BS EN 1993-1-10 2005), for example for beam–column connections made
from full steel billet/sections, steel according to the required quality class in BS EN 10164
(BS EN 10164 2004) is used.
Hot dipped galvanised steel is used for exposed connections, usually of secondary struc-
tural significance, such as dovetail channels for brick ties. The basic plate is grade S275 steel,
and grade S355 is used in the more highly stressed plates.
Exposed steelwork is cleaned by blasting to BS EN ISO 8501 Part 1 (BS EN ISO 8501-1
2007) to Swedish standard SA2, where most of the mill-scale rust and paint, etc. are removed
and any remaining is very well adhered, or may be by hand-powered tools to standard ST2,
Yield and ultimate tensile strength of steel base plates
S235 235 360

43 S275 275 430
50 S355 355 510
S420 420 520
S450 450 550
Design the same as yield strength = /γ 0 .γ 0 = 1.0.
a Included for comparison of BS notation.
with poorly adhering rust, mill-scale, etc. being removed, leaving surface contamination
that is well adhered. Steelwork entirely cast in concrete is cleaned by blasting to ISO 8501-1
to standard SA1, where poorly adhering mill-scale, rust and old paint and foreign matter
are removed, but well-adhered contaminants remain. No primer is used. Boundaries with
exposed steelwork are shown on production drawings.
Rolled steel sections and bent or flat steel plates are welded to form steel connectors in many
highly stressed support situations where direct contact between concrete surfaces is to be
avoided. Welding conforms to BS EN 1011, Part 1 (BS EN 1011-1 2009). Welding equip-
ment conforms to BS EN 60974, Part 1 (BS EN 60974-1 2005) and may use proprietary
plant such as Murex, Thermadyne, Rand, Migatronic, Cebora, etc. This BS was published
in January 2006 but is currently being revised.
Welding electrodes conform to BS EN 1011 Parts 2 and 3 (BS EN 1011-2 2001, BS EN
1011-3 2000). Rod sizes should vary between 2.5 and 6 mm. Welded electrodes are mostly
grade E275. When used to join grade S275 steel (previously known as grade 43), this gives
the yield strength of the weld as fyw = 275 N/mm2 and ultimate tensile strength as fuw = 430
N/mm2. When used in combination with grade S355, steel grade E355 electrodes give fyw and
fuw = 355 and 510 N/mm2. The values are divided by γM2 = 1.25, so that the design strengths
are pywd = 220 and 284 N/mm2, respectively. Intermittent fillet welds are rarely used as weld
lengths tend to be short. Weld strength is according to Table 2.15.
Welding of ferrous reinforcing bars and ferrous cast in fixings such as plates, rolled and
hollow steel sections use either metal inert gas (MIG) welding or metal active gas (MAG)
welding. The welding electrode rod for a high tensile bar is a low hydrogen type to cover a
wide range of applications for welding carbon manganese and low alloy steels.
Yield and ultimate tensile strength of welds
S235 188 235 360

S275 220 275 430
S355 284 355 510
Design = / 2 . 2 = 1.25.
Welding of reinforcing steel bars to structural steel in load-bearing joints is in accordance

with BS EN ISO 17660 Part 1 (BS EN ISO 17660-1 2006). The processes used are shielded
metal arc, gas metal arc. The welding electrode rods for B500 rebar is low hydrogen type
electrodes to cover a wide range of applications for welding carbon manganese and low
alloy steels. Spot welding of non-load-bearing bars is according to BS EN ISO 17660 Part 2
(BS EN ISO 17660-2 2006).
Butt and fillet welds are fully fused fillet welds to make lap joints, corner joints, and
T joints, having a convex profile and may be either continuous or intermittent. Rebar T-joints,
that is bars of equal diameter crossing at right angles, are a flared V-groove weld. Butt
welds may be used along a single edge in a single plane. The throat thickness of the weld,
run-out lengths, under cutting, cracks and inaccuracies of position are inspected for com-
pliance, recorded and if necessary rejected. When thicker materials are welded, the edges
are bevelled to form a single or double V-groove. No backup weld or backing strip is used.
Preheating is not required, and postweld heat treatment is normally not required unless
specified on the production drawings. Site welding is carried out using an intermediate rein-
forcing bar that is provided in V-grooved butting joints. The diameter of the bar is equal to
0.7 × the depth of V-groove.
Cast-in fixings, such as threaded sockets, lifting loops and sockets, spherical head anchors,
rotating anchors, flat steel, double wall and tail anchors, etc. are used according to manufac-
turers’ instructions, including edge distances, depth of embedment, anchorage reinforcement,
concrete strengths, etc.
Lifting systems are designed according to their intended use and the mode of lifting,
for example tilting, flat, vertical, number of lifters and angle of lifting chains, concrete
strength, element dimensions, type of lifting equipment, speed of lifting and ground condi-
tions. Spherical head lifting anchor and plate anchors are available in steel minimum grade
S355 or stainless steel 304 or hot dip spun galvanized. Anchors are manufactured, batch-
checked and tested by an independent testing house. Lifting loops are manufactured using
galvanised steel wire of minimum grade S500 and supplied with anchorages. Polypropylene
loops may be used where steel is not appropriate. Care is taken not to pass these loops over
sharp edges. Anchors have a minimum factor of safety of 3. The factor of safety against con-
crete failure is taken as 2.5. Bespoke lifting devices cast into concrete may be designed and
manufactured by the manufacturer or frame erector. Load tests can be carried out according
to BS 5080 Part 1 (BS 5080-1 1993) and evaluated according to BS EN 1990 Part 1 (BS EN
1990-1 2002).
Metric bolts, set screws and nuts conform to BS 3692 (BS 3692 2001). These are current
documents with no BS EN. The minimum diameter for a structural bolt or threaded rod is
12 mm. Clearance holes should not be more than 2 mm greater than the nominal bolt or rod
diameter. Holes should not be enlarged.
Mating surfaces are cleaned to standard SA2, where most of the mill-scale rust and paint
etc. are removed and any remaining is very well adhered, and (excluding high-strength
friction grip [HSFG] bolts) receive zinc or lead paint or similar priming unless the fixing is
to be cast in concrete.
Strength grades for carbon steel are preferably limited to Grade 4.6, 5.6, 6.8, 8.8 (both ≤
16 mm and >16 mm) and 10.9. Grade strength of nuts is 4, 5, 6, 8 and 10, respectively. Black
Yield and ultimate tensile strength of holding down bolts
4.6 192 240 400

5.6 240 300 500
6.8 384 480 600
8.8 512 640 800
10.9 720 900 1000
Design = /γ 2 .γ 2 = 1.25. Shear = 0.6 .
Code, e.g. 4.6: 4 means = 400 N/mm2; 6 means = 0.6 × 400 = 240 N/mm2.
bolts of grade 4:6 (pybk = 240 N/mm 2 in tension and in shear = 0.6 × 240 = 144 N/mm 2) and
8:8 (pybk = 480) are used in many connections. Bolt strengths are given in Table 2.16.
HSFG bolts are used in special circumstances where the integrity and safety (both tem-
porary and permanent) of connections made with ordinary bolts in clearance holes cannot
be guaranteed. HSFG bolts conform to the UK NA to BS EN 1993 Part 1–8 (NA to BS EN
1993-1-8 2005) using grade 8.8, 10.9 and 12.9. Preloading of HSFG bolts conforms to BS
EN 14399 Part 1 (BS EN 14399-1 2005). HSFG bolts are tightened using power tools show-
ing specified torque or tightened by hand using spanners according to the load-indicating
washer.
Cast-in sockets do not have a BS EN but are specified according to manufacturer’s details.
The grade may be 4.6 or 8.8 and may be stainless steel or bright zinc plated (ref UK manu-
facturer) to BS 7371 Part 12 (BS 7371-12 2008). Cast-in channels do not have a BS EN but
are specified according to manufacturer’s details. The grade of steel may be S235 or S275
and be hot dipped galvanised to BS EN ISO 1461 (BS EN ISO 1461 2009), or pregalvanised
grade Z275 (according to manufacturer’s details as the BS for this is withdrawn). Stainless
steel channels may be of grade 1.4301. The T-head bolts may be hot rolled steel in grade 4.6
or 8.8 or stainless steel.
Epoxy-based mortars are used to make, either partially or completely, connections where a
rapid gain in strength is required, for example up to 40 N/mm 2 in 2 to 3 h. Care is taken to
ensure that these materials have not exceeded their shelf life, are being used correctly and for
the right application. The thermal expansion of epoxy materials is seven times that of con-
crete, and this is accounted for in design. Epoxy compounds comprise two parts to be site
mixed - epoxy resins and hardeners. They are occasionally used as pressure injections for
crack filling or to restore tensile strength. Manufacturer’s procedures are strictly adhered to.
Concrete substrates must be prepared either by use of scabbling, grit blasting or needle
gunning, or similar, and surfaces contaminated with oil or grease require suitable prepara-
tion such as steam cleaning in conjunction with a suitable detergent.
Epoxy resins conform to BS ISO 18280 (BS ISO 18280 2005). They may be used as
structural adhesives for connections, together with the requirements for confinement and
Non-cementitious materials in precast construction
Elastomeric bearings, Bearing pads Comp. strength = 7–10 N/mm2

e.g. neoprene, rubber Shear strength = 1 N/mm2
Compressive strain = 15%
Bitumen impregnated Backing strip to concrete joints Compressibility = 85%
sealing strip
Polysulphide sealants Expansion joints Elongation strain < 50%
Epoxy resin mortars Compression, shear or tension joints Comp strength = 55–110 N/mm2
Tensile strength = 9–20 N/mm2
Elongation strain < 15%
Polyester resin mortars Compression, shear or tension joints Comp strength = 55–110 N/mm2
Tensile strength = 6–15 N/mm2
Elongation strain < 2%
Polystyrene Filler, backup blocks
PTFE (Teflon) Frictionless bearings Compression = 8 N/mm2 virgin and
15 N/mm2 reinforced.
Coefficient of friction 0.04
ductility using reinforcement crossing the joint line. They may be used as the chemical
for resin anchor bolts according to the proprietary information given by suppliers. Acrylic
polymer–modified cementitious materials such as high build structural repair mortars may
be used for repair and infilling in joints of minimum width according to manufacturer’s
requirements, typically 50 mm.
Neoprene, rubbers and mastics are used for soft bearings, backing strips etc. The PCI
Manual on Architectural Precast Cladding (PCI 1989) gives extensive guidance on the use
of these materials. Although they are not used extensively in precast structures, a typical
range of applications is given in Table 2.17
BS EN 197-1. 2011. Cement – composition, specifications and conformity criteria for common
cements, BSI, London, UK.
BS EN 206. 2013. Concrete – Specification, performance, production and conformity, BSI, London, UK.
BS EN 206-9. 2010. Concrete – Additional rules for self-compacting concrete, BSI, London, UK.
BS EN 480-1. 2006. Admixtures for concrete, mortar and grout. Test methods. Reference concrete
and reference mortar for testing, +A1:2011, BSI, London, UK.
BS EN 934-1. 2008. Admixtures for concrete, mortar and grout. Common requirements, BSI,
London, UK.
BS EN 934-2. 2009. Admixtures for concrete, mortar and grout. Concrete admixtures. Definitions,
requirements, conformity, marking and labelling, +A1:2012BSI, London, UK.
BS EN 934-3. 2009. Admixtures for concrete, mortar and grout. Admixtures for masonry mortar.
Definitions, requirements, conformity, marking and labelling, +A1:2012BSI, London, UK.
BS EN 1008. 2002. Mixing water for concrete, BSI, London, UK.
BS EN 1097-6. 2000. Tests for mechanical and physical properties of aggregates, BSI, London, UK.
BS EN 1011-1. 2009. Welding. Recommendations for welding of metallic materials. General guidance
for arc welding, BSI, London, UK.
BS EN 1011-2. 2001. Arc welding of ferritic steels, BSI, London, UK.
BS EN 1011-3. 2000. Arc welding of stainless steels, BSI, London, UK.
BS EN 1168. 2005. Precast concrete products – Hollow core slabs, +A3:2011, BSI, London, UK.
BS EN ISO 1461. 2009. Hot dip galvanized coatings on fabricated iron and steel articles – Specifications
and test methods, BSI, London, UK.
BS EN 13369. 2013. Common rules for precast concrete products, BSI, London, UK.
BS EN 1744-1. 2009. Tests for chemical properties of aggregates – Chemical analysis, BSI, London, UK.
BS EN 1990-1. 2002. Eurocode 0, Basis of Structural Design, BSI, London, UK.
BS EN 1992-1-1. 2004. Eurocode 2: Design of concrete structures. Part 1-1: General rules and rules
NA to BS EN 1992-1-1. 2004. UK National Annex to Eurocode 2: Design of concrete structures.
Part 1-1: General rules and rules for buildings, BSI, London, UK.
BS EN 1992-1-2. 2004. Eurocode 2: Design of concrete structures, Part 1-2: General rules – Structural
fire design, BSI, London, UK.
NA to BS EN 1992-1-2. 2004. UK National Annex to Eurocode 2: Design of concrete structures,
Part 1-2: General rules – Structural fire design, BSI, London, UK.
BS EN 1993-1-1. 2004. Eurocode 3, Design of Steel Structures – Part 1-1: General Rules and Rules
for Buildings, BSI, London, UK.
BS EN 1993-1-10. 2005. Eurocode 3. Design of steel structures – material toughness and through-
thickness properties, +A2:2009, BSI, London, UK.
NA to BS EN 1993-1-8. 2005. UK National Annex to Eurocode 3. Design of steel structures. Design
of joints, BSI, London, UK.
BS 3692. 2001. ISO metric precision hexagon bolts, screws and nuts. Specification, BSI, London, UK.
BS EN ISO 3766. 2003. Construction drawings. Simplified representation of concrete reinforcement,
BSI, London, UK., 2004.
BS 4483. 2005. Steel fabric for the reinforcement of concrete, BSI, London, UK.
BS 5080-1. 1993. Structural fixings in concrete and masonry – Part 1: Method of test for tensile load-
ing, BSI, London, UK.
BS 5896. 2012. High tensile steel wire and strand for the prestressing of concrete, BSI, London, UK.
BS 6744. 2001. Stainless Steel Bars for the Reinforcement of and Use in Concrete, BSI, London, UK.
BS 7371-12. 2008. Coatings on metal fasteners. Requirements for imperial fasteners, BSI, London, UK.
BS 8110-1. 1997. Structural Use of Concrete – Part 1: Code of practice for design and construction,
BSI, London, UK., August 2007.
BS 8500-1. 2015. Concrete. Complementary British Standard to BS EN 206-1. Part 1. Method of
specifying and guidance for the specifier, BSI, London, UK.
BS EN ISO 8501-1. 2007. Preparation of steel substrates before application of paints and related prod-
ucts. Visual assessment of surface cleanliness. Rust grades and preparation grades of uncoated
steel substrates and of steel substrates after overall removal of previous coatings, BSI, London, UK.
BS 8666. 2005. Scheduling, dimensioning, bending and cutting of steel reinforcement for concrete,
BSI, London, UK.
BS EN 10025-1. 2004. Hot rolled products of structural steels. General technical delivery conditions,
BSI, London, UK.
BS EN 10025-2. 2004. Hot rolled products of structural steels, Part 2: Technical delivery conditions
for non-alloy structural steels, BSI, London, UK.
BS EN 10080. 2005. Steel for the reinforcement of concrete, BSI, London, UK.
BS EN 10088-5. 2009. Stainless steels. Technical delivery conditions for bars, rods, wire, sections
and bright products of corrosion resisting steels for construction purposes, BSI, London, UK.
BS EN 10138-1. 2015 under technical review. Prestressing steel – Part 1: General requirements, BSI,
London, UK.
BS EN 10138-2. 2015 under review (accepted). Prestressing steel – Part 2: Stress relieved cold drawn
wire, BSI, London, UK.
BS EN 10138-3. 2015 under technical review. Prestressing steel – Part 3: strand, BSI, London, UK.
BS EN 10164. 2004. Steel products with improved deformation properties perpendicular to the sur-
face of the product. Technical delivery conditions, BSI, London, UK.
BS EN 10210-1. 1994. Hot finished structural hollow sections of non-alloy and fine grain structural
steels – Part 1: Technical delivery requirements, BSI, London, UK.
BS EN 12620. 2002. Aggregates for concrete, BSI, London, UK.

BS EN 13139. 2002. Aggregates for mortar, BSI, London, UK.
BS EN 13224. 2004. Precast concrete products – Ribbed floor elements, BSI, London, UK.
BS EN 13670. 2009. Execution of concrete structures, BSI, London, UK.
BS EN 14399-1. 2005. High-strength structural bolting assemblies for preloading. General require-
ments, BSI, London, UK.
BS EN ISO 17660-1. 2006. Welding of reinforcing steel – Part 1: Load-bearing welded joints, BSI,
London, UK.
BS EN ISO 17660-2. 2006. Welding of reinforcing steel – Part 2: Non-load-bearing welded joints,
BSI, London, UK.
BS ISO 18280. 2005. Plastics. Epoxy resins, BSI, London, UK.
BS EN 60974-1. 2005. Arc welding equipment – welding power, BSI, London, UK.
NSCS. 2010. National Structural Concrete Specification Edition 4, CONSTRUCT, London, UK.
NSCS website. National Structural Concrete Specification (NSCS), http://www.concretecentre.com/
codes_standards/nscs.aspx, May 2016.
PD 6687-1. 2010. Background paper to the UK National Annexes to BS EN 1992-1 and BS EN 1992-
3, BSI, London, UK.
PD 6682-3. 2003. Aggregates for mortar- Guidance on the use of BS EN 13139, BSI, London, UK.
Prestressed Concrete Institute. 1989. Manual for the Structural Design of Architectural Precast
Concrete, 2nd ed., PCI, Chicago, IL, 340pp.
Chapter 3
This chapter introduces the basic principles and some of the design and analysis procedures,
involved in the design of precast concrete skeletal structures, essentially a beam–column
framework possibly braced using walls and/or cores, as well as briefly discusses precast
portal frames and wall frames. Eurocodes EC0, EC1 and EC2 used to determine the com-
binations and arrangement of gravity and horizontal loads acting on floors, beams and
structures are introduced. The design of reinforced and prestressed concrete elements, con-
nections and structures will follow in later chapters.
Preliminary structural design, which many people refer to as the feasibility stage, is more
often a recognition of the type of structural frame that is best suited to the form and func-
tion of a building than the structural design itself. The creation of a large ‘open plan’ accom-
modation giving the widest possible scope for room utilisation clearly calls for a column
and slab structure, as shown in Figure 3.1, where internal partitions could be erected to suit
any client’s needs. The type of structure used in this case is often referred to as ‘skeletal’ –
resembling a skeleton of rather small but very strong components of columns, beams, floors,
staircases, and sometimes structural (as opposed to partition) walls. Of course, a skeletal
structure could be designed in cast in situ concrete and structural steelwork, but here we
will consider only the precast concrete version.
The basis for the design of precast skeletal structures has been introduced in Figures 1.11
and 1.13. The major elements (the precast components) in the structure are shown in
Figure 3.2. Note that the major connections between beams and floors are designed and con-
structed as ‘pinned joints’, and therefore the horizontal elements (slabs, staircases, beams)
are all simply supported. They need not always be pinned (in seismic zones, the connections
are made rigid and very ductile) but in terms of simplicity of design and construction it is
still the preferred choice. Vertical elements (walls, columns) may be designed as continuous,
but because the beam and slab connections are pinned there is no global frame action and
no requirement for a frame stiffness analysis, apart from the distribution of some column
moments arising from eccentric beam reactions. The stiff bracing elements such as walls are
designed either as a storey height element, bracing each storey in turn, or as a continuous
element bracing all floors as tall cantilevers.
In office and retail development, distances between columns and beams are usually in the
range of 6–12 m (Figures 1.7 and 3.3) depending on the floor loading, method of stability
and intended use. In multi-storey car parks, where the imposed loading (vehicle gross weight
<30 kN according to the NA to BS EN 1991-1-1, Table NA.6) is 2.5 kN/m 2 it is around 16 m
for floor spans × 7.2 m for beams, giving three parking bays between columns (Figure 1.6).
The exterior of the frame – the building’s weatherproof envelope – could also be a skel-
etal structure, in which case the spaces between the columns would be clad in brickwork,
65
Precast skeletal structure showing large unobstructed spaces for the benefit of construction
workers and the client.
Definitions in a precast skeletal structure.
precast concrete panels, sheeting, etc. Alternatively, the envelope might be constructed in
solid precast bearing walls, which dispenses with the need for beams, and is referred to as
a ‘wall frame’ (Figure 1.14).
Examples of residential buildings where a precast wall frame would be the obvious
choice are shown in Figures 3.4 through 3.7 – the walls are all load-bearing and they
support one-way spanning floor slabs. There is less architectural freedom compared
to the skeletal frame, for example walls should (preferably) be arranged on a rectan-
gular grid and of fixed modular distance, usually 300 mm, which is quite important
Precast frame analysis 67
Precast skeletal structure in Portugal. (Courtesy of Ergon, Belgium.)
Wall frames are best suited for apartments, hotels, schools, shopping units, as in this example at
Rhodes, near Sydney, Australia.
Precast wall and slab frames in Kuala Lumpur, Malaysia.
Cross-wall system in precast wall frames.
economically. A wall frame may be more economical and may often be faster to build,
especially if the external walls are furnished with thermal insulation and a decorative
finish at the factory. Figures 1.14 and 1.22 are good examples of this. Distances between
walls may be around 6 m for hotels, schools, offices and domestic housing, and 10–15
m in commercial developments. Given this description, wall frames appear to be very
simple in concept, but in fact are quite complicated to analyse because the walls have
Precast wall and slab frame at Strijkijzer, Den Haag, the Netherlands.
very large in-plane rigidity whilst the connections between walls and floors are more
flexible. Differential movement between wall panels and between walls and floors has
resulted in major serviceability problems over a 25+ year life, often leading to a break-
down in the weatherproof envelope and the eventual condemnation of buildings, which
are structurally adequate.
The third category of precast building is the ‘portal frame’ used for industrial buildings
and warehouses where clear spans of some 25–40 m I-section or T-section prestressed raf-
ters are necessary; Figures 3.8 and 3.9. Although portal frames are nearly always used for
single-storey buildings, they may actually be used to form the roof structure to a skeletal
frame, and as this book is concerned with multi-storey structures it gives us a reason to men-
tion them. The portal frame looks simple enough and in fact is quite rudimentary in design,
providing that the flexural rotations at the end of the main rafters, which we can assume
will always cause cracking damage to the bearing ledge, are catered for by inserting a flex-
ible pad (e.g. neoprene) at the bearing. As mentioned before, pinned connections between
the rafter and column are the preferred choice – they are easy to design and construct. But
the columns must be designed as moment-resisting cantilevers, which might cause a problem
in some structures as explained later in Section 3.6.2. A moment-resisting connection is
equally possible allowing some moment continuity into the column at the eaves. However,
unless the columns are particularly tall, say more than about 8 m, it is not worth the extra
effort.
Precast portal frames with flat (or shallow inclination) roof structures comprising pre-
stressed or reinforced beams of 6–8 m span supporting long-span precast folded plate roof
elements, spanning around 20 m. This is a popular option for industrial buildings, and in
the case of Figure 3.10 used in laboratory buildings. The overhang beam is an option for
sun or rain shading.
Table 3.1 reviews the various types of precast structures with respect to their possible
applications.
Definition of a precast portal frame.
Precast portal frame. (Courtesy David Fernandez-Ordoñez, Escuela Técnica Superior de

Ingeniería Civil, Madrid, Spain.)
Portal frame with folded plate roof units, the University of Sao Carlos, Brazil.
Application and types of precast concrete frames
Office 2–0 6–15 ✓

2–50 6–15 ✓
Retail, shopping 2–10 6–10 ✓ ✓
complex
Cultural 2–10 6–10 ✓
Education 2–5 6–10 ✓ ✓
Car parking 2–10 15–20 ✓
Stadia 2–4 6–8 ✓
Hotel 2–30 6–8 ✓
Hospital 2–10 6–10 ✓
Residential 1–40 4–6 ✓
Industrial 1 25–40 ✓
Warehouse 2–3 6–8 ✓
with office 25–40 ✓
a Typical values, depending on the location, terrain, requirements, etc.
One of the most frequently asked questions is … how is a precast concrete structure ana-
lysed compared to a monolithic cast in situ one? The first response is to say that a precast
concrete structure is not a cast in situ structure cut up into little pieces making it possible
to transport and erect. It was mentioned in Chapter 1 that the passage of forces through
the prefabricated and assembled components in a precast structure is quite different to a
continuous (monolithic) structure. This is certainly true near to connections. It is therefore
2-D simplification of a 3-D skeletal structure.
possible to begin a global analysis by first considering the behaviour of a continuous frame
and identifying the positions where suitable connections in a precast frame may be made.
A two-dimensional in-the-plane simplification is appropriate in the first instance. This is
defined in Figure 3.11 where there are no structural frame components, only simply sup-
ported floor units, connecting the 2-D in-plane frames together.
Figure 3.12 shows the approximate bending moments and deflected shape in a three-
storey continuous beam and column frame subject to vertical (gravity) patch loads and
horizontal (wind) pressure. Call this frame F1. The beam–column connections have equal
strength and stiffness as the members. The stability of F1 is achieved through the combined
action of the beams, columns and beam–column connections in bending, shear and axial.
This is called an ‘unbraced’ frame. There are points of zero moment (‘contraflexure’) in F1,
which depend on the relative intensity of the two load cases. If gravity loads are dominant,
beam contraflexure is near to the beam–column connection, typically 0.1 times the span of
the beam as shown in Figure 3.13; but if the horizontal load is dominant (more rare), con-
traflexure is at mid-span, with the final location for combined loading at about 0.15 × span.
In the column, contraflexure is always at mid-storey height, and this is a good place to make
a pinned (notionally = small moment capacity) connection between two precast columns.
Now, if the strength and stiffness of the connection at the end of the beam are reduced to
zero, whilst the column and the foundation are untouched, the resulting moments and deflec-
tions in this frame, called F2, are as shown in Figure 3.14. The columns alone achieve the
stability of F2 – the beams transfer no moments, only axial forces and shear. The foundations
must be moment-resisting (rigid). This is the principle of a pinned jointed unbraced skeletal
frame. In taller structures, > three storeys or about 10 m, the large sizes of the columns
Deformation and bending moment distribution in a continuous structure subjected to (a) gravity
loads and (b) horizontal sway load.
become impractical and uneconomic leading to bracing. The bracing may be used in the full
height, called a ‘fully braced’ frame, or up to or from a certain level, called a ‘partially braced’
frame. The differences are explained in Figure 3.15. The bracing could be located in the upper
storeys, providing the columns in the unbraced part below the first floor are sufficiently stable
to carry horizontal forces and any second-order moments resulting from slenderness.
Pinned connections may be formed at other locations. Referring back to frame F1, if the
flexural stiffness of the members at the lower end of a column is greater than that at the
upper end, the point of contraflexure will be near to the lower (stiffer) end of the column.
If the strength and stiffness of the lower end of the column are reduced to zero, whilst the
beam and beam–column connections are untouched, the resulting moments and deflections
in this frame, called F3, are as shown in Figure 3.16a. The stability of F3 is achieved by the
portal frame action of inverted U frames – clearly not a practical solution for factory cast
Beam half-joints at 0.1× span close to points of contraflexure in a continuous beam.
large spans so that this method is used for repetitious site casting. Therefore, a practical solu-
tion is to prefabricate a series of L-frames as shown in Figure 3.16b for long-span beams
and small-storey height columns in a parking structure. Foundations to F3 may be pinned,
although most contractors prefer to use a fixed base for safety and immediate stability.
The so-called H-frame is a variation on F3. Referring back to frame F1, if pinned con-
nections are made at the points of column contraflexure, structural behaviour is similar
Deformation and bending moment distribution in a pinned jointed structure subjected to

(a) gravity loads. ( )
( ) Deformation and bending moment distribution in a pinned jointed structure subjected

to (b) horizontal sway load.
Partially braced structures.
to a continuous frame as explained in Figure 3.17. Connections between frames are made
at mid-storey height positions. Although in theory the connection is classed as pinned, in
reality there will be some need for moment transfer, however small. Therefore, H-frame
connections are designed with finite moment capacity, this also gives safety and stability
to the H-frames, which by their nature tend to be massive. The foundation to half-storey
height ground floor columns must be rigid. The connection at the upper end of the column
may be pinned if it is located at a point of contraflexure. If not the connection must possess
flexural strength as shown in Figure 3.17, where the H-frame has been used in a number of
multi-storey grandstands.
Structural systems for (a) portal U-frames and (b) portal L-frames.
The object of analysis of a structure is to determine bending moments, shear and axial
forces throughout the structure. Monolithic two-dimensional plane frames are analysed
using either rigorous elastic analysis, for example moment distribution or stiffness method,
either manually or using a computer program. Moment redistribution may be included in
the analysis if appropriate. However, often it is only required to determine the moments
and forces in one beam or one column, so codes of practice allow simplified substructuring
H-frames (a) structural system, (b) deformation and bending moments.
techniques to be used to obtain these values. Figure 3.18 gives one such substructure, called
a ‘subframe’ – refer to (Bhatt et al. 2014) for further details. If the frame is fairly regular,
that is spans and loads are within 15% of each other, substructuring gives 90%–95% agree-
ment with full frame analysis.
Substructuring is also carried out in precast frame analysis, except that, where pinned con-
nections are used, no moment distribution or redistribution is permitted. Figure 3.19 shows
Substructuring method for internal beam in a continuous frame.
Substructuring methods for internal beam and columns in a pinned jointed frame (a) internal beam.
( )
( ) Substructuring methods for internal beam and columns in a pinned jointed frame
(b) upper floor column and (c) ground floor column.
subframes for internal beam and upper and ground floor columns where all beam–column
connections are pinned. For rigid connections, refer to Figure 3.18. Horizontal wind loads
and sway forces due to imperfections are not considered in subframes because the bending
moments due to horizontal loads in an unbraced frame (there are no column moments due
to horizontal loads in a braced frame) are additive to those derived from subframes. Elastic
analysis is used to determine moments, forces and deflections, but a plastic (ultimate) section
analysis is used for the design of the components. Clearly, some inaccuracies must be accepted,
but according to ‘Designer’s Guide to EN 1992-1-1 and EN 1992-1-2’ (Narayanan and Beeby
2005), a design using the partial safety factors (PSFs) and methodologies in the Eurocodes
(design philosophy and materials) is “likely to lead to a structure with a reliability index
greater than the target value of 3.8 stated in the code for a 50-year reference period.”
The primary Eurocodes used in the design of precast concrete structures are
• Eurocode 0 ‘Basis of design’ (BS EN 1990 2002).

• Eurocode 1 ‘Actions on Structures – Part 1-1: General Actions – Densities, self-weight,
imposed loads for buildings’ (BS EN 1991-1-1 2002) plus other parts dealing with
fire, snow, wind, thermal, execution (during construction) and accidental (explosion,
impact, etc.) actions.
• Eurocode 2 ‘Design of Concrete Structures – General rules and rules for buildings’ (BS
EN 1992-1-1 2004) including coefficients for sway loads due to imperfections (lack of
plumb, built-in curvature, etc.).
• Eurocode 3 ‘Design of steel structures. General rules and rules for buildings (BS EN
1993-1-1 2005) where steelwork, steel inserts, welding, etc. is required in the additional
parts of EN 1993.
Each pan-European document is accompanied by national annexes (NAs) appropriate to

national working practices, regional conditions and established/historical precedence, for
example stability ties for robustness are the same as in the British code BS 8110: 1997.
This book will refer to the UK NAs to Eurocodes EC0 (NA to BS EN 1990 2002), EC1
(NA to BS EN 1991-1-1 2002), EC2 (NA to BS EN 1992-1-1 2004) and briefly to the
NA to Eurocode 3 where steelwork, inserts, welding, etc. is required (NA to BS EN 1993-
1-1 2005). Appendix 3A (at the end of this chapter) summarises the content of Eurocodes
EC2 Parts 1-1 and 1-2, together with the specific clauses related to precast and prestressed
concrete elements in the NA to BS EN 1992-1-1. Reference will also be made to the UK’s
Published Document PD 6687-1 (PD 6687-1 2010) that gives guidance on some specific
items that were not published in the concrete Eurocodes or were in need of additional or
noncontradictory additional information. The main items in the PD relating to the design of
precast concrete structures are listed in Appendix 3B.
These documents give the magnitude and combinations of loads, loading patterns, and
PSFs γf (in BS EN 1990) for gravity and horizontal loads in frames and beams. Four conditions
are considered, each with their own values of γf follows
a. Serviceability limit state (stress, cracking, deflection, dynamic, fatigue)

b. Ultimate limit state (ULS) (strength, buckling)
c. Instability limit state (for over-turning)
d. Accidental limit state (fire, robustness, progressive collapse)
However, each condition varies depending on the nature of the loads. These are called
‘actions’ in the Eurocodes, and those applicable to the super-structure are as follows:
i. Permanent actions: self-weight, dead loads of toppings, finishes, services, perma-

nent walls Gk; prestressing forces P; settlement of supports, sway loads due to
imperfection.
ii. Variable actions: imposed floor live loads; demountable partitions, snow loads Qk;
wind loads Wk; temperature effects.
iii. Accidental actions: fire, impact, loss of support, collapse, explosion, Ad, etc.
Dead, live and wind loads are based on the 95% characteristic value for uniformly distrib-
uted load (UDL) known as gk, qk and wk [kN/m 2] and for line/beam loads and point loads
as Gk, Qk and Wk [kN/m or kN].
The self-weight of plain concrete made with normal-weight aggregates (approx.
2600 kg/m3) is taken as 24 kN/m3, according to BS EN 1991-1-1, Table A.1, unless it
is shown by the manufacturer that the characteristic self-weight of elements is different.
An additional 1 kN/m3 is made for reinforcement and prestressing tendons, although it is
unlikely that tendons will add this amount, for example 10 no. 9.3 mm strands in a 1200 ×
150 deep solid slab add only 0.22 kN/m3. The density of wet concrete is taken as 25 kN/m3.
The densities or self-weight of other building materials and stored materials in warehouses,
etc. are given in BS EN 1991-1-1, Tables A.2 through A.12. Note that the self-weight of
masonry units are given in BS EN 771 (BS EN 771 2011) and not in the masonry code (BS
EN 1996-1-1 2005).
The design values of actions for each of the limit states depend on the nature of the
load (i) to (iii), the use of the floor slabs (e.g. residential, parking, storage) and the num-
ber and location of the variable loads. Statistically, it is improbable that all imposed loads
will be acting at their characteristic value Qk1, Qk2 … Qki and at the same time, that is full
live loads will not act at all floor levels in a multi-storey building, or live, wind and snow
loads will not act at the same time. Exceptions to this obviously apply and the designer
must be aware of the certain simultaneous combination, such as full live loads acting on
a staircase and landing at the same time, in which case the characteristic load will be
taken for both elements.
3.3.2.1 Serviceability limit state

Historically, the ‘characteristic’ imposed (live) load Qk was used in all serviceabil-
ity calculations of service stresses in prestressed concrete, crack spacing and widths,
and short- and long-term deflections using viscoelastic deformations due to creep and
other effects such as the relative shrinkage between concrete cast at different times.
The Eurocodes consider this too severe for long-term effects of cracking and deflection,
and, with the exception of calculating service stresses in prestressed concrete in order
to avoid sudden rupture after cracking, reduced values of Qk are permitted as shown
in Figure 3.20. This is an illustration of the representative values for the characteristic
Qk, combination ψ0 Qk, frequent ψ1Qk, and quasi-permanent ψ2 Qk values of imposed
loading over a period of time, which can of course be extended to the whole life of the
structure. In fact, clause A1.4.2 of EN 1990 allows the serviceability criteria to be speci-
fied for each project and agreed with the client, but the actual circumstantial definitions
recommended to be used with particular serviceability requirements in clause A1.4.2 of
the NA to BS EN 1990 are
a. for function and damage to structural and non-structural elements (e.g. partition
walls, etc.), the characteristic combination, for example stress, strength
b. for comfort to user, use of machinery, avoiding ponding of water, etc. the frequent
combination
c. for appearance of the structure, the quasi-permanent combination, for example defor-
mation, deflections
Illustration of variable actions.
Partial load factors for live loads

ψ ψ ψ
Domestic, residential, offices 0.7 0.5 0.3
Shopping, congregation 0.7 0.7 0.6
Storage 1.0 0.9 0.8
Traffic area < 3 vehicle weight 0.7 0.7 0.6
Traffic area > 3 vehicle weight 0.7 0.5 0.3
Roof 0.7 0 0
Snow at altitude > 1000 m 0.7 0.5 0.2
Snow at altitude < 1000 m 0.5 0.2 0
Wind pressure 0.5 0.2 0
Source: From NA to BS EN 1990, Table NA.A1.1.
ψ0 is used for load combinations of ultimate strength, ψ1 is used for checking crack widths and
decompression stresses for certain durability requirements, ψ2 is used for calculating deflections.
These are according to Expressions 6.14b, 6.15b and 6.16b of EN 1990 as follows.
The design service moment M s, shear force Vs and end reaction Fs are based on the design
service load = characteristic load × set of load factors ψ as follows:
1. ‘Characteristic’ combination according to BS EN 1990, Exp. 6.14b.
Gkj “+” P “+” Qk,1 “+” ψ0 Qk,1 with j ≥ 1 and i > 1 (3.1)
where ψ0 is the characteristic load factor according to the UK NA to BS EN 1991, repro-

duced here in Table 3.2, for example for domestic usage, ψ0 = 0.7.
The symbol “+” means in combination with. For example referring to Figure 3.21, if the
stresses at the bottom of a prestressed beam due to prestress = +12.0 N/mm 2 (compression),
due to dead load = −5.0 N/mm 2 (tension), and due to two independent live loads = –8.0
and –4.0 N/mm 2 , the characteristic stress f b = +12.0 – 5.0 – 8.0 – 0.7 × 4.0 = −3.8 N/mm 2
(tension).
2. ‘Frequent’ combination according to BS EN 1990, Exp. 6.15b.
Gkj “+” P “+” ψ1 Qk,1 “+” ψ2 Qk,1 with j ≥ 1 and i > 1 (3.2)
Example of the characteristic, frequent and quasi-permanent combinations of service stresses

in a fictitious prestressed concrete section.
where ψ1 and ψ2 are the frequent and quasi-permanent load factors in Table 3.2, for example
for domestic usage, ψ1 = 0.5 and ψ2 = 0.3.
Continuing the example mentioned earlier, the final frequent stress fb = +12.0 – 5.0 – 0.5 ×
8.0 – 0.3 × 4.0 = +1.8 N/mm2 (effectively remains in compression).
3. ‘Quasi-permanent’ combination according to BS EN 1990, Exp. 6.16b
Gkj “+” P “+” ψ2 Qki with j ≥ 1 and i ≥ 1 (3.3)
Continuing the example mentioned earlier would not be meaningful as the ‘quasi-permanent’
combination is used for calculating deflections, but for completeness fb = +3.4 N/mm2 (com-
pression). It is clear from these three examples that the conditions of stress are less onerous with
each successive combination, and this is a reflection of the diminishing effect of viscoelastic
deformations according to the use of buildings and the effect of specific creep. Note that in
Table 3.2 for storage the ψ factors are between 0.8 and 1.0, indicating a higher specific creep.
3.3.2.2 Ultimate limit state

This limit state is known as ‘structure STR’. The design ultimate moment MEd, shear force
VEd and end reaction FEd are based on the design ultimate load Ed = characteristic load x
set of PSFs γ. Loads are called ‘favourable’ or ‘unfavourable’ in creating the worst possible
effects in an element, frame or subframe. The ultimate load combination is according to BS
EN 1990, Exp. 6.10, or for STR limit state the least favourable (greater) of Exp. 6.10a and
6.10b, which will always be less than Exp. 6.10. The NA to BS EN 1990, Table NA.A1.2 (B) –
Design values of actions (STR) (Set B) give the PSF values. The combinations are
6.10a γG,j Gk,j “+” γP P “+” Σ γQ,1 ψ0,1 Qk,1 (3.4)
6.10b ξ γG,j Gk,j “+” γP P “+” γQ,1 Qk,1 “+” Σ γQ,i ψ0,i Qk,I (3.5)
with j ≥ 1 and i > 1

where γG,j = 1.35 unfavourable, γG,j = 1.0 favourable, γQ,1 = 1.5, ξ = 0.925
for prestress γP = 0.9 (used for ultimate shear capacity)
NA to BS EN 1990, Table NA.A1.2 (B) notes: “Either expression 6.10, or expression 6.10a
together with and 6.10b may be made, as desired. The characteristic values of all permanent
actions from one source are multiplied by γG,sup = 1.35, if the total resulting action effect is
unfavourable and γG,inf = 1.0, if the total resulting action effect is favourable. For example all
actions originating from the self-weight of the structure may be considered as coming from one
source; this also applies if different materials are involved. When variable actions are favour-
able Qk should be taken as 0.” In other words, in frame or subframe analysis 1.35 Gk,j + 1.5
ψ0, Qk (6.10a or b) is carried on one element and 1.0 Gk,j on the adjacent element. Each live
load is taken as the ‘dominant’ in turn, with all of the others as ‘accompanying’ in turn, until
the maximum combination is found. Therefore, if there are two live loads present there will be
three load combinations, that is 6.10a; 6.10b Qk1 dominant; and 6.10b Qk2 dominant.
If the building is domestic with ψ0 = 0.7, use the greater of
6.10a 1.35 Gk,j + 1.5 × 0.7 Qk,1
6.10b 0.925 × 1.35 Gk,j + 1.5 Qk,1 + 1.05 Qk2
6.10b 0.925 × 1.35 Gk,j + 1.05 Qk1 + 1.5 Qk2
To satisfy the ULS design, the three load combinations must be used to determine the maxi-
mum end reactions, and bending moments and shear forces at all points along the span.
Finally, it is necessary to define the effective span of floor slabs (simply supported, cantilevers)
and beams (on dry bearings or mechanical connectors). The clear span of floors ln = distance
between beam centres – sum of half breadth of beams. Referring to BS EN 1992-1-1,
Exp. 5.8 leff = ln + min{h/2; Lb1/2}+ min{h/2; Lb2 /2} (3.6)
where
h is the depth of slab (including topping)
Lb2 and Lb1 are the bearing lengths at either end
For cantilevers, Lb2 is also the width of the support. For continuous elements after comple-
tion of the continuity (i.e. stage 2 loading), leff = distance between beam centres. If a bearing
medium (pad, plate) is provided, leff is to the centre of the pad, and this is also the case for
beams supported on steel inserts, cleats and plates, etc.
Example 3.1
Calculate the maximum ultimate end reaction FEd in a simply supported beam of effective
span 6.0 m subjected to Gk = 30 kN/m, Qk1 = 20 kN/m, and Qk2 = 40 kN point load at
mid-span. ψ0 = 0.7.
Solution
Exp. 6.10a wEd = 1.35 × 30 +(0.7 × 1.5) × 20 = 61.5 kN/m
PEd = 1.05 × 40 = 42 kN.
Then F Ed = 61.5 × 6.0/2 + 42/2 = 205.5 kN.
Exp. 6.10b = 1.25 × 30 + 1.5 × 20 = 67.5 kN/m + 42 kN (UDL dominant)
Then F Ed = 67.5 x 6.0/2 + 42/2 = 223.5 kN. Maximum.
Exp. 6.10b = 1.25 × 30 + 1.05 × 20 = 58.5 kN/m + 1.5 × 40 = 60 kN (point load
dominant)
Then F Ed = 58.5 × 6.0/2 + 60/2 = 205.5 kN.
3.3.2.3 Instability limit state

This limit state, known as ‘equilibrium EQU’, is used to check uplift in the back-span of
cantilevers, and over-turning of frames including the effect of horizontal wind pressure or
other forces. The equilibrium load combination is according to BS EN 1990, Exp. 6.10. The
PSFs are given in NA to BS EN 1990, Table NA.A1.2 (A) – Design values of actions (EQU)
(Set A) Exp. 6.10 as follows:
6.10 γG,j Gk,j “+” γQ,1 Qk,1 “+” Σ γQ,i ψ0,i Qk,I (3.7)

γG,j = 1.1 (unfavourable) and 0.9 (favourable)
γQ = 1.5 (unfavourable) and 1.5 ψ0 accompanying
γQ = 0 (favourable)
At installation refer to BS EN 1990, Table A2.4(A) as follows:
γG,j = 1.05 (unfavourable) and 0.95 (favourable)
γQ,i = 1.35 (unfavourable) and 0 (favourable)
Example 3.2
Calculate the minimum end reaction FEd in a simply supported beam of effective span 6.0
m with a 3.0 m span overhanging cantilever at one end subjected to Gk = 30 kN/m and
Qk = 20 kN/m.
Solution
Exp. 6.10. Main span = 0.9 × 30 = 27 kN/m

Exp. 6.10. Cantilever span = 1.1 × 30 + 1.5 × 20 = 63 kN/m.
Over-turning moment at cantilever = 63 × 3.02 /2 = 283.5 kN
Then F Ed = 27 × 6.0/2 – 283.5/6.0 = 81.0 – 47.25 = 33.75 kN > 0. No uplift.
3.3.2.4 Accidental limit state

The accidental load combination is according to BS EN 1990, Exp. 6.11. The PSFs are all
γ = 1.0. ψ values are given in NA to BS EN 1990, Table NA.A1.3 – Design values of actions
for use in accidental combinations of actions, Exp. 6.11 as follows:
6.11 Gk,j “+” P “+” Ad “+” ψ1,1 Qk,1 “+” Σ γQ,i ψ2,i Qk,I (3.8)

where Ad is the value of the accidental action. ψ1,1 is applied to the dominant action and
ψ2,i to the others. However, in accidental situations, it may not be obvious which is which
and therefore ψ1 is applied to all.
3.3.3.1 Permanent, variable and wind actions

In frame and subframe analysis without sway, the critical gravity load combinations with
their associated PSFs γG and γQ are
1. All spans loaded with the maximum ultimate load wEd,max = γG Gk + γQ Qk (for BS EN
1990, Exp. 6.10a or b).
2. Alternate (‘pattern’) spans loaded with wEd,max on one span and the minimum wEd,min = 1.0 Gk
on the adjacent span.
For frame analysis with sway, horizontal loads Wk are combined with gravity Gk and Qk
load combinations 6.10a and 6.10b for three situations:
1. Permanent “+” imposed actions (gravity dead + live)

2. Permanent “+” wind actions (gravity dead + wind)
3. Permanent “+” imposed “+” wind actions (all)
Partial load factors and safety factors for gravity and horizontal loads
6.10a Permanent + imposed 1.35 1.0 ψ0 1.5 0 N/A

Permanent + wind 1.35 1.0 N/A N/A ψ0 1.5 = 0.75
All 1.35 1.0 ψ0 1.5 0 ψ0 1.5 = 0.75
6.10b Permanent + imposed 1.25 1.0 1.5 0 N/A
Permanent + wind 1.25 1.0 N/A N/A 1.5
All 1.25 1.0 1.5 0 ψ0 1.5 = 0.75
Source: From NA to BS EN 1990, Table NA.A1.2 (B).
Alternative terminology to BS EN 1990: beneficial = favourable, adverse = unfavourable.
For wind load, ψ0 = 0.5 according to Table NA.A1.1 of NA to BS EN 1990 (see Table 3.2).
The fundamental PSF for wind load (notation used here γW) is as NA to BS EN 1990, Table
NA.A1.2 (B) – Design values of actions (STR) (Set B) γW = 1.5, and is modified by ψ0 = 0.5
(see Table 3.2) in the same way as for gravity loads. The values for γ and ψ are summarised
in Table 3.3.
Example 3.3
Calculate the maximum ultimate bending moment M Ed at the lower end of the columns
of height h = 4.0 m in Figure 3.22. The beam–column connections are pinned, and the
foundation is rigid. The distance from the edge of the column to the centre of the beam
end reaction is 100 mm. Characteristic beam loading is Gk = 40 kN/m and Qk = 30 kN/m,
and the wind pressure equates to a horizontal load Wk = 12 kN. The carry-over moment
at the lower end of the column is equal to 50% of the upper end moment due to beam
eccentricity. Let ψ0 (gravity load) = 0.7.
Solution
Eccentricity of beam reaction R from the centre of column e = 300/2 + 100 = 250 mm.
Moment at the lower end of column due to R Ed , M Ed = 0.5 R Ed e
Detail to Example 3.3.
Moment at the lower end of each column due to wind load, M Ed = W Ed h/2 (because
there are two columns)
Ultimate load combinations and moments are summarised in the following table.
6.10a P+I 85.5 342.0 42.75 — — 42.75

P +W 54.0 216.0 27.0 9.0 18.0 45.0
All 85.5 342.0 42.75 9.0 18.0 60.75
6.10b P+I 95.0 380.0 47.5 — — 47.5
P +W 50.0 200.0 25.0 18.0 36.0 61.0
All 95.0 380.0 47.5 9.0 18.0 65.5
P, permanent (dead load); I, imposed (live load), W, wind load.
Then M Ed,max = 65.5 kNm using Exp. 6.10b for all loads (it is interesting to note that
according to BS 8110 values for all loads, wult = 1.2 × 70 = 84 kN/m, Mu (gravity) = 42 kNm,
Wu = 1.2 × 12 = 14.4 kN, M u (wind) = 28.8 kNm. Total M u = 70.8 kNm).
3.3.3.2 Horizontal forces due to imperfections

All buildings, including precast concrete structures built with the greatest practical accu-
racy, will contain imperfections due to construction methods, errors or natural effects.
Some of these are unavoidable, for example over-turning moments due to balconies, hang-
ing façade panels, etc. resulting in horizontal deflection and curvature in columns and walls.
The reactions to the precast frame from the inclined staircase shown in Figure 3.23 are not
Inclined staircase imposing horizontal forces to the structure at Bella Sky Hotel, Denmark.
(Courtesy Ramboll, Denmark.)
exactly imperfections but demonstrate the point of transferring inclined gravity loads into
horizontal forces.
BS EN 1992-1-1 defines imperfections as ‘possible deviations’ in geometry and ‘positions’
of loads in Section 5.2 and as quantified by code Exp. 5.1 through 5.4 as an ultimate hori-
zontal force Hi = Nθi due the inclination θi according to the height l and number m the
elements contributing to the imperfection, recognising the improbability that imperfection
will be the same in all elements. The value attributed to θi is related to ‘Class 1 execution
deviations’ according to BS EN 13670 (BS EN 13670 2009) and is taken into account at the
ULS and accidental design situations, but not at serviceability. Deviations in cross section
dimensions are taken into account in material safety factors. Imperfections and deviations
should not be included in structural analysis as their effect is additional to first-order bending
moments and shear forces, but not when checking deflections.
BS EN 1992-1-1 distinguishes between (a) whole braced or unbraced structures, known
as ‘global analysis’, (b) isolated columns in braced or unbraced structures, and (c) floor and
roof diaphragm action (see Chapter 8) as follows:
(a) For structures, the horizontal force Hi is applied to the bracing system, for example
cores or shear walls, at each floor level as the horizontal component of the total ulti-
mate gravity load at that floor level. Referring to Figure 3.24a (adapted from BS EN
1992-1-1, Fig. 5.1b).
Hi = θi (Nb – Na) = θi Σ V Ed at the floor level connections (3.9)
where θi = θ 0 αh αm (3.10)
θ0 = 1/200 (NA BS EN 1992-1-1)
Examples of the effect of geometric imperfections. (a) Bracing system and (b) isolated column
in unbraced structure. (Adapted from BS EN 1992-1-1. 2004, Eurocode 2: Design of Concrete
Structures – Part 1-1: General rules and rules for buildings, BSI, London, February 2014,
Fig. 5.1a1 and b.)
αh = 2/3 ≤ (2/√l) ≤ 1 (3.11)
l = total height of braced structure (m)
αm = √0.5(1 + 1/m) (3.12)
m = number of vertical members contributing to the horizontal force on the bracing

system.
(b) For isolated columns, Hi is applied to unbraced columns as an eccentricity ei as shown
in Figure 3.24b (adapted from BS EN 1992-1-1, Fig. 5.1a and Exp. 5.2). This is used
because columns in a precast structure are statically determinate and there is no
moment continuity between the rows of columns.
ei = θi l 0/2 (3.13)
where l0 is effective length (m) of the column at the floor level that ei is considered,
that is at the second floor level l0 is based on the height to the second floor, etc.
αh = 2/3 ≤ (2/√l) ≤ 1, where l = actual length of column at the level that ei is acting
(m) and m = αm =1
(c) For isolated columns (and minor axis of walls) in a braced structure αh is simplified
such that αh = m = αm = 1
ei = l 0/400 (3.14)
Example 3.4
Calculate the horizontal forces at each roof and floor level and the over-turning moment at
the foundation due to imperfection in the braced skeletal structure shown in Figure 3.25.
The number of columns in each line is six, and there are five rows of column. There are
two sets of shear walls in each of the external rows of columns. The total ultimate gravity
load per floor = 15,000 kN and at the roof = 7,000 kN.
Detail to Example 3.4 (braced) and Example 3.5 (unbraced).
Solution
l = 10.5 m, then αh = 2/√10.5 = 0.617 use 2/3
m = 30 columns. αm = √0.5(1 + 1/30) = 0.719
θi = (1/200) × (2/3) × 0.719 = 0.002396 (or 1 in 417)
At roof level, Hi,roof = θI N/2 sets of shear walls = 0.002396 × 7000/2 = 8.38 kN
per wall
At floor level, Hi,floor = 0.002396 × 15,000/2 = 17.97 kN per wall
The over-turning moment M i due to Hi = 8.38 × 10.50 + 17.97 × (7.25 + 4.00) =
290.1 kNm per wall.
Example 3.5
Calculate the horizontal forces at each roof and floor level and the over-turning moment
at the foundation due to imperfection in one of the internal columns, if the same structure
shown in Figure 3.25 is unbraced. The effective length factor for the columns may (in this
example) be taken as 2.2. The total ultimate gravity load per column per floor = 900 kN
and at the roof = 500 kN.
Solution
At the roof level. l = 10.50 m, then αh = 2/√10.50 = 0.617 use 2/3. l 0 = 2.2 × 10.5 =
23.1 m
ei = (1/200) × (2/3) × 23.1/2 = 0.039 m
Then M i due to ei = 500 × 0.039 = 19.5 kNm which equates to Hi = 19.5/10.5 =
1.85 kN
At the second floor level. l = 7.25 m, then αh = 2/√7.25 = 0.742. l 0 = 2.2 × 7.25 =
15.95 m
ei = (1/200) × 0.742 × 15.95/2 = 0.030 m
3.72 kN
At the first floor level. l = 4.00 m, then αh = 2/√4.00 = 1. l 0 = 2.2 × 4.00 = 8.80 m
ei = (1/200) × 1 × 8.80/2 = 0.022 m
4.95 kN
Total Mi = 19.5 + 27.0 + 19.8 = 66.3 kNm per column.
(Note that M i for isolated columns is greater per column than if the total M i for the walls
was divided over the total number of columns).
Figure 3.19a. The subframe consists of the beam to be designed of span L1, and half of the
adjacent beams of span L 2 and L 3. The eccentricity of the beam end reaction from the cen-
troidal axis of the column is e. Alternate pattern loading is used. The height of the column
above and below the beam is actually of no consequence to beam. It is assumed that the
cross section and flexural stiffness of the column is constant.
The maximum moment in the beam is: M1 = wEd,max(L 2 – 2e)2 /8 (3.15)
The beam end reaction is R1 = wEd,max L 2 /2 (3.16)
(Note the shear force in the beam is V Ed = wEd,max (L 2 – 2e)/2).
End reactions in the adjacent beams are R1 = wEd,minL1/2 and R3 = wEd,minL3/2 (3.17)
The resulting maximum bending moment in the column is given by
Mcol = (R 2 – R1) e h3/(h2 + h3) (3.18)
assuming that R1 < R 3 and h1 > h3. Figure 3.26a shows the final moments.
Bending moments in a pinned jointed frame for (a) internal beams, (b) upper floor columns.
( )
( ) Bending moments in a pinned jointed frame for (c) ground floor columns.
Figure 3.19b. The subframe consists of the column to be designed of height (distance between
centres of beam bearing) h2 , and half the adjacent columns of heights h1 and h3. Because the
column is continuous, the cross section and flexural stiffness EI of each part of the column
is considered as shown in the figure. The beams are pattern loaded as mentioned earlier,
of span L 4/2 and L 5/2, and the eccentricity of each beam end reaction from the centroidal
axis of the column is e 4 and e5, respectively. The moment at the upper end of the designed
column is given by
and at the lower end is
(3.19)
where R4 and R 5 are given in Equations 3.2 and 3.3. Figure 3.26b shows the final moments.
Note that patch loading produces single curvature in the columns.
Figure 3.19c The subframe consists of the column to be designed of height (distance between
the centre of first floor beam bearing and 50 mm below top of foundation (see section 9.4))
h1, and half the adjacent column of height h2. All other details are as before. If the foundation
is rigid (moment resisting), the moment at the upper end of the designed column is given by
Equation 3.5 with appropriate notation. The carry-over moment at the lower end is equal to
50% of the upper end moment. If the foundation is pinned, the upper end moment is given by
(3.20)
and the lower end moment is zero. Figure 3.26c shows the final moments. Patch loads pro-
duce single curvature in the columns.
Example 3.6
Determine, using substructuring techniques, the bending moments in the beam X and
columns Y and Z identified in Figure 3.27. The beam–column connections are pinned,
and the foundation is rigid. The distance from the edge of the column to the centre of
the beam end reaction is 100 mm. Characteristic beam loading is Gk = 40 kN/m and
Qk = 30 kN/m.
Solution
wEd,max = max{1.35 × 40 + 1.05 × 30; 1.25 × 40 + 1.5 × 30} = max {85.5; 95.0} = 95.0
kN/m; wEd,min = 40 kN/m.
Beam subframe
e = 450/2 + 100 = 325 mm
Equation 3.1. M1 = 95.0 × (8.000 – 2 × 0.325)2 /8 = 641.5 kNm
Column subframe
Beam end reactions R1 = 95.0 × 8.000/2 = 380.0 kN; R2 = 40 × 6.000/2 = 240.0 kN
e 1 = e 2 = 300/2 + 100 = 250 mm
but EI1/h1 = EI 2 /h 2
Equation 3.5. At upper and lower ends, Mcol = (380.0 – 240.0) × 0.250 × 0.5 =
17.5 kNm
Column subframe
Beam end reactions as before. e 1 = e 2 = 450/2 + 100 = 325 mm
Given that E is constant
Equation 3.5. At upper end, Mcol,upper = (380 – 240) × 0.325 × 451/(451 + 211) = 31.0 kNm
At lower end, Mcol,lower = 50% × 31.0 = 15.5 kNm.
Connections form the vital part of precast concrete design and construction. They alone can
dictate the type of precast frame, the limitations of that frame, and the erection progress.
It is said that in a load-bearing wall frame the rigidity of the connections can be as little as
1/100 of the rigidity of the wall panels −200 N/mm 2 per mm length for concrete panels ver-
sus 2.7–15.0 N/mm 2 per mm length for joints (Straman 1990). Moreover, the deformity of
the bedding joint, that is the invisible interface where the panel is wet bedded onto a mortar,
between upper and lower wall panels can be 10 times greater than that of the panel.
The previous paragraph contained the words connections and joints to describe very simi-
lar things. Connections are sometimes called ‘joints’ – the terminology is loose and often
interposed. The definition adopted in this book is as follows:
Connection: is the total construction between two (or more) connected components: it
includes a part of the precast component itself and may comprise several joints.
Joint: is the part of a connection at individual boundaries between two elements (the ele-
ments can be precast components, in situ concrete, mortar bedding, mastic sealant, etc.)
For example in the beam–column assembly shown in Figure 3.28, a bearing joint is made
between the beam and column corbel, a shear joint is made between the dowel and the
angle, and a bolted joint is made between the angle and column. When the assembly is
completed by the use of in situ mortar/grout, the entire construction is called a connection.
This is because the overall behaviour of the assembly includes the behaviour of the precast
components plus all of the interface joints between them. Engineers prove the capacity of the
entire connection by assessing the behaviour of the individual joints.
Structurally, joints are required to transfer all types of forces – the most common of these
being not only compression and shear, but also tension, bending and occasionally torsion.
The combinations of forces at a connection can be resolved into components of compressive,
tensile and shear stress, and these can be assessed according to limit state design. Steel (or
other materials) inserts may be included if the concrete stresses are greater than permissible
Moment and shear transfer at a bearing corbel.
values. The effects of localised stress concentrations near to inserts and geometric disconti-
nuities can be assessed and proven at individual joints. However, connection design is much
more important than that because of the sensitivity of connection behaviour to manufactur-
ing tolerances, erection methods and workmanship.
It is necessary to determine the force paths through connections in order to be able to
check the adequacy of the various joints within. Compared with cast in situ construction,
there are a number of forces which are unique to precast connections, namely frictional
forces due to relative movement causes by shrinkage, etc pretensioning stresses in the con-
crete and steel, handling and self-weight stresses. In the example shown in Figure 3.29, a
reinforced concrete column and corbel support a pretensioned concrete beam. The figure
shows that there are 10 different force vectors in this connection as follows:
1. A: diagonal compression strut in corbel

2. B: horizontal component reaction to force at A
3. C: vertical component reaction to force at A
4. D: internal diagonal resultant to forces B and C
5. E: diagonal compression strut in beam
6. F & G: horizontal component reaction to force E
7. H: tension field reaction to forces E & F
8. J: horizontal friction force caused by the relative movement of beam and corbel
9. K: horizontal membrane reaction to beam rotation due to eccentric prestressing
The structural behaviour of the frame can be controlled by the appropriate design of connec-
tions. In achieving the various structural systems in Section 3.3.2, it may be necessary to design
and construct either/both rigid and/or pinned connections. Rigid monolithic connections can
only truly be made at the time of casting, although it is possible to site cast connections that
have been shown to behave as monolithic, for example cast in situ filling of prefabricated soffit
Force paths in beam to column corbel connection.
beams before and after casting as shown in Figure 3.30a and b. The advantages lost to in situ
concreting work (cold climates in particular), the delayed maturity, the increase in structural
cross section, and the reliance on correct workmanship, etc. detract this solution in favour of
bolted or welded mechanical devices. Rigid connections may be made at the foundation where
there is less restriction on space as shown in Figure 3.31. In very simple terms, a bending
moment is generated by the provision of a force couple in rigid embedment, that is no slippage
when the force is generated. Pinned connections are designed by an absence of this couple,
although many connectors designed in this way inadvertently contain a force couple, giving
rise to spurious moments which often cause cracking in a region of flexural tension.
To gain an overview of the various types, Figure 3.32 and Table 3.4 show the locations,
classification and basic construction of connections in a precast structure.
In theory, no connection is fully rigid or pinned – they all behave in a semi-rigid manner,
especially after the onset of flexural cracking. Using a ‘beam-line’ analysis, Figure 3.33, we
can assess the structural classification of a connection. Although the beam-line approach was
developed for structural steelwork in c1936, research carried out since 1990 has shown that
the method is appropriate to precast connections (Elliott et al. 1998, Elliott et al. 2003a,b,
Ferriera et al. 2003, Elliott and Jolly 2013).
The moment–rotation (M-θ) diagram in Figure 3.33 is constructed by considering the two
extremes in the right hand part of Figure 3.33. The hogging moment of resistance of the beam
at the support is given by MRd > wL2/12, and the rotation of a pin ended beam subjected to a
UDL of w is θ = wL3/24EI. The gradient of the beam line is 2EI/L. The M-θ plot for plots 1
and 2 give the monolithic and pinned connections, respectively. In reality, the behaviour of a
connection in precast concrete will follow plots 3, 4 or 5, etc. If the M-θ plot for the connec-
tion fails to pass through the beam-line, that is plot 5, the connection is deemed not to possess
sufficient ductility and should be considered in design as ‘pinned’. Furthermore, its inherent
(a)
(b)
Cast concrete topping over precast soffit beams forms a fully rigid connection.
(a) Prepared for casting and (b) as cast and awaiting structural topping to also cover the slabs.
stiffness (given by the gradient of the M-θ plot) is ignored. Conversely, if the M-θ behaviour
follows plot 3 (the gradient must lie in the shaded zone and the failure takes place outside the
shaded zone), the effect of the connection will not differ from a monolithic by more than 5%.
Structural stability and safety are necessary considerations at all times during the erection
of precast concrete frames. The structural components will not form a stabilising system
until the connections are completed – in some cases, this can involve several hours of matu-
rity of cast in situ concrete/grout joints, and several days if structural cast in situ toppings
Site placed infill
50 ignored
F
1.5h approx
F
Precast column to pocket connection.
Types of connections in a precast structure.
are used to transfer horizontal forces. A stabilising system must comprise two things as
shown in Figure 3.34:
1. A horizontal system, often called a ‘floor diaphragm’ because it is extremely thin in

relation to its plan area
2. A vertical system in which the reactions from the horizontal system are transferred to
the ground (or other substructure)
Types of connections in precast frames
Beam – column head 1 Pinned Dowel

Beam – column head 2 Rigid Dowel plus continuity top steel
Rafter – column head 3 Pinned Dowel
Rafter – column head 4 Rigid Bolts (couple)
Column splice 5 Pinned Bolts/dowel
Rigid Bars in grouted sleeve (couple)
Threaded couplers
Steel shoes
Beam – column face 6 Pinned Bolts
Welded plates
Notched plates
Dowels
Beam – column corbel 7 Pinned Dowel
Beam – column corbel 8 Rigid Dowel plus continuity top steel
Beam – beam 9 Pinned Bolts
Dowels
Slab – beam 10 Pinned Tie bars
Slab – wall 11 Pinned Tie bars
Column – foundation 12 Pinned Bolts
Column – cast 13 Pinned or rigida Bolts
beam or retaining wall Rebars in grouted sleeve
a Depending on the design of the cast substructure.
2 1
MR = wL Rigid M
12
3
Pinned M=0
Failure of
connection θ
Semi
Be
rigid M
am
5
-li
θ
ne
2
θ
3
θR = wL
24EI
Definition of moment–rotation characteristics.
Loa
din
gx
Ch
n gy ord
adi x
Lo rd
y
o
Ch
y
rd
cho
Tie Bracing x
Strut x
Strut y
Bracing y
Reactions x
Reactions y
Stabilising systems in braced frames.
The horizontal system is considered in detail in Chapter 8 where reference is also made to
the many code regulations on this topic. When subjected to horizontal wind or lack-of-
plumb forces, the floor slab acts as a deep beam and is subjected to bending moments Mh
and shear forces Vh (h being the subscript used for horizontal diaphragms). The basic design
method is shown in Figure 3.35. The design is a three-stage approach:
1. The floor diaphragm is analysed as a long, deep beam which is supported by a number
of shear walls, shear cores, deep columns (wind posts), or other kinds of bracing such
as steel cross bracing. Figure 3.35a.
2. If there are only two supports (bracing), the analysis is statically determinate and MEd,h
and VEd,h may be calculated directly. If there are more than two supports, irrespec-
tive of where they are positioned, the analysis is statically indeterminate. The support
reactions must first be found by a technique which considers the relative stiffness and
position of each support, and the horizontal (e.g. wind load) pressure distribution.
The derivation is given in Section 8.1 after which MEd,h and VEd,h may be calculated.
3. The area of reinforcement required to resist M Ed,h and V Ed,h is determined as follows:
A sh = M Ed,h γm /0.8 B f yk (3.21)
where
0.8 B is the assumed lever arm between the compression zone and the tie steel (the
assumption is known to be conservative)
f yk /γm is the design stress in the tie steel with γm = 1.15
High tensile rebar with fyk = 500 N/mm2 or standard helical strand with fpk = 1770 N/mm2 is
used (super strand or Dyform tend to be too stiff to handle) – the reasons are given in Section 8.4.
A shd = V Ed,h γm /0.6 μ f yk (3.22)
where μ is the coefficient of friction as given in BS EN 1992-1-1, clause 6.2.5(2). Hollow core
slabs are considered as being untreated and smooth, then μ = 0.6 with no special, that is ex-
factory, edge preparation (see Section 8.2). For ex-steel mould with smooth surfaces, μ = 0.5.
4. The tie steel A shd must be placed everywhere moments occur. The tie steel A svh must
be placed only where the shear force is greater than a certain value. This is found by
checking that the interface shear stress vEdi = V Ed,h /B (D – 30 mm) does not exceed
vRdi ≤ 0.15 N/mm2 for smooth and rough surfaces (as in the case of machine cast hollow
core units and as-cast precast planks) or vRdi ≤ 0.10 N/mm 2 for very smooth surfaces
cast against steel moulds, according to BS EN 1992-1-1, clause 10.9.3(12). (The reason
for the deduction of 30 mm is explained in Section 8.4.1.)
Horizontal force (pressure)
Compression field Reactions

(diagrammatic only) (bracing elements)
(a) Tension chord
Section at welded joints
Mesh typically A142 or A193
Structural concrete topping

screed – 40 mm min. depth
at highest point in slab
Site welded connections

are ignored in diaphragm
calculations
(b)
Diaphragm floor action. (a) Deep beam analogy. (b) Reinforced structural topping in double-
tee floors. ( )
30 mm minimum
Site placed reinforcement into

slot formed in precast slabs
In situ concrete edge

(or internal) beam = chord
element in diaphragm
500 mm
Section through slot

(c) in hollow core slab
( ) Diaphragm floor action. (c) Perimeter reinforcement in hollow core floors.
Diaphragms may be reinforced in several ways. In Figure 3.35b, a reinforced cast in situ top-
ping transfers all horizontal forces to the vertical system – the precast floor plays no part but
for restraining the topping against buckling. In Figure 3.35c, there is no cast in situ topping.
Perimeter and internal tie steel resists the chord forces resulting from horizontal moments.
Coupling bars are inserted into the ends of the floor units, and together with the perimeter
steel provides the means for shear friction generated in the concrete-filled longitudinal joints
between the units.
Example 3.7
Determine the shear wall reactions and diaphragm reinforcement in the floor shown in
Figure 3.36a. The precast units are 150 mm deep hollow cored and have an ex-factory
edge finish. The characteristic wind pressure on the floor wk = 3 kN/m. Tie steel is high
tensile ribbed bar f yk = 500 N/mm 2 . Suggest some reinforcement details.
Floor units
3 kN/m
Shear walls
or bracing
5.0 m
24.0 m 6.0 m
(a)
4.2 kN/m
47.2 kN 78.8 kN
Mmax at V = 0
11.24 m
(b) Shear force diagram
Detail to Example 3.7. (a) Plan view of floor diaphragm, (b) Wind loading diagram and shear
force diagram in the floor diaphragm.
Solution
NA to BS EN 1990, Table NA.A1.2 (B) – Design values of actions (STR) (Set B) γW = 1.5
Design ultimate wind load = 1.5 × 3.0 = 4.5 kN/m.
From Figure 3.36b, support reaction R1 = 50.62 kN; R 2 = 84.38 kN.
Shear span from LHS (distance to zero shear and hence point of maximum
moment) = 50.62/4.5 = 11.25 m.
MEd,h,max = 50.62 × 11.25/2 = 284.74 kNm; VEd,h,max = 57.38 kN at RHS of 24 m span.
A shd = 284.74 × 106/0.8 × 5000 × (500/1.15) = 164 mm 2 . Use 2 no. H12 bars (226).
Interface shear stress vEdi = 57.38 × 103/5000 × (150 – 30) = 0.096 N/mm 2 < 0.15
N/mm 2 allowed. No shear reinforcement needed.
Vertical stabilising systems are dictated by the necessary actions of the structural system,
that is skeletal, wall or portal frame. Column effective lengths depend on the type and direc-
tion of the bracing. However, there is a broad classification as the structure is
1. Unbraced frame, Figure 3.37, where horizontal force resistance is provided either by
moment resisting frame action, cantilever action of columns, or cantilever action of
wind posts (deep columns)
2. Braced frame, Figure 3.38, where horizontal force resistance is provided either by
cantilever action of walls or cores, in-plane panel action of shear walls or cores, infill
walls, cross bracing, etc. or
3. Partially braced frame, Figure 3.39, which is some combination of (1) and (2)
Column lo = 2.083 l
Rigid joints Column lo ≈ 1.5 l Pinned joints Column lo = 1.0 l
lo ≈ 2.2 l
Frame action Cantilever columns Cantilever wind post
Alternative sway mechanisms and resulting column effective length factors.
All columns lo = 1.0 l
Alternative full-height bracing mechanisms and resulting column effective length factors.
The type of stabilising system may be different in other directions. The floor plan arrange-
ment and the availability of shear walls/cores will dictate the solution. The simplest case is a
long narrow rectangular plan where, as shown in Figure 3.40a, shear walls brace the frame
in the y direction only, the x direction being unbraced. In other layouts, shown for example
in Figure 3.40b, it is nearly always possible to find bracing positions. Precast skeletal frames
Column lo = 2.083 l
Column lo = 1.0 l Column lo = 1.09 l Column lo = 2.0 l
Alternative partial height bracing mechanisms and resulting column effective length factors.
Floor units
y Shear walls
Column and
or other bracing
(a) x beam frame
Stairs
Service core
lifts etc.
Stairs
(b)
Positions of shear walls and cores in alternative floor plan layouts. (a) Positions of shear walls,
(b) positions of shear cores or walls around stairs and lift shafts. ( )
Braced Unbraced
Shear core
Construction and
movement gap
(c)
( ) Positions of shear walls and cores in alternative floor plan layouts. (c) bracing meth-
ods and positions in partially braced irregular or non-symmetrical buildings.
of three or more storeys in height are mostly braced or partially braced. This is to avoid hav-
ing to use deep columns to cater for sway deflections, which give rise to large second-order
bending moments. Section 6.2.6 refers in more detail.
It is not wise to use different stabilising systems acting in the same direction in different
parts of a structure. The relative stiffness of the braced part is likely to be much greater than
in the unbraced part, giving rise to torsional effects due to the large eccentricity between the
centre of external pressure and the centroid of the stabilising system, as explained in Figure
3.40c. The different stabilising systems should be structurally isolated – Figure 3.40d.
In calculating the position of the centroid of a stabilising system, the stiffness of each com-
ponent of thickness t and length L is given by Ecm,long I, where Ecm,long = long-term Young’s
modulus (usually taken as 15 kN/mm 2) and I = tL 3/12. First moments of stiffness are used
to calculate the centroid as explained in Example 3.8.
Example 3.8
Propose stabilising systems for the five-storey skeletal frame shown in Figure 3.41a.
The beam–column connections are all pinned, and the columns should be the minimum
possible cross section to cater for gravity loads. Wind loading may be assumed to be uni-
form over the entire façade. Use only shear walls for bracing.
Hint: the grid dimensions around the stairwell may be taken as 4 m × 3 m, and at the lift
shaft 3 m × 3 m.
Solution
A braced frame is required up to the fourth floor, after which a one-storey unbraced
frame may be used. It would not otherwise be possible to satisfy the requirement of
minimum column sizes for gravity loads. To avoid torsional effects (see Figure 3.40c),
the centroid of the stabilising system should be as close as possible to the centre of exter-
nal pressure, that is at x ≈ 24 m and y ≈ 16 m. It is necessary to first consider the two
orthogonal directions.
Stability in -direction
The centroid of the stability walls x′ ≈ 24 m.
36.0 m 12.0 m No visual obstruction

except for lift shaft
Stairs
Columns
4 bays at 8.0 m = 32.0 m

shown thus
Opaque obstruction
Lift shaft
allowed only in lift

and stair wells
Span
Floor
Stairs
6 bays at 8.0 m = 48.0 m

Pinned or rigid
(a) foundation
4.0
16.0
x΄ = 26.3
3.0
12.0
y΄ = 16.0
Shear centre
4.0
3.0 33.0 3.0 6.0 3.0

Origin
(b) 48.0
Detail to Example 3.8 (dimensions in metres). (a) Plan view and cross-section of framed struc-
ture, (b) Plan view showing positions of shear walls.
Select walls as shown in Figure 3.41b. On the assumption that the material and con-
struction of all walls is the same, Young’s modulus and thickness of wall are common to
all walls and need not be used in the calculation.
´ + ´ + ´ + ´ + ´ + ´
Centroid of stiffness ¢= = ,
´ + ´
which is sufficiently close to the required point to eliminate significant torsional effects.
Stability in -direction
The centroid of the stability walls y′ ≈ 32/2 = 16 m.
´ + ´ + ´
Centroid of stiffness ¢= = , which is at the correct point.
´
To assist the transition between the British code BS 8110:1997 and the Eurocodes EC0, EC1
and EC2, the following precast reinforced and prestressed concrete elements are designed:
1. Reinforced concrete rectangular beam

2. Reinforced concrete rectangular column
3. Prestressed concrete solid floor slab
Clauses and tables in the codes are indicated within ‘{}’.
A 600 mm deep × 300 mm wide r.c. beam carries uniformly distributed dead and live load-
ing of 40 and 30 kN/m over a simply supported clear span of 5.85 m with bearing lengths of
150 mm. The beam carries office loading. The exposure is internal, and the fire resistance is
60 min. The design life is 50 years.
Design the main reinforcement at mid-span and the shear reinforcement at the support
and position where nominal links are required. Calculate the crack width and the long-term
deflection using the appropriate creep factor and check the span/depth ratio for the area of
rebars designed. Use fck /fcu = 32/40, high tensile main bars and links grade f yk /fy = 500 N/mm 2 ,
and normal-weight concrete with a 20 mm coarse aggregate.
BS 8500-1. Table A.4 for 50 years

Exposure XC1. Cover = 15 + Δ Exposure XC1. = 15 + Δ
{7.3} Δ = 5 mm {4.4.1.3(3)} Δ = 5 mm
Cover to links = 20 mm Cover to links = 20 mm
. 1 h . R60. BS EN 1992-1-2
{Table 3.4} = 20 mm {Table 5.5} for = 300 mm, axis = 25 mm
Let links ϕ = 8 mm ∴ = max{20 + 8 = 28 mm; 25 mm}
Assume main bars ϕ = 32 mm
= 300 mm; = 600 – 20 – 8 – 16 = 556 mm = 300 mm; = 600 – 28 – 16 = 556 mm
{BS EN 1991-1-1, Table A1.1}

= 0.6 × 0.3 × 24 = 4.32 kN/m = 0.6 × 0.3 × 25 = 4.5 kN/m
{BS EN 1990, Table A1.2(B) and
{Table 2.1} Table A1.1} ψ0 = 0.7
( )
= 1.4 × 44.32 + 1.6 × 30 = 110.1 kN/m {Exp. 6.10a; 6.10b} = max{1.35 × 44.5 + 0
0.7 × 1.5 × 30; 1.25 × 44.5 + 1.5 × 30} = 100.7 kNm
{3.4.1.2} {5.3.2.2, Fig. 5.4a}
= min{5.85 + 0.15; 5.85 + 0.46} = 6.0 m = min{5.85 + 0.15; 5.85 + 0.46} = 6.0 m
= 110.1 × 6.02/8 = 495.5 kNm = 100.7 × 6.02/8 = 453.1 kNm
{3.4.4.4} = 495.5 × 106/40 × 300 × 5562 {3.1.7(3)} = 453.1 × 106/32 × 300 × 5562
= 0.134 < 0.156 for / ≤ 0.5 = 0.153 < 0.206 for / ≤ 0.6
/ = 0.5 + √0.25 – /0.9 = 0.82 < 0.95 / = 0.5 + √0.25 – /1.133 = 0.84 < 0.95
= 0.82 × 556 = 455 mm = 0.84 × 556 = 467 mm
= 495.5 × 106/455 × 0.87 × 500 = 2503 mm2 = 453.1 × 106/467 × 0.87 × 500 = 2230 mm2
Use 2 no. H32 ± 2 no. H25 bars (2,590) Use 3 no. H32 bars (2,412)
Spacing = (300 – 88 – 114)/3 = 33 mm Spacing = (300 – 88 – 96)/2 = 58 mm
{3.12.11.1} > 20 + 5 = 25 mm OK {Table 7.3N} < 100 mm for any value of σ
{Table 7.1, NAD Table NA.4} = 0.3 mm
Although compression steel is not required
{Table 3.25} Min ′ = 0.2% = 360 mm2 {9.2.1.1(1)} Min ′ = 0.13% = 234 mm2
Use 2 no. H16 (402) at ′ = 36 mm Use 2 no. H16 (402) at ′ = 36 mm
Comments. EC2 requires 11% less area of rebar.
{3.4.5.3} = 5.85 – 2 × 0.556 = 4.74 m {6.2.1(8)} = 5.85 – 2 × 0.556 = 4.74 m

= 110.1 × 4.74/2 = 261.0 kN = 100.7 × 4.74/2 = 238.7 kN
{3.4.5.3} = 261.0 × 103/300 × 556 {6.3.2. Exp. 6.9} max = 1 0.5 sin 2θ
= 1.56 N/mm2 < 0.8√ = 5.06 N/mm2 1 = 0.6 (1 − /250), = 0.9 and = /1.5
> = 0.79 × 1.51/3 × 1.17/1.25 = 0.84 N/mm2 θ = 0.5 sin−1 (238,700/(0.5 × 0.523 × 300 × 501
where 100 / = 1.5, ( /25)1/3 = 1.17 × (32/1.5)) = 8.2o < 22.5o ∴ use cot θ = 2.5
{Table 3.7} > {6.3.2. Exp. 6.8}
/ = 300 × (1.56 – 0.84)/(0.87 × 500) / = 238,700/501 × 0.87 × 500 × 2.5
= 0.5 mm2/mm = 250 mm2/m/leg = 0.438 mm2/mm = 219 mm2/m/leg
{3.4.5.5} s < 0.75 × 556 = 417 mm {9.2.2(6)} ≤ 0.75 = 417 mm
Use H8 links at 200 mm c/c (250) Use H8 links at 225 mm c/c (222)
Nominal where = + 0.4 = 1.24 N/mm2 {Exp. 9.5N} min/ = 0.08 × √32 × 300/500
or where shear force = 0.272 mm2/m
∴ = 1.24 × 300 × 556 × 10–3 = 207 kN ∴ min = 238.7 × (0.272/0.438) = 148.0 kN
at 1,530 mm from centre of support. at 1,120 mm from the centre of support
.
EC2 requires 13% less area of links, and nominal links to EC2 start at 1.36 times the distance
for BS 8110.
Short-term Young’s modulus

{2.5.4} = 200 kN/mm2 {3.2.7(4)} = 200 kN/mm2
{Part 2, 7.2} = 20 + 0.2 × 40 = 28 kN/mm2 {Table 3.1} = 22 (40/10)0.3 = 33.34 kN/mm2
α = 200/28 = 7.14 α = 200/33.34 = 6.00
Long-term Young’s modulus
Bottom and sides exposed = 2 / = 360,000/
(300 + 2 × 600) = 240 mm
( )
Age at loading = 28 days. Indoor exposure

RH = 50%
{Part 2, Fig. 7.1} φ = 2.45 {Fig. 3.1a} φ(∞ ) = 2.4
{Part 2, 3.6} = 28/3.45 = 8.11 kN/mm2 {7.20} = 33.34/3.4 = 9.80 kN/mm2
∴α = 200/8.11 = 24.64 ∴ α = 200/9.80 = 20.39
α − 1 (uncracked concrete) = 23.64 α − 1 = 19.39
Not required in BS 8110 + Σ(α – 1) = 234,587 mm2

= 300 × 6002/2 + 19.39 × (2,590 × 566 +
402 × 36)/234,587 = 342.3 mm from top
= 8,590 × 106 mm4
= 8,590/257.5 = 33.36 × 106 mm4
{7.4.3} = 33.36 × 3.02 = 100.8 kNm
{Table 3.1} = 0.3 × 322/3 = 3.02 N/mm2
= (44.5 + 0.3 × 30) × 6.02/8 = 240.7 kNm
> use partially cracked
{Part 2, 3.6} Instantaneous value

Solving first m.o.a. 2/2 + (α − 1) ′( – ′) = α
( – )
= 202.2 mm
= 3211 × 106 mm4
Long-term value
Solving first m.o.a. 2/2 + (α ×1) ′( – ′) = α
( – )
= 302.0 mm = 279.4 mm
= 7545 × 106 mm4 = 6,407 × 106 mm4
1
{Part 2, 3.3.3) ψ = 0.25 offices {Exp. 7.18 and 7.19} better presented as
Instantaneous curvature (×10–6 mm−1 units) = + [( – ) 0.5 ( / )2]
δ = 44.32 + 0.25 × 30 = 51.82 kN/m = 6407 + [(8,590 – 6,407) × 0.5 × (100.8/240.7)2] =
6599 × 106 mm4
= 51.82 × 6.02/8 = 233.2 kNm
1 233.2/(28,000 × 3,211) = 2.59
= 44.32 × 6.02/8 = 199.4 kNm (dead)
1 199.4/(28,000 × 3,211) = 2.22
Long-term curvature Long-term curvature (×10–6 mm−1 units)
1 199.4/(8,116 × 7,545) = 3.26 1 /
1 3.26 + (2.59 − 2.22) = 3.63 1 240.7/9,806 × 6,599 = 3.72
{Part 2, 3.7.2} Deflection δ = 2 (1/ ) δ= 2 (1/ )
{Part 2, Table 3.1} = 0.104

δ = 0.104 × 60002 × 3.63 × 10–6 = 13.6 mm δ = 0.104 × 6,0002 × 3.72 × 10–6 = 13.9 mm
{3.4.6.3} /δ = 440 > 250 OK {7.4.1(4)} /δ = 431 > 250 OK
{3.4.6.1} / = 20 × 0.744 = 14.87 {7.4.2(2)} / = 11 + 1.5√32 × 0.432 = 14.67

( )
{Table 3.9} Basic / = 20 where ρ /ρ = 0.432; ρ = 0

{3.4.6.5} MF = 0.55 + [(477 – 332) ρ = 10–3 √32 = 0.00566
/(120 (0.9 + 5.34))] = 0.744 ρ = 2,230/300 ×566 = 0.0131 > ρ {Exp. 7.16b}
{Table 3.10} = (2/3) × 500 × (2,503/2,590) (Table 7.4N} =1
= 332 N/mm2
/ 2 = 495.5 × 106/300 × 5562 = 5.34
∴ > 6,000/14.87 = 404 mm < 556 OK ∴ > 6,000/14.67 = 409 mm < 556 OK
.
Deflections and span/depth ratios are the same in spite of different approaches in the two codes and that
EC2’s = 1.19 . Changing the creep coefficient φ from 2.4 to 1.5 results in only a 1.3 mm reduction in
deflection, showing that φ is not a sensitive parameter.
{Part 2, 3.8.3} /2 = 14 kN/mm2 {7.3.4(1)} = max (ε − ε )

Spacing = 3 × 28 = 84 mm {7.3.4(3)} max = 3.4 × 28 + 0.8 × 0.5 × 0.425 ×
32/0.075 = 168 mm
= 84 × 736 × 10–6 = 0.06 mm = 168 × 913 × 10–6 = 0.15 mm
where ε = ε1 − ε′ = 804 – 68 = 736 × 10–6 where ε − ε = [206 – (0.4 × 3.02/0.075) ×
ε1 = ( – )/ /2 = 334.4 × (1 + 6.0 × 0.075)]/200,000 = 913 × 10–6
(556 – 302)/(14,000 × 7545) = 804 x 10–6 but > 0.6 × 206/200,000 = 618 × 10–6
where = 74.32 × 6.02/8 = 334.4 kNm {7.3.4(2)}ρ = 2412/300 × 107 = 0.075
= 44.32 + 30 = 74.32 kN/m {7.3.2(3)}Minimum is = ( – )/3
{Part 2, Eq. 13} ε′ = 300 × 298 × 302/(3 × = (600 – 279.4)/3 = 107 mm
200,000 × 2,590 × 254) = 68 x 10–6 {7.1(2)} = = 0.3 × 322/3 = 3.02 N/mm2
σs = α ( – )/ = 20.39 × 240.7 × (556 −
279.4)/6599 = 206 N/mm2
{3.2.4} < max = 0.3 mm {Table NA.4} < max = 0.3 mm for XC1
.
Crack width is greater in EC2 due to increased crack spacing, that is 84 and 168 mm in the two codes.
The final effective strains are also greater in EC2, possibly because of the assumption in BS 8110 using
/2 = 14 kN/mm2 compared to the long-term value in EC2.
A two-storey 300 mm × 300 mm edge column supports beams on one side only in an
unbraced sway frame as shown in Figure 3.42a. The exposure, fire resistance, design life
and beam end reactions are as given in 3.6.1, except that the roof beam may be taken as
60% of the floor beam reactions. The beam reactions act at 80 mm from the face of the
column. The flexural stiffness of the beam-to-column connection may (in this exercise) be
taken as 1/10 of that of the column. The characteristic horizontal wind load is shown in
Figure 3.42a.
Design the main reinforcement and specify the shear links. Use fck/fcu = 40/50, fyk/fy =
500 N/mm2, and normal-weight concrete. Effective creep factor φef = 1 (used in EC2).
Moment distribution factors at the first floor (upper end of column is pinned, lower end
at foundation is fixed) = 4EI/3.5 /(4EI/3.5 + 3EI/3.0) = 54% with 50% carryover (c/o) to
the foundation.
Pinned joints with 1/10

column stiffness
Wk = 7 kN
Floor beams
3000
designed in 3.6.1
Clear 6100
Wk = 14 kN
All columns 300 × 300
3500
Clear 2900
(a)
e Vroof
54% M = Veℓat first floor
Voor
50% carry over

to foundation
(b) First order frame moments Due to wind loads Second order deflection Second order moments
Detail to code comparison for reinforced concrete column. (a) Frame arrangement, (b) bending moments
due to beam eccentricity, horizontal loads (wind and imperfections) and second-order deflections.
is the same as for the beam, ∴ cover to links

20 mm
. 1 h . R60. BS EN 1992-1-2
{Table 3.4} = 20 mm {BS EN 1990, Table A1.1} In fire ψ2 = 0.3 for
offices beam load and ψ2 = 0 for wind load
( )
= 44.5 × 6.0/2 = 133.5 kN per beam

= 30 × 6.0/2 = 90 kN per beam
{2.4.2} max = 133.5 + 0.3 × 90 = 160.5 kN
Per floor plus 60% at roof = 96.3 kN
Self-weight = 0.3 × 0.3 × 25 × 3 = 6.75 kN
at first floor level = 424 kN
Eccentricity of reaction = 150 + 80 = 230 mm
= 160.5 × 0.23 = 36.9 kNm × 54% as
distributed = 19.9 kNm
{5.3.2} = 19.9/424 = 0.047 m = 47 mm
{5.3.2. Method B, Table 5.2b}as 0 > 3.0 m
/ = 47/300 = 0.16 < 0.25
Then = /(0.7( s ))
Try 4 H25 bars = 1963 mm2
= 424/(0.7 (3002 × 26.67 +1963 × 435) x 10–3
= 0.19 and ω = 1963 × 435/3002 × 26.67 = 0.36
{Table 5.2b} R60 requires min/ = 300/25
But 25 mm < + ϕ + ϕ /2 = 20 + 8 + 12
= 40 mm ∴ not critical
Assume main bars ϕ = 25 mm Assume main bars ϕ = 20 mm

{3.12.7.1} Links ≥ ϕ /4 use 8 mm {9.5.3(1)} Links ≥ ϕ/4 use 8 mm
= 300 mm; = 300 − 20 − 8 − 12 = 260 mm = 300 mm; = 300 − 40 = 260 mm
/ = 0.87. Use design chart in Figure 3.43a / = 0.87. Use design chart in Figure 3.43b
Self-weight = 0.3 × 0.3 × 24 × 6.5 = 14.0 kN Self-weight = 0.3 × 0.3 × 25 × 6.5 = 14.6 kN
{Part 2, clause 2.5} β = 2 + 0.3 α ,min where {5.8.3.2, Exp. 5.16}

α = 1.0 for foundation and 10 given for beam 1 = 0.1 and 2 = 1/( /10) = 10
1
β = 2 + 0.3 × 1.0 = 2.3 β = (1 + 0.1/1.1) × (1 + 10/11) = 2.08

Clear first floor 0 = 3500 – 600 = 2,900 mm
= 2.3 × 2,900 = 6,670 mm. / = 22.2 0 = 2.08 × 2,900 = 6032 mm. λ = l0/ = 69.6
{3.8.1.3} For unbraced / > 10 ∴ slender where = 300/√12 = 87 mm
{3.8.1.7} / < 60 limit {5.8.3.1(1)} λ = 20 ABC/ = 33.4
{3.8.3.1} = 22.22 × 300 /2,000 = 73.9 where = 1/(1 + 0.2φ ) = 1/1.2 = 0.83
= + w= 1.31 where ω = 0.36
= 1.7 − = 0.7 where = +1 for unbraced
= 502 × 103*/3002 × 26.67 = 0.209
* = 502 kN (see case 1 in the following)
λ >λ 69.6 > 33.4 ∴ slender
{5.8.8.2(3)} 2 = 19.4 × 10–6 × 6,0322/10 =
71 mm
{5.8.8.3(1)} where 1/ = φ /0.45 = 1.05 ×
435/0.45 × 260 × 200,000= 19.4 × 10–6 mm−1
{5.8.8.3(3)} Figure 3.43b = 1 (because
< 0.25 see the following)
( )
{5.8.8.3(4)} φ = 1 + (0.35 + /250 − λ/150) φ

= 1 + (0.35 + 0.16 − 69.6/150) × 1 = 1.05
Clear height to roof 0 = 6,500 – 400 = 6,100 m
= 2.3 × 6,100 = 14,030 mm. / = 46.7 < 60 0= 2.08 × 6,100 = 12,688 mm. λ = 0/ = 146
{3.8.3.1} = 46.7 2 × 300 /2,000 = 327 2= 18.6 × 10–6 × 12,6882 /10 = 299 mm
{3.8.3.8}All columns have equal stiffness where φ = 1 and 1/ = 18.6 × 10–6 mm−1
{3.8.5} Column deflection checked as / > 30
Refer to Figure 3.42b

= (7.0 × 6.5 + 14.0 × 3.5)/3 = 31.5 kNm = 31.5 kNm
{3.1.4.2} Total for floor beams + self-weight {5.2(5)} Imperfection at first floor
= 44.32 × 12 + 3 × 14 = 574 kN α = 2/√3.5 = 1.069 use 1, α = 1
= 1.5% = 0.015 × 574 = 8.6 kN {5.2(7)} = (1/200) × 1 × 6.032 /2 = 0.015 m
<14 kN wind load = 302.1* × 0.015 = 4.6 kNm *see the
following
At roof = 0.6 × 8.6 = 5.2 kN At roof α = 2 /√6.5 = 0.784, α = 1
<7 kN wind load = (1/200) × 0.784 × 12.688 /2 = 0.025 m
= 181.2* × 0.025 = 4.5 kNm *see the
following
= 4.6 + 4.5 = 9.1 kNm (added to γ )
1
= 110.1 × 6.0/2 = 330.3 kN = 100.7 × 6.0/2 = 302.1 kN
= 0.6 × 330.3 = 198.2 kN = 0.6 × 302.1 = 181.2 kN
= 330.3 + 198.2 + 1.4 × 14.0 = 548 kN = 302.1 + 181.3 + 1.25* × 14.6 = 502 kN
Refer to Figure 3.42b
Node moment 0 = , where = 150 + 80 *γ = 1.25 is for critical beam load Exp. 6.10b
= 230 mm from the centre of column = 230 mm
0 = 330.3 × 0.23 = 76.0 kNm × 54% as 0 = 302.1 × 0.23 × 54% × 50% c/o
distributed x 50% c/o = 20.5 kNm = 18.8 kNm
{Eq. 35} = = 330.3 × 0.0739 {5.8.8.2(3)} 2 = 2 = 302.1 × 0.071
+ 198.2 × 0.327 = 89.2 kNm + 181.3 × 0.299 = 75.7 kNm
= 20.5 + 89.2 = 109.7 kNm = 18.8 + 75.7 + 9.1 = 103.6 kNm
{3.8.2.4} min = 548 × 0.020 = 11.0 kNm {6.1(4)} min = 502 × 0.020 = 10.0 kNm
= 548 × 103/50 × 3002 = 0.12 = 502 × 103/40 × 3002 = 0.14
2 = 109.7 × 106/50 × 3003 = 0.081 2 = 103.6 × 106/40 × 3003 = 0.096
Figure 3.43a gives = 1 Figure 3.43b confirms = 1

∴ = 0.12 × 50 × 300 × 300/500 = 1080 mm2 ∴ = 0.15 × 40 × 300 × 300/500 =
1080 mm2
2
γ = all 1.2 Table 3.3. Exp. 6.10b is critical over 6.10a
γ = 1.25, γ = 1.5, γ = 0.75
= 1.2 × 74.32 × 6.0/2 = 267.5 kN as case 1. as case 1 + wind
= 0.6 × 267.5 = 160.5 kN
= 267.5 + 160.5 + 1.2 × 14.0 = 445 kN
( )
0 = 267.5 × 0.23 × 54% × 50% = 16.6 kNm

= 267.5 × 0.0739 + 160.5 × 0.327 = 72.2 kNm
= 1.2 × 31.5 = 37.8 kNm = 0.75 × 31.5 = 23.6 kNm
= 16.6 + 72.2 + 37.8 = 126.6 kNm = 103.6 + 23.6 = 127.2 kNm
= 0.10 and 2 = 0.094 = 0.14 and 2 = 0.118
∴ = 0.19 × 50 × 300 × 300/500 = 1710 mm2 ∴ = 0.21 × 40 × 300 × 300/500 = 1,512 mm2

3
γ = 1.0 and 1.4 γ = 1.25, γ = 1.5
= 1.0 × 44.32 × 6.0/2 = 133.0 kN = 44.5 × 6.0/2 = 133.5 kN
= 0.6 × 133.0 = 79.8 kN = 0.6 × 133.5 = 80.1 kN
= 133.0 + 79.8 + 14.0 = 226.8 kN = 133.5 + 80.1 + 14.6 = 228.2 kN
0 = 133.0 × 0.23 × 54% × 50% = 8.3 kNm 0 = 133.5 × 0.23 × 54% × 50% = 8.3 kNm
= 133.0 × 0.0739 + 79.8 × 0.327 2 = 133.5 × 0.071 + 80.1 × 0.299 = 33.4 kNm
= 35.9 kNm = 133.5 × 0.015 + 80.1 × 0.025 =
4.0 kNm
= 1.4 × 31.5 = 44.1 kNm = 1.5 × 31.5 = 47.2 kNm
= 8.3 + 35.9 + 44.1 = 88.3 kNm = 8.3 + 33.4 + 4.0 + 47.2 = 92.9 kNm
= 0.05 and 2 = 0.065 = 0.063 and 2 = 0.086
∴ = 0.13 × 50 × 300 × 300/500 = 1170 mm2 ∴ = 0.19 × 40 × 300 × 300/500 = 1368 mm2

Maximum = 1710 mm2 (Case 2) Maximum = 1,512 mm2 (Case 2)
Use 4 no. H25 bars (1963) Use 4 no. H25 bars (1963)
{3.12.7.1} Links ≥ ϕ /4 use 8 mm {9.5.3(1)} Links ≥ ϕ /4 use 8 mm
Spacing s = 12ϕ = 300 mm {9.5.3(3)} s = min {20ϕ, b, h, 300} = 300 mm
{9.5.3(4)} = 0.6s within at top and bottom
{3.12.6.2} max = 0.08 × 3002 = 7,200 mm2 {9.5.2(3)} max = 0.04 × 3002 = 3,600 mm2
min = 0.004 × 300 = 360 mm {9.5.2(2)} min = 0.002 × 3002 = 180 mm2 or
2 2
0.1 × 502 × 103/435 = 115 mm2
The Eurocodes require 12% less , due mainly to reduced γ (1.25 & 1.5 versus 1.4 & 1.6) which reduces
by 9%. First- and second-order bending moments are roughly equal.
Calculate the service and ultimate moment of resistance, and the ultimate shear capacity of a
1200 mm wide × 200 mm deep prestressed concrete solid floor slab. The slab is pretensioned
using 6 no. 12.5 mm plus 6 no. 9.3 mm diameter standard 7-wire helical strands at 30 mm
bottom cover, plus 4 no. 5 mm diameter wires at 25 mm top cover. The tendons are Class 2
relaxation and stressed initially at 70%. The exposure is internal with relative humidity = 50%,
fire resistance is 60 minutes, and design life is 50 years. Bearing length is 100 mm. The slab
is designed as a Class 2 member for permissible tension to BS 8110. The floor carries office
loading for characteristic dead loads of 1.5 kN/m2 finishes and 0.6 kN/m2 services/ceiling, a
superimposed live load of 4.0 kN/m2 and demountable partitions of 1.0 kN/m2.
The section properties are cross-sectional area Ac = 232,040 mm2; second m.o.a Ix-x = 779 ×
106 mm4; depth = 200 mm; centroid from bottom yb = 99.4 mm; breadth at top, bottom and
at centroid = 1154, 1197 and 1134 mm, respectively. Height to the centroid of all tendons
ys = 47.0 mm. Height to the centroid of tendons in tension zone yT = 35.7 mm.
(a) Reinforced concrete column design chart to BS 8110. (b) Reinforced concrete column design
chart to BS EN 1992-1-1.
Use fck /fcu = 45/55, at transfer fck(t)/fci = 30/35, f pk /f pu = 1770 N/mm 2 , cement CEM I class
52.5R and a 10 mm coarse gravel aggregate. The self-weight of the precast concrete is
determined by the manufacturer as 24.5 kN/m3 giving a self-weight of 4.90 kN/m 2 . In this
exercise, ignore the reduced compression acting at the level of the strands due to self-weight
and dead loads in the calculation of creep losses.
BS 8500-1. Table A.4 for 50 years BS 8500-1. Table A.4 for 50 years
Exposure XC1. Cover = 15 + Δ Exposure XC1. = 15 + Δ
{7.3} Δ = 5 mm {4.4.1.3(3)}Cover controlled Δ = 5 mm
Cover to tendons ≥ 20 mm Cover to tendons ≥ 20 mm
. 1 h . R60. BS EN 1992-1-2
{Table 3.4} = 20 mm {Table 5.8} Depth ≥ 80 mm < 200 mm
{5.2(5)}Axis = 25 + 15 − Δa = 29 mm
{Exp. 5.3} Δ = 0.1 (500 – θ ) = 10.7 mm
{Fig. 5.1, curve 3}θ = 390°C for
{Exp. 5.2} (θ ) = 805/1,770 = 0.455
σ = 0.523 × 1,770/1.15 = 805 N/mm2
where / = (7.0 + 0.3 × 5.0)/(1.25 × 7.0 +
1.5 × 5.0) = 0.523
{4.3}using ψ = quasi-permanent ψ2 = 0.3
with = 4.9 + 1.5 + 0.6 = 7.0 kN/m2 and
= 5.0 kN/m2
Centroid to steel tendons in fire zone = =
35.7 mm > 29 mm
{Part 2, 7.2} = 20 + 0.2 × 55 = 31 kN/mm2 {Table 3.1} = 22 (53/10)0.3 = 36.3 kN/mm2

{4.8.3.1} = 20 + 0.2 × 35 = 27 kN/mm2 ( ) = 22 (38/10)0.3 = 32.8 kN/mm2
{BS 5896, Table 6} = 195 kN/mm2 strand {3.3.6(3)} = 195,000 N/mm2 strand
Although for wire = 205 kN/mm2 use same as for
strand
= 195/27 = 7.22. = 195/31 = 6.29 ( ) = 195/32.8 = 5.94. = 195/36.3 = 5.37
= 779 × 106/99.4 = 7.836 × 106 mm3

= 779 × 106/101.6 = 7.744 × 106 mm3
= 99.4 – 47.0 = 52.4 mm
1
1 = 5.29 -1 = 4.37
= 98.3 mm; = 799.6 × 106 mm4 = 98.5 mm; = 796.0 × 106 mm4
= 8.134 × 106 mm3; = 7.862 × 106 mm3 = 8.082 × 106 mm3; = 7.842 × 106 mm3
= 6 × 52 + 6 × 93 + 4 × 19.6 = 948.5 mm2
Initial σ = 0.7 × 1770 = 1239 N/mm2 σ = 0.7 × 1770 = 1,239 N/mm2
= 1239 × 948.5 = 1,175,241 N = 1239 × 948.5 = 1,175,241 N
{4.8.2.1} relaxation loss = 1.2 × 2% = 0.024 {3.3.2(7)} Relaxation at = 20 h
{Exp. 3.29} for Class 2 μ = 0.7; ρ1000 = 2.5%
Loss Δσ = 1239 × 0.66 × 2.5 × (9.1 × 0.7)
(20/1000)0.75(1 – 0.7) = 4.95 N/mm2
after relaxation loss = +8.99 N/mm2 σ after relaxation loss = +9.17 N/mm2
Shortening loss = 8.99 × 7.22/1,239 = 0.0524 {Exp. 3.29} Δσ = 9.17 × 5.94 = 54.46 N/mm2
σ 0 = 1,239.0 – 4.95 – 54.46 = 1,179.6 N/mm2
Transfer = 1 – 0.024 – 0.0524 = 0.924 Transfer = 1,179.6/1,239 = 0.952
Transfer = 0.924 × 1,175,241 = 1,085,467 N 0 = 0.924 × 1,175,241 = 1,085,467 N
( )
{Table 3.1} = 0.3 × 452/3 = 3.80 N/mm2

{4.3.5.2} = 0.45√35 = 2.66 N/mm2 {Exp. 3.4} ( ) = (38/53) × 3.80 = 2.72 N/mm2
= +11.94 N/mm2 < 0.5 × 35 = 17.5 OK σ ( ) = +12.31 N/mm2 < 0.6 × 30 = 18.0 OK
= −2.67 N/mm2 > −2.66 say OK σ ( ) = −2.75 N/mm2 > −2.72 say OK
{4.8.5.2} Creep φ = 1.8 {Exp. B.2} φ = 2.31 × 1.39 × 0.62 × 0.99
=1.98
where {Exp. B.4} β ( ) = 2.31
{Exp. B.3b/B.8c} φ = 1.39, with = 400 mm
{Exp. B.5}β ( ) = 0.62, where = 7.6 days at
50oC curing for 20 h
{Exp. B.7}β ( ) = 0.99 for = 20,833 days and
{Exp. B.8b}β (days) = 803
σ after initial loss = +8.77 N/mm2
{4.8.2.1} Loss = 1.8 × 0.924 × 0.0524 = 0.087 {Exp. 5.46} Δσ = 1.98 × 8.77 × 5.37/1.103 =
84.40 N/mm2
where denominator in code Exp. 5.46 = 1.103
{4.8.4} Shrinkage strain = 300 × 10–6 {Exp. 3.8} ε ε ε = 404 +88= 492 × 10–6
where {Exp. 3.9} ε = 0.985 × 0.73 × 566 × 10–6
= 404 × 10–6
β ( ) = 0.985 with as earlier, = 1 day
{Table 3.3} = 0.73 for = 400 mm
{Exp. B.11} ε = 0.85 × (220 + 110 × 6) ×
e-(0.11 × 53/10) × 1.356 = 566 × 10–6
β = 1.55 × [1 − (50/100)3] = 1.356
{Exp. 3.12} εca = 2.5 × (45 − 10) × 10–6 = 88 × 10–6
Loss = 300 × 10–6 × 195,000/1,239 = 0.0472 {Exp. 5.46} Δσ = 492 × 10–6 × 195,000/1.103
= 87.0 N/mm2
{5.10.6(1b)} Long-term relaxation
{Exp. 3.29} μ = 1,179/1,770 = 0.666
{Exp. 3.29} σ = 1,179 × 0.66 × 2.5 × e(9.1 × 0.666)
(500,000/1,000)0.75(1 − 0.666) = 39.65 N/mm2
{Exp. 5.46} Δσ = 0.8 × 39.65/1.103
= 28.7 N/mm2
σ = 1,179.6 − 84.4 − 87.0 − 28.7 = 979.5 N/mm2
= 0.924 − 0.087 − 0.0472 = 0.789 (21.1%) = 979.5/1,239 = 0.7905 (20.95%)
= 0.789 × 1,175,241 = 927,621 N = 0.791 × 1,175,241 = 929,076 N
= +10.20 N/mm2 < 0.33 × 55 = 18.15 OK σ = +10.22 N/mm2 < 0.45 × 45 = 20.25 OK
= −2.28 N/mm2 > 0.45 √55 = −3.34 OK σ = −2.28 N/mm2 > −3.80 OK
Btm = (10.20 + 3.34) × 8.134 = 110.1 kNm = (10.22 + 3.80) × 8.082 = 113.3 kNm
Top = (2.28 + 18.15) × 7.862 = 160.6 kNm = (2.28 + 20.25) × 7.842 = 176.7 kNm*
{4.3.7.1} {3.3.6(7)}
Refer to this book first ed. for full analysis Refer to Section 4.3.4 for full analysis
= 0.45 = 24.75 N/mm2 and λ = 0.9 = 0.567 = 25.5 N/mm2 and λ = 0.8
( )
in tension zone = 870 mm2; = 164.3 mm

Prestrain after losses ε = 0.005015 {5.10.9} ε = 0.004946
Refer to stress versus strain diagrams in the
following
Strain ε = 0.012970; stress = 1539 N/mm2 ε = 0.012700; stress = 1,442 N/mm2
{4.3.7.3} = 50.1 mm; = 141.8 mm = 51.2 mm; = 143.8 mm
= 870 × 1539 × 141.8 × 10–6 = 189.9 kNm = 870 × 1442 × 143.8 × 10–6 = 180.5 kNm
{4.3.8.1} {6.2.2(2)}
{4.3.8.4} = 100 + 99.4 = 199.4 mm {6.2.2(2)} = 100 + 99.4 = 199.4 mm
Mean diameter of strands = 10.9 mm {Exp. 8.16 and 8.18} 2 = 1.2 × 0.19 × 979.5 ×
= 10.9 × 240/√35 = 442 mm 10.9/4.06 = 721 mm
{Exp. 8.15} = 3.2 × 1.0 × 1.27 = 4.06 N/mm2
{8.10.2.2} = 0.7 × 2.72/1.5 = 1.27 N/mm2
/ = 199.4/442 = 0.451 α 2 = 199.4/721 = 0.276
= 927,621/232,040 = 4.00 N/mm2 σ = 0.9 × 929,076/232,040 = 3.60 N/mm2
= 4.00 [0.451 × (2 − 0.451)] = 2.79 N/mm2 where {2.4.2.2(1)}γ = 0.9
= 0.24√55 = 1.78 N/mm2 {3.1.6.2(P)} =0.3 × 452/3 × 0.7/1.5 = 1.77 N/mm2
first m.o.a. = 5.8326 × 106 mm3
{Eq. 54} = 0.67 × 1134 × 200 × {Exp. 6.4} = (779 × 1,134/5.8326) ×
√(1.782 + 0.8 × 2.79 × 1.78) = 406.1 kN √(1.772 + 0.276 × 3.60 × 1.77) = 335.3 kN
1. to the Eurocodes is greater than Class 2 BS 8110 because is 13% greater than . However, there
is no greater stress than allowed in the Eurocodes, such as Class 3 (0.2) in BS 8110.
2. < because is not allowed to reach maximum design stress in the Eurocodes stress versus
strain idealisation.
3. < because 2 is 63% greater than , and the build-up of prestress is linear in the Eurocodes
rather than parabolic in BS 8110.
* See Section 4.3.3 for the application of in the top surface in design.
This code effectively replaces BS 8110, Parts 1 to 3, although the execution of work (toler-
ances, setting out, etc.) is found in BS EN 13670:2009, Execution of concrete structures.
The division between commonplace and special design work separated in BS 8110 Parts 1
and 2 no longer exists, and there are no N-M interaction charts for column design. The last
point reflects the fact that EC2 is a limit state code of principles rather than methods. The
current amendment was published in February 2014. The UK Technical Committee B/525
(sub-committee 2) is currently engaged in a revision of the code.
Precast concrete is not treated as a separate design and construction method although, as
with BS 8110, there are certain aspects of design, such as bearings, anchorage at supports,
bursting, floor systems, compression/tension/shear joints, connections, pocket foundations,
and corbels, collected in a separate section, in this case Section 10.
The format of BS EN 1992-1-1, as with all material based on the Eurocodes, is as
follows:
Section 1 Scope – references; assumptions; definitions; symbols. Note that symbols are often only
defined here and not in the text
Section 2 Basis of design – requirement related to BS EN 1990, Annex B; requirements related to BS EN
1991-1; material properties; PSFs γ and γ , load combinations and equilibrium
Section 3 Materials – (concrete, rebar, tendons) strength, stress – strain models, deformation, shrinkage
and creep; fatigue; anchorage; prestressing
Section 4 Durability – environmental and exposure classes; cover to reinforcement
Section 5 Structural analysis – load cases; imperfections, sway; structural models; linear elastic, plastic and
non-linear analysis; redistribution; second-order effects with axial load (columns, walls);
prestressing – stressing; forces; losses; service and ultimate; fatigue
Section 6 ULS – bending, shear, torsion and punching shear; strut-and-tie models; anchorage and laps;
partially loaded areas (localised bearings); fatigue
Section 7 Serviceability limit state – crack control, spacing and crack width, deflections
Section 8 Detailing in general – rebars – bar spacing, anchorage, laps, links details; prestressing tendons
– anchorage, transmission length, development length
Section 9 Detailing in particular – maximum and minimum areas; anchorage at supports; shear, torsion
and surface reinforcement; solid and flat slabs, columns and walls, deep beams and stability ties
Section 10 Precast concrete elements and structures – materials; losses of prestress; bearings; anchorage
at supports; bursting; floor systems; compression/tension/shear joints; half joints; pocket
foundations; corbels
Section 12 Plain and lightly reinforced concrete – reduction factors for strength; precast walls and infill
shear walls, construction joints, strip and pad footings
Informative annexes – (A) improved PSFs; (B) creep and shrinkage strains in detail; (C)
reinforcement properties; (D) prestressing tendons losses; (E) strength classes for durability;
(F) tensile stresses in rebars in biaxial and shear stress fields; (G) soil-structure; (H) second-
order effects; (I) flat slab and shear walls; (J) regions of discontinuity
The code is not prescriptive, and it is necessary to turn to calculation methodology given in
documents published for example by The Concrete Centre, for example calculation of area
of flexural and shear reinforcement in beams and N–M charts for r.c. columns.
The main issues relating to the design of precast concrete structures in NA to BS EN

1992-1-1 are
2.4.2.2(1) Partial factor for prestress at ULS γ = 0.9.

3.1.2(2)P Value of max. Shear strength of concrete classes higher than C50/60 should be determined
by tests, etc.
3.1.6(1)P Value of αcc = 0.85 for compression in flexure and axial loading and 1.0 for other
phenomena, that is in bending = 0.85 /1.5 = 0.567 but in shear = 0.667 .
4.4.1.3(3) Δ under controlled conditions, such as steel mounts known as soldiers in front of
hollow core slabs machines, may be reduced to 10 mm > Δ > 5 mm.
5.1.3(1)P Simplified load arrangements. Consider the two following arrangements for ‘all spans’ and
alternate spans: (i) all spans carrying γ γ ; and (ii) alternate spans carrying
γ γ , other spans carrying only γ ; the same value of γ should
be used throughout the structure. For one-way spanning slabs, use the ‘all spans’
loaded if (i) area of each bay > 30 m2; (ii) / ≤ 1.25; and (iii) < 5 kN/m2 excluding
partitions.
5.5(4) Moment redistribution formula: values for steels with yk ≤ 500 N/mm2, 1 = 0.4, 2 = 0.6 +
0.0014/εcu2. Then if εcu2 = 0.0035, 2 = 1. Code Exp. 5.10a for ≤ 50 N/mm2, δ = 1 + 2
/ for zero moment redistribution δ = 1 = 0.4 + / , then ≤ 0.6, that is for the balanced
section, the limiting depth of the neutral axis is 0.6 .
5.10.9(1)P For pre-tensioning, = 1.0 and = 1.0, that is there are no modifications to the action of
prestress.
6.2.3(3) Values of ν1 = ν unless the design stress of the shear reinforcement < 0.8 yk, ν1 is modified.
7.2 The different limits of compressive stress in service depending on durability requirements
and the avoidance of non-linear creep in prestressed sections in flexure.
7.3.1(5) Limitations of crack width max. Use Table National Annex NA4. This reduces wmax in r.c.
sections to 0.3 mm and, in prestressed sections, the limiting permissible tension in service
to zero for exposure class greater than XC1, although the value of the imposed live load
may be reduced.
7.4.2(2) Values of basic span/depth ratios. Use Table NA5 which gives additional information and limits.
8.3(2) Minimum mandrel diameter Ф min. Use in Table NA6a and Table NA6b, which contain
additional information regarding scheduling reinforcement.
8.8(1) Additional rules for large diameter bars Ф > 40 mm.
9.5.2(1) Minimum diameter of longitudinal reinforcement in columns Фmin = 10 mm.
9.5.2(3) Maximum area of longitudinal reinforcement in columns. The designer should consider the
practical upper limit taking into account the ability to place the concrete around the rebar,
that is when casting columns horizontally mould, the maximum area is often around 8 to
10% of the area of concrete. This issue is considered further in the PD 6687-1:2010 (PD
6687-1 2010).
9.7(1) Minimum area of distribution reinforcement in deep beams = 0.2 % in each face.
9.10.2.2(2) Force to be resisted by peripheral tie. 1 = (20 + 4 0) where 0 is the number of storeys;
2 = 60 kN.
9.10.2.3(3) Minimum tensile force internal tie. = [( + )/7.5]( /5)( ) ≥ kN/m, with full
definitions. Maximum spacing of internal ties = 1.5 .
9.10.2.3(4) Internal ties on floors without screed. = [( + )/7.5]( /5) ≥ kN/m.
9.10.2.4(2) Horizontal ties to external columns and/or walls at each floor level: = the greater of 2
≤ s/2.5 and 3% at that level.
12.3.1(1) Values of α = 0.6 and α = 0.6 (plain concrete).
Annex J Design of corbels use PD 6687-1:2010.
This code provides the following two alternatives for designing r.c. and prestressed concrete
elements and structures for the actions of fire:
1. Performance-based design procedures in Section 2 to 4

2. Prescriptive rules, that is design aids such as tables and diagrams in Section 5.
The design procedure gives an analytical procedure taking into account the behaviour
of the structural system at elevated temperatures, the potential heat exposure and the
beneficial effects of active and passive fire protection systems, together with the conse-
quences of failure. The main text, together with informative annexes A to E, includes
most of the principal concepts and rules necessary for structural fire design of concrete
structures.
Section 1 Scope – references; assumptions; definitions; symbols

Section 2 Basis of design – requirements; actions; material properties; verification methods
Section 3 Material properties – at elevated temperatures; concrete with siliceous and
calcareous aggregates; thermal elongation of bars and tendons
Section 4 Design procedures – simplified and advanced calculation methods; shear, torsion
and anchorage; spalling; joints; protective layers
Section 5 Tabulated data – columns, walls, beams and slabs; fire thickness and axis distance
to bars
Section 6 High-strength concrete – calculation models and tabulated data for columns, walls,
beams and slabs
Informative annexes (A) Temperature profiles; (B) Simplified calculation methods; (C) Buckling of
columns under fire conditions; (D) Calculations for shear, torsion and anchorage;
(E) Simplified calculations for beams and slabs
Section 1 Scope – references; assumptions; definitions; symbols.

The main issues relating to the design of precast concrete structures in NA to BS EN
1992-1-2 are
3.2.3(5) Values for the parameters of the stress–strain relationship of reinforcing steel at elevated
temperatures. Use Class N (Table 3.2a).
3.2.4(2) Ditto cold worked (wires and strands) prestressing steel at elevated temperatures.
Use Class A.
5.6.1(1) Web thickness. Use dimensions for Class WA.
This PD gives guidance on some specific items that were not published in the concrete
Eurocodes or were in need of additional or noncontradictory additional information.
Background research is cited in many cases. It is not to be regarded as a British standard.
This PD gives noncontradictory complementary information for use with BS BS EN 1992
Parts 1-1 and 1-2 and their UK NAs.
The main items in the PD relating to the design of precast concrete structures are
2.5 Bond stress for mild steel cl. 8.4.2(2). The PD gives = η1η2 (0,36√f )/γ .
2.11 Calculation of effective length of columns, cl. 5.8.3.1, 5.8.3.2 (4) and (5). In the
calculation of the flexibility at the ends of the column, the stiffness of the
beam(s) attached to the column is taken as 2 ( / ) to allow for the effects of
cracking.
2.11.3 Calculation of limiting slenderness ratio, λ . where adjacent spans of beams do not
differ by more than 15%, columns may be assumed to be in double curvature
bending for the calculation of λ (i.e. value of moment ratio r < 0).
2.12 Design moment in columns, cl. 5.8.7.3 and 5.8.8.2. For braced structures,
= maximum of ( 0 + 2), ( 02) or (M01+ 0.5 2).
2.14 Design shear – point loads close to support. cl. 6.2.2 (6). Point loads close to
support will need to be considered in conjunction with other loads on the
member. Design shear between the point load and the support is
= ,other + β ,point-load, and therefore the reduction factor β cannot be applied
to the total . The clause explains how to deal with this situation.
2.20 Stress limitation in serviceability limit state. cl. 7.2(5). A modular ratio of 15 should be
used when calculating tensile stresses in rebars (≤ 0.8 and tendons ≤ 0.75 k)
under the characteristic combination of loads.
2.21.1 Control of cracking without direct calculation. cl. 7.3.3. Where the assumptions
relating to Table 7.2N and Table 7.3N are not met, crack width is verified using
the calculation procedure.
2.21.2 Calculation of crack widths, cl. 7.3.4.Values for from Fig. 6 of the PD are
proposed. It is unsure how Fig. 6a is interpreted.
2.22 Crack widths for non-rectangular tension zones and irregular bar layouts. Based
on BS 8110, the recommendation is = 3 εm/[1+ 2( – )/( – )].
2.23.2 Span/depth ratio. Exp. 7.17 in cl. 7.4.2(2).Values of ( s,prov/ s,req) or (310/σs) should
be limited to 1.5.
2.23.3 The value for ζ in Exp. 7.18 in cl. 7.4.3, and hence the value of σ or , should be
based on the frequent (not quasi-permanent) combination of loading.
2.26.1 Vertical ties, cl. 9.10.2. For notes that vertical ties are required in framed as well as
load-bearing structures, see Chapter 11 of this book.
2.26.2 Anchorage of precast floor and roof units and stair members. BS EN 1992-1-
1:2004 or BS EN 1991-1-7 does not cover this. All precast floor, roof and stair
members should be effectively anchored whether or not such members are used
to provide other ties required in BS EN 1992-1-1:2004, cl. 9.10.2. The anchorage
should be capable of carrying the dead weight of the member to that part of the
structure that contains the ties.
2.28 Detailing rules for particular situations, Annex J. NA to BS EN 1992-1-1:2004
declares that this is not applicable in the United Kingdom. Alternative versions for
frame corners and corbels are given in Annex B of this PD.
3 BS EN 1992-1-2:2004, Structural fire design. The tabular methods for assessing
the fire resistance of columns are limited to braced structures. However, at the
discretion of the designer, the methods given in BS EN 1992-1-2:2004 for
columns may be used for the initial design of unbraced structures. In critical
cases, the chosen column sizes should be verified using BS EN 1992-1-2:2004,
Annex B.
Bhatt, P., MacGinley, T. J. and Choo, B. S. 2014. Reinforced Concrete Design to Eurocodes: Design
Theory and Examples, 4th ed., CRC Press/Taylor & Francis, London, UK., 880pp.
BS EN 771-1. 2011. Specification for masonry units, clay masonry units, BSI, London, UK.
BS EN 1990. 2002. Eurocode 0: Basis of structural design, BSI, London, UK., March 2006.
NA to BS EN 1990. 2002. UK National Annex to Eurocode 0: Basis of structural design, BSI, London,
UK., December 2004.
BS EN 1991-1-1. 2002. Eurocode 1: Actions on Structures – Part 1-1: General Actions – Densities,
self-weight, imposed loads for buildings, BSI, London, UK., December 2004.
NA to BS EN 1991-1-1. 2002. UK National Annex to Eurocode 1: Actions on Structures – Part
1-1: General Actions – Densities, self-weight, imposed loads for buildings, BSI, London, UK.,
December 2005.
BS EN 1992-1-1. 2004. Eurocode 2: Design of Concrete Structures – Part 1-1: General rules and rules
NA to BS EN 1992-1-1. 2004. UK National Annex to Eurocode 2: Design of Concrete Structures –
Part 1-1: General rules and rules for buildings, BSI, London, UK., December 2005.
BS EN 1993-1-1. 2005. Eurocode 3. Design of steel structures. General rules and rules for buildings,
BSI, London, UK., September 2006.
NA to BS EN 1993-1-1. 2005. UK National Annex to Eurocode 3. Design of steel structures. – Part
1-1: General rules and rules for buildings, BSI, London, UK.
BS EN 1996-1-1. 2005. Eurocode 6: Design of masonry structures – Part 1-1: General rules for rein-
forced and unreinforced masonry structures, BSI, London, UK.
BS EN 13670. 2009. Execution of concrete structures, BSI, London, UK.
Elliott, K. S., Davies, G., Ferreira, M., Mahdi, A. A. and Gorgun, H. 2003a. Can precast concrete
structures be designed as semi-rigid frames? – Part 1, The experimental evidence, The Structural
Engineer, London, 81(16), 14–27.
Elliott, K. S., Davies, G., Ferreira, M., Mahdi, A. A. and Gorgun, H. 2003b. Can precast concrete
structures be designed as semi-rigid frames? – Part 2, Theoretical equations and applications for
frame design, The Structural Engineer, London, 81(16), 28–37.
Elliott, K. S., Davies, G., Mahdi, A. A., Gorgun, H., Virdi. K. and Ragupathy, P. 1998. Precast concrete
semi-rigid beam-to column connections in skeletal frames. COST C1 International Conference
on Control of the Semi-rigid Behaviour of Civil Engineering Structural Connections, Liege,
Belgium, pp. 45–54.
Elliott, K. S., Ferreira, M., de Aranjo, D. and El Debs, M. 2005. Analysis of multi-storey precast
frames considering beam-column connections with semi-rigid behaviour. Keep Concrete
Attractive, FIB Symposium, Budapest, Hungary, pp. 496–501.
Ferreira, M. A., El Debs, M. K. and Elliott, K. S. 2003. Analysis of multi-storey precast frames with
semi-rigid connections. 45th Brazilian Concrete Congress, Brazilian Conference on Concrete,
IBRACON 2003, Brazilian Concrete Institute – IBRACON, Vitoria, Brazil.
Narayanan, R. S. and Beeby, A. 2005. Designer’s Guide to EN 1992-1-1 and EN 1992-1-2, Thomas
Telford, London, UK.
PD 6687-1. 2010. Background Paper to the UK National Annexes to BS EN 1992-1 and BS EN
1992-3, BSI, London.
Straman, J. P. 1990. Precast concrete cores and shear walls. Prefabrication of Concrete Structures,
International Seminar, Delft, the Netherlands, 25–26 October 1990, pp. 41–54.
Chapter 4
Precast concrete flooring offers an economic and versatile solution to ground and suspended
floors in any type of building construction. Worldwide, approximately half of the floors
used in commercial and domestic buildings, and more than three-fourths in car parks, are of
precast concrete. Precast floors offer both design and cost advantages over traditional meth-
ods such as cast in situ concrete, steel–concrete composite and timber floors. There are a
wide range of flooring types available to give the most economic solution for all loading and
spans. The floors give maximum structural performance with minimum weight and may be
used with or without structural toppings, non-structural finishes (such as tiles, granolithic
screed) or with raised timber floors.
Precast concrete floors offer the twin advantages of
1. Off-site production of high-strength, highly durable units

2. Fast erection of long-span floors on site
Figure 4.1 shows some 10 m long × 1.2 m wide floors positioned at the rate of 1 unit every
10–15 min, depending on the height of hoisting – equivalent to covering an area of the size of a
soccer field in 20 days. Each vehicle carries about 20 tonnes of flooring, approximately four to
six units and so erection rates are slowed down more by the problems of getting vehicles onto
restricted sits than in erecting the units. These particular units are called ‘hollow core floor units’,
or hollow-core planks in Australia and the United States, and are manufactured wet-cast as
shown in Figure 2.4 or by machine extrusion or slip forming as shown in Figure 2.13. The term
‘unit’ is used to describe the precast element before it is grouted into position on site, after which
it is termed a ‘floor slab’. Figure 4.2 shows the moment when a hollow core unit is lifted from
the steel casting bed and illustrates the principle of a voided unit. Consequently, the self-weight
of a hollow core unit is about one-half of a solid section of the same depth. It is said to have a
‘void ratio’ of 50%. Deeper hollow core units, such as the newly developed 1000 mm deep units
shown in Figure 4.3 from Italy, have void ratios nearer to 70%. Units this deep have a limited
market often for cut-and-cover tunnels. The most common depths range from 150 to 400 mm.
Most units are 1200 mm wide; Figure 4.4 shows a very efficient profile comprising seven cores
that are stacked in huge quantities in many countries as shown in Figure 4.5. A 1200 mm section
can be split into 400 or 600 mm wide units as shown in Figure 4.6. Figure 4.7 shows the storage
of some 2400 mm wide units, manufactured in Barbados, sometimes used in large/regular floor
areas in North America. The solid cores locate the lifting points in which a transverse rebar is
laid across the unit to facilitate lifting from the bed. Russian factories manufacture 1500 mm
wide units, mainly because 1.5 m was a traditional width for wet-cast floor slabs and architects
prefer to keep to this modular dimension.
125
Prestressed concrete hollow core floor units.
Prestressed concrete hollow core floor units are lifted from the steel casting bed after about
16 h curing.
Table 4.1 lists a range of hollow core units from 100 to 500 mm depth based on units
containing a smaller number of larger cores, for example 4 to 6 no. per 1200 mm width,
Figure 4.2, and a larger number of oval-shaped cores, for example 9–11, Figure 4.4.
Details of how to calculate the moments of resistance and shear capacities are given in
Section 4.4.
Precast concrete floors 127
A 1000 mm deep hollow core unit developed by Nordimpianti of Italy, at Sao Paulo exhibition in 2014.
Slipformed hollow core units with efficient ‘pear’ shaped cores, narrowing to allow additional
prestress where splitting might otherwise occur.
The height of voids should not exceed h − 50 mm, where h is the overall depth of the unit.
The diameter of circular voids is usually h − 75 mm. The minimum flange thickness depends
on the overall depth of the unit h, given by 1.6 . However, because of cover requirements
it is usually necessary for the bottom flange to be at least 30 mm thick. The minimum width
of a web should not be less than 30 mm.
Hollow core units were developed in the 1950s when the dual techniques of long-line
prestressing and concrete production through machines were being developed by companies
such as Spiroll in the United States and Roth in Europe. Precast concrete engineers contin-
ued to optimise the cross section of the units leading to the so-called double-tee unit, shown
in Figure 4.8, achieving even greater spans and reduced mass compared with hollow core
units. Although the finer points of detail of double-tees vary in many different countries, the
Stacking 1200 mm wide hollow core floor units awaiting delivery within seven days after casting.
Two 600 mm wide hollow core floor units made from 1200 mm wide casting bed, 400 mm wide
units are also possible.
unit comprises two deep webs, reinforced for strength, joined together by a relatively thin
flange, for stability. Two major drawbacks with double-tees are
1. Not being able to provide sufficient pretensioning force to satisfy the available section
as strands can only be placed in pairs at the bottom of the webs, forcing more strands
to be positioned higher in the web-reducing eccentricity.
The 2.4 m wide hollow core floor units manufactured by Preconco Ltd., Barbados.
Structural properties of prestressed hollow core floor units
150 11 × 60 2.69 67.5 100.1 120.5

200 6 × 140 3.33 123.7 179.0 101.1
11 × 60 3.33 120.8 175.7 158.5
250 6 × 140 3.71 183.9 258.1 131.1
11 × 60 4.03 180.9 256.8 200.1
300 6 × 135 4.22 249.0 342.3 174.0
11 × 60 4.75 246.8 341.7 238.2
350 6 × 135 4.62 326.9 448.0 212.0
400 6 × 135 5.03 407.9 551.2 247.0
450 6 × 135 5.43 507.6 677.8 283.0
500 6 × 135 5.84 598.7 792.7 317.1
Note: Exposure = XC1 and fire resistance = R60 minutes.
Cover or average cover to pretensioning strands = 30 mm (9.3 mm dia.) and 35 mm (12.5 mm dia.). Final prestress at bot-
tom (excluding self-weight), σ = 12.5–13.0 N/mm2 depending on the availability of the positions of tendons. Self-weight
includes infilled joints between units. Exposure = XC1, = 45 N/mm2, = 3.80 N/mm2.
2. The centroidal axis is high in the section, typically yb = 0.7h × depth from the bottom,
resulting in a low value of the section modulus at the bottom Zb relative to the cross-
sectional area, for example Zb /Ac h ≈ 0.12 compared with 0.24 for hollow core units.
This offers the potential for composite action with a structural topping.
Deflected or debonded tendons are used in some cases to overcome transfer stress problems
in long-span units. The cross-sectional profile and end detail at the half-joint is shown in
Figure 4.9. Typical widths are 2.4–3.0 m, and depths range from 400 to 1200 mm. The void
Prestressed double-tee floor units.
Double-tee end profile – the half joint raises the bearing level and reduces structural floor depth.
ratio is about 70%, allowing the unit to span over longer spans and with less weight per
area than the hollow core unit. The rate of erection is comparable to hollow core units, but
most double-tee floors require a structural topping (see Section 4.4) to be site cast, together
with a reinforcing mesh, thus reducing the overall benefit gained by the greater spans and
reduced weight. Figure 4.10 shows a variation of the double-tee in that the keyed end section
is made solid to facilitate continuous and composite action between spans, the splayed end
is reinforced to resist large negative bending moments on either side of the interior beam(s).
Unlike hollow core units, double-tees do not have ‘trade names’ as their manufacture is not
a proprietary method.
Table 4.2 lists some types of prestressed double-tee units having 2 no. webs 160 mm
wide at the bottom, together with their moments of resistance and shear capacity.
Multi-rib units with tapered half joint and special end detail suitable for continuous design, at
Samsung, South Korea.
Structural properties of double-tee floor units (without structural toppings)
300 75 180 2.60 69.8 116.4 68.0

400 75 180 2.96 124.1 201.2 89.6
500 75 180 3.32 190.4 303.0 106.5
600 75 180 3.68 271.2 427.6 120.5
700 75 180 4.04 362.3 566.4 133.2
800 75 180 4.40 471.0 733.8 144.9
a Web breath refers to the breadth of each web near the centroidal axis.
Axis distance to pretensioning strands = 50, 100, 150 and 200 mm (12.5 mm dia.). Final prestress at bottom σ (excluding
self-weight) = 18.0 – 19.0 N/mm2 depending on the availability of positions of tendons. Some tendons debonded at the
support. Exposure = XC1, data based on = 45 N/mm2, ( ) = 30 N/mm2, 25% final losses of pretensiong force.
Comparison with Table 4.1 is interesting, for example service moment of resistance MsR for the
400 mm deep hollow core and double-tee units as follows (see Section 4.3.3 for terminology):
σ
Hollow core 5.03 1266 12.86 4646 24.44 339.9
Double-tee 3.32 1116 18.70 3776 13.24 124.1
M sR = (σb + fctm) Zb /b, where fctm = 3.8 N/mm 2 , b = 1.2 m and 2.4 m for hollow core units
and double-tee units.
This comparison highlights the issue of low values of Zb for double-tee units, even though
the unit can be prestressed to a higher level at mid-span (18.7 N/mm 2) thanks to debonding
at the support. If this same double-tee unit is designed compositely with a 75 mm topping
M sR increases to 170 kNm/m width, closing the gap with the hollow core unit relative to
their self-weight.
Both hollow core unit and double-tee floors are restricted, certainly in economical terms,
to a rectangular plan shape. It is possible to make trapezoidal or splayed ended units to suit
non-rectangular building grids, but the detailing of these units would be difficult and not
economical. Some companies quote 20%–50% surcharges for manufacturing nonstandard
units. A precast flooring method which enables non-rectangular layouts is the ‘composite
beam and plank floor’. This is a tertiary system in which a composite floor is produced as
shown in Figure 4.11; primary beams (r.c., precast, steel etc.) support long span beams,
reinforced or prestressed depending on structural requirements and manufacturing capabil-
ity. These carry precast concrete planks that may be shaped to suit non-rectangular, even
curved, building layouts. The planks are relatively inexpensive to produce in a range of
moulds of different sizes. It is usual for a structural topping to be applied to the floor, and
this is reinforced using a mesh. The final constructed floor resembles a double-tee floor in
structural form and has a similar void ratio of about 70%, but the way in which each of
these has been achieved may be tailored to suit the building requirements.
The precast planks described earlier may be used in isolation of the precast beams, span-
ning continuously between brick walls, steel or r.c. beams. The cross section of composite
planks is as shown in Figure 4.12a. To speed erection rates, the planks may be up to 3 m
wide (1.2 and 2.4 m are common). The lattice floor slab, commonly known as ‘Omnia’ is
a semi-precast option of using a lightweight, yet versatile in terms of shape, precast soffit
strengthened to span up to about 6 m using top reinforcement in compression. Figure 4.12b
and c show the two stages of precast production from a UK manufacturer and Figure 4.12d a
Composite beam and plank flooring comprising precast primary and secondary beams, precast
planks and structural topping.
(a) Composite plank floor options. (b) Reinforced lattice girder and rebar cage for composite
plank floor. ( )
(c) Precast concrete stage of composite plank lattice floor. (d) Construction
stage of composite plank lattice floor. (Courtesy of Pfeifer Seil und Hebetechnik
Memmingen, Germany.)
similar variation ideal for making both floors and beams continuous in Germany. The tops of
the beams may be provided with interface shear loops to make a composite beam. Lightweight
infill blocks (e.g. dense polystyrene) are sometimes placed on to the tops of the planks to
reduce weight by about 25%, but the weight saving blocks may cost more than the displaced
concrete. It isrelatively easy to form large-size voids in this floor and to add site reinforcement
to cater for stress raisers at corners etc.
Prestressed concrete composite plank, often known as ‘half-slab.’
Statically reinforced concrete may be limited in terms of span and depth, mainly due to
deflection or crack width criteria which can be overcome by prestressing. The other type of
half-slab is shown in Figure 4.13, a prestressed solid plank, reinforced with up to 12 no. 5 or
7 mm wires, or between 4 and 9 no. 7-wire helical strands, depending on depth and loading
requirements. Solid slabs, ‘often called solid planks,’ are usually designed compositely with
an in situ topping as discussed in Section 4.4.1. The precast soffit section is typically 100 mm
deep, but depths of between 75 and 200 mm are possible. Solid planks are often used in
domestic properties where the requirement for 300 kg/m 2 sound insulation can easily be met
using a 75 mm precast soffit (12 no. 7 mm wires at 20 mm cover) and 50 mm topping, and
for an imposed live load of 1.5 kN/m 2 plus 1.2 + 0.6 kN/m2 for finishes, services and ceiling,
spans of 5.0 m are possible without temporary propping, or 5.5 m with propping. Propping
helps to optimise the design where the limiting span is controlled almost equally by service
stress, final deflection and deflection after finishes (see Section 4.6.2). In fact, a limiting span
of 4.5 m (length/depth = 4500/75 = 60) due to lifting and handling may be imposed by some
manufacturers.
Each of the flooring systems introduced earlier has successively eroded the major advan-
tages in the use of precast concrete floors over competitors such as timber or cast in situ
floors. Since the 1960s, the hollow core market in Europe recently passed 1000M m2 – about
300 soccer pitches per month! The advantages with precast are
1. Manufacture of units simply and economically

2. Safe and rapid erection of floors
3. Possiblity to create a structurally complete precast floor
4. Use of minimum amounts of in situ reinforcement and wet concrete
However, these may be in competition with other criteria depending on-site access, structural
design requirements, interface with other trades, availability of expensive or cheap labour,
services requirements, etc. Specifiers must therefore study all available options.
A floor slab may comprise of a large number of individual units, each designed to cater for
specified loads, moments etc. or it may comprise a complete slab field where the loads are
shared between the precast units according to the structural response of each component. It
is first necessary to define the following:
‘Floor unit’: a discrete element designed in isolation of other units, Figure 4.4
‘Floor slab’: several floor units structurally tied together to form a floor area, with each
unit designed in isolation usually for uniformly distributed loading, Figure 4.1
‘Floor field’: a floor slab where each floor unit is designed as part of the whole floor,
allowing lateral spreading of isolated point or wall loads, and spreading of section
properties due to the loss of a hole, for example See Figures 4.14 and 4.15 later.
Most floor units, for example hollow core unit and double-tee, are one-way spanning, sim-
ply supported units. Composite plank may be designed to span in two directions, but the
distances between the supports in the secondary direction may be prohibitively small to suit
manufacturing or truck restrictions of about 3 m width. Structural toppings will enable
slabs to span in two directions, although this is ignored in favour of one-way spans. Hollow
core units may be used without a structural topping because the individual floor units are
keyed together over the full surface area of their edges – the longitudinal joints between the
units shown in Figure 4.1 are site filled using flowable mortar to form a floor ‘slab’. Vertical
and horizontal load transfer is effective over the entire floor area. This is not the case with
Mechanism for lateral load distribution of hollow core floors.
Steel trimmer angles are used to support units around large voids in hollow core slabs.
all the other types of precast floor where a structural (i.e. containing adequate reinforce-
ment) topping must be used for horizontal load transfer, flexural and shear strength or
simply to complete the construction.
The most common situation is a uniformly distributed floor load acting on one-way span-
ning units with no secondary supports. Each unit will be equally loaded and there is no
further analysis required of the slab, only the design of each unit according to Section 4.3.
Where line loads or point loads occur, unequal deflections of individual units will cause
interface shear in the longitudinal joints between them, and load sharing will result in a slab
‘field’ as shown in Figure 4.14.
Hollow core units are not provided with transverse reinforcement in the precast units
or in the joints between the units. The line load produces a shear reaction in the longi-
tudinal edge of the adjacent units, and this induces torsion in the next slab. The capac-
ity of the hollow core slab to carry the torsion is limited by the tensile capacity of the
concrete. The magnitude of the shear reaction depends on the torsional stiffness and
the longitudinal and transverse stiffness of all the adjacent units, low stiffness result-
ing in low load sharing. The precast units are assumed to be cracked longitudinally
in the bottom flange, but shear friction generated by transverse restraints in the floor
plate ensures integrity at the ultimate state. The deflected profile of the total floor slab
is computed using finite strips and differential analysis. The cross section of each floor
element is considered as a solid rectangular element, and the circular (or oval) voids are
ignored. As the result is unsafe, reduction factors of about 1.25 are applied to the shear
reactions (Van Acker 1984). Interesting results and further analysis may be found in
The Structural Engineer, ACI Journal and PCI Journal (e.g. Pfeifer and Nelson 1983,
Stanton 1987, Moss 1995).
Standard edge profiles have evolved to ensure an adequate transfer of horizontal and
vertical shear between adjacent units. The main function of the joint is to prevent rela-
tive displacements between units. In hollow core units, these objectives are achieved using
structural grade in situ concrete (C20/25 minimum) compacted by a small-diameter poker

in dampened joints. The edges of the slab are profiled to ensure that an adequate shear key
of in situ concrete (6–10 mm size aggregate), rather than grout, is formed between adjacent
units. The manufacturing process is not sympathetic to providing projecting reinforcement
across the joint. The capacity of the shear key between the units is sufficient to prevent the
adjacent slabs from differential movement. Despite a slight roughening of the surfaces dur-
ing the manufacturing process where indentations of up to 2 mm are present, the surface is
classified in BS EN 1992-1-1, clause 10.9.3(12) for smooth ex-cast edge surfaces as opposed
to being ‘roughened.’ The design ultimate horizontal shear resistance is given as vRdi = 0.15
N/mm2. Vertical shear capacity is based on single castellated joint design with minimum
root indentation 40 mm × 10 mm deep with a 40 mm gap between the units. The shear
force (kN/m) is according to clause 10.9.3(5) as vEd = qEd be/3, where qEd is the imposed
variable (live) load and be is the breadth of the unit, for example for 1.5 × 5 = 7.5 kN/m2 load
acting on 1200 mm wide units = 3.0 kN/m, but not be less than 0.5 qEd = 3.75 kN/m.
The shear stress capacity for an indented profile is given in clause 6.2.5(2) (Exp. 6.25)
vRdi = c fctd = 0.5 × 1.03 = 0.515 N/mm2, where fctd = 0.7 × 0.3 × 202/3/1.5 = 1.03 N/mm2
(for fck = 20 N/mm2). The depth of the infilled joint is taken as h – 30 mm (for reasons given
in Section 8.4.1). Therefore, in this example, if h = 150 mm, vRd = 0.515 × 1000 × 120 ×
10–3 = 61.8 kN/m.
The transverse moments and shear forces may be distributed over an effective width equal to
the total width of three 1.2 m wide precast units, or one quarter of the effective span (leff/4)
either side of the loaded area. Where a structural reinforced topping is present, the maxi-
mum number of units is four ≤ leff/4 as mentioned earlier. This information is not present
in the Eurocodes and so is taken from BS8110, Part 1, clause 5.2.2.2 and 5.2.2.3. If a free
edge is present, the load spread is over two units, unless there are edge loads (e.g. from trim-
mer angles as shown in Figure 4.15) where the load spread is over 1½ units. The equivalent
uniformly distributed loading on each slab unit may thus be computed. This is a conserva-
tive approach as data given in the Informative Annex C in the hollow core product standard
(BS EN 1168) show that for spans exceeding 4 m, up to five units are effective, given by
α factors in Figure 4.16a through c. The data also show that for edge elements, for example
adjacent to a large void or free edge, only two slabs contribute significantly in carrying
the load. Figure 4.15d shows the assumptions used in the distribution of vertical shear in
joints between adjacent units due to point, line and partial line loads. The α factors given in
Figure 4.16 are used for service loads. However, at the ultimate limit state, α should be mul-
tiplied by 1.25 for the directly loaded unit, with the total percentage share being reduced by
the same amount in the adjacent units.
Line loads parallel to the span and ≤5 kN/m may be replaced by UDL over a width
equal to l/4 on both sides of the load. If the available width next to the load is < leff/4, the
load is distributed over a width equal to the available width on one side plus leff/4 on the
other side. If a steel trimmer angle is used around a void adjacent to the loaded unit, load
distribution cannot take place into the unit supported by the trimmer angle; therefore, the
distribution takes place on the opposite side only. The maximum panel width w ≤ 0.25 leff +
100 mm but not more than 1.5 units (because the trimmer is positioned at the far edge of
a unit). If trimmer angles are present on both sides of a unit, or trimmer angles are present on
one side and the other is a free edge, then w = one unit width = 1.2 m. Line loads of length
≥ leff/2 should be considered as linear loads, otherwise are considered as linear loads if the
4 5
4
60
3
α1 α1 α2 α3 α4 α5
50
α1 α2 α3 α4 α5
40 x = ½l
40
5 α3
30 l
1 30
α2 α2
α3 α4
20 α2 = α4 1 20
α1 = α5 α1
α3 α5
10 10
α4
α5
4 6 8 10 12 14 m 4 6 8 10 12 14 m
(a) 2 (b) 2
1.0 0
0.5b
1.5b
0.8
2.5b
3.5b
4.5b
0.6
1
0.4
s 3 s
Fd
0.2
b
0.0
4 6 8 10 1 12 14 16 m
(c) 2
b b b b
b
l/2
b
l
a
l/2
b
a
a
(d) Point loads Line loads
(a) Load distribution percentage factors α (1) for 1200 mm wide hollow core floor slabs based
on span (2). (a) line loads at edge (4) and in field (5) parallel with the span, (b) point loads (3) at
mid-span according to BS EN 1168:2005+A3:2011, Annex C. (c) Reaction force distribution coef-
ficients (1) for three-edge supported 1200 mm wide hollow core units for line loads (3) parallel
with the span (2) according to BS EN 1168:2005+A3:2011, Annex C. = (1)/(3). (d) Distribution of
vertical shear in joints between adjacent units according to BS EN 1168:2005+A3:2011, Annex C.
(Courtesy of British Standards Institute, London.)
centre of the load is at mid-span and as point loads in the centre of the load if the centre
is not at mid-span.
In Figure 4.16b, α is given for point loads at mid-span (l/x = 2). For loads near the support,
l/x ≥ 20, the loading percentages of the directly loaded unit should be taken as 100% and of
the non-loaded slabs as 0%. For l/x values between 2 and 20, the loading percentages may
be derived from linear interpolation. For point loads not at mid-span (and linear loads as
defined earlier considered as point loads), the effective length of the joint transmitting the
shear force should be chosen equal to two times the distance from the centre of the load to
the nearest support (see Figure 4.16d).
In case of a linear load F (kN/m) and one supported lateral edge, the resultant FR of the
reaction distributed in the lateral support is given by F R = k F where the distribution factor
k is given in Figure 4.16c as a function of l (in m) and of the distance s of the load from the
nearest lateral support. If the number n of units > 5, FR is multiplied by 1 – (n – 5)s/50 b,
where b is the breadth of the elements. For example a hollow core slab field comprising 7
units of l = 8 m span supports a full span line load of 10 kN/m acting at 3.0 m (s = 2.5b)
from a parallel supported edge. Determine the final reaction due to the line load in the
supported lateral edge unit. Figure 4.16a gives α1 = 0.16 × 10.0 = +1.60 kN/m. Figure 4.16c
gives k = 0.47. Then FR = 10.0 × 0.47 × (1 – 2 × 3.0/50 × 1.2) = −4.23 kN/m. Of this reaction,
Figure 4.16a for edge load gives α1 = 0.32 × −4.23 = −1.35 kN/m reacted by unit 1. The final
load carried by unit 1 w 1 = +1.60 – 1.35 = +0.25 kN/m. Similar data for reactions due to
point loads are given in BS EN 1168.
Welded connections between adjacent double-tee units, or between the units and a
supporting member, are shown in Figure 4.17. Electrodes of grade S275 are used to
Plan on joint
Connectors at
Anchor bars as 2.0–2.4 m centres
shown below
75–100
Steel angles/plates
as shown below
Mild steel angle

with anchor bars
Mild steel plate

with anchor bars
Larger diameter mild

steel anchor bar
Mild steel plate

with anchor bars
Welded plate connectors in top flanges between double-tee slabs.
form short continuous fillet welds between fully anchored mild steel plates (stainless
steel plates and electrodes may be specified in special circumstances). A small saw cut
is made at the ends of the cast-in plate to act as a stress reliever to the heated plate dur-
ing welding.
Double-tee units are either designed compositely with a structural topping, in which case
the flange thickness is 50–75 mm, or are self-topped with thicker flanges around 120 mm.
In the former, the top of the unit is a roughened profile as given in clause 6.2.5(2). The hori-
zontal shear is transferred in the in situ topping using a design value for interface shear stress
vRdi = cfctd = 0.4 × 1.03 = 0.41 N/mm2 for C20/25 topping.
Figure 4.18a gives an example of a hollow core floor slab field comprising 7 × 1.2 m wide ×
300 mm deep units × 9.0 m clear span subjected to
• Uniformly distributed loading: finishes, services/ceiling and imposed live 1.5, 0.6 and
4.0 kN/m 2
• 100 mm wide wall (partial line load) of 5 kN/m
• 150 mm cross-wall of 8 kN/m
• Two 10 kN points loads
100 bearing lengths

215 × 215 brick pillar
50
1 2 3 4 5 6
Edge of
2 no. 10 kN point slab field
dead loads
300
100 Cross wall 8 kN/m
9000
Edge of 300 × 200 clear l = 9 100
slab field 1500 hole span
6000
Void Line
wall
1200 5 kN/m
3000
50
200 × 200 end notch 100 × 100 end notch
Floor loads UDL
Self weight of 300 mm depth hcu = 3.95 kN/m2

Finishes = 1.50 kN/m2
Services, ceiling = 0.60 kN/m2
Dead load = 6.05 kN/m2
Imposed live load = 4.00 kN/m2
(a) Total = 10.05 [ultimate 13.56] kN/m2
(a) General arrangement of floor slab field. ( )
Area of trimmed slabs =

1 4.55 × 1.2 = 5.46 m2
3.05 × 1.2 = 3.66 m2
Free edge
8 kN/m × 0.6 m
Trimmer angle point loads.
Dead = 6.05 × 5.46/4 = 8.28 kN
Live = 4.0 × 1.365 = 5.46 kN
1500 4550 + 100/2 = 4600
void 3050 + 1500 = 4550
Dead = 6.05 × 3.66/4 = 5.54 kN
Live = 4.00 × 0.915 = 3.66 kN
From centre of support
3000 + 50 = 3050
Trimmer angles have 200 ×

150 mm bearing onto 1000
edges of supporting units 250 × 200 wide notch (notch length = 200 + 50)
(b) Panel width = 1.2 m
8 kN/m × 2.1 m
300
300 × 200 wide
4600
4550
Trimmers loads, footprint and

4250
position same as unit no. 1 5.0/3 = 1.67 kN/m is distributed to unit 2.

Then load acting on 1.8 m width panel =
1.67 × 1.8/1.2 = 2.50 kN/m.
1100
150 long × 100 wide notch
(c) Panel width = 1.8 m
3
8 kN/m × 1.5 m
10 kN
300 × 200 wide hole

6050
4600
4550
4250
Trimmed edge. 5 kN/m

Spreading cannot pass
beyond this edge.
1100
150 long × 100 notch
(d) Panel width = 3.6 m
( ) (b) Loads and voids in Unit 1. (c) Loads and voids in Panel 2-3 for Unit 2. Maximum
panel width to the right of Unit 2 based on load spreading of trimmer loads = 0.25 ×
9.1 + 0.15 = 2.435 m > 1.8 m, not critical. (d) Loads and voids in Panel 2-3-4 for Unit
3. Maximum panel width to left + to right of Unit 3 = 1.8 + (0.25 × 9.1 + 0.1) = 4.175
m > 3.6 m, not critical. ( )
8 kN/m × 0.3 m
10 kN 5 kN (because 50% of this
load spreads 1½ units
towards no. 4, the
300 × 200 wide hole remainder to no. 6)
6050
4600
4550
4250
5 kN/m
(e) Panel width = 3.6 m
6
10 kN 10 kN
Free edge
10 kN
Free edge
6050
6050
Panel width = 2.4 m

Panel width = 3.6 m
(f) (g)
( ) (e) Loads and voids in Panel 3-4-5 for Unit 4. Maximum panel width to both sides of
Unit 4 = 0.5 × 9.1 + 0.1 = 4.65 m, not critical. Note: this would be limiting if a struc-
tural topping was present, where 4 × 1.2 = 4.8 m is permitted. (f) Loads and voids in
Panel 4-5-6 for Unit 5. Maximum panel width to left + to right of Unit 5 = (0.25 × 9.1 +
0.1) + 1.8 (free edge) = 4.175 m > 3.6, not critical. (g) Loads and voids in Panel 5–6 for
Unit 6. Maximum panel width to left + half of Unit 6 = (0.25 × 9.1 + 0.1) + 0.6 = 2.975
m > 2.4 m, not critical. ( )
350 MRd (including 200 mm wide hole)
300
MEd in unit no. 2 MsR (including 200 mm wide hole)

250
Moment (kNm)
200
150
100
Ms in units no. 2, 1, 3, 4, 5 and 6
in decending order
50
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
(h) Distance from left hand support (m)
( ) (h) Final bending moment diagrams for (all units) and (critical Unit 2 only) and
resistances and for Unit 2 in Panel 2-3.
The hollow core units used in this exercise are similar to the 4-void unit shown in Figure 4.2
comprising up to 10 no. 12.5 mm strands at 40 mm cover. Each strand (Figure 2.27) has an
area of 93 mm 2 .
The bearing length Lb = 100 mm, then effective span leff = min{9.0 + 0.3; 9.0 + 0.1} = 9.1 m.
A 1.5 m long × 1.2 m wide trimmed void (steel trimmer angles 200 mm bearing length,
see Figure 4.15) is situated in Unit 2, together with a 300 × 200 mm wide hole and two
notches 200 × 200 mm and 100 × 100 mm located in the corners of units. The task is to
consider lateral load spreading in a series of floor ‘panels’ over a maximum of three × 1.2 m
width or 0.25l + 0.1 m, unless there is an obstruction (such as a free edge or trimmed void)
to the load spreading. Figure 4.18b through g show the direct and distributed loads to each
of the 6 loaded units (the trimmed unit is excluded).
In design, each unit is selected for design in turn, the panel width w is determined,
and the point, line and cross-wall loads are added. A fixed datum is required, usually
the left hand edge of the panel, and the position of all holes and notches starts from
there. Where a parallel line load or a point load is positioned within the panel width,
the full line or point load is used. However, if the load is positioned at the far edge of an
adjacent unit, that is at the extremity of the panel, only half the load will be distributed
to the designed unit.
Unit 1. Figure 4.18b. w = one unit = 1.2 m (free edge and trimmed units). Load = UDL plus
three point loads (2 no. trimmer reactions, cross-wall x ½ unit length). The 200 mm notch
has an effective length of 200 + ½ bearing length 50 = 250 mm length. The magnitudes of
the dead and live loads are shown in Figure 4.18b. This case need not be considered under
the rules given in BS EN 1168 as there is no load spreading. The service loads per 1.2 m
width [ultimate 6.10(b) in brackets] are as follows:
13.74 + 4.80 kN
9.20 kN [18.54 + 6.00]
[12.42]
gk = 7.26 kN/m. qk = 4.8 kN/m [16.27]
70.20 kN
3.00
[94.44]
4.60
9.100
Then: M s1 = 180.3 kNm, M Ed1 = 242.2 kNm and V Ed1 max = 91.1 kN/unit (note V Ed is
calculated at Lb /2 + yB = 50 + 153 = 0.203 m from centre of support).
Panel 2-3 for Unit 2. Figure 4.18c. w = 1½ unit = 1.8 m (because of the limitation of load
spread from the trimmer loads) <leff /4 + 0.15 = 2.425 m. Only ½ of 5.0 kN/m line load will
be distributed over a 1.8 m panel. The loads [ultimate] per 1.8 m panel width are as follows:
9.2 kN
13.74 + 16.8 kN
[12.42]
[18.54 + 21.0]
gk = 2.5 kN/m [3.13]

gk = 10.89 kN/m. qk = 7.20 kN/m [24.41]
73.94 kN
3.00
[98.68]
4.55
4.60
9.100
Then: M s2 = 184.8 kNm/unit, M Ed2 = 246.3 kNm/unit and V Ed2 max = 95.4 kN/unit.
Panel 2-3-4 for Unit 3. Figure 4.18d. w = 3 units = 3.6 m < 1½ × 1.2 (to left of unit 3) + (leff/4 +
0.1) (to right of unit 3) = 4.175 m. The loads [ultimate] per 3.6 m panel width are as follows:
12.0 kN 10.0 kN
[15.0] [12.5]
gk = 5.0 kN/m [6.25]

gk = 21.77 kN/m. qk = 14.40 kN/m [48.82]
63.64 kN
6.05
[84.96]
4.55
4.60
9.100
Panel 3-4-5 for Unit 4. Figure 4.18e. w = 3 units = 3.6 m < leff /2 + 0.1= 4.65 m. Only ½ of
the 10 kN point load straggling Units 5 and 6 will be distributed into this panel. The loads
[ultimate] per 3.6 m panel width are as follows:
10.0 + 5.0 kN
2.4 kN [12.5 + 6.75]
[3.0]
gk = 5.0 kN/m [6.25]

gk = 21.77 kN/m. qk = 14.40 kN/m [48.82]
62.62 kN
6.05
[83.69]
4.55
4.60
9.100
Panel 4-5-6 for Unit 5. Figure 4.18f. w = 3 units = 3.6 m < leff /4 + 0.1 (to left) + 1.8 (to
right, limited at free edge) = 4.175 m. However, the position of the point loads (x = 6.05 m)
may yield a narrower spread width than 0.25leff. Using the parabolic load spread equation
w′/leff = 1.2 (x/leff) (1 – x/leff) = 1.2 × 0.665 × 0.335 = 0.267 > 0.25, therefore not critical.
The loads [ultimate] per 3.6 m panel width are as follows:
20.0 kN
[25.0]
gk = 21.77 kN/m. qk = 14.40 kN/m [48.82]
59.29 kN
x = 6.05 [79.53]
l = 9.100
Panel 5-6 for Unit 6. Figure 4.18g. w = 2 units = 2.4 m < leff /4 + 0.1 (to left) + 0.6 (to right,
limited at free edge) = 2.975 m. The loads [ultimate] per 2.4 m panel width are as follows:
10.0 kN
[12.5]
gk = 14.52 kN/m. qk = 9.60 kN/m [32.54]
58.19 kN
6.05 [78.15]
9.100
The final bending moment diagrams for the six units is shown in Figure 4.18h, in which
M s2 in Unit 2 is clearly critical, a conclusion that may have been reached by inspection of
the slab field due to the reactions from the trimmer angles and that Panel 2-3-4 contains a
200 mm hole close to mid-span. The moment capacity of Unit 2 appears to be reduced by
the presence of the 200 mm wide hole in Unit 3, but the bending moment diagram actually
represents the distributed capacity of Units 2 & 3 within Panel 2-3, where the hole reduces
the panel width from 1.8 to 1.6 m, with the consequential loss of strands and hence moment
capacities M sR and M Rd. The build-up (sloping lines) is due to the gradual (assumed linear)
increase in prestress and bond strength within the transmission and anchorage lengths on
either side of the hole where the strands are cut.
Lateral spreading of prestress and section properties where holes are present is not covered
in codes of practice or product standards but may be considered using the same approach for
lateral distribution of loads within the panel width. This has been carried out in Panel 2-3,
where reduced section properties and prestress lost at the 200 mm wide hole in Unit 3 have
been averaged across the panel width of 1½ units. In the final design of the floor slab field,
Unit 2 contained 10 no. strands (Ap2 = 930 mm 2), but two strands lost at the hole in Unit 3
(Ap3 = 8 × 93 = 744 mm 2) means that the moment capacity at the position of the hole in Unit
3 is calculated using a mean area = (744 + ½ × 930)/1½ = 806 mm 2 . Similarly, mean val-
ues of the cross-sectional properties Ac and Zb used in the calculation of prestress are used
to calculate M sR at the hole, and mean values of b′ = (1000 + ½ × 1200)/1½ = 1067 mm,
d′ and to calculate . Therefore, the moment capacity of Unit 3 without hole spreading
would be M sR = 180.5 kNm, and with hole spreading is = 189.7 kNm > M s2 = 184.8
kNm, that is Units 2 & 3 fail without hole spreading. Note that in the stress analysis the
bending stresses due to the self-weight of the units are not distributed, only the stresses after
the joints are in situ filled and the floor field is completed. Similarly, without hole spreading
M Rd = 261.4 kNm and with hole spreading = 281.2 kNm > M Ed2 = 246.3 kNm, that is
units still pass without hole spreading.
More than 90% of all precast concrete used in flooring is prestressed, the remainder being
statically reinforced. Slabs are designed in accordance with national codes of practice
together with other selected literature, which deals with special circumstances (Walraven and
Mercx 1983, FIP Recommendations 1988, PCI Manual 1991, Girhammer 1992, Pajari 1998,
Concrete Manufacturers’ Association (South Africa) 2013, fib Recommendations 2000, fib
Recommendations 2015). It is necessary to check all possible failure modes, shown diagram-
matically in Figure 4.19. These are, from short to long spans, respectively as follows:
• Bearing capacity FRd

• Shear capacity V Rd,c and V Rd,cr
• Flexural capacity, service M sR and ultimate M Rd
• Deflection limits, δ total and δ after finishes
• Handling restriction (imposed by manufacturer)
Hollow core units and slab fields are also designed against local failures due to punching
shear and partial line load in the mid-width and at edges of hollow core units. See BS EN
1168, Section 4.3.3.2.4 and 4.3.3.2.5 for design methods and equations (however, note that
these may be moved to the informative annexes in the revision to this standard).
Bearing
Shear
Applied load
Flexure
Deflection
Handling
limit
Span
Schematic representation of applied load vs span characteristics of flexural elements.
Standardised cross sections and reinforcement quantities are designed to cater for all combi-
nations of floor loading and spans. Section sizes are selected at incremental depths, usually
50 mm, and a set of reinforcement patterns are selected. For example in the 1200 mm
wide × 300 mm deep unit shown in Figure 4.4, there are seven voids and eight webs where
reinforcement may be placed. Possible combinations of strand patterns are as follows:
• 8 no. 9.3 mm strands, total area = 8 × 52 = 416 mm 2

• 8 no. 12.5 mm strands, total area = 8 × 93 = 744 mm 2
• 8 no. (row 1) plus 2 no. (row 2) 12.5 mm strands, total area = 10 × 93 = 930 mm 2
Up to a maximum of 8 no. (row 1) plus 6 no. (row 2) 12.5 mm strands, total area =
14 × 93 = 1302 mm 2 at which point the demand on either the pretensioning bed and equip-
ment is exceeded or the permissible surface stress(es) when the strands are detensioned
within the first 24 h after casting is violated.
Moment resistance, shear force resistance and flexural stiffness, that is deflection limits,
are first calculated and then compared with design requirements. Designers usually have
between 4 and 8–10 options of different depths and reinforcements to choose from – the
economical one being the shallowest and most heavily reinforced unit, although unaccept-
able deflections may rule this one out. The additional advantage is that the depth of the
‘structural floor zone’ is kept to a minimum.
The flexural behaviour of precast prestressed concrete is no different to any other type of
prestressed concrete. In fact, improved quality control of factory cast concrete may actually
improve things and certainly helps to explain the excellent correlation between test results
and theory found in precast units. The flexural behaviour of reinforced precast is certainly no
different to cast in situ work, all other things being equal. Thus, it is only necessary to discuss
further the parameters, both material and geometric, unique to precast concrete.
The major difference in behaviour in precast units is due more to the complex geometry
found in voided units such as hollow core and bubble units, which have rapidly reducing
web thickness near the neutral axis. Subjected to a bending moment M s, the concrete in the
tension face should not exceed the mean tensile strength of concrete, M s/Zb > fctm , where Zb
is the section modulus at the tension face and fctm = 0.3 fck2/3 (for concrete fck ≤ 50 N/mm 2 ,
otherwise 2.12 ln (1 + fcm /10)). However, with a margin of about 1.5 actual measured values
are closer to ¢ = , for example for C45/55 fctm = 3.80 N/mm 2 vs f t′= 5.5 N/mm 2 .
The limit of fctm is based on a requirement in BS EN 1992-1-1, clause 7.3.2(4) to guard
against sudden rupture by not being able to specify the minimum area of reinforcement in
a prestressed element, thereby restricting the tensile stress to σct,p = fctm in clause 7.3.2(2).
Furthermore, fctm is appropriate only for durability class XC1 (internal exposure low humidity
or permanently wet) according to BS EN 1992-1-1, Table 7.1N when M s is the characteristic
combination of loading gk + qk. For durability classes XC2 to XC4, the limiting tension is
zero (called ‘decompression’ in Table 7.1N) and M s is the quasi-permanent load gk + ψ2 qk.
For durability, XD1/XS1 or greater fctm = 0 and M s is the frequent load gk + ψ1 qk. The crack
width wk for prestressed concrete is limited wmax ≤ 0.2 mm. The loading and tensile stress
and crack width limitations are summarised in Table 4.3.
After cracking, tension stiffening of the concrete (due to the elasticity of the reinforce-
ment) allows reduced tensile stress in this region, but when the tensile stress reaches the nar-
row part of the web, cracks extend rapidly through the section and the flexural stiffness of
the section reduces to a far greater extent than in a rectangular section. Figure 4.20a shows
this behaviour in a flexural test carried out on a 320 mm deep hollow core unit. The service-
ability limiting state must be checked to prevent this type of behaviour, hence the need for
clause 7.3.2(4).
A second reason why the service condition is calculated in this manner is that the ratio
of the ultimate moment of resistance M Rd to the serviceability moment of resistance M sR is
usually about 1.6–1.8. Thus, with the use of the partial load factors (γG = 1.25 or 1.35 and
γQ = 1.50 or 1.5 ψ0), the serviceability condition will always be critical provided the amount
of prestress is sufficient. Finally, the problem of cracking in the unreinforced zones is par-
ticularly important with regard to the uncracked shear resistance. It is therefore necessary
to ensure that tensile stresses are not exceeded.
Serviceability limit state loading, tensile stresses and crack width limitations
XC0, XC1 Characteristic 0.2 Quasi-permanent 0.3 (Note 1)

+ + ψ2
XC2, XC3, XC4 Characteristic 0.2 0.3
+
Note 2 Quasi-permanent Zero N/A
+ ψ2
XD1, XD2, XD3 Frequent Zero N/A
XS1, XS2, XS3 + ψ1
Source: This table is an interpretation of Table 7.1N in BS EN 1992-1-1.
Note 1. May be increased if appearance is not limiting.
Note 2. Both conditions should be checked, although it is rare that the service stresses under characteristic load with
will be critical unless is very low and high.
N/A, not applicable as no cracking is allowed.
(a)
(b)
(a) Flexural bending testing of a hollow core floor unit. (b) Ultimate failure in a hollow core floor unit.
M sR is calculated by limiting the flexural compressive and tensile stresses in the concrete
both in the factory transfer and handling condition and in service. Figure 4.21 shows the
stress conditions at these stages for applied sagging moments – the diagrams may be inverted
for cantilever units subject to hogging moments.
The limiting compressive stress is according to BS EN 1992-1-1, clause 7.2(3) as ≤0.45 fck,
a limit taken in order to avoid the need for non-linear creep in crack width and defor-
mation calculations using quasi-permanent loads. In the absence of direct information
for the characteristic load ‘+’ prestress, the same value is taken although clause 7.2(3)
Principles of serviceability stress limitations for prestressed concrete elements.
suggests this can be taken to ≤ 0.6 fck even for the worst durability conditions. Note that
0.45 fck is about the same as the BS 8110 limit of 0.33 fcu. As it happens, this limit is
rarely critical in slabs other than the temporary condition in prestressed solid slab units.
The limiting flexural tensile is given in Table 4.3. To optimise the design it is clear from
Figure 4.21, that the limiting stresses at transfer should be equally critical with the limit-
ing service stress, and that the top and bottom surface stresses should attain maximum
values simultaneously. In practice, this is impossible in a symmetrical rectangular section
such as a hollow core unit but can be better achieved in a double-tee section. Also, the
balance between the limiting concrete stresses at transfer and in service is dictated by the
maturity of concrete and the need to de-tension the reinforcement within 12–18 h after
casting.
The transfer stress is expressed in the usual manner as the characteristic strength fck(t)
(where (t) symbolises early strength < 28 days) either as a function of the final concrete
strength fck according to clause 3.1.2.(5) and (6) or is measured by the manufacturer con-
verting cube strength to fck(t) according to BS EN 1992-1-1, Table 3.1. The limiting compres-
sive strength at the bottom fibre at transfer is according to clause 5.10.2.2.(5) ≤ 0.6 fck(t),
and more importantly at the top fibre by the tensile strength fctm(t), where defined by clause
3.1.2.(6) and (9) as
(4.1)
For example fck = 45 N/mm 2 , fcm = 45 + 8 = 53 N/mm 2 , fctm = 0.3 × 452/3 = 3.80 N/mm 2 ,
fck(t) = 30 N/mm 2 , fcm = 30 + 8 = 38 N/mm 2 , then fctm(t) = 3.80 × 38/53 = 2.72 N/mm 2 .
For fck = 45 N/mm 2 (the typical strength), fck(t) should be about 28–30 N/mm 2 (35–40 cube
strength). The use of rapid hardening cements, semi-dry mixes and humid indoor curing
conditions are conducive to early strength gain.
A typical hollow core slab production factory is shown in Figures 2.13 and 4.2. Steel
reinforcement, ‘tendons’ such as wire or helical strands of total area Ap, is stretched between
jacking equipment at either end of long steel beds, about 100–150 m long, after which
concrete is cast around the tendons. The tendons are positioned eccentrically relative to the
centroid of the section to produce the desired pretensioning stresses shown in Figure 4.21.
The initial prestress (which is set by the manufacturer) is around 70%–75% of the ultimate
strength f pk = 1670–1770 N/mm 2 for wire and f pk = 1750–1860 N/mm 2 for strand. The many
different types of tendons available (See Section 2.3.4) simplify to either 9.3, 10.9 and
12.5 mm diameter 7-wire helical strand, or 5 and 7 mm diameter plain or indented wire.
Table 2.12 lists their properties.
The tendons cannot sustain the initial stress for the following reasons:
1. During tensioning the tendons immediately relax, about 0.5%–1% of the initial stress
is lost, some of this may be due to slight slippage of the tendons in the grips. Over
the long term, the tendons lose a further 2%. The relaxation classification is given
as ρ1000 = 2.5% for relaxation class 2 according to BS EN 1992-1-1, clauses 3.3.2.(6)
and (7). The 1000-hour relaxation test value is provided by the manufacturer (or as
given in BS 5896). The initial relaxation loss at the time of transfer is according to
code equations 3.28 through 3.30. The initial relaxation loss is about 0.5% and is
subtracted from the initial prestress before further losses are calculated. The final
relaxation loss is about 2%.
2. After the concrete has hardened around the reinforcement and the bars are released
from the jacking equipment, the force in the bars is transferred to the concrete by bond.
The concrete shortens elastically – this may be calculated knowing Young’s modulus
of the concrete at this point in time transfer. This is called “elastic shortening” and
because the reinforcement is obliged to shorten the same amount as the concrete has,
the stress in it reduces too, by about 4%–6% depending on the total prestressing force.
Losses 1 and 2 are called “transfer losses.”
3. Desiccation of the concrete follows to cause a long-term shrinkage loss. This is the
product of the shrinkage per unit length (about 400 × 10 –6 for indoor and 250 × 10 –6 for
outdoor exposure) and modulus of elasticity of the tendons Ep = 195 kN/mm 2 (strand)
and 205 kN/mm 2 (wire). This gives a shrinkage loss between 55 and 70 N/mm 2 , about
4.0%–5.5%.
4. Finally, creep strains are allowed for using a specific creep strain (i.e. creep per unit
length per unit of stress) of about 2.0 for indoor curing and loading at 90 days in the
United Kingdom. Creep affects the tendons in the same manner as elastic shortening
because its effect is not only measured at the centroid of the bars, but also is influenced
by the shape of the element, called notional dimension (or depth or thickness) ho (area-
to-perimeter ratio) and fck. Hence, the creep loss is about 1.5 times the elastic shortening
loss, 7%–9%.
Total losses range from about η = 18%–25% for minimum to maximum levels of prestress.
The initial prestress = σpi, after initial losses = σpm0, after short-term losses at installation = σpmi
and the design effective prestress after all losses = σpo (note EC2 uses σp∞ for the final prestress,
and some texts use f for these stresses). The final force in the tendons Ppo = Ap σpo.
The full analysis is as follows (BS EN 1992-1-1, clause references are LHS):
Depth of unit = h; Cover to tendons = c
Height to the centroid of strands ys = c + ϕ/2; Height to the centroid to concrete = yb
Gross concrete area (exclude infill) Ac; Eccentricity of strands zcp = yb − ys
The second moment of area Ic; Section modulus at bottom Zb = Ic /yb
Section modulus at top Zt = Ic /(h − yb); Z at the tendon centroid Z z = Ic /zcp
Notional dimension = ho
Step 1
5.10.3 Losses at transfer:
3.3.2(7) Immediate relaxation of tendon at t hours
Exp. 3.29 μ = ratio of initial prestress η (typically 0.7)
Initial prestress after friction and wedge slippage σpi = η f pk (4.2)
Exp. 3.29 For Class 2. (4.3)
Prestress at release σr = σpi − Δσpr

Prestress force at release Pr = σr Ap
Step 2
5.10.4 Instantaneous deformation due to elastic shortening determined after
relaxation loss.
Concrete stress at the level of tendons σc = Pr/Ac + Pr zcp/Zz {−M self /Zz if self-weight is
considered for losses at mid-span in the stockyard} (4.4)
Exp. 5.44 Loss Δσel = σc Ep/Ecm(t)
5.10.3(2) After initial losses, σpm0 = σpi − Δσpr − Δσel ≤ kT f pk = 0.75 f pk (4.5)
Prestress force at transfer Ppm0 = σpm0 Ap

Rtr = σpm0/σpi
No partial factor for possible variations in prestress is applied to Ppm0 according to the UK
NAD to BS EN 1992-1-1, clause 5.10.9(1)P, rsup = 1.0 (whilst EN 1992-1-1 gives rsup = 1.05).
Some texts (e.g. Bhatt 2011) apply γsup = 1.05 from EN 1992-1-1 to Ppm0. Furthermore, it is
common for manufacturers to directly measure prestress by elongation, vibrating wire or
cross-bow tension meters, allowing rsup = 1.0 to EN 1992-1-1 itself.
Check maximum surface stress at transfer:
At bottom σb(t) = Ppm0/Ac + Ppm0 zcp /Zb (4.6)
5.10.2.2(5) limit σb(t) ≤ σc = 0.6 fck(t)
Note if σb(t) > 0.45 fck(t) is considered for calculation of a camber up to the date where σb(t) =
0.45 fck, which is approximately three days, and also just at the ends of the unit where tensile
stress due to self-weight is small. In these circumstances, non-linear creep is negligible.
At top (4.7)
The positive (compression) stress due to the self-weight of the slab at the end of the trans-
mission zone may be deducted from the above stress, unless of course the slab is held down
during transfer by weight packs, thereby nullifying the self-weight stress as the unit cambers
upwards. Without the weight pack, Equations 4.7 becomes
(4.8)
where M sw = w 0 L lpt /2 – w lpt 2 /2, where w 0 = self-weight, lpt = basic transmission length =
lpt2 /1.2 (see Section 4.3.7.1).
Step 3(a)
Short-term losses can be calculated up to installation time ti using relative humidity
RH = 70% with all faces exposed, and to the long-term 500,000 h (57 years) using service
RH s = 50% (indoor) or 70% (outdoor) with the exposed sides only exposed (other sides
protected by finishes, grouting, etc.).
5.10.6(1a) Loss due to creep to installation. See Annex B.1. Although the strength of concrete at
one day will be the transfer strength fck(t) N/mm2, after a few days it will reach the 28-day strength
fck N/mm2, and so the mean strength fcm is taken for the strength factors in this calculation.
Notional depth during this period is for all faces exposed ho = 2Ac /u (ignoring the cores
in hollow core slabs)
Exp. B.1 Creep coefficient φ(ti,to) = φRH β (fcm) β (to) βc(ti,to) (4.9)
for ti = installation, to = transfer age (days) and RH = 70%
Exp. B.3b/B.8c Relative humidity factor φRH = [1 + (1 – RH/100)/(0.1 ho1/3) α1]α2 (4.10)
where α1 = (35/fcm)0.7 and α2 = (35/fcm)0.2
Exp. B.4 Strength factor β (fcm) = 16.8/√fcm (4.11)
Exp. B.5 Age at release loading factor β (to) = 1/(0.1 + toT 0.2) (4.12)
a
æ ö
Exp. B9 = ç ÷ ³ (4.13)
è ø
where α = 1, 0, −1 for Class R, N, S cement
Exp. B10 Equivalent age at transfer tT = to e−[4000/(273+T) – 13.65] (4.14)
where T = mean curing temperature °C during curing time in days, taken as 50°C
Exp. B.7 βc(ti,to) = [(ti – to)/(βH + ti – to)]0.3 (4.15)
where to is transfer age
Exp. B.8b RH factor βH (days) = 1.5 [1 + (0.012 RH)18] ho + 250 α3 (4.16)
where α3 = (35/fcm)0.5
After initial losses σc = Ppm0/Ac + Ppm0 zcp /Zz {−M self /Zz} (4.17)
Exp. 5.46 (4.18)
Prestress after short-term creep losses σpmi = σpi − Δσpci
Prestress force at installation Ppmi = σpmi Ap
Step 3(b)
5.10.6(1a) Loss due to creep to t = 500,000 h; service RH s = 50%.
Notional depth is for bottom only ho = 2Ac /b
Exp. B.2 Creep coefficient φ(t,to) = φRH β (fcm) β (to) βc (t,to) (4.19)
Exp. B.3b/B.8c φRH = [1 + (1 – RH s /100)/(0.1 ho1/3) α1]α2 (4.20)
Exp. B.7 βc (t,to) = [(20,833 – to)/(βH +20,833 – to)]0.3 (4.21)
Exp. B.8b RH factor βH (days) = 1.5 [1 + (0.012 RH s)18] ho + 250 α3 (4.22)
σc,q-p = Ppm0/Ac + Ppm0 zcp /Zz {−(M self + MG + ψ2 MQ) Zz self-weight plus
dead loads and quasi-permanent live loads} (4.23)
Exp. 5.46 (4.24)
Step 4
5.10.6(1a) Loss due to shrinkage from ts = transfer age to t = 500,000 h. See Annex B.2
RH during the service period of shrinkage = RHs
Notional depth is for bottom only ho = 2Ac /b
Exp. B.12 RH factor βRH = 1.55 [1 – (RHs/100)3] (4.25)
Type of cement (use rapid hardening) = Class R (αds1 = 6, αds2 = 0.11)
Exp. B.1 (4.26)
Table 3.3 = - - £ <
- - £ <
- - £ <
Exp. 3.10 Age factor βds (t,ts) = (20,833 – ts)/(20,833 – ts + 0.04 ho3/2) (4.27)
Exp. 3.9 Drying shrinkage strain εcd = βds(t,ts) kn εcd,o

10.3.1.2(3) Autogenous shrinkage strain is taken as zero if heat curing is used.
Exp. 3.8 Total shrinkage strain εcs = εcd
Creep coefficient φ(t,to) = φRH β (fcm) β (to) βH using the aforementioned values
Exp. 5.46 (4.28)
Step 5
5.10.6(1b) Long-term loss due to tendon relaxation is based on σpm0
Exp. 3.29 (4.29)
Exp. 3.29 (4.30)
Exp. 5.46 (4.31)
Prestress after final losses σpo = σpi – (Δσpr + Δσel + Δσp,c + Δσp,s + Δσp,r)
Final prestress force Ppo = σpo Ap
\ Total loss factor Rwk = 1 − σpo /σpi
No partial factor for variations in prestress and losses is applied to Ppo according to the UK
NAD to BS EN 1992-1-1, clause 5.10.9(1)P, rinf = 1.0 (whilst EN 1992-1-1 gives rinf = 0.95).
Some texts (e.g. Bhatt 2011) apply γinf = 0.95 from EN 1992-1-1 to Ppo.
Check maximum surface stress in service:
At bottom σb = Ppo /Ac + Ppo zcp /Zb (4.32)
7.2(3) Limit σb ≤ σc,p = 0.45 fck
At top σt = Ppo /Ac − Ppo zcp /Zt (4.33)
7.3.2(4) & (2) Limit σt ≥ −σct,p = −fctm (or zero according to BS EN 1882-1-1, Table 7.1N)
To calculate M sR , the section is considered uncracked and the net cross-sectional area Ac
and the second moment of area Ic are used to compute maximum fibre stresses σb and σt at
the bottom and top of the section. M sR is given for XC1 exposure by the lesser of:
M sR = (σb + fctm) Zb,co (4.34)

or
M sR = σb Zb,co (if durability > XC1) (4.35)
or
M sR = (σt + 0.45 fck) Zt,co (4.36)
M sR in Equation 4.34 is the capacity to be checked against the characteristic service load
gkx + qk. M sR in Equation 4.35 is checked against the quasi-permanent load gk + y 2 qk for
durability exposure classes XC2 to XC4, and the frequent load gk + y1qk for XD and XS. M sR
in Equation 4.36 is checked against the quasi-permanent load for all classes of durability
in order to avoid non-linear creep (BS EN 1992-1-1, clause 7.2.(3)), and if the top surface is
exposed long-term to XD or XS M sR has the capacity ≤ (st + 0.6 fck) Zt,co for the characteristic
service load (BS EN 1992-1-1, clause 7.2.(2)).
Where Zb,co and Zt,co are the compound section modulii using the transformed area of
tendons based on a modular ratio, without creep effects, m = Ep /Ecm. This applies only to
bending stresses, not to prestress.
Area of compound section Ac,co = Ac + (m − 1) Ap
yb,co = (Ac yb + (m − 1) Ap ys)/Ac,co
Ic,co = Ic + Ac (yb,co – yb)2 + Σ (m − 1) Ap (ys – yb,co)2 per layer of tendons
Zb,co = Ic,co /yb,co
Zt,co = Ic,co /(h − yb,co)
Double-tee slabs present a special case. Because of its cross section, the centroid of the unit lies
close to the top flange, and therefore the section modulus Zt to the top fibre is very large, typ-
cally 3 times Zb. Consequently, as the top fibre does not give a limiting value to MsR , the influ-
ence of fck is very small, as given in Equation 4.1. As the controlling influence in Equation 4.1
is σb, the stress at transfer becomes very important. It is therefore necessary with double-tee
units to try to achieve the maximum possible transfer strength, say about fck(t) = 35 N/mm2.
Example 4.1
Calculate M sR at the support and mid-span for the 200 mm deep prestressed hollow core
unit shown in Figure 4.22. The initial prestressing force may be taken as 70% of charac-
teristic strength of ‘standard’ 7-wire helical strand. Manufacturer’s data give relaxation
Class 2 detensioned at 20 h after 50°C mean temperature curing. Exposure is XC1. Span
of the unit = 8.0 m. Floor finishes = 1.5 kN/m 2 . Geometric and material data given by the
manufacturer are as follows:
Area = 152 × 103 mm 2; I = 697 × 106 mm4; yb = 99 mm; fck = 45 N/mm 2; fck(t) = 30 N/mm 2;
f pk = 1770 N/mm 2; Ep = 195 kN/mm 2; Ap = 52 and 93 mm 2 per 9.3 and 12.5 mm dia.
1154
40
140
200
40 51
35
70 (see Example 4.8)
35 cover 1197
Cross-sectional hollow core floor unit to Example 4.1. Strands represented by solid dots
(9.3 mm dia.) and open dots (12.5 mm dia.)
strand; cover to strand = 35 mm. Concrete density = 24.5 kN/m3. Gravel aggregates.
Cement CEM I grade 52.5R. Area of infilled joints = 7500 mm 2 .
Is the critical fibre stress at the top or bottom of the unit?
Solution
Material properties
fcm = 45 + 8 = 53 N/mm 2; fcm(t) = 30 + 8 = 38 N/mm 2
Ecm = 22 × (53/10)0.3 = 36,283 N/mm2; Ecm(t) = 36,283 × (38/53)0.3 = 32,837 N/mm2
fctm = 0.3 × 452/3 = 3.80 N/mm 2; fctm(t) = 3.8 × 38/53 = 2.72 N/mm 2
Ap = 6 × 93 + 4 × 52 = 766 mm 2
Section properties
Self-weight of unit = 152 × 103 × 24.5 × 10 –6 = 3.724 kN/m.
Zb = 697 × 106/99 = 7.040 × 106 mm3; Zt = 697 × 106/101 = 6.901 × 106 mm3
ys = (6 × 93 × 41.25 + 4 × 52 × 39.65)/766 = 40.8 mm
zcp = 99 – 40.8 = 58.2 mm; Zz = 697 × 106/58.2 = 11.979 × 106 mm3
Section properties of compound section with the transformed area of tendons

m − 1 = (195,000/36,283) − 1 = 4.37
Ac,co = 152,000 + 4.37 × 766 = 155,351 mm2; yb,co = 97.7 mm; Ic,co = 708.1 × 106 mm4
Zb,co = 7.244 × 106 mm3; Zt,co = 6.924 × 106 mm3; Zz,co = 708.1 × 106/58.2 = 12.17
× 106 mm3
These values are used only for calculating stresses due to loading (not for prestress), and
for M sR and deflections.
Prestress
Initial prestress σpi = 0.7 × 1770 = 1239.0 N/mm 2
Initial Ppi = 1,239.0 × 766 × 10 –3 = 949,074 kN
Δσpr = 1239.0 × 0.66 × 2.5 × e(9.1 × 0.7) × (20/1000)(0.75 × (1–0.7)) × 10 –5 = 4.95 N/mm 2
Pmo = (1,239.0 − 4.95) × 766 = 945,281 N
σcp = (945,281/152,000) + (945,281 × 58.2/11.979 × 106) = 10.81 N/mm 2
At mid-span, bending moment due to self-weight M s0 = 3.724 × 8.02 /8 = 29.79 kNm
Due to M s0, σcp0 = −29.79 × 12.17 = −2.45 N/mm 2

Then at mid-span σcp =10.81 – 2.45 = 8.36 N/mm 2
\ Δσel = 195,000 × 10.81/32,837 = 64.20 N/mm 2. At mid-span = 49.66 N/mm 2
σpm0 = 1169.9 N/mm 2 and at mid-span σpm0 = 1184 N/mm 2 < 0.75 × 1770 =
1328 N/mm 2 OK
Ppm0 = 1,169.9 × 766 = 896,106 N
Check transfer stresses the support
σb(t) = (896,106/152,000) + (896,106 × 58.2/7.040 × 106) = 13.30 N/mm 2 < 0.6 ×

30 = 18.0 OK
σt(t) = (896,106/152,000) − (896,106 × 58.2/6.901 × 106) = −1.66 N/mm2 > −2.72 OK
(Note that the self-weight of the unit at the end of the transfer length may be considered
if σb(t) and/or σt(t) are violated).
Maturity of concrete during curing
Mean temperature during 20 h curing = 50°C
tT = (20/24) e−[4000/(273 + 50)) − 13.65] = 2.96 days
Factor for Cement Class R = 1
toT = 2.96 × [9/(2 + 2.961.2) + 1]1 = 7.65 days
Long-time losses to life using RH = 50% with the bottom only exposed
ho = 2 × area/bottom = 2 × 152,000/1,200 = 254.0 mm
φRH = [1+ (1 − (50/100) × 0.75/(0.1 × 254.01/3))] × 0.92 = 1.464
β (fcm) = 16.8/√53 = 2.308
β (to) = 1/(0.1+ 7.650.2) = 0.624
βH = ((1.5 × (1+ (0.012 × 50)18) × 254.0) + (250 × 0.81)) = 584 days
βc(ti,to) = [(20,833 − 1)/(584 + 20,833 – 1]0.3 = 0.992
φ(t,to) = φRH β (fcm) β (to) βc (t,to) = 1.464 × 2.308 × 0.624 × 0.992 = 2.091
σcp = (896,106/152,000) + (896,106 × 58.22 /697 × 106) = 10.25 N/mm 2 .
At mid-span after subtracting self-weight (3.72 + 7500 × 24.5 × 10 –6 = 3.91 kN/m) and
dead UDL (1.5 × 1.2 = 1.8 kN/m)
At mid-span, bending moment due to self-weight M s0 = 5.71 × 8.02 /8 = 45.68 kNm

Due to M s0, σcp0 = −45.68 × 12.17 = −3.75 N/mm 2
Then σcp = 10.25 − 3.75 = 6.62 N/mm 2
Denominator in Exp. 5.46 = 1.126
Δσp,c = 195,000 × 2.091 × 10.25/(36,283 × 1.126) = 102.3 N/mm 2 . At mid-span =
66.12 N/mm 2
Concrete shrinkage
kn = 0.80
βRH = 1.55 × (1 − (50/100)3) = 1.356
βds (t,ts) = (20,833 − 1)/[(208 33 − 1) + 0.04 × 254.01.5] = 0.992
εcs = βds(t,ts) kn εcd,o = 0.992 × 0.80 × 0.85 × (220 + 110 × 6) × e –0.11 × 53/10 × 1.356
× 10 –6 = 0.000447
φ(t,to) = 2.091
Δσp,s = 0.000447 × 195,000/1.126 = 77.5 N/mm 2
Strand relaxation
μ = 1169.9/1770 = 0.661. At mid-span = 1184/1770 = 0.669
σpr = 1,169.9 × 0.66 × 2.5 × e(0.661 × 9.1) × [(500,000/1,000)(0.75 × (1–0.661))] × 10 –5
= 38.37 N/mm 2
At mid-span σpr = 40.29 N/mm 2
Δσp,r = 0.8 × 38.37/1.126 = 27.27 N/mm 2
At mid-span Δσp,r = 28.63 N/mm 2 (slightly greater due to smaller losses)
Final prestress
σpo = 1169.9 – 102.3 – 77.5 – 27.3 = 962.8 N/mm2. At mid-span σpo = 1012.2 N/mm2
Ppo = 962.8 × 766 = 737,511 N. At mid-span = 775,318 N
Rwk = 737,511/949,074 = 0.777. At mid-span ratio Rwk = 0.817
Final prestress at mid-span

σb = (775,318/152,000) + (775,318 × 58.2/7.040 × 106) = 11.51 N/mm2 < 0.45 × 45
= 20.25 N/mm2
σt = (775,318/152,000) – (775,318 × 58.2/6.900 × 106) = –1.44 N/mm2 > –3.80 N/mm2
M sR at mid-span is the lesser of:

At the bottom fibre. M sR,b = (11.51 + 3.80) × 7.244 = 110.9 kNm
At the top fibre. M sR,t = (20.25 + 1.44) × 6.924 = 150.2 kNm > 110.9 kNm
7.2(3) But in design M sR,t need only be checked against quasi-permanent service
loads.
Final prestress at support

σb = (737,511/775,318) × 11.51 = 10.95 N/mm 2
σt = (737,511/775,318) × –1.44 = –1.37 N/mm 2
M sR at the support is the lesser of:

M sR,b = (10.95 + 3.80) × 7.244 = 106.8 kNm
M sR,t = (20.25 + 1.37) × 6.924 = 149.7 kNm > 106.8 kNm
The bottom fibre is critical in both cases.
The distribution of M sR between support and mid-span is taken as parabolic to mimic

the parabolic distribution of bending stress due to self-weight and dead loads.
Example 4.2
Find the required compressive fck in Example 4.1 that would equate the service moment
based on the top and bottom limiting service stress conditions, thus optimising the
strength of concrete.
Solution
Solving M sR = (σb + fctm) Zb,co = (σt + 0.45 fck) Zt,co
Then (+10.95 + fctm) 7.244 × 106 = (−1.37 + 0.45 fck) 6.924 × 106
Hence, fck = 29.3 N/mm 2
And M sR = 100.9 kNm. The result is less than the transfer strength suggesting an
impractical solution. This result further demonstrates that increasing fck to say 50 N/mm 2
would have little effect on the value of M sr. It is therefore necessary to modify the section
properties Zb,co and Zt,co to obtain comparability, as follows.
Example 4.3
Find the required values of Zb,co and Zt,co in Example 4.1 necessary to equate the value
of M sR obtained from limiting stresses. Calculate the new value of M sR . Study the cross
section and check whether the new values of Zb,co and Zt,co can be achieved practically.
Solution
Solving M sR = (σb + fctm) Zb,co = (σt + 0.45 fck) Zt,co
Then (11.51 + 3.80) Zb,co = (1.44 + 20.25) Zt,co
Then Zb,co /Zt,co = 21.69/15.31 = 1.417, that is yt /yb = 1.417
Also yb + yt = 200 mm
Solving yb = 82.7 mm and yt = 117.3 mm
Then Zb,co = 708.1 × 106/82.8 = 8.557 × 106 mm3 and Zt,co = 6.037 × 106 mm3
and M sR = (11.51 + 3.80) × 8.557 × 106 × 10 –6 = 131.0 kNm > 110.9 kNm in
Example 4.1.
To achieve this condition, the geometric centroid must be lowered by 99.0 – 82.7 =
16.3 mm. To achieve this, the voids must be repositioned or modified in shape. It is not
possible to raise the position of the voids by twice this distance (the core area is about
50%) by making the thickness over the top of the cores too shallow hft = 40 – 32 = 8 mm.
In calculating the ultimate resistance, material partial safety factors should be applied as per
usual, γs = 1.15 for steel and γc = 1.5 for concrete in flexure. The ultimate flexural resistance
M Rd when using bonded tendons is limited by one the following:
1. Ultimate compressive strength of concrete, 0.567 fck

2. The design tensile stress in the tendons, f p
The ultimate failure of prestressed concrete beams and slabs is considerably ductile, but in
spite of sections being underreinforced flexurally due to the small areas of reinforcement
(compared to comparable r.c. sections), the final mode of failure is sudden and often gov-
erned by compressive failure of the concrete as shown in Figure 4.20b. It should be noted
that the cross-line load applied in this test is not representative of uniformly distributed
loading and will have contributed to the localised compressive failure.
The depth of the (strain-responsive) neutral axis X is obtained by considering the equilibrium
of the section. The tensile strength of the steel depends on the final prestress σpo in the tendons
after all losses and initial prestress levels have been considered. In most hollow core produc-
tion (and sections where the tendons are all near to the bottom), the ratio σpo /fpk = 0.53–0.6,
depending on the area of tendons. For double-tee units where most of the tendons are posi-
tioned further from the bottom, σpo /fpk = 0.45–0.5. The stress history for pretensioned tendons
is shown in Figure 4.23a, where the effect of prestrain after losses εpo (A), decompression Mk =
σb Zb,co (B), cracking Mk > fct Zb,co (C) and ultimate strength Mk = fp Ap z (assuming tendons fail
before concrete) (D) are noted, where Mk is the applied bending moment at each stage.
≤ 0.85 fck/γm design
X
d
Tension
stiffening
σb Zero fct ( fctm design)
A, Prestress B, Decompression C, Cracking D, Ultimate
Ultimate
fp
Loading
σpi Prestress
Cracking
σpo σpo + Mk/Zz
(a) A C Time and increasing loading D
Limiting stress and strain according

to clause 3.3.6(7)
Design value
fpd = fpk/γs 0.87fpk
fp,max
fp
0.9 fpk/γs
Ep
(b) εLOP = 0.9fpk/γsEp εp εud = 0.0200 εuk = 0.0222
(a) Stress history of pretensioned tendons. (b) Stress versus strain relationship for pretension-
ing tendons according to BS EN 1992-1-1, Fig. 3.10. ( )
0.0035 fcd
dn = 0.4 X
X 0.8X
Fc = 0.8 fcd b X
εp Fs = fp Ap
(c)
Compression zone
dn
x
(d)
( ) (c) Force equilibrium in flexural design. (d) Compressive stress zones in top flange
and above hollow cores.
Values for X/d and f p may be obtained from strain compatibility εp = εpo + εcu3 (d/X – 1),
equilibrium Fs = f p Ap = Fc = 0.567 fck 0.8 b X, and the stress versus strain relationship shown
in Figure 4.23b (From EN1992-1-1, Fig. 3.10), solving for X as follows:
Fig. 3.4, Table 3.1 εcu3 = 0.0035 for fck ≤ 50 N/mm 2

Otherwise εcu3 = 0.0026 + 0.035 [(90 – fck)/100]4 for fck ≤ 90 N/mm 2
3.3.6(7), Fig. 3.10 εLOP = 0.9f pk /(γs Ep) at which stress = 0.9 f pk /γs (4.37)
3.3.6(7), Fig. 3.10 εuk = 0.0222 at which design stress f pd = f pk /γs

5.10.9 Pre-strain due to prestress after losses, εpo = rsup σpo /Ep but rsup = 1.0
Effective depth of tendons in the tension zone, d = h – ysT
Let X = depth to neutral axis, and first assuming that 0.8X < hft
Total strain = pre-strain + compatibility concrete strain, εp = εpo + εcu3 (d/X – 1) (4.38)
Refer to Figure 4.23b. If εLOP < εp < εuk in the inclined branch of Figure 4.23b
æe -e ö
= + ç ÷ (4.39)
èe - e ø
but where the strain is limited εp ≤ εud = 0.9εuk = 0.02
Figure 4.23c. Force equilibrium f p Ap = fcd b 0.8X (4.40)
where fcd = 0.85 fck /γc and b = breadth of unit plus infill = 1200 mm
Note that for 50 < fck ≤ 90 N/mm 2 , the stress block depth is [0.8 – (fck – 50)/400]X.
Combining Equations 4.38 through 4.40
0.8 fcd b (εuk – εLOP) X2 − [0.9(εuk – εLOP) + 0.1(εpo – εcu3 − εLOP)] Ap fpd X
− 0.1εcu3 d Ap fp = 0 (4.41)
Solving yields X
Depth to the centroid of concrete area, dn = 0.5 × 0.8X = 0.4X
If the compression zone lies beneath the level of the top flange, that is 0.8X > hft, the com-
pression force is the sum of the force in the top flange above the cores and webs between the
cores as shown in Figure 4.23d.
Thus, Equation 4.40 is amended to f p Ap = bc hft fcd + 0.8 fcd (b – bc) X
where bc is the total breadth of cores, such that Equation 4.41 is amended to
0.8 fcd (b – bc) (εuk – εLOP) X 2 – {[0.9(εuk – εLOP) + 0.1(εpo – εcu3 – εLOP)] Ap f pd
– fcd bc hft (εuk – εLOP)} X – 0.1εcu3 d Ap f pd = 0 (4.42)
Solving yields X > hft /0.8
+ -
Depth to centroid of concrete area, =
+ -
Lever arm = d – dn
From Equation 4.38, εp = εpo + 0.0035 (d/X – 1) ≤ εud = 0.02 (4.43)
æ e -e ö
f p is found from Equation 4.39, but f p,max ≤ 0.9 f pd + 0.1 f pd ç ÷ (4.44)
è e -e ø
Ultimate moment of resistance, M Rd = f p Ap (d − dn) (4.45)
A simplified approach for rectangular sections is to obtain f p /f pk and X/d from a table of
data (similar to BS8110’s familiar Table 4.4) based on compatibility of ultimate strain and
equilibrium of forces:
Fs = Ap f p = Ap f pk (f p /f pk)
Fc = 0.567 fck b 0.8X = 0.453 fck bd (X/d)
If εp ≤ εLOP (elastic branch of σ-ε curve), from Equations 4.38 and 4.43 × Ep /f pk
Values of / and / for prestressed design of rectangular sections
0.6 0.5 0.4 0.6 0.5 0.4

0.10 0.86 0.85 0.85 0.19 0.19 0.19
0.15 0.83 0.82 0.82 0.27 0.27 0.27
0.20 0.81 0.80 0.80 0.36 0.35 0.35
0.25 0.80 0.79 0.79 0.44 0.44 0.43
0.30 0.79 0.79 0.78 0.52 0.52 0.52
0.35 0.79 0.78 0.78 0.61 0.60 0.60
0.40 0.78 0.78 0.77 0.69 0.69 0.68
Data calculated using = 1770 N/mm2. If this is changed to say 1500 N/mm2, the differences are 0.2%.
Values of ε /ε and / 2 for prestressed design of rectangular sections

ε ε
0.6 0.5 0.4 0.6 0.5 0.4

0.10 2.82 2.76 2.65 0.079 0.079 0.079
0.15 2.08 1.96 1.84 0.110 0.110 0.109
0.20 1.65 1.53 1.42 0.139 0.138 0.137
0.25 1.39 1.27 1.15 0.164 0.164 0.163
0.30 1.21 1.09 0.188 0.187 0.186
0.35 1.09 0.208 0.207 0.207
0.40 0.226 0.226 0.225
Values of ε < ε in italics.
f p /f pk = f po /f pk + εcu3 Ep (d/X – 1)/f pk (4.46)
Then X/d = [Ap f pk /0.453 fck bd] [f po /f pk + εcu3 Ep (d/X – 1)/f pk] (4.47)
Solving the quadratic gives X/d, dn = 0.4X, f p /f pk from which M Rd is determined. If the final
strain εp = εpo + εcu3 (d/X – 1) > εLOP, Equation 4.47 is not valid and Ep is replaced by the
functions in Equation 4.41. The results for f p /f pk and X/d are given in Table 4.4 in terms of
three values of final prestress f po /f pk = 0.4–0.6. To check whether εp ≤ εLOP ratios of εp /εLOP
are given in Table 4.4, with elastic values in italics being in the minority. Finally, M Rd /fck bd2
is given for all cases (Table 4.5).
Example 4.4
Calculate M Rd for the section used in Example 4.1. Is the unit critical at the service or
ultimate limit state?
Solution
fcd = 0.85 fck /1.5 = 0.567 × 45 = 25.5 N/mm 2
ApT = 766 mm 2
ysT = ys = 40.8 mm (all in tension); d = 200 – 40.8 = 159.2 mm
εpo = 962.8/195,000 = 0.004937

εp = 0.004937+ 0.0035 (159.2/X – 1) Equation 1
Fc = 0.8 × 25.50 × 1,200 X = 24,480 X; Fs = f p × 766
Then for equilibrium: X/f p = 766/24,480 = 0.0313 Equation 2
εLOP = 0.9 × 1770/1.15/195,000 = 0.007104
If εp > 0.007104, then f p = 1385 + [154 × (0.0222 − εp)/(0.0222 − 0.007104)]
Equation 3
Combining Equations 1 through 3 gives: 370 X 2 −15,350 X −65,686 = 0
Then X = 45.4 mm, but as 0.8X < hft = 40 mm compression block is rectangular
dn = 0.4 × 45.4 = 18.2 mm; z = 159.2 − 18.2 = 141 mm
From Equation 1. εp = 0.013696
From Equation 3. f p = 1452 N/mm 2 < limit at εs = 0.02 = 1516 N/mm 2
M Rd = 1452 × 766 × 141 × 10 –6 = 156.9 kNm
Anchorage bond length lbpd to full M Rd = 736 + 0.19 × 11.2 × (1452 − 962.8)/2.13 =
1228 mm
where lpt2 = 736 mm (see Section 4.3.7.1), α2 = 0.19, average diameter = 11.2 mm
and f bpd = 1.2 × 0.7 × 3.80/1.5 = 2.13 N/mm 2
To check whether the unit is critical at ultimate, the ratio M Rd /M sR = 156.9/110.9 = 1.42.
This ratio is greater than the ratio of the M Ed to M s using γG = 1.25 and γQ = 1.5 where, in
this example, M Ed /M s = 1.37. Thus, the unit is not critical at ultimate strength.
Deflection calculations are always carried out for prestressed members – it is not sufficient
to check span-effective depth ratios as in a reinforced section. This is because the strength-
to-stiffness ratio of a prestressed section is considerably greater than in a reinforced section.
The effects of strand relaxation, creep etc. have greater effects as the degree of prestress
increases. The general method of curvature–area may be adopted in prestressed design. For
nondeflected strands, the curvature diagram is rectangular.
Net deflection is found by superposition of upward cambers due to pretensioning and
downward gravity loads. Calculations are based on a flexurally uncracked stiffness Ecm Ic,co
using the transfer value Ecm(t) for initial camber due to prestress and the final value E cm and
appropriate creep factors for long-term total deflections and deflections due to imposed dead
and live loads after finishes are applied. The latter are referred to as ‘movement’ to differen-
tiate them from total deflections. The compound value Ic,co is used, making a difference of
about 2%–3% compared to Ic.
Young’s modulus (kN/mm 2) is according to BS EN 1992-1-1, Table 3.1, Ecm = 22 (fck +
8/10)0.3 for gravel and reduced by 10% for limestone. The same equations may be used with
the transfer strength. When considering the gain in Young’s modulus over time between the
age at transfer (typically 16 hours) and at installation, say 28 days, a factor may be used
representing the approximate integration of Ec over that period, that is Ecm(t)/0.5(Ecm(t) +
Ecm). This factor may be used to adjust the short-term creep coefficients when calculating
deflections at installation.
Precamber deflection comprises the following four parts:
1. Immediate camber after detensioning (and cutting of hollow core slabs) due to the
immediate prestressing force Ppm0 after initial strand relaxation, elastic shortening
2. Short-term value due to Ppmi after Ppm0 plus short-term creep losses
3. Long-term value due to Ppo after all losses
4. Self-weight deflection
In-service long-term deflections are calculated in the usual manner taking into consideration
the support conditions, loading arrangement and creep. Dead plus quasi-permanent live
loads Gk + ψ2 Qk are used. Final deflection, including upward camber, is limited according
to BS EN 1992-1-1, clause 7.4.(4) to span/250. Movement, or deflection after finishes are
applied, is limited to span/500 where brittle finishes are to be applied or to span/350 for
non-brittle finishes. Architects may also impose a limit of 20 mm where the span/350 or 500
limit for longer spans may lead to damage of finishes.
Deflections are determined for the immediate elastic deformation, plus the viscoelastic
effects of creep, which is reflected in the changing values of pretension force and Young’s
modulus with time. The creep coefficients ψ are obtained from the Italian Concrete
Association ASSAP Manual (ASSAP 2002) used in recent fib publications, and the Eurocode
design guide by the Precast Concrete Federation (British Precast 2008). Because the stress
conditions at the top and bottom of the unit are different, the values for ψ determined for
the effect of dead and imposed loads are not the same as the internal strain-related creep
coefficients φ(t,to) (long term), φ(ti,to) (up to installation) and φ(t,ti) (after installation) used
for calculating losses in prestress.
Long-term deformations are based on a single creep coefficient ψ∞ = 2.5 together with
a concrete aging coefficient to allow for stress increments after initial loads X = 0.8 (greek
C, not X). The term ‘coefficient’ is used to distinguish this from creep ‘factors’ φ used in
stress analysis. Creep coefficients over time intervals 1 day to 12 months are as follows
(ASSAP 2002):
Coefficient of development at:
Transfer = 0.1 15 days = 0.3

28 days = 0.4 2 months = 0.5
3 months = 0.6 6 months = 0.7
12 months = 0.8 ∞ = 1.0
Let L = effective span (assume the same as the actual length of unit for calculating the
camber), Ppm0 = prestressing force at transfer, Ppmi = prestressing force at installation, Ppo =
final prestressing force, zcp = eccentricity, Ecm and Ecm(t) = Young’s modulus at 28 days and
at transfer, Ic,co = the second moment of the area of compound section, ψ1 = creep coefficient
from transfer to installation, ψ28 = creep coefficient from installation (assume 28 days) to
long-term, and ψ∞ = creep coefficient at long-term, wo, w1, w2 and w3 = uniformly distrib-
uted loads due to self-weight of unit, self-weight of slab after grouting, dead loads and
imposed live loads, respectively and ψ2 = quasi-permanent live load factor.
a. At transfer
Immediate camber δ1 = −Ppm0 zcp L 2 /8 Ecm(t) Ic,co (4.48)
Deflection due to self-weight
δ2 = 5 wo L 4/384 Ecm(t) Ic,co (4.49)
Resultant deflection at transfer = δ1 + δ2
b. At installation
Referring to the creep development coefficients in Section 4.3.6, the effective creep
coefficient ψ1 between 1 and 28 days takes into account the mean change in Ecm as
ψ1 = {Ecm(t)/0.5 (Ecm + Ecm(t))} × 2.5 × − 0.1) (4.50)
Camber at installation is due to the camber at transfer, plus further viscoelastic move-
ment ψ1 δ1 minus a (optional) small reduction due to the reduction in pretensioning
force from transfer Ppm0 to installation Ppmi. The second term is often ignored as it
makes about a 2%–3% difference in δ3.
δ3 = (1 + ψ1) δ1 + (Ppm0 – Ppmi) zcp L 2 /8 Ecm Ic,co (4.51)
Deflection at installation due to self-weight is due to the static deflection plus further
viscoelastic movement after transfer:
δ4 = + (1 + ψ1)δ2 (4.52)
Resultant deflection at installation = δ3 + δ4
c. Long term
Effective creep coefficient after installation takes into account the increase in strength
given by X = 0.8, and Young’s modulus after 28 days.
ψ28 = 0.8 × 2.5 × (1.0 − 0.4) (4.53)
Long-term camber is due to the camber at installation plus further viscoelastic move-
ment due to the pretensioning force at installation Ppmi, minus a (optional) reduction
due to the reduction in pretensioning force from installation Ppmi to long-term Ppo
δ5 = δ3 – [ψ28 Ppmi – (Ppmi – Ppo)] zcp L 2 /8 Ecm Ic,co (4.54)
Long-term deflection due to self-weight and infill w 1 is due to deflection at installation

plus viscoelastic movement of w 1 using ψ28:
δ6 = + δ4 + ψ28 5 w 1 L 4/384 Ecm Ic,co (4.55)
Static plus viscoelastic deflection due to UDL dead w2 added after installation using ψ28:
δ7 = + (1 + ψ28) 5 w 2 L 4/384 Ecm Ic,co (4.56)
Static plus viscoelastic deflection due to UDL live load ψ2 w 3 using 0.8ψ∞
δ8 = + (1 + 0.8ψ∞) 5 ψ2 w 3 L 4/384 Ecm Ic,co (4.57)
plus similar expressions for imposed point loads and line loads, for example point dead
load PG at distance a from support, where a < L – a. If a > L – a, replace a with L – a:
+y æ ö
d = ç - ÷ (4.58)
è ø
plus ditto point live load PQ:
y + y¥ æ ö
d = ç - ÷ (4.59)
è ø
Final long-term deflection = δ5 + δ6 + δ7 + δ8 + δ9 + δ10.
d. Deflection after installation

Long-term movement due to viscoelastic self-weight and dead load deflections after
installation are due to the post-installation creep factor ψ28, static and viscoelastic live
load deflections are due to 1 + ψ28 (not 1 + 0.8ψ∞), and changes in the camber after
installation. It appears from some experimental data on composite hollow core slabs
(Elliott and Ibrahim 2006, Ibrahim and Elliott 2009, Ibrahim et al. 2016) that changes
in deflection due to imposed load are closer to the effect of 1 + ψ28 than 1 + 0.8ψ∞.
Movement is the greater of δ11 or δ12
δ11 = 5 [(1 + ψ28) ψ2 w 3] L 4/384 Ecm Ic,co (4.60)
or
δ12 = 5 [ψ28 (w 1 + w 2) + (1 + ψ28) ψ2 w 3] L 4/384 Ecm Ic,co + (δ5 − δ3) (4.61)
or
δ12 = δ11 + 5 [ψ28 (w 1 + w 2)] L 4/384 Ecm Ic,co + (δ5 − δ3) (4.62)
plus deflections due to point and line dead and live loads, where (1 + ψ28) over the long
term is replaced by ψ28 after installation.
Summary deflections
7.4.1(4) Final long-term deflection = δ5 + δ6 + δ7 + δ8 + δ9 + δ10 < /250

7.4.1(5) Variable long-term movement due to live load only = δ11 < /500 or /350
7.4.1(5) Overall long-term movement, due to δ11 plus creep of camber and dead loads =
δ12 < /500 or /350
Example 4.5
Calculate the long-term deflection and the deflection after installation for the hollow core
unit in Example 4.1. The floor slab is for use in offices and is subjected to imposed dead
loads of 1.5 and 0.6 kN/m 2 for finishes and services/ceiling, and a live load of 5 kN/m 2 ,
respectively over a simply supported span of L = 8.0 m. Check the code limits if the floor
finishes are non-brittle. The self-weight of the precast unit alone is 3.72 kN/m run. After
the joints are grouted on site, the self-weight of the precast floor is 3.91 kN/m.
Solution
From Example 4.1, considering the prestress at the support (not mid-span), the initial loss
at transfer is 5.6% and final losses are 22.3%. Moreover (but not included in Example 4.1),
the losses at installation at 28 days after transfer are 9.1%.
Then Pi = 949.1 kN, Ppm0 = 896.1 kN, Ppmi = 862.3 kN and Ppo = 737.5 kN.
L = 8,000 mm, Ecm(t) = 32,837 N/mm 2 , Ecm = 36,283 N/mm 2 , zcp = 58.2 mm and
Ic,co = 708.1 × 106.
Deflection at transfer
δ1 = −896,106 × 58.2 × 8,0002 /(8 × 32,837 × 708.1 × 106) = −17.9 mm
δ2 = 5 × 3.72 × 8,0004/(384 × 32,837 × 708.1 × 106) = +8.5 mm
Net deflection at transfer = −9.4 mm
Deflection at installation
ψ∞ = 2.5; ψ (1 day transfer) = 0.1; ψ (28 days) = 0.4
ψ1 = [32,837/(0.5 × (32,837 + 36,283))] × 2.5 × (0.4 − 0.1) = 0.71
Ppm0 – Ppmi = 33,843 N
δ3 = 1.71 × −17.9 + [33,843 × 58.2 × 8,0002/(8 × 36,283 × 708.1 × 106)] = −30.1 mm
δ4 = 1.71 × 8.5 = +14.6 mm
Net deflection at installation = −15.5 mm
Long-term deflections
Xψ ∞ = 0.8 × 2.5 = 2.00; ψ28 = 2.00 × (1 − 0.4) = 1.20
Ppmi – Ppo = 124,752 N
δ5 = −30.1 −[(862,263 × 1.20) − 124,752] × 58.2 × 8,0002/(8 × 36,283 × 708.1 × 106)
= −46.6 mm
Self-weight of slab, w 1 = 3.91 kN/m
Floor dead load per unit width, w 2 = (1.50 + 0.60) × 1.2 = 2.52 kN/m
Offices ψ2 = 0.3, then q-p live load per unit width ψ2 w3 = 0.3 × 5.00 × 1.2 = 1.80 kN/m
δ6 = 14.6 + [(1.20 × 3.91 + (1 + 1.20) × 2.52 + (1 + 2.00) × 1.80] × [5 × 8,0004/(384 ×
36,283 × 708.1 × 106)] = 14.6 + 15.63 × 2.076 = +47.1 mm
Final deflection = −46.6 +47.1 = ±0.5 mm < 8000/250 = 32 mm
Long-term movement due to live load only δ11 = (1 + 1.20) × 1.80 × 2.076 = ±8.2 mm
< 8000/350 = 22.9 mm
δ12 = 8.2 + [1.20 × (3.91 + 2.52) × 2.076] + (−46.60 − −30.11) = +7.7 mm < 8.2 mm
Calculating the shear capacity of a precast concrete section is no different to any other type
of reinforced or prestressed section, provided that localised forces and stresses at end con-
nections are dealt with (see Sections 8.6 and 9.2) and do not interrupt the distribution of
shear. Shear capacity is determined only at the ultimate state – unlike flexure, there are no
limiting service conditions for shear. However, there is a paradox here because design equa-
tions derive from an elastic analysis.
Ultimate shear capacity is the sum of the various actions of the concrete (aggregate inter-
lock), longitudinal bars (dowel action) and shear reinforcement (stirrup action). However,
many reinforced and prestressed units, such as hollow core units and prestressed planks,
have no shear reinforcement due to their manufacturing methods, the only reinforcement
being longitudinal rebars or pretensioning strands. The shear capacity of these units there-
fore depends on the shear resistance of the concrete in combination with dowel action alone.
The calculation of shear resistance carries numerous partial safety factors, that is on
tensile strength fctd = 0.7 fctm /1.5, on the precompression γp,fav = 0.9, on the strength due to
aggregate interlock CRd,c /1.5, on the transmission length 1.2 and γc = 1.5 on bond strength.
There are good reasons for this. Experimental tests have found that shear resistance can
vary widely for nominally the same section. Variations in the order of 20% are quite com-
mon, especially where there are no shear stirrups and the shear resistance relies on the
diagonal strength of the concrete. A lot depends on compaction of concrete in the webs and
their breadth. In some units, such as hollow core units, small variations in the breadth of
webs can lead to premature failure of the entire section. The failure mode is also quite sud-
den, depending on the relationship between the development of a flexural crack before shear
cracking propagates, as shown in Figure 4.24a in a 150 mm deep hollow core unit, or pure
shear-compression cracks as shown in Figure 4.24b in a 300 mm deep unit. In the latter, the
shear resistance is about two times that of the former, but the bending (cracking) moment
of resistance is about six times. Note how failure has occurred due solely to one large crack
through the webs. The position and inclination of this crack, typically 30°–40° (depending
on the position of the shear load) has a major influence of the ultimate shear capacity.
In prestressed sections, shear capacity is calculated for two conditions; the uncracked sec-
tion and the cracked section in flexure. Figure 4.25. The uncracked ultimate shear resistance
V Rd,c (BS EN 1992-1-1, clause 6.2.2, Exp. 6.4) is greater than the ultimate cracked value
V Rd,cr (BS EN 1992-1-1, Exp. 6.2) because the full section properties are considered and a
small amount of diagonal tension in the concrete is permitted. Note that the code notation
is V Rd,c for both – this book uses V Rd,cr for the cracked capacity.
4.3.7.1 Shear capacity in the uncracked region, V Rd,c

The term uncracked refers to flexural cracking, and where this exists V Rd,c must not be used.
The ultimate shear capacity is given in BS EN 1992-1-1, clause as 6.2.2(1), Exp. 6.4 as
Exp. 6.4 = +a g s (4.63)
3.1.6.2(P) fctd = 0.7 fctm /1.5 = 0.14 fck2/3
6.2.2(2) αl = lx /lpt2 (defined later) (4.64)
6.2.2(2) σcp = Ppo /Ac = compressive stress at the centroidal axis due to
prestress after all losses (4.65)
2.4.2.2(1) γp,fav = 0.9 as this is a favourable effect
For flanged sections, bw is taken as the web width (narrower part). In flanged sections
where the centroidal axis is in the flange, Equation 4.63 should be applied at the junction
of the web-flange, that is σcp is calculated there. For cross sections where the width varies
(a)
(b)
(a) Shear-tension failure in a 150 mm deep hollow core unit (with permission from Bison
Manufacturing, the United Kingdom). (b) Shear-compression failure in a 300 mm deep hollow
core unit. (With permission from Bison Manufacturing, the United Kingdom.)
over the height, the maximum principal stress may occur on an axis other than the cen-
troidal axis. In such a case, the minimum value of the shear resistance should be found by
calculating V Rd,c at various axes in the cross section. This is particularly relevant to I beams
and some types of hollow core slabs with tall oval-shaped cores, where the principal shear-
tension stress occurs below the centroidal axis. See Section 4.3.7.3.
Principles of shear resistance for prestressed concrete elements.
Development of stresses in the transmission zone 2 and anchorage zone of prestressed

members at (1) release of tendons, (2) after all losses and (3) ultimate. (Based on BS EN 1992-1-1,
Design of concrete structures – Part 1-1: General rules and rules for buildings, BSI, London,
Figs. 8.16 and 8.17.)
Design rules recognise the fact that the critical shear plane may occur in the prestress
development zone where σcp is not fully developed. It is known that prestressing forces
develop somewhere between linearly and parabolically, although BS EN 1992-1-1,
Figs. 8.16 and 8.17 (combined here in Figure 4.26), adopts a linear development of
stress in service and bi-linear at ultimate. Therefore, a reduced value α1σcp is used up to σpo,
where α1 = lx/lpt2. Known as the ‘critical distance’ to the shear point, lx is measured from
the very end of the unit (not the centre or edge of bearing) to a shear plane drawn at 45°
from the inner edge of the bearing, then lx = lb + yb. lpt2 is the transmission length as follows:
Exp. 8.18 Design transmission length lpt2 = 1.2 lpt (4.66)
Exp. 8.16 Basic transmission length lpt = α1 α2 σpm0 Ф/f bpt (4.67)
8.10.2.2 Transmission length coefficients, α1 = 1.0 for gradual release; α2 = 0.19 for 7-wire
strand, α2 = 0.25 for wire
Ф = diameter of the strand or wire, or averaged if different diameters and of the same type
and same f pk
8.10.2.2. Design tensile strength at transfer fctd(t) = 0.7 fctm(t)/1.5
Exp. 8.15 Bond stress f bpt = 3.2 × 1.0 fctd(t) (4.68)
Recent tests (Elliott 2014) have shown that the code expression 6.4 for αl may be multi-
plied by 1.5 for hollow core units manufactured using the long-line manufacturing method
in which the units are cut to length, rather than cast from a bare end. This comes from
the results of full-scale shear tests together with measurements of lpt in units pretensioned
using 5 and 7 mm wires. No results are available for strands. However, a new bond–slip
relationship for the determination of transfer length lpt in prestressed beams using strain
gauges on strands has been proposed by Park and Cho (2014) followed by a finite element
analysis by Oh et al. (2014). The results indicate that fck, σpm0 and Ф as well as concrete
cover (not included in code equations) are the important parameters in lpt.
Example 4.6
Calculate the uncracked shear capacity V Rd,c for the hollow core unit in Example 4.1.
Nominal bearing length lb = 100 mm. Check the ultimate shear force V Ed at the face of
the support using the loads given in Example 4.5.
Solution
fctd(t) = 0.7 × 2.72/1.5 = 1.27 N/mm 2; f bpt = 3.2 × 1.27 = 4.06 N/mm 2
Average diameter = 11.2 mm; σpm0 = 1169.9 N/mm
lpt2 = 1.2 × 0.19 × 1169.9 × 11.2/4.06 = 736 mm
Ppo = 962.8 kN (from Example 4.1)
Prestress at NA (including γp,fav) γp,fav σcp = 0.9 × 962.8 × 103 × 766/152,000
= 4.37 N/mm2
fctd = 0.7 × 3.80/1.5 = 1.77 N/mm 2
lx = lb + yb = 100 + 99.0 = 199.0 mm
αl = 199.0/736 = 0.27
Ic = 697.0 × 106 mm4, bw = 303 mm, Sc = 4.80 × 106 mm3
V Rd,c = (697.0 x 106 x 303/4.80 x 106) x [√(1.772 + 0.27 x 4.37 x 1.77)] = 100.6 kN
Ultimate shear force
From Example 4.5, wEd = 1.25 × (3.91 + 1.2 × 2.1) + 1.5 × 1.2 × 5.0 = 17.04 kN/m
Shear span = l – lb – 2yb = 8000 – 100 – 2 × 99 = 7702 mm

V Ed = 17.04 × 7.702/2 = 65.6 kN < 100.6 kN OK
End reaction FEd = 17.04 × 8.000/2 = 68.1 kN (see Example 4.8)
4.3.7.2 Shear capacity in the flexurally cracked region V Rd,cr

The shear capacity VRd,c extends to a point where the service moment Ms < cracking moment
Mc = (σb + fctk,0.05/γc) Zb,co according to clause 6.2.2(2). See Figure 4.25. In the flexurally cracked
region, shear resistance is a function of both the concrete and dowel action, as in reinforced
concrete, plus a contribution of the level of compressive stress 0.15 σcp causing shear friction in
the compressive zone. VRd,cr is calculated using the following (semi-empirical) equations:
Exp. 6.2a V Rd,cr = [CRd,c k (100 ρ1 fck)1/3 + 0.15 σcp] bw d (4.69)
6.2.2(1) (4.70)
Steel ratio extends beyond the shear plane ρ1 = Ap /(bw d) ≤ 0.02 (4.71)
C Rd,c = 0.18/1.5 = 0.12
Exp. 6.2b Minimum V Rd,cr = [vmin + 0.15 σcp] bw d (4.72)
Exp. 6.3. vmin = 0.035 k3/2 √fck (4.73)
Example 4.7
Calculate V Rd,cr for the slab in Example 4.1. Determine the cracking moment of resistance
where the flexurally cracked V Rd,cr is used.
Solution
k (≤2.0) = 1 + √(200/159.2) = 2.12, use 2.0
ρ1 (≤0.02) = 766/(303 × 159.2) = 0.0159
Maximum V Rd,cr = [(0.18/1.5) × 2.0 × (100 × 0.0159 × 45)1/3 + 0.15 × 4.37)] × 303 ×
159.2 × 10 –3 = 79.5 kN
vmin = 0.035 × 2.03/2 √45 = 0.66 N/mm 2
Minimum V Rd,cr = [0.66 + 0.15 × 4.37] × 303 × 159.2 × 10 –3 = 63.5 kN
Critical V Rd,cr = 79.5 kN
Zbc,co = 7.244 × 106 mm3, σb = 10.95 N/mm 2 at the support
Use V Rd,cr, where M s > Mc = 7.244 × 106 × (10.95 + 1.77) × 10 –6 = 92.14 kNm
4.3.7.3 Shear capacity of hollow core units due to combined stress in the webs
The calculation for VRd,c has always been carried out on a 45° shear plane at the centroidal
axis, where it is understood shear cracks will first propagate. This may not be the case in
parallel sided webs, for example oval- or pear-shaped cores as in Figures 4.2 and 4.4, where
the combined effects of the transfer of prestress and the principal shear stress may cause
Principles and definitions used in the calculation of ( ) at different heights in the webs of
hollow core units.
shear failure to initiate at a point below the centroidal axis. BS EN 1992-1-1 recognizes
this in clause 6.2.2.(2) “For cross sections where the (web) width varies over the height,
the maximum principal stress may occur on an axis other than the centroidal axis. In such
a case the minimum value of the shear resistance should be found by calculating VRd,c at
various axes in the cross section.” The European Product Standard for hollow core units BS
EN 1168:2005+A3(2011) (BS EN 1168 2005), clause 4.3.3.2.2.1 presents Equations 4.74
through 4.77 to calculate VRd,c at various points up the web along a 35° shear plane as shown
in Figure 4.27. Then VRd,c(y) at distance lx from the end of the unit and at height y from the
bottom of the section, where lx ≥ lb + 0.5h, is given according to clause 4.3.3.2.2.2 as
(4.74)
where the compressive stress at height y due to prestress minus bending at distance lx is
(4.75)
The bending stress term may be ignored, if lx – lb /2 is small. BS EN 1168 does not mention
the favourable effect of prestress γp,fav = 0.9 to be applied to Pt(lx) in Equation 4.75. The con-
crete shear stress γcp(y) at height y due to transmission of prestress at lx is
(4.76)
where Sy = first moment of area above y and about the centroidal axis; Ac(y) = concrete
section area above height y; Pt (lx) = prestressing force at distance lx such that Pt(lx)/ x =
gradient of the prestressing force in the transmission zone, assuming a linear development
of stress (Figure 4.26) = P po /lpt2 ; and C pt (y) = a factor according to position of tendons.
C pt (y) = −1 if y ≤ Y Pt, otherwise = 0. Other definitions are shown in Figure 4.27.
As an alternative to Equation 4.74, the shear capacity at the centroidal axis may be accord-
ing to Equation 4.63 but factored as follows according to clause BS EN 1168, 4.3.3.2.2.3:
= + a g s (4.77)
where 0.9 is a reduction factor referred to the transmission length. BS EN 1168 states that
VRd,c(y) need not be checked at points less than 0.5h from the edge of the support, lx < lb +
0.5h. This clause effectively rules out checking VRd,c at the bottom of the webs, where the
greatest principal stresses actually occur. For example in a 200 mm deep unit with a 40 mm
bottom flange, 100 mm bearing length, lb + 0.5h = 200 mm, whereas lx = 100 + 40/tan35° =
157 mm. In other words, VRd,c is not checked until the height y = (200 – 100) tan35° = 70 mm,
which is only about 29 mm below the centroidal axis. For further information for other
depths of hollow core units, see Elliott and Jolly 2013.
Example 4.8
Calculate V Rd,c at y = 75 mm from the bottom of the unit where the total web breadth is
303 mm for the slab in Example 4.1. See Figure 4.22. Section analysis has found that at
height y: A c(y) = 87,000 mm 2 and Sy = 4.111 × 106 mm3. The end reaction of the slab in
Example 4.5 is FEd = 68.1 kN due to ultimate load wEd = 17.04 kN/m.
Solution
lx at y = 70 mm = 100 + 70 cot 35° = 200 mm
Pt (lx) = 737.5 × 200/736 = 200.4 kN
dPt(lx)/dx = Pt /lpt2 = 737.5 × 103/736 = 1002 N/mm
Yc – yPt = zcp = 58.2 mm
C Pt(y) = 0 because Yc > yPt = 40.8 mm
τcp = 1,002 × (89,000/152,000 − 4.111 × 58.2/697)/303 = 0.76 N/mm 2
M Ed at x = 200 – 100/2 = 150 mm from the centre of bearing = F Edx – wEdx 2 /2 =
68.1 × 0.15 – 17.04 × 0.152 /2 = 10.02 kNm
Yc – y = 99 – 70 = 29.0 mm
σcp(y) = 200,400 (1/152,000 + 29.0 × 58.2/697 × 106) = 1.80 N/mm 2 minus 10.02 ×
106 × 29.0/697 × 106 = 0.42 N/mm 2
V Rd,c(y) = (697.0 × 303/4.111) × [√(1.772 + (0.9 × 1.80 – 0.42) × 1.77) – 0.76] × 10 –3
= 79.0 kN
Using simplified Equation 4.77

V Rd,c(y) = 0.8 × (697.0 × 303/4.80) × [√(1.772 + 0.9 × 0.27 × 4.37 × 1.77)] = 78.9 kN
This is in close agreement with Equations 4.74 through 4.76 and represents a huge 21%
reduction of (100.6 – 79.0)/100.6 on V Rd,c. However, using Equations 4.74 through 4.76 at
the centroidal axis y = Yc = yb = 99 mm and lx = 241 mm on the 35° plane still gives σcp(y) =
1.59 N/mm 2 , γcp = 0.49 N/mm 2 and hence V Rd,c(y) = 87.8 kN, already a 12% reduction on
V Rd,c suggesting some ambiguity in Equations 4.74 through 4.76.
The completion of the design enables us to summarise the imposed load versus clear span
relationships for a range of prestressed hollow core units as shown in Figure 4.28. The limit-
ing criteria will be either service moment of resistance (governed by fctm for durability class
XC1) or deflection for the longest spans (kink in the plot where imposed load = about 3 kN/m 2).
Shear and bearing are unlikely to be critical except for heavily loaded short spans, or for large
point loads located at about 0.25l where V Rd.cr may be critical. The prestressed units used in
the derivation of this graph were pretensioned such that after losses σb = approx 12.5–13.0
N/mm2, with fck = 45 N/mm2, fck(t) = 30 N/mm2 and fpk = 1770 N/mm2 stressed to 70%, cover to
Load vs span graph for prestressed hollow core floor units for domestic loading, exposure XC1
and R90 (except R60 for 150 mm depth) fire resistance.
strands = 30 mm (although 25 mm could have been used except for the problem of longitu-
dinal splitting due to radial stresses beneath strands having too little cover) and lb = 100 mm.
In general, the limiting span/depth is around 40–45 for an imposed live load of 5 kN/m2.
In this context, ‘bearing’ capacity refers to the contact bearing pressure at the bearing ledge.
We make this distinction as certain literature refers to bearing capacity in calculating shear
resistance. It is first necessary to distinguish between ‘isolated’ and ‘non-isolated’ compo-
nents because the bearing allowances differ. Non-isolated components are connected to
other components with a secondary means of support, even if this is only a grouted shear
key. Isolated components rely entirely on their own bearing for total support. BS EN 1992-
1-1 clause 10.9.5.3 states that (1) The nominal length shall be 20 mm greater than for
non-isolated members. (2) If the bearing allows movements in the support, the net bear-
ing length a1 shall be increased to cover possible movements. (3) If a member is tied other
than at the level of its bearing, a1 shall be increased to cover the effect of possible rotation
around the tie.
Where wide slab units are dry-bedded, the full width of the unit is not to be taken as the
bearing length to allow for uneven bearings where daylight gaps are sometimes seen in the
contact plane. The effective breadth b1, clause 10.9.5.2.(3) is taken as the actual breadth if a
bedding material is used, such as mortar, neoprene strips, etc. but not greater than 600 mm
for wide slabs such as hollow core units on dry bearings.
Nominal bearing length lb (termed a in the code) is typically 75–100 mm on concrete,
75 mm on steelwork and 100–150 mm on masonry depending on the quality and thick-
ness of the wall. The net bearing length, the structurally effective part, a1 is less than a to
account for manufacturing and construction tolerances, according to clause 10.9.5.2(1),
Exp. 10.6 as
(4.78)
where the ineffective lengths for line supports are a2 = 0–25 mm {Table 10.3}, a3 = 5 mm
{Table 10.4 for tendons exposed at ends of members}, Δa2 = 10 mm ≤ l/1200 < 30 mm
onto steel or precast or Δa2 = 15 ≤ l/1200 < 40 mm onto brickwork or in situ concrete
{Table 10.5} and Δa3 = ln/2500. For concentrated reactions such as the webs of double-
tee units, a2 = 5–35 mm and a3 = 15 mm. The data are summarised here in Tables 9.2
through 9.4.
The ultimate bearing capacity is
(4.79)
The ultimate bearing stress fRd is according to clause 10.9.5.2(2) for a dry bearing fRd =
0.4 fcd, where fcd = fck/1.5 is the weaker strength. Otherwise fRd = 0.85 fcd, where fcd is the
weaker strength of the slab, support or bedding material, such as elastomeric strip, or
mortar wet bedding on plastic (not steel) levelling shims according to the contractor’s
preference. Note that fRd = 0.3 fcd given in clause 10.9.4.3(3) applies to compressive axial
forces in walls or columns, etc. on dry bearings. Normally, spans up to about 10 m do not
require bedding materials on steel or precast concrete supports. Note that these values are
not applicable to concentrated forces at “nodes,” see Section 9.3.2 and BS EN 1992-1-1,
Section 6.5.4.
The calculation of ‘a’ is iterative procedure because the bearing stress σEd = FEd /a1 b1
is required before a1 is calculated. However, for back calculation to FRd, the length a1 is
already known. There is a danger of entering a loop in the calculation of a1 when σEd /fcd is
close to the limits of 0.15 and 0.4 in Tables 10.2 and 10.3; in this situation, the greater value
for a1 is used. Minimum values of a1 are given in Table 10.2, where ‘line supports’ include
wide slabs such as hollow core, plank, Omnia slab, etc. ‘Ribbed’ floors include double-tee
units, beams in beam & block. ‘Concentrated’ supports include isolated beams in a beam &
composite slab floor system (Figure 4.11).
The procedure is as follows, for example the required nominal bearing length for the
hollow core unit in Example 4.1, spanning 8.0 m and subject to FEd = 68.1 kN, with a dry
bearing onto a precast concrete beam grade C32/40 is as follows:
3.1.6(1) fcd = 32/1.5 = 21.33 N/mm 2 (αcc = 1 for bearing)
10.9.5.2(2) f Rd = 0.4 × 21.33 = 8.53 N/mm 2
10.9.5.2(3) b1 = min{600, 1200} = 600 mm
Table 10.2 Try σEd /fcd = 0.15 − 0.4, then for ‘line’ support a1 ≥ 30 mm
10.9.5.2(1) a1 = 68.1 × 103/(600 × 8.53) = 13.3 mm < 30 mm
Table 10.2 σEd /fcd = (68.1 × 103/600 × 30)/21.33 = 0.17 < 0.40, then a1 = 30 mm OK
Table 10.3 r.c. ≥ C30/37, for ‘line’ support a2 = 10 mm
Table 10.4 Tendons exposed at the end of unit, a3 = 5 mm
Table 10.4 Δa2 = min{8000/1200; 10} = 10 mm
10.9.5.2(1) Δa3 = 8000/2500 = 3.2 mm
Exp. 10.6 Then a = 30 + 10 + 5 + √102 + 3.22 = 55.5 mm < 100 mm provided.
Example 4.9
Calculate the bearing capacity F Rd of the hollow core unit in Example 4.1. Nominal
bearing length a = 100 mm dry bearing onto a cast in situ reinforced concrete beam of
fck = 25 N/mm 2 . The strands in the hollow core unit are exposed at their ends. What is
the minimum allowable value of a.
Solution
fcd = 25/1.5 = 16.66 N/mm 2 for the supporting beam

f Rd = 0.4 × 16.66 = 6.66 N/mm 2
For r.c. < C30/37, for ‘line’ support, try σEd/fcd = 0.15 – 0.4 and a2 = 15 mm; a3 = 5 mm
Δa 2 = min{8000/1200 + 5; 15} = 15 mm; Δa3 = 8000/2500 = 3.2 mm
Then a1 < 100 – 15 – 5 – 15.4 = 64.6 mm
σEd/fcd = (68.1 × 103/600 × 64.6)/16.66 = 0.11 < 0.15, then a2 = 10 mm and so a1 =
69.6 mm > 40 mm maximum in Table 10.2. No further requirement to check σEd/fcd.
b1 = min{600, 1200} = 600 mm
FRd = a1 b1 f Rd = 69.6 × 600 × 6.66 × 10 –3 = 278.3 kN > F Ed = 68.1 kN
To determine minimum a, try a1 = 30 mm
σEd/fcd = (68.1 × 103/600 × 30)/16.66 = 0.23 > 0.15, then a1 = 30 mm and a2 = 15 mm
a ≥ 30 + 15 + 5 + 15.4 = 65.4 mm, but practically use 75 mm.
The depth of the double-tee section is often reduced at the ends of the units to facilitate a recessed
bearing on to the supporting beam or wall (see Figures 4.8 through 4.10). The concrete is heavily
reinforced at this point because it is subjected to shear, bending and bearing. A truss analogy or
the strut and tie method of analysis is used to calculate the forces. Figure 4.29 shows a typical
end shear cage in which the vertical bars are provided for the ultimate shear resistance, and the
horizontal U-bars are for the lateral bursting resistance.
Shear and bursting reinforcement at ends of double-tee units.
Example 4.10
Calculate the bearing capacity FRd and the nominal length a for a double-tee floor
unit of 15.0 m span supported on a precast concrete beam of fck = 32 N/mm2 using a
150 × 150 mm neoprene bearing pad of compressive strength fbed = 10 N/mm2. The
2 no. webs in the double-tee are each 250 mm wide. The strands at the end of the unit
are exposed.
Solution
bw = 250 mm > b1 = 150 mm
For the end bearing design for the web of the unit, as a concentrated bearing see
Section 9.3.2.
10.9.5.2(2) f Rd = 0.85 × 32/1.5 = 18.13 N/mm 2 > f bed = 10 N/mm 2
FRd = a1 b1 f bed = 150 × 150 × 10 x 10 –3 = 225.0 kN
For precast beam C32/40 > C30/37, for ‘concentrated’ support σEd /fcd = (225.0 ×
103/150 × 150)/18.13 = 0.55 > 0.4, then for precast > C30/37 a 2 = 25 mm
For concentrated load with exposed tendons, a3 = 15 mm
Δa 2 = min{15,000/1,200; 10} = 12.5 mm; Δa3 = 15,000/2,500 = 6.0 mm
a ≥ 150 + 25 + 15 + 13.9 = 203.9 mm round off 210 mm
The structural capacity of precast floor units may be increased by adding a layer of
structural reinforced concrete to the top of the unit. Providing that the topping concrete
is fully anchored and bonded to the precast unit, the two concretes – precast and cast
in situ – may be designed as monolithic. The section properties of the precast unit plus
the topping are used to determine the structural performance of the composite floor. A
composite floor may be made using any type of precast unit, but clearly there is more to
be gained from using voided prestressed units, such as hollow core unit or double-tee,
which are lightweight and therefore cheaper to transport and erect than solid reinforced
concrete units. Figure 4.30 shows the details for the most common types of composite
floors.
The minimum thickness of the topping should not be less than 40 mm (50 mm is more
practical). There is no limit to the maximum thickness, although 75–100 mm is a practical
limit. When calculating the average depth of the screed allowances for camber, allowing
span/300 will suffice. The grade of in situ concrete is usually C25 or C30, but there is no
reason why higher strength cannot be used except that the increased strength of the com-
posite floor resulting from the higher grade will not justify the additional costs of materials
and quality control. See Table 2.1 for the concrete data. The topping must be reinforced,
but, as explained later, there need only be tie steel at the interface between the precast and
in situ topping if the design dictates. Mesh reinforcement of a minimum of 0.13% x con-
crete area is the preferred choice – see Section 2.2 for the data.
Composite floor profiles.
The main benefit from composite action is in increased bending resistance and flexural
stiffness – shear and bearing resistance is barely increased. There are however a number of
other reasons why a structural topping may be specified, such as
• To improve vibration, thermal and acoustic performance of the floor

• To provide horizontal diaphragm action (see Chapter 7)
• To provide horizontal stability ties across floors (see Chapter 11)
• To provide a continuous and monolithic floor finish (e.g. where brittle finishes are
applied)
Composite floor design is carried out in two stages, before and after the in situ topping
becomes structural. In prestressed concrete, the transfer stress condition must also be satis-
fied. Therefore, the precast floor unit must carry its own weight plus the self-weight of the
wet in situ concrete (plus a construction traffic allowance of 1.5 kN/m 2). The composite
floor (precast + hardened topping) carries imposed loads. In the final analysis, the stresses
and forces resulting from the two cases (minus the construction traffic allowance, which is
temporary) are additive. Relative shrinkage stresses should be added, see Section 4.4.2.2.
Figure 4.31 shows the bending load capacity for double-tee floors achieved using a 75 mm
thick structural topping. The slab is propped at two points, at one-third and two-third span,
Load vs span graph for composite double-tee slabs, using 75 mm structural topping, propped at
2 points and exposure XC1.
during the construction of the topping. The benefit from the topping decreases as (1) the
span increases because the self-weight of in situ topping nullifies the additional capacity,
and (2) the unit depth increases, because the section modulus of the composite section is
proportionately less.
The following design procedures are used.
4.4.2.1 Serviceability state of stress

Permissible service stresses are checked at two stages of loading – Stages I and 2 before and
after hardening of the in situ topping as follows:
• Stage 1 for the self-weight of the precast slab plus the self-weight of in situ concrete
screed, plus an allowance for temporary construction traffic of up to 1.5 kN/m 2 .
The section properties of the precast unit alone are Zb and Zt for stress analysis and
Zb,co for moment of resistance.
• Stage 2 for superimposed loading. The section properties of the composite section
Zbc,co and Icc,co for deflections are used (See Section 4.3.3 Step 5 for definitions).
Stage 1: Referring to Figure 4.32, a precast prestressed unit has the final prestress after
losses of σb and σt according to Equations 4.32 and 4.33. It is likely that σt will be negative =
tension. Let Ms1 be the maximum service bending moment due to the stage 1 load. The flexural
stresses at the bottom and top of the precast floor unit must first satisfy
f b1 = −M s1/Zb,co + σb < +0.45 and > −fctm (4.80)
f t1 = +M s1/Zt,co + σt > − and < +0.45 fck (4.81)
Principles of serviceability stress limitations for prestressed concrete composite elements.
(the most likely condition in bold) where Zb,co and Zt,co are the stage 1 section compound
modulii for the precast unit alone. Of course, the transfer condition must be checked first.
Stage 2: Let M s2 be the maximum service bending moment due to the Stage 2 load. The flex-
ural stresses at the bottom and top of the precast unit and at the top of the composite floor
are derived from the composite section as follows:
f b2 = +M s2 /Zbc,co (4.82)
f t2 = −M s2 /Ztc,co (4.83)
(4.84)
where Zbc,co, Ztc,co and Z′tc,co are the Stage 2 composite compound section modulii for the
composite section at the bottom of the precast unit, the top of the precast unit and the top
of the topping, respectively.
To calculate these values, the second moment of area Icc,co of the composite section must
be determined using the modular ratio approach for fcki and Ecmi of the in situ topping.
This is to allow for the different Young’s modulus of the precast and in situ concrete, typi-
cally (25/50)0.3 = 0.81. The breadth of the in situ topping is reduced in the proportion to
Ecm such that the effective breadth of the topping is as follows:
beff = b Ecmi /Ecm (4.85)
Note that in this case, b = full breadth of the precast unit (not the manufactured top breadth).
Then, the stresses from prestressing and stage 1 loads are added to those from Stage 2
loads and must satisfy
f b = σb − M s1/Zb,co – M s2 /Zbc,co > - (4.86)
f t = σt + M s1/Zt,co + M s2 /Ztc,co < 0.45 fck (4.87)
f t′ = + M s2 /Z′tc,co < 0.567 fcki (4.88)
where in Equation 4.87 M s2 is the quasi-permanent load and is therefore rarely critical.
Equation 4.88 is rarely critical and does not affect the design of the precast unit.
The critical situation nearly always occurs in the bottom of the slab because the new
position of the neutral axis is often close to the top of the precast unit. Equation 4.86 can
be written as
M sR2 = M s2 > (fctm + σb) Zbc,co − [M s1 (Zbc,co/Zb,co)] (4.89)
However, the situation at the top of the precast unit should also be checked for completeness,
as follows:
M sR2 = M s2 > (fctm − σt) Ztc,co − [M s1 (Ztc,co/Zt,co)] (4.90)
Total M sR = minimum M sR2 + M s1 (4.91)
The construction traffic loading needs to only be considered as part of M s1 when checking
that the service stresses do not exceed 0.45 fck and -fctm. It does not have to be included in
the final calculation in Equation 4.89 because it is not a permanent load (nor does if affect
the ultimate limit state).
It is not possible to prepare in advance standard calculation for composite floor slabs. This
is because the final stresses in Equations 4.86 and 4.87 depend on the respective magnitudes
of the stage 1 and Stage 2 loads and moments, that is the same precast concrete unit may
have different load-bearing capacity when used under different conditions. The following
examples will show this.
Example 4.11
Calculate the Stage 2 service bending moment M sR2 that is available if the hollow core
unit in Example 4.1 has a 75 mm minimum thickness structural topping. The top surface
of the topping is power floated (as shown in Figure 4.33) so that floor finishes can be laid
directly. Therefore, floor finishes, ceiling and services may be taken as 0.4 + 0.6 = 1.0
kN/m 2 . The floor is simply supported over an effective span of 8.0 m. The precamber of
the hollow core unit at installation may be taken from Example 4.5 as −15 mm. Use fck
for the topping = 25 N/mm 2 . Self-weight of in situ concrete (including mesh) = 25 kN/m3.
What is the maximum imposed service loading?
Power floating the structural topping to produce a finished floor.
Before commencing the solution, a technical note that the creep loss is smaller because
of the tensile bending stress due to self-weight {−M self /Zz} in Equation 4.17, which serves
to reduce the compressive strain causing creep, now includes the self-weight of the top-
ping. As a result, σb increases slightly from 11.51 to 11.594 N/mm 2 .
Solution
Section properties
Young’s modulus for topping Ecmi = 22 × (25 + 8)0.3 = 31,476 N/mm 2

Self-weight of topping = 25 × [75 + (50% camber at install) 0.5 × 15.0] × 10 –3 =
2.06 kN/m 2
m in situ/precast = 31,476/36,283 = 0.868
beff (serviceability) = 1200 × 0.868 = 1041 mm
ybT = 200 + 0.5 × 75 = 237.5 mm
Acc,co = 155,351 + 1,041 × 75 = 233,426 mm 2
ybc,co = [155,351 × 99.0] + [1,041 × 75 × 237.5]/233,426 = 144.5 mm
Icc,co = 1759.6 × 106 mm4
Zbc,co = 1759.6 × 106/144.5 = 12.178 × 106 mm3
Ztc,co = 1759.6 × 106/(200 – 144.5) = 31.698 × 106 mm3
Then Zbc,co /Zb,co = 12.178/7.244 = 1.681
Stage 1 stresses
Self-weight of hollow core unit w 0 = 3.91 kN/m
Self-weight of topping w 1 = 2.06 × 1.2 = 2.47 kN/m for 1.2 m wide unit
w 0 + w 1 = 3.91 + 2.47 = 6.38 kN/m; M s1 = 6.38 × 8.0002 /8 = 51.1 kNm
f b1 = 51.1/7.244 = −7.05 N/mm 2 tension
f t1 = 51.1/6.925 = +7.37 N/mm 2 compression
Stage 2 stresses
After topping, w2 = 1.2 × 1.0 = 1.2 kN/m; w3 = 1.2 × 5.0 = 6.0 kN/m; Ms2 = 57.6 kNm
f b2 = 57.6/12.178 = −4.73 N/mm 2 tension
f t2 = 57.6/31.698 = +1.82 N/mm 2 compression
Check at the top of topping, f′t2 = 57.6/13.48 = +4.27 N/mm2 < 0.45 × 25 = 11.25
N/mm2.
Final stresses
Bottom = 11.594 + −7.05 + −4.73 = −0.184 N/mm 2 > fctm = −3.80 N/mm 2
Top = −1.447 + 7.37 + 1.82 = +7.744 N/mm 2 < 0.45 fck = 20.25 N/mm 2
Therefore, additional moment capacity available due to bottom stress = −0.184 −
(−3.80) = +3.616 N/mm 2
M sR – M 2 = 3.616 × 12.178 × 106 x 10 –6 = 44.0 kNm
At the top, MsR – M2 = (20.25 – 7.744) × 31.698 × 106 × 10–6 = 396.4 kNm, clearly
not critical
Alternatively Equations 4.89 and 4.91, M sR = (3.80 + 11.59) × 12.178 – 51.1 ×
1.681 + 51.1 = 101.5 + 51.1 = 152.6 kNm
M sR – (M s1 + M s2) = 152.6 – (51.1 + 57.6) = 43.9 kNm (the same result after
rounding off)
The allowable imposed load = 8 × 101.5/8.02 = 12.69 kN/m = 10.57 kN/m2 > 6.0 kN/m2
Note that the total M sR = 152.6 > 110.9 kNm is 38% greater than that for the basic unit
in Example 4.1. Using the original design loading, the maximum effective span for this
composite slab could be increased from 8.00 to 9.20 m, or using the original effective
span the imposed live load could be increased from 5.00 to 9.58 kN/m 2 . In both cases,
the service resistance would be critical over the ultimate limit states of bending and shear.
4.4.2.2 Relative shrinkage between in situ topping to precast unit

In calculating service stress and deflections, the effects of the relative shrinkage of the topping
to that of the precast unit should be added to those resulting from loads (and prestressing
if applicable). Shrinkage-induced curvature is according to clause 7.4.3(6). Unlike BS 8110,
which recommended relative shrinkage = 100 × 10–6 between precast and in situ concrete,
EN 1992-1-1 does not give a value preferring to calculate the shrinkages of the in situ and
precast separately, = εshi − εsh. EN 1992-1-1, clause 2.2.2 states that shrinkage (and creep)
should ‘generally’ be taken into account in the serviceability limit state and should only be
considered at ultimate where their effects are significant, for example in the verification of
ultimate limit states of stability where second-order effects are important, which rules out
floor slabs. If the topping is cast in bays, possibly a few days apart, the shrinkage effect will
be considerably reduced due to the localised curvatures beneath each pour.
The effect of relative movement between precast and in situ concrete is well documented
in text books and fib documents [e.g. fib 1982, Bhatt 2011] and may be summarised as
follows:
Free shrinkage stress = Ecmi
Restoring force = Aci = Ecmi Aci (4.92)
where Ecmi Aci is the axial stiffness of the in situ concrete topping of area Aci
Lever arm from the centroid of topping (depth t) and composite section a = hs + 0.5t − ybc,co
Restoring stresses at the top of the precast slab σsh,t and bottom σsh,b are (compression
positive):
σsh,t = + /Acc,co + a/Ztc,co (4.93)
σsh,b = + /Acc,co − a/Zbc,co (4.94)
For example assuming that the topping is cast in a single pour, and subjecting the compos-
ite slab in Example 4.11 to a relative shrinkage strain = 150 × 10 –6. = 150 × 10 –6 ×
31,476 × 75,000 × 10 = 354.1 kN. a = 237.5 – 144.5 = 93.0 mm.
–3
σsh,t = 354.1 × 103 × (1/233,426) + (93.0/31.698 × 106) = +1.51 + 1.04 = +2.55 N/mm 2
σsh,b = 354.1 × 103 × (1/233,426) − (93.0/12.178 × 106) = +1.51 – 2.70 = −1.19 N/mm 2
In effect, the shrinkage-induced stress reduces M sR by 1.19 × 12.178 = 14.5 kNm. In the long
term, the stress reduces due to the exponential creep growth factor ψ = 1 – e−φ, for example
φ = 2.0, ψ = 0.86.
4.4.2.3 Ultimate limit state of bending

The design at ultimate limit state is also a two-stage process, with the flexural stresses
resulting from the self-weight of the precast element plus any wet in situ concrete being
carried by the precast unit alone. The lever arm is the same as in a non-composite design,
that is d. The method is to calculate the area of steel Ap1 required in stage 1, and to add the
area Ap2 required in Stage 2 using an increased lever arm. See Figure 4.34a. The effect of the
structural topping is to increase the lever arm to the steel reinforcement by an amount equal
to the thickness of the screed hs, proving that the depth to the neutral axis is less than hs. See
Figure 4.34b. In Stage 2, the effective breadth of the topping is
bef = b × fcki (in situ)/fck (precast) (4.95)
where b is the full breadth of the precast unit, not the manufactured breadth (note beff is used
for effective breadth in service).
Most design engineers choose not to separate the design into two stages, using the com-
posite section properties alone. This was a satisfactory approach in BS 8110 where the
plateau in the stress–strain relationship meant that the ultimate stress f p was not dependent
on the design strain εp. However, the inclined branch of BS EN 1992-1-1, Fig. 3.10 (Figure
4.23b) means that f p is still increasing with εp. Adopting a two-stage approach, Figure 4.34a,
equilibrium in the section due to stage 1 moment M Ed1 is determined first, followed by the
compatibility and equilibrium due to Stage 2 moment M Ed2 as follows.
Stage 1. M Ed1 = due to self-weight, topping, propping (and any trimmer angle dead loads
acting as a point load).
K1 = M Ed1/fck(t) b1 d12 (4.96)
where b1 = breadth of precast unit, d1 = effective depth from the top of precast unit
Principles of ultimate strength for prestressed concrete composite elements. (a) stage 1 in
precast unit, (b) stage 2 in composite slab.
fck(t) = strength of precast concrete at installation at t days < 28 days, but not less than
fck(t) at transfer
Exp. 3.2 fcm(t) = exp{s [1− √(28/t)]} fcm (4.97)
fcm = fck + 8 N/mm 2

fck(t) = exp{s[1− √(28/t)]}(fck + 8) – 8 (units of N/mm 2 only)
Cement s = 0.2 (Class R), 0.25 (Class N) or 0.38 (Class S)
z1 = [0.5 + (0.25 – K1/1.134] d1 ≤ 0.95 d1
x1 = (d1 – z1)/0.4
If 0.8x1 ≤ hft then Ap1 = M Ed1/z1 f p1
Fig. 3.10 where f p1 = stress in tendons at strain εs = 0.02

If 0.8 x1 > hft compression is in webs of width bw
Restart analysis using K1w = M Ed1/fck(t) bw d12 (4.98)
z1 = (0.5 + (0.25 – [K1w /1.134 – β]) d1
where β = (b1 – bw) hft (d1 – 0.5 hft)/(2 bw d12) (4.99)
Then Ap2 = Ap – Ap1
The composite M Rd,c2 is calculated according to Section 4.3.5 but using Ap2 instead of Ap
Total M Rd,c = M Rd,c2 + M Ed1
It is seen that there is a difficulty in calculating standard values for MRd for specific units
because the stage 1 moments and area of steel must be first known. As these are span-
dependent, the superimposed moment capacity MRd,c2 is a function of span and stage 1
loads.
Example 4.12
Calculate the ultimate bending moment of resistance M Rd,c in Example 4.11. In this
example only, let the age at installation = 7 days. What is the maximum ultimate imposed
load? Is the composite slab critical at service or at ultimate?
Solution
Effective depth in hollow core unit d1 = 159.2 mm and b1 = 1154 mm
Effective depth in composite section d2 = 159.2 + 75 = 234.2 mm
Consider stage 1 moment acting on precast alone at installation at ti = 7 days
fck(ti) = [38 e0.2 (1 − √(28/7))] − 8 = 35.4 N/mm 2
M Ed1 = 1.25 × 51.1 = 63.8 kNm where γG = 1.25 from Exp 6.10b
K1 = 63.8 × 106/(35.4 × 1154 × 159.22) = 0.062
z1 = 150.0 mm
Ultimate stress f p is not known until the Stage 2 M Rd is found, therefore taking the limit-
ing strain as 0.02 then f p1 = 1516 N/mm 2
Ap1 = 63.8 × 106/(150.0 × 1516) = 280 mm 2
Ap2 = 766 − 280 = 486 mm 2
εpo = 0.00494, εp = 0.00494 + 0.0035 (234.2/X – 1) = 0.001437 + 0.820/X Equation 1
Effective breadth of topping bef = 1200 × 25/45 = 667 mm
Fc = 0.567 × 45 × 667 × 0.8 X = 13,600 X; Fs = 486 f p
X/f p = 486/13,600 = 0.0357 Equation 2
Then X = 53.0 mm < 75 mm depth of topping
dn2 = 0.4 × 53.0 = 21.2 mm
z 2 = 234.2 − 21.2 = 213.0 mm
From Equation 1, εp = 0.016892
From Equation 3, f p = 1485 N/mm 2
M Rd,c = f p Ap2 z 2 + M Ed1 = 1485 × 486 × 213.0 × 10 –6 + 63.8 = 217.4 kNm
Stage 2 M Rd,c2 = 217.4 −63.8 = 153.6 kNm
Ultimate imposed load wEd = 8 × 153.6/8.02 = 19.2 kN/m = 16.0 kN/m2 > 6.0 kN/m2
Refer to Example 4.11. Ratio of ultimate/service resistance = 217.4/152.6 = 1.424. As
this value is less than γQ = 1.5 and greater than γG = 1.25, the composite slab may be
critical under either limit states. In this loading case, γ = 1.360 < 1.424; therefore, the
composite slab is just critical at service. Using the original design loading, the maximum
effective span for this composite slab could be increased from 8.00 to 9.51 m, or using the
original effective span the imposed live load could be increased from 5.00 to 9.83 kN/m 2 .
However, in both cases, the service limit is first critical.
4.4.2.4 Ultimate shear capacity of composite sections

The design for ultimate shear capacity for composite hollow core slabs is defined in BS EN
1168 as a two-stage design but is not mentioned as such in BS EN 1992-1-1. It is clearly
technically correct to deduct the elastic shear stress vEd1 = V Ed1 S c /bw Ic consumed in stage 1
loading before the topping hardens, but it highlights the paradox that V Rd,cc is an ultimate
strength based on elastic analysis. However, there is some ductility if the shear-tension fail-
ure mode is critical (as is the case of UDL) even in hollow core units with narrow webs, and
so adopting this method is valid. Referring to BS EN 1168, Annex F:
Clause F2.2. (4.100)
where
V Ed1 is the ultimate dead load acting on the unit, including self-weight of wet topping
Sc Ic is the first and second moments of area of the non-composite section
Scc is the first moment of area for the composite section above the centroidal axis of the
composite section
Scc = beff t (h + 0.5t − yb,c) + b (h − yb,c)2 /2
where beff is the elastic effective breadth (not ultimate bef)

If the centroidal axis is below top flange, yb,c < h – hft subtract for the webs
Scc = beff t (h + 0.5t − yb,c) + b (h − yb,c)2 /2 − (b − bw) (h − hft − yb,c)2 /2
6.2.2(2) Use V Rd,cc where M s = M s1 + M s2 < composite cracking moment Mcr,c for Stage 2
loading as follows. Section is racked at ultimate when
(4.101)
That is Mcr,c varies along the span according to stage 1 service moment, including the effects
of propping. Where M s1 = 0, Mcr,c = (σb + fctd) Zbc,co.
For the ultimate cracked shear capacity V Rd,cr,c:
Clause F4. The same equations as non-composite units except composite effective depth
d + t replaces d.
Example 4.13
Calculate the ultimate shear capacity V Rd,cc and V Rd,crc in Examples 4.11 and 4.12. To
remind the reader, for the non-composite section V Rd,c = (697.0 × 106 × 303/4.80 × 106) ×
[√(1.772 + 0.27 × 4.37 × 1.77)] = 100.6 kN. Derive the expression from Equation 4.100
for the conditions where V Rd,cc and V Rd,crc are used in the span.
Solution
Uncracked capacity V Rd,cc
Sc = 4.80 × 106 mm3, Ic (not compound) = 697.0 × 106 mm4
V Ed1 at shear plane = 1.25 × 6.38 × 7.702/2 = 30.72 kN
where to c.g. shear plane lv = 8.000 – 2 × 0.099 – 0.100 = 7.702 m
vEd1 = 30,725 × 4.800 × 106/(303 × 697 × 106) = 0.70 N/mm 2
Scc = 8.801 × 106 mm3, Icc = 1723.0 × 106 mm4
V Rd2 = (1723.0 × 106 × 303/8.801 × 106) × [√(1.772 + 0.270 × 4.37 × 1.77) − 0.70] =
94.20 kN
V Rd,cc = 94.20 + 30.72 = 124.92 kN
Cracked capacity V Rd,crc

d + t = 234.2 mm (Example 4.12), k = 1 + √(200/234.2) = 1.92
vc,min = 0.035 × 1.921.5 × √45 = 0.63 N/mm 2
ρ = 766/(303 × 234.2) = 0.0108
Maximum VRd,crc = [(0.18/1.5) × 1.92 × (100 × 0.0108 × 451/3 + 0.15 × 4.37)] × 303 ×
234.2 × 10–3 = 106.3 kN
Minimum V Rd,crc = (0.63 + 0.15 × 4.37) × 303 × 234.2 × 10 –3 = 90.9 kN
Use VRd,cc = 124.92 kN when Ms < Mcr,c = (10.95 + 1.77) × 12.177 − Ms1 [(12.177/7.244) − 1] =
154.9 – 0.68 Ms1 kNm, otherwise VRd,crc = 106.3 kN. The conditions exist when x = 2.85 m
from the centre of support, where Ms = 128.7 kNm and Ms1 = 39.5 kNm.
Propping is a technique which is used to increase the service moment capacity by reversing
the stage 1 stresses particularly at mid-span. This is achieved by placing a rigidly founded sup-
port, load bearing props, such as Peri or similar, in the desired place whilst the in situ concrete
topping is hardening. See Figures 4.35 and 4.36. To ensure that the props are always effective,
many contractors prefer to use two props rather than one – just in case the foundation to one
of the props is ‘soft.’ However, the following analysis will consider a single mid-span propped
floor slab. The reader can easily extend the same analysis to multiple props.
The benefit derives from the fact that the stage 1 moments due to the self-weight of the wet
concrete screed are determined over a continuous double span, each of leff /2. When the props
are removed, the prop reaction R creates a new moment which is carried by the composite
section. Finally, the superimposed loads are added as shown in Figure 4.36.
When rigidly secured, the prop reaction is R = 0.625 w1 leff, where w1 = self-weight of the wet
in situ topping (allowing for precamber of the slab) and leff = effective span of slab. The negative
moment at mid-span due to the prop M = −0.625 w1 leff × leff/4 = −5w leff l2/32. The additional
Stage 2 moment following the removal of the prop is M = +5w1 leff 2/32. Equation 4.89 is there-
fore modified as
M sR2 = M 2 > (fctm + σb) Zbc,co − [M s1 (Zbc,co /Zb,co)] – 5 w 1 leff 2 /32 (4.102)
For two props positioned at one-third and two-third span:
M sR2 = M 2 > (fctm + σb) Zbc,co − [M s1 (Zbc,co /Zb,co)] – 11 w 1 leff 2 /90 (4.103)
Propping of prestressed floor slab. (Courtesy of Bison Manufacturing, Swadlincote, UK.)
Bending moments resulting from propping.
When prop(s) is removed M 2 = + 5w 1 leff/32 or +11 w 1 leff/90, respectively, plus all imposed
dead and live loads.
The economical and practical benefits of propping wide slabs such as hollow core unit and
double-tees should be carefully considered. Propping can be quite expensive and may slow
down site erection rates.
Example 4.14
Repeat Example 4.11 with the hollow core unit propped at mid-span. The effective span is
now increased to 9.50 m to reflect the increased service moment capacity due to propping.
The increased length of slab has also reduced creep losses such that σb = 11.86 N/mm2 and
σt = −1.48 N/mm2
Solution
Stage 1
M s1 = 6.38 × 9.5002 /8 = 71.97 kNm minus moment (due to propping) = 5 × 2.47 ×

9.5002 /32 = −34.86 kNm
M s1 = 37.1 kNm
f b1 = 37.1/7.244 = −5.12 N/mm 2; f t1 = 37.1/6.925 = +5.36 N/mm 2
Stage 2. Prop removed
M s2 = 34.86 + 7.2 × 9.5002 /8 = 34.86 + 81.23 = 116.1 kNm
f b2 = 116.1/12.178 = −9.54 N/mm 2; f t2 = 116.1/31.698 = +3.66 N/mm 2
Diagram of service and bending moments after propping.
Final stresses
f b = 11.86 + −5.12 + −9.54 = −2.80 N/mm 2 > 3.80 N/mm 2
f t = −1.48 + 5.36 + 3.66 = +7.54 N/mm 2 < 20.25 N/mm 2
Additional moment of resistance due to propping MsR – Ms = (3.80 – 2.80) × 12.178 ×
106 × 10 –6 = 12.18 kNm
M sR = 71.97 + 81.23 + 12.18 = 165.4 kNm

M s2 = 81.23 + 12.18 = 93.41 kNm
Imposed service load in Stage 2, ws2 = 8 × 93.41/9.502 = 8.28 kN/m = 6.90 kN/m 2
This is the situation at mid-span. However, the negative bending moment due to prop-
ping (as shown in Figure 4.36) causes a shift in the position of critical stress from mid-span
x/leff = 0.5 to about 0.47. At this point, ws2 = 6.77 kN/m2 , a reduction of only 1.9% compared
to the mid-span calculation, but as the depth of topping increases relative to the depth of
the prestressed unit, this difference is larger, for example if the topping depth increases to
150 mm, the critical stress is at x/L = 0.43 and difference in ws2 is 5.1% as shown in Figure
4.37. The reader should note that although propping increases MsR,c, it does not increase MRd,c
(except for a small increase due to lower M Ed1) or V Rd,c, and as M Rd,c becomes critical, the full
extent of propping is not advantageous.
4.4.3.1 Deflections in composite slabs

The calculation of deflections in composite slabs differs from that in noncomposite units
in the following ways: the second moment of area after the topping has hardened is the
transformed composite compound value Icc,co; creep coefficients for viscoelastic movement
are taken after the topping has hardened; and the effect of propping is to eliminate stage 1
deflection due to the topping (zero deflection at the actual position of the prop and small
deflections between the props and the supports).
Transfer deflections δ1 and δ2, and deflections at installation due to camber and self-weight δ3
and δ4 are as for the non-composite unit using Equations 4.48 through 4.52. After installation,
the static deflection due to the self-weight of topping w1:
δ5 = +5 w 1 leff4/384 Ecm Ic,co (4.104)
The equivalent upward deflection due to single propping at mid-span of the wet in situ
topping. The prop reaction force is 0.625 w1 leff, then:
δ6 = −0.625 w 1 leff4/48 Ecm Ic,co (4.105)
Ditto due to twin propping at ⅓ and ⅔ points due to reaction forces 11 w 1 leff/90, then:
δ6 = −w 1 leff4/76.84 Ecm Ic,co (4.106)
Resultant deflection at installation if propped = δ3 + δ4 + δ5 + δ6

In the long term, imposed dead loads are taken as being applied immediately after instal-
lation at ti days). The creep coefficient for the in situ concrete topping is 0.8ψ∞ = 2.0 (because
in situ concrete is fresh at installation). This controls the long-term deflection for imposed
dead and live loads after the topping has hardened. Note that Ecm for the precast is used, not
Ecmi because Icc,co is the transformed section based on Ecm.
Long-term camber is due to the camber at installation, plus further viscoelastic movement
based on Icc,co minus a reduction due to the reduction in pretensioning force from installa-
tion Ppmi to the long-term Ppo
δ7 = δ3 – [ψ28 Ppmi − (Ppmi – Ppo)] zcp leff 2 /8 Ecm Icc,co (4.107)
Long-term deflection due to self-weight and infill of the slab w 0 is due to the previous deflec-
tion δ4 plus further viscoelastic movement, and the static and viscoelastic deflections due
to topping w 1, using a net creep factor ψ∞,ti = 0.8ψ∞ − ψ1 after installation at time t i. But,
because the former is a function of Icc,co and the latter a function of Ic,co before hardening of
the topping and then Icc,co afterwards, this particular creep coefficient is multiplied by the
ratio of the respective moments of area Ic,co /Icc,co:
ψ∞,ti = (0.8 ψ∞ − ψ1) Ic,co /Icc,co (4.108)
δ8 = δ4 + [5 (ψ∞,ti w 0 + (1 + ψ∞,ti) w 1) leff4/384 Ecm Ic,co] (4.109)
Deflection due to imposed dead w 2 and quasi-permanent live load ψ2 w 3:
δ9 = + 5 [(1 + ψ28) w 2 + (1 + 0.8 ψ∞) ψ2 w 3] lef4/384 Ecm Icc,co (4.110)
of which live load only
δ10 = + 5 (1 + 0.8 ψ∞) ψ2 w 3 leff4/384 Ecm Icc,co (4.111)
Overall long-term movement due to live load, visco self-weight and topping using ψ28 and
dead loads using ψ∞,t, and changes in camber:
δ11 = δ10 + 5 [(w 0 + w 1) ψ28 + w 2 0.8 ψ∞] leff4/384 Ecm Icc,co + (δ7 − δ3) (4.112)
There is no effect of creep due to the removal of the prop, that is instantaneous downward
deflection, acting on composite section is δ12 = + 0.625 w 1 leff4/48 Ecm Icc,co and ditto for twin
props δ12 = + w 1 leff4/76.84 Ecm Icc,co.
Example 4.15
Calculate the long-term deflection and movement after installation for the composite hol-
low core unit in Example 4.13 propped at mid-span.
Solution
From Examples 4.5 and 4.13, Ppm0 = 896.1 kN, Ppmi = 862.3 kN and Ppo = 737.5 kN.
l = 9500 mm, Ecm(t) = 32,837 N/mm2, Ecm = 36,283 N/mm2, zcp = 58.2 mm, Ic,co = 708.1 × 106
and Icc,co = 1759.6 × 106 mm4. Self-weight slab and topping = 3.91 and 2.475 kN/m.
Transfer: Net deflection (pro-rata Example 4.5) δ1 + δ2 = −25.3 + 17.0 = −8.3 mm
Installation: Net deflection (pro-rata Example 4.5) δ3 + δ4 = -42.4 + 29.1 = −13.3 mm
δ5 + δ6 at mid-span = zero. The deflection due to the topping is eliminated by propping,
except that at one-fourth span the net deflection of self-weight and prop reaction = +0.28 mm.
Long-term final deflection
ψ1 = 0.71; ψ28 = 0.8 × 2.5 × (1.0 – 0.4) = 1.20

ψ∞,t = (0.8 × 2.5 − 0.71) × 708.1 × 106/1759.6 × 106 = 1.29 × 0.402 = 0.521
δ7 = 42.4 − [(1.20 × 862,263) − (862,263 − 722,114)] × 58.2 × 9.5002/(8 × 36,283 ×
1,759.6 × 106) = −51.6 mm
δ8 = 29.1 + 5 × [0.521 × 3.91 + (1+ 0.521) × 2.47] × 9,5004/(384 × 36,283 × 708.1 ×
106) = +53.0 mm
Imposed dead δ9 = 5 × (1 + 1.20) × 1.20 × 9,5004/(384 × 36,283 × 1,759.6 × 106) =
+4.4 mm
Imposed quasi-permanent live δ10 = 5 × (1 + 2.00) × (0.3 × 6.00) × 9,5004/(384 ×
36,283 × 1,759.6 × 106) = +9.0 mm
Net deflection due to addition and removal of the prop δ6 + δ12 = (0.625 × 2.48 ×
9,5004/48 × 36,283 × 708.1 × 106) × −1 +708.1 × 106/1759.6 × 106) = −11.0 + 4.4 =
−6.6 mm
Final deflection δ7 + δ8 + δ9 + δ10 +(δ6 + δ12) = +8.2 mm
Movement after installation and after the prop is removed
Visco camber δ7 + δ3 = − (51.6 – 42.4) = −9.2 mm
Visco self-weight, topping, imposed dead and live* = 5 × [1.29 × 6.38 + 1.20 × 1.20 +
2.20* × 1.80] × 9,5004/(384 × 36,283 × 1,759.6 × 106) = 13.7 + 2.4 + 6.6 = +22.7 mm
Movement after installation = +13.5 mm
Under the action of vertical flexural shear, the horizontal interface between the precast unit
and in situ topping will be subjected to a horizontal shear force vEdi. Figure 4.38. Often
termed ‘shear flow,’ because it is measured in force per linear length, this shear derives from
the equilibrium of the vertical shear at a section. It is the result of imposed loads present only
after the in situ concrete topping has hardened. The distribution of interface shear is identical
to the imposed shear force distribution and must therefore be checked at all critical sections.
Interface shear need only be checked for the ultimate limit state. The design method is based
on partly experimental evidence for the empirical terms in BS EN 1992-1-1, clause 6.2.5(1)
and Exp. 6.25 and will ensure that serviceability conditions are satisfied.
* See commentary to Equations 4.60 and 4.61.
Shear slip must be

restored by interface force
Zero interface shear
stress at mid span
Unbonded
interface
In situ topping
Precast unit
(a) Interface shear links
Fc,t
z2 vEdi
VEd at support
(b)
(a) Interface shear stress and forces in composite elements. (b). Interface shear stress. (Based
on BS EN 1992-1-1, Design of concrete structures – Part 1-1: General rules and rules for build-
ings, BSI, London, Figs. 8.16 and 8.17.)
Providing that the in situ topping is fully bonded to the precast unit, full interaction is
assumed, that is there is no relative slippage between the two concretes. The horizontal shear
force Fc,t at the interface is equal to the total force in the in situ topping due to imposed loads.
It is therefore necessary to have carried out a calculation for the ultimate limit state in flexure
and to have determined the depth to the neutral axis xc before this is attempted. This is a
benefit from having carried out a two-stage approach to ultimate flexure.
The section is flexurally uncracked at the support where the interface shear is a maximum.
Therefore only Stage 2 loads, after hardening of the topping, contribute to vEdi. Let MEd2 =
maximum ultimate moment due to Stage 2 loads, including dead and live point and line loads:
Fc = 0.567 fcki bi 0.8 xc
where xc = (dc – z 2)/0.4 < 0.6dc
dc = effective depth from the top of composite section
If 0.8Xc > topping thickness t, Fc,t = 0.567 fcki b t (4.113)

otherwise Fc,t = 0.567 fcki b 0.8 Xc (4.114)
Fc,t only acts at the point of maximum bending moment – elsewhere it is less than this and
may even change sign in a continuous floor. Therefore, the distance over which this force
is distributed along the span of the floor is taken as the distance from the maximum to the
zero or minimum moment.
The average ultimate shear stresses at the interface may be calculated according to
Figure 4.38b for ultimate interface force Fc,t due to a change in moment ∂M over an incre-
mental length Δx:
¶ ¶
= D ; but = then = D
¶ ¶
Stress = force/contact area = vEdi = Fc,t /bi Δx = V Ed /bi z 2
6.2.5(1) (4.115)
where
Fc,t is the compression in topping due to ultimate Stage 2 loading
Fc is the total compression in topping and precast unit due to ultimate Stage 2 loading
V Ed is the ultimate vertical shear force at support (not on shear plane) due to loads after
the topping has hardened
z 2 is the lever arm of rebars-topping due to Stage 2 loads
bi is the width of interface = 1200 mm (or breadth of unit)
vEdi is then distributed in accordance with the magnitude of the vertical shear V Ed at
any section
The design interface shear stress vEdi is first checked against the cohesive and frictional shear
resistance according to BS EN 1992-1-1, Exp. 6.25:
6.2.5 (1) vRdi = c fctd + μ σn (4.116)
where σn = normal stress = 0 (the vertical pressure of imposed load and self-weight of topping
is ignored).
3.1.6.2(P) Design in situ tensile strength fctdi = 0.7 × 0.3 fcki2/3/1.5
6.2.5(2) c = 0.025–0.1, μ = 0.5 (‘very smooth’ against a steel, plastic or prepared

wooden mould)
c = 0.2, μ = 0.6 (‘smooth’ after casting, slipformed or extruded)
c = 0.4, μ = 0.7 (‘rough’, > 3 mm rough ridges at 40 mm spacing by raking etc.)
If vEdi is still greater than vRdi, then shear reinforcement is added (per 1 m run) projecting
from the precast unit into the structural topping:
6.2.5(1) vRdi = c fctd + μ σn + μ A s f yd /Aj (4.117)
where
A s is the area interface shear reinforcement*
Aj is the area of interface joint
* Fig. 6.10. Distribution of A s should respond to the distribution of VEd.
Mild steel
bent loop Approx. 25 mm
projection
Embedment to
half slab depth
Interface shear links (or loops or dowels) in joints between hollow core floor units.
If the normal pressure σn is zero, the amount of steel required is
æ ö
=ç - ÷ (4.118)
è øm
but not less than 0.13% or 0.26 fctmi /f yk of the contact area as the bars are required for
confinement. The reinforcement should be adequately anchored on both sides of the inter-
face. If loops are used, as shown in Figure 4.39, the clear space beneath the bend should be
at least 5 mm + size of aggregate. The spacing of links should not be too large, 1.2 – 1.5 m
being typical for hollow core slabs. If vEdi < vRdi no interface shear reinforcement is required,
although some contractors choose to place H10 or H12 loops (as shown in Figure 4.39) at
about 1.2 m intervals anyway. The loops should pass over the top of the bars in the struc-
tural topping and be concreted into the joints between the precast units.
Example 4.16
Calculate the shear reinforcement necessary to satisfy the ultimate horizontal shear force
at the precast – in situ interface in Example 4.11 and 4.12 if
1. UDL loading as Example 4.11

2. Increased live load of 10 kN/m 2
The top surface of the hollow core unit is ‘as extruded’ finish. Use high tensile reinforce-
ment for interface links or loops f y = 500 N/mm 2 .
Solution
(i) UDL loading only
M Ed1 = 1.25 × 6.38 × 8.0002 /8 = 63.8 kNm
M Ed2 = 1.25 × 7.20 × 8.0002 /8 = 84.0 kNm
K 2 = 84.0 × 106/(25 × 1200 × 234.22) = 0.051; then z 2 = 222.5 mm
x2 = (234.2 – 222.5)/0.4 = 29.3 mm; thus 0.8 x2 < topping depth 75 mm, then
Fct/Fc = 1.0, that is the ratio of flexural force in topping/total force due to Stage 2
bending = 1.00
V Ed1 at the support = 1.25 × 6.38 × 8.0/2 = 31.9 kN, and due to Stage 2 loads
V Ed2 = 42.0 kN
vEdi = 1.00 × 42.0/1.2 × 222.5 = 0.16 N/mm 2
vRdi = 0.2 × 0.7 × 0.3 × 252/3/1.5 = 0.24 N/mm 2 < 0.5 × 25/1.5 = 8.33 N/mm 2
No interface shear loops required
(ii) Imposed live load = 10 kN/m 2

M Ed2 = 156.0 kNm
K 2 = 156.0 × 106/(25 × 1200 × 234.22) = 0.095; then z 2 = 212.6 mm
x 2 = 53.9 mm; thus Fct /Fc = 1.0 and V Ed2 = 78.0 kN
vEdi = 1.00 × 78.0/1.2 × 212.6 = 0.306 N/mm2 > 0.24 N/mm2 use interface loops
Smooth surface profile μ = 0.6
Interface shear loops required A s = 1000 × (0.306 − 0.24) × 1200/(0.6 × 0.87 ×
500) = 302 mm 2 /m run
Use H10 at 500 mm spacing (2 × 78.5 × 1000/500 = 314 mm 2 /m)
Prestressed concrete elements designed with a limiting tensile stress of zero or fctm should
not, by definition, require a check for crack width, and this has been the practice for many
years. Unacceptable cracking may be avoided if the level of prestress does not exceed 0.75 f pk
in service, typically 1330 N/mm 2 , which is uncommon and means that the initial prestress
needs to as high as 0.85 f pk.
BS EN 1992-1-1, clause 7.1(2) states that for the purpose of calculating crack widths, service
stresses should be for the flexurally uncracked section and the flexural tensile stress ≤ fctm.
Furthermore, clause 7.3.2(4) states that where crack control is necessary no minimum rein-
forcement is required provided that, under characteristic loading, the tensile stress is less than
fctm. This is always the case as the only time characteristic loading is not used, the limiting
tensile stress is zero for exposure classes greater than XC1, according to UK NAD Table NA4,
Table 4.3 shows the part of Table 7.1N applicable to prestressed concrete with bonded tendons
(Figure 4.40). Table 7.1N suggests that crack widths are required by calculation and should be ≤
0.2 mm under the frequent load combination crack spacing. Therefore, crack width calculations
are undertaken according to clause 7.3.4 as follows:
Exp. 7.8 Design crack width wk = sr,max (εsm − εcm) (4.119)
s - +a r
r s
Exp. 7.9 e -e = ³ (4.120)
σt Mś /Zt,co
x
+ =
h–x h–a–x
fctm
Determination of stress in tendons in the calculation of crack width in prestressed concrete

sections.
σs = stress variation in tendons after decompression, taken at the level of tendons

- -
nearest to soffit = (4.121)
-
where
a is the axis distance = minimum cover + tendon diameter
x is the depth to zero stress after decompression when the bottom stress is limited to
é ¢ -s ù
fctm = ê ú (4.122)
êë ¢ -s + úû
For composite slabs
é ¢ + ¢ -s ù
=ê ú (4.123)
êë ¢ + ¢ -s + úû
kt = 0.4 for long-term loading

fct,eff = fctm
αe = Ep /Ecm
ρeff = Ap /Ac,eff
7.3.2(2) Ac,eff = minimum of b (h − x)/3 or (less likely) 2.5 b (h − d)

= service moment due to frequent load combination = 1.0 Gk + ψ1 Qk
= stage 1 ditto = 1.0 Gk self-weight and topping
= Stage 2 ditto 1.0 Gk imposed dead + ψ1 Qk
Exp. 7.11 Crack spacing (mm) sr,max = k3 c + k1 k 2 k4 ϕ/ρeff (4.124)
where k1 = 1.6, k 2 = 0.5, k3 = 3.4 and k4 = 0.425

ϕ = mean tendon diameter based on area (if mixed sizes at minimum cover)
c = minimum cover
Example 4.17
Calculate the crack spacing and crack width in the composite hollow core slab in Example 4.13.
Solution
Depth x to zero stress at M s = 200 × 7.54/11.34 = 133.0 mm
where top stress = −1.48 + 5.36 + 3.66 = 7.54 N/mm2, and bottom stress fctm =
3.80 N/mm2
Axis height a = 40.8 mm, cover c = 35 mm, and mean dia. strands = 11.4 mm
Act,eff = min{1197 × 200 − 133.0/3; 2.5 × 1197 × 40.8; 0.5 × 152,000} =

26,716 mm 2
Ap /Act,eff = 766/26,716 = 0.0287
s = 3.4 × 35 + (1.6 × 0.5 × 0.425 × 11.4/0.0287) = 255 mm
σs = [(200 − 40.8 − 133.0)/(200 − 133.0)] × −3.80 × 195,000/36,283 = −5.9 N/mm 2
εsm − εcm = 0.000018
wk = 0.000018 × 255 = 0.005 mm < 0.2 mm
As expected, wk is not critical as the tensile stress is limited to fctm.
Shallow precast slabs, hence the name ‘planks,’ are laid between supports and are used as
permanent formwork for an in situ concrete topping. The precast plank is between 65 and
100 mm thick, depending on the span, and the depth of the complete floor is between 150
and 200 mm. See Figure 4.12a through d. Steel bars, wires or tendons placed in the precast
plank act as the flexural sagging reinforcement, and a light steel mesh (e.g. A142, A193) in
the in situ concrete acts as hogging reinforcement. The diagonal bars in the lattices provide
shear resistance during the construction stage. The planks can also be pretensioned using
longitudinal strands or wires only as shown in Figure 4.13.
The composite floor slab may be designed as either simply supported or continuous as
shown in Figure 4.30. Note that there should also be continuity bars at the bottom of a
continuous slab at the support to ensure rigidity in the compression zone and to cater for
reversals of bending moments due to creep, shrinkage, temperature effects, etc. when no (or
small) superimposed load is present. The floor slab is designed as one-way spanning, despite
the introduction of transverse reinforcement in the topping. Two-way spanning capabilities
still have not been fully exploited, for example by using high bond steel–fibre reinforced
concrete infill to connect short lengths (about 10 ϕ long) of rebar projecting from the sides
of the precast plank unit (Figures 4.41 and 4.42).
Lightweight void formers – typically, but not always made of expanded polystyrene – are
pressed into the top surface of the precast plank during manufacture. The adhesion and
friction between the polystyrene and wet precast concrete are sufficient to hold the block in
position throughout the erection and in situ concreting operations.
In reinforced planks (i.e. nonprestressed), deflections are catered for either by checking
that the effective span/effective depth l/d ratio is within the allowable limit according to BS
Cast in situ topping Continuous top steel

in topping Interface shear reinforcement
Continuity bars in
bottom and ends of
precast concrete unit
Continuity in composite plank floor.
EN 1992-1-1, clause 7.4.2(2). However, where the precast plank is propped during casting
of the topping, l/d ratio checks may not be valid, and, as clause 7.4.3(3), Exp. 7.18 allows
the partially cracked stiffness to be used (not permitted in BS 8110), direct calculation may
give a better approximation of deflection. In prestressed planks, the same procedures given
in Section 4.4.3.1. are adopted.
A΄s,l
sl
Asw,d
hl
hp β
b
Cover is to
As,l Transverse As,t transverse bars
(a) β = slope of diagonal
As,mesh Aś if required

d´
In situ
h topping d
3 mm gap
(b) As
Structural details of composite plank floor. (a) Prefabricated cros-section details. (b)
cross-section. ( )
Pitch spacing of
diagonals sl A΄s,l
Asw, d α = inclination of diagonals
α α
As,l
(c) Transverse As,t
Asw
added s
links Aś curtailed
As full length or 50% curtailed
Clear span
Lb
Effective span = clear + min{0.5 Lb ; 0.5h} each end
(d)
( ) Structural details of composite plank floor. (c) Elevation of lattices. (d) End eleva-
tions of composite slab.
In the temporary stage, the precast plank is simply supported. The unit may be designed
so that unpropped simply supported spans of up to 4.0–5.0 m are possible, usually by
increasing the number of lattices to increase shear stiffness, but not necessarily increas-
ing the number of bottom bars. The top bar in the lattice is in compression but is firmly
restrained both vertically and horizontally by the inclined bars making the lattice. The unit
is most critical when the self-weight of the wet in situ concrete is added to the self-weight
of the precast plank. An allowance for construction traffic of upto 1.5 kN/m2 is added to
the temporary loading when calculating the sizes of bars required. In the permanent situa-
tion, the hardened in situ concrete provides the compressive resistance. The flexural sagging
resistance at mid-span is governed by the strength of the bottom reinforcing bars, as in an
ordinary underreinforced situation.
The lattice girder is manufactured from a 5 or 7 mm diameter hard drawn wire, or 8–20 mm
diameter ribbed high tensile bar with a characteristic 0.2% proof stress of 500 N/mm2. It
consists of small diameter bars in pairs, typically 5, 6 or 7 mm diameter bars, bent into a
zig-zag girder where the inclination of the links is between 45° and 65° to the longitudinal
direction, and about 70o to the transverse direction. The lattice has a single top bar, typically
H8 to H20, depending on the span of the floor and the propping methods used, and two
bottom bars of smaller size, depending on the additional area of reinforcement in service
at ultimate. The lattices are cut so that a complete stirrup is left intact at both ends of the
girder. The buckling strength of the diagonal link and the top bar is checked according to
the steelwork Eurocode BS EN 1993-1-1, clause 6.3.1.3.
Stage 1 during hoisting and propping

During hoisting, the precast plank is reinforced using the lattices in tension and by com-
pression in the top bars, and later by additional rebars in the bottom designed for ultimate
moment. Hoisting is specified by the manufacturer at two points (two brothers = four
lifters) at one-fourth length from each end (Figure 4.43a), except for length (about) > 6.5 m
where three points (six lifters) are used at 1/7th length and at mid-length (Figure 4.43b).
zh
0.25 L 0.25 L
Total L = clear span + 2 Lb
(a)
0.32 whL 0.36 whL 0.32 whL
0.143 L 0.357 L
Total L = clear span + 2 Lb
(b)
MRd,l here
x
R1 R2 R1
Effective span leff
(c)
P P P
X X X X
(d)
(a) Two-point hoisting of precast plank unit for length/lattice lever arm, / < 50. (b) Three-
point hoisting of precast plank unit. (c) Maximum bending moment over multiple props.
(d) Multiple back propping of upper floors.
These positions minimise the positive bending moments during lifting that will cause
compression in the top bars.
For 2 point lifting, let wh = self-weight plank (25 kN/m3 density) +25% crane impact,
L= actual total length of plank, then using γG = 1.35:
Negative M Ed,h = −1.35 wh (0.25L)2 /2 = −0.0422 wh L 2 (4.125)
Positive M Ed,h = −1.35 wh (0.5L)2 /8 −0.0422 wh L 2 = 0 (4.126)
For 3 point lifting
M Ed,h = 1.35[−wh (0.5 L)2 /2 + 0.32wh 0.357 L 2] = −0.011 wh L 2 (4.127)
Lever arm zh = hl – 0.5(ϕs,l + ϕs,l ′), where ϕs,l and ϕs,l ′ = lattice bar diameter at the bottom
and top, hl = overall depth of lattices.
The top bars area is
A s,l ′ = M Ed,h /f yk,b zh (4.128)
where f yk,b is buckling strength due to slenderness as follows:

Bottom lattice bars A sl (usually set by the manufacturer as ϕ = 6 mm) = A s,l ′ is not critical
due to the added bars A s in Stage 2.
M Rd,l of lattices = A s /l ′ zh (4.129)
The buckling capacity f yk,b of rebars is according to Eurocode 3, BS EN 1993, Part 1-1 (BS
EN 1993-1-1 2005).
Top bars. Effective length lo = 0.7 × distance between diagonals
Diagonal bars. Effective length lo = 0.7 × distance between top and bottom bars
Radius of gyration for round bars i = ϕ/4
6.3.1.3 Slenderness λ = lo /i
Critical λcrit = 93.3 √(235/f yk) (4.130)
where f yk generally = 500 N/mm 2

Nondimensional slenderness λ′ = λ/ λcrit
Table 6.2 α is from Curve c for round bars
Table 6.1 α = 0.49
6.49 Stress reduction factor χ = 1/[Ф + √(Ф2 − λ’2)] ≤ 1 (4.131)
where Ф = 0.5 [1 + α (λ′ − 0.2) + λ′2] (4.132)

f yk,b = χ f yk /γMo where γMo = 1.0 (this over rides γm = 1.15 at ultimate)
The design may therefore enter into a loop, switching from smaller to larger bars, as the
diameter of the bar influences λ, which controls χ and hence f yk,b.
During propping, the distance between props x is given by the MRdl of the lattices at the first
interior prop where the bending moment at the prop is greatest. The number of props N = leff/x.
The total mass during propping wh = self-weight of plank w0 + self-weight topping w1 + tem-
porary construction traffic wc (taken as 1.5 kN/m2). Referring to Figure 4.43c, Mprop at the first
interior prop = wh x2/8 (conservative as for more than one prop, the denominator is 9.3 to 10)
\ = (4.133)
The prop reaction R is determined for each situation. The prop reaction is transferred to
the composite slab due to the self-weight of concrete, excluding the temporary construction
traffic load.
Back propping of upper floor
After the removal of props, back propping of the next upper floor may be considered
if the props beneath the designed slab are not continuous to a foundation. Referring to
Figure 4.43d, let X = equidistance X between props and the supports. Self-weight of the
upper precast plank and topping = w u. Then the moment due to upper prop reactions P:
M Ed,u = due to P is later checked against the design value of M Ed
Ultimate flexural design in Stage 2 after hardening of in situ topping
M Ed2 is due to Stage 2 imposed loads and removal of props
d2 = h – c – distribution bar dia. – main bar radius
K 2 = M Ed2 /fcki b d22 ≤ 0.206 for singly reinforced sections
z 2 = [0.5 + (0.25 – K 2 /1.134] d2 ≤ 0.95 d2
x 2 = (d2 – z 2)/0.4
A s2 = M Ed2 /z 2 0.87 f yk
Minimum and maximum areas and spacing
9.2.1.1(1) A s,min ≥ max{0.0013 b d2; 0.26 fctm b d2 /f yk}
9.2.1.1(3) A s,max ≤ 0.04b h
Area A s,min mesh in topping = max{0.0013 b (h − hp); 0.26 fctm b (h − hp)/f yk}
9.3.1.1(2) Area of transverse (distribution) steel As,dist ≥ 0.2 As are provided at
nominal cover
9.3.1.2 Bars at simple support A s ≥ 50% A s in span
8.2.(2) Spacing > max{bar diameter; aggregate size + 5 mm; 20 mm)
Bond and anchorage lengths
8.4.4 lbd = α1 α2 α3 α4 α5 lb,rqd > max{0.3 lb,rqd; 10Ф; 100 mm) (4.134)
Table 8.2 α1 = 1.0 for straight bars, and α3 = α4 = α5 = 1
α2 = 1 – 0.15(cd – Ф)/Ф (4.135)
Fig. 8.3 cd = min{spacing/2; cover}
8.4.3.(2) lb,rqd = 0.25 σsd Ф/f bd (4.136)
where σsd = actual design stress = 0.87 f yk (4.137)
8.4.2 f bd = 2.25 η1 η 2 fctd (4.138)
Fig. 8.2 η1 = 1 for good casting conditions; η2 = 1 for bars ϕ ≤ 32 mm.
Without shear reinforcement

Refer to BS EN 1992-1-1, clause 6.2.2(1) and code Fig. 6.3. This is calculated at the plane
where the main bars A sl extend lbd + d2 beyond section considered. If the bars extend one
bond length lbd beyond face of support, the shear is calculated at d2 from the face of the
support.
6.2.2.(1) V Rd,c = 0.12 k (fck 100 A sl /b d2)1/3 b d2 (4.139)
but V Rd,c ≥ 0.035 k3/2 √fck b d2 (4.140)
where k = 1+ √(200/d2) < 2 (mm units)
With shear reinforcement if V Ed > V Rd,c
Refer to BS EN 1992-1-1, clause 9.3.2(1). Use shear reinforcement only if total depth h ≥
200 mm. In shallower sections, it is necessary to increase h to nullify need for shear links.
Links should only be required in cases of narrow width, for example at notches, or for
large point loads close to support. Provide links as double leg or single leg L bars enclosing
main bars.
Refer to Figure 4.42 for the contribution from the diagonal bars in the lattices. The diago-
nals are of area A sl = area per diagonal × n number lattices, inclined at α and (180-α) along
span and β to the vertical, with spacing sl and effective depth dl = hl – top and bottom radii
of lattice’s main bars.
a b
6.3.2. Exp. 6.13 = éë( q+ a) + ( q+ - a ) ùû
(4.141)
If VEd > VRd,s,l additional links of area Asw at spacing s required to resist the excess VEd − VRd,s,l
6.3.2. Exp. 6.8 - = q (4.142)
a
6.3.2. Exp. 6.9 - = (4.143)
q+ q
where αcw = 1, v1 = 0.6 (1 − fck /250), z = 0.9d
- æ - ö
Rearranging q = ç ÷ (4.144)
è ø
cot θ is determined, but 1.0 ≤ cot θ ≤ 2.5
-
= (4.145)
q
Exp. 9.4 > (4.146)
9.3.2(4) s ≤ 0.75d
Example 4.18
Design a composite reinforced concrete plank floor to carry a characteristic imposed dead
load of 1.0 + 0.6 kN/m2 for finishes and ceiling/services, and a superimposed live load of
5.0 kN/m2 over a simply supported effective span of 5.5 m. Construction traffic load may
be taken as 1.5 kN/m2. The floor is to be used for offices. Check the design for:
(i) No propping
(ii) Props at one-third and two-third span positions
Use fck = 40 N/mm2 for the precast, fcki = 25 N/mm2 for the in situ topping. High tensile
bars in the precast plank and mesh in the plank and in situ topping, fyk = 500 N/mm2,
fywk = 250 N/mm2 (diagonals). Cover to all reinforcement = 25 mm. Bearing lengths =
100 mm. Use polystyrene blocks 300 mm wide × 70 mm depth of density 10 kg/m3 to
save weight (Figure 4.44).
Carry out the design at both the temporary and permanent stages for flexure and vertical
shear only.
Solution
Try a 1200 mm wide × 80 mm deep precast plank unit with 3 no. lattices. The standard
lattice from a proprietary manufacturer has a pair of 6 mm diameter bottom bars 70 mm
As,d
Fc,t
z2 vEdi
α
VEd2
Interface shear due to flexural shear and shrinkage of topping.
apart (to outer face of bars), a top bar at 130 mm total height, and the pitch of diagonal
links (parallel with span) = 200 mm. Using 300 mm wide polystyrene blocks, there is
sufficient width for only two rows of blocks between 3 no. lattices.
Self-weight of precast plank w 0 = 0.08 × 25 = 2.00 kN/m 2 = 2.40 kN/m run
Self-weight of topping and polystyrene blocks w 1 = 2.554 kN/m run
Construction traffic load wc = 1.80 kN/m run
Distribution bars for precast plank using Mesh

A s,min = max{0.13% × 1000 × 80; 0.26 × 3.51 × 1000 × 80/500} = 146 mm 2 /m run
where fctm precast = 0.3 × 402/3 = 3.51 N/mm 2
Use A193 mesh at 25 mm bottom cover. Transverse bar diameter = 7 mm.
Mesh in topping
As,min = max {0.13% × 1200 × 120; 0.26 × 2.57 × 1200 × 120/500 = 192 mm2/
unit width
where fctmi in situ = 0.3 × 252/3 = 2.57 N/mm 2
Use A193 mesh at 25 mm top cover. Mesh provides 6 no. 7 mm bars =
232 mm2 per unit.
Hoisting
Stage 1 design later will show ϕ = 20 mm top bars in lattices.
Bottom bars specified as 6 mm dia. Lattice hl = 130 mm.
zh = 130 − 0.5 × (20 + 6) = 117 mm
L/zh = 5600/117 = 48 < 50
Hoisting with four lifters at L/4 = 1.400 m from ends

wh = 2.40 kN/m plus 25% for impact = 3.00 kN/m
γG = 1.35 (when acting alone)
M Ed,h = 0.0422 wh L 2 = 0.0422 × 3.00 × 5.6002 = 3.97 kNm per unit
(i) No propping
Stage 1: Precast plank and topping with construction traffic
l = 5.500 m, γG = 1.25 when combined with imposed loading
Stage 1 wG1 = 2.40 + 2.55 + 1.80 = 6.75 kN/m, wEd1 = 1.25 × 6.75 = 8.44 kN/m
M Ed1 = 8.44 × 5.5002 /8 = 31.89 kNm
Cover to bottom lattice bars = nominal cover + mesh bar dia. = 25 + 7 = 32 mm
Slenderness of top bars λ = lo /i = 0.7 × 200/5.0 = 28.0, where i = 20/4 = 5.0 mm
λcrit = 93.3 √(235/500) = 64.4, λ′ = 28.0/64.4 = 0.435
α = 0.49, Ф = 0.5 (1 + 0.49 × 0.235 + 0.4352) = 0.652
χ = 1/[Ф + √(Ф2 − λ′2)] ≤ 1 = 1/[0.652 + √(0.6522 – 0.4352)] = 0.879
f yk,b = χ f yk /γMo = 0.879 × 500/1.0 = 439.3 N/mm 2
As1 = 31.89 × 106/(117 × 439.3) = 620 mm2 per top lattice bar = 620/3 = 207 mm2.
Use 3 no. lattices with 3 no. H20 (942) top bars plus 6 no. H6 (170) bottom bars
Flexural design Stage 2 using additional bottom bars
Stage 2 wG2 = 1.2 × (1.0 + 0.6) = 1.92 kN/m, wQ2 = 1.2 × 5.0 = 6.00 kN/m
wEd2 = 1.25 × 1.92 + 1.5 × 6.00 = 11.40 kN/m
M Ed2 = 11.4 × 5.5002 /8 = 43.10 kNm
h = 200 mm, b = 1200 mm, and due to blocks bw = 1200 – 2 × 300 = 600 mm, hf = 200 –

80 – 70 blocks = 50 mm
Try 10 mm dia. additional bottom bars between lattices resting on transverse bars of
the mesh
d2 = h – c − Ф/2 = 200 − 32 − 5 = 163 mm
K 2 = 43.10 × 106/(25 × 1200 × 1632) = 0.054
z2 = 155 mm, x 2 = 20 mm < hf /0.8 design as rectangular
A s2 = 43.10 × 106/(155 × 0.87 × 500) = 640 mm 2 > A s,min = 357 mm 2
Then As,l = 170 mm2, As,mesh = 231 mm2, As2 required = 640 − 170 − 231 = 240 mm2
Use 4 no. H10 bars (314) either side and between lattices
Total area of bottom bars provided A s,prov = 314 + 170 + 231 = 715 mm 2
Mesh plus top lattice A′s,prov = 232 + 942 = 1174 mm 2
Capacity of lattice diagonals during construction, incl. construction traffic
V Ed,h during hoisting = 1.35 × 3.0 × 1.4 = 5.67 kN per unit
Vs1 including traffic = 6.75 × 5.500/2 = 18.56 kN
V Ed1 = 1.35 × 18.56 = 25.1 kN
Length of diagonal = √(1302 + 70/22 + 200/22) = 167.7
Try 6 mm diameter. λ = L e /i = 0.7 × 167.7/1.5 = 78.3. λcrit = 91.0. λ′ = 0.860
f yk,b = 0.625 × 250 = 156.2 N/mm 2
Diagonals α = tan−1 130/100 = 52°, β = tan−1 130/35 = 75°
A sw per diagonal leg = 25.1 × 103/(2 × 3 × sin 52° × sin 75° × 156.2) = 35 mm 2
Use R7 diagonals (38) at 200 mm pitch
Ultimate V Rd,c without links
bw after deduction for blocks = 600 mm
k = min{2: 1+ √(200/163) = 2.0
A s /bw d2 = min{0.02; 715/(600 × 163) = 0.007
V Rd,c = 2.0 × 0.12 (100 × 0.007 × 25)1/3× 600 × 163 × 10 –3 = 61.81 kN
V Rd,c min = 0.035 × 2.01.5 × √25 × 600 × 163 × 10 –3 = 48.41 kN
lw = 5500 – 100 – 2 × 163 = 5074 mm
V Ed = (6.2 + 11.4) × 5.074/2 = 44.5 kN
No additional shea.r reinforcement required

(ii) Props at ⅓ and ⅔ span positions
Props at 1.833 m from supports. Total load due to self-weight of plank and topping = 4.95
× 5.500 = 27.2 kN. Prop reactions (each) = 9.99 kN and support reactions = 3.63 kN.
M Ed1 max occuring at 0.77 m from support = +2.26 kNm and −2.56 kN at props is
less critical than M Ed,h = −3.97 kNm per unit
Try 8 mm dia. top bars (smallest recommended in compression) in lattices
zh = 130 − 0.5 × (8 + 6) = 123 mm
λ = lo /i = 0.7 × 200/2.0 = 70.0, λcrit = 64.4, λ′ = 1.087
f yk,b = 0.491 × 500 = 245.5 N/mm 2
A s1 = 2.26 × 106/(123 × 245.5) = 75 mm 2
and A s1 = 3.97 × 106/(123 × 0.87 × 500) = 74 mm 2
Use 3 no. lattices with 3 no. H8 (150) top bars plus 6 no. H6 (170) bottom bars
Flexural design Stage 2 using additional bottom bars
M Ed2 due to imposed loads and removal of props = 11.4 × 5.5002 /8 + 1.25 × 9.99 ×
1.833 = 65.97 kNm
Try 16 mm dia. additional bottom bars, d2 = 200 − 32 − 8 = 160 mm
K 2 = 65.97 × 106/(25 × 1200 × 1602) = 0.086
z 2 = 147 mm, x 2 = 33 mm < hf /0.8 design as rectangular
A s2 = 1034 mm 2 > A s,min = 350 mm 2
A s2 required = 1034 – 170 lattice − 231 mesh = 633 mm 2
Use 4 no. H16 bars (804) either side and between lattices
Total A s = 804 + 170 + 231 = 1205 mm 2
Capacity of lattice diagonals during construction, incl construction traffic and propping
Vs = 6.93 kN, V Ed1 = 1.35 × 6.93 = 9.36 kN > V Ed,h = 5.67 kN
Slenderness and f yk,b = 156.2 N/mm 2 as before
A sw per diagonal leg = 9360/(2 × 3 × sin 52° × sin 75° × 156.2) = 13 mm 2
Use R6 diagonals (28) at 200 mm pitch
Ultimate V Rd,c without links
k = min{2: 1+ √(200/160) = 2.00
A s /bw d2 = min{0.02; 1205/(600 × 160) = 0.013
V Rd,c = 2.000 × 0.12 (100 × 0.013 × 25)1/3 × 600 × 160 × 10 –3 = 72.66 kN
V Rd,c min = 0.035 × 2.001.5 × √25 × 600 × 160 × 10 –3 = 47.52 kN
Shear span lw = 5500 – 100 – 2 × 160 = 5080 mm
V Ed = (6.2 + 11.4) × 5.080/2 = 44.7 kN
No additional shear reinforcement required.
m
6.2.5(1) = £ = + ms + + a b+ (4.147)
where
Fc,t is the compression in topping due to ultimate Stage 2 loading
Fc is the total compression in topping and precast unit due to ultimate Stage 2 loading
V Ed2 is the ultimate vertical shear force at support (not on shear plane)
z 2 is the lever arm of rebars-topping due to Stage 2 loads
bi is the width of interface = bw if infill blocks, otherwise = b
σn is the normal stress = 0 (neglecting self-weight or other imposed pressures)
A s,d is the area of diagonals in lattices
A swi is the area of additional interface shear reinforcement (vertical links)*
Aj is the area of interface joint
α, β, is the inclination of diagonals along and perpendicular to span
f yd = f ywk /γs
3.1.6.2(P) fctdi = 0.14 fcki 2/3
6.2.5.(2) c = 0.2, μ = 0.6 (a free surface left without treatment after casting)
æ ö
=ç - - m a a q÷ (4.148)
è øm
Example 4.19
Check the interface shear stress due to Stage 2 loads and the relative shrinkage between
the precast plank and topping in Example 4.18 Case (ii) with props (this case has the
greatest V Ed2).
Solution
Example 4.17 Case (ii) shows that γG = 1.25 and γQ = 1.5
V Ed1 = 1.25 × (4.95 × 5.500/2 – 9.99) = 4.54 kN
V Ed2 = 11.4 × 5.500/2 + 1.25 × 9.99 = 43.82 kN
From Example 4.18 Case (ii), z 2 = 147 mm, x 2 = 33 mm < hf, then Fc,t /Fc = 1.0
vEdi = 1.00 × 43,820/(600 × 147) = 0.50 N/mm 2
vRdi = min{0.2 × 1.20 = 0.24 N/mm2; 0.5 × 25/1.5 = 8.33 N/mm2} = 0.24 < 0.5 N/mm2
where fctdi = 0.14 × 252/3 = 1.20 N/mm 2
Smooth surface profile, μ = 0.6
Interface links A si /s = 1000 × (0.50 − 0.24) × 600/(0.6 × 0.87 × 250) = 1187 mm 2 /m
No. of lattices = 3. No. of diagonals per pitch = 4. No. of diagonals = 12
A sw per 6 mm dia. diagonal = 28.3 mm 2
A s,d /s = 12 × (0.6 sin 52° + cos 52°) sin 75° × 28.3 × 1000/200 = 1782 mm 2 /m
A s,d /s > A si /s, no additional links required.
* Fig. 6.10. Distribution of A swi should respond to the distribution of V Ed2 .
ASSAP. 2002. The Hollow Core Floor Design and Applications, Manual, 1st ed., Association of
Manufacturers of Prestressed Hollow Core Floors (ASSAP), Verona, Italy.
Bhatt, P. 2011. Prestressed Concrete Design to Eurocodes, Routledge, Oxford, UK., 596pp.
British Precast. 2008. Precast Eurocode 2: Worked Examples, British Precast Concrete Federation,
Leicester, UK., 77pp.
BS EN 1992-1-1. 2004. Eurocode 2, Design of concrete structures – Part 1-1: General rules and rules
for buildings (with National Application parameters), BSI, London.
BS EN 1992-1-2. 2004. Eurocode 2, Design of concrete structures, Part 1-2: Structural fire design
(with National Application parameters), BSI, London.
BS EN 1993-1-1. 2005. Eurocode 3, Design of steel structures – Part 1-1: General rules and rules for
buildings, BSI, London.
Concrete Manufacturers’ Association. 2013. Hollow Core Slab Systems Information Manual,
Concrete Manufacturers’ Association, Johannesburg, South Africa, 36pp.
Elliott, K. S. 2014. Transmission length and shear strength of hollow core slabs, Magazine of Concrete
Research, 66, 12.
Elliott, K. S. and Ibrahim, I. 2006. Composite construction in prestressed precast concrete Floors.
Ninth International Conference on Concrete Engineering (CONCET 2006), Institution of
Engineers Malaysia, Kuala Lumpur, Malaysia, pp. 104–116.
FIP Recommendations. 1988. Precast Prestressed Hollow Cored Floors, FIP Commission on
Prefabrication, Thomas Telford, London, UK., 31pp.
Girhammer, U. A. 1992. Design principles for simply supported prestressed hollow core slabs,
Structural Engineering Review, Oxford, UK., 4(4), 301–316.
Ibrahim, I. and Elliott, K. S. 2009. Long-term behaviour of precast prestressed hollow core slabs with
concrete toppings. Seventh Asia Pacific Structural Engineering and Construction Conference
(APSEC) and Second Asian European Civil Engineering Forum (EACEF), Kuala Lumpur,
Malaysia.
Ibrahim, I.S, Elliott, K. S., Abdullah, R., Kueh, A. B. H. and Sarbini, N. N. 2016. Experimental study
on the shear behaviour of precast concrete hollow core slabs with concrete topping, Engineering
Structures, 125, 80–90.
Moss, R. M. 1995. Load testing of hollow plank concrete floors, The Structural Engineer, 73(10),
161–168.
Oh, B. H., Lim, S. N., Lee, M. K. and Yoo, S. W. 2014. Analysis and prediction of transfer length in
pretensioned, prestressed concrete members, ACI Structural Journal, 111(3), 549–560.
Pajari, M. 1998. Shear resistance of PHC slabs supported on beams, Journal of Structural Engineering,
124(9), Part 1: Tests, 1050–1061. Part 2: Analysis, 1062–1073.
Park, H. and Cho, J-Y. 2014. Bond-slip-strain relationship in transfer zone of pretensioned concrete
elements, ACI Structural Journal, 111(3), 503–514.
PCI. 1991. PCI Manual for the Design of Hollow Core Slabs, Precast/Prestressed Concrete Institute,
Chicago, IL, 88pp.
PD 6687-1. 2010. Background Paper to the UK National Annexes to BS EN 1992-1, BSI, London.
Pfeifer, D. W. and Nelson, T. A. 1983. Tests to determine the lateral distribution of vertical loads in a
long span hollow core floor assembly, PCI Journal, 23(6), 42–57.
Stanton, J. 1987. Proposed design rules for load distribution in precast concrete decks, ACI Structural
Journal, September–October, 371–382.
Walraven, J. C. and Mercx, W. 1983. The Bearing Capacity of Prestressed Hollow Core Slabs, Heron,
28, No. 3, University of Delft, Delft, the Netherlands, 46pp.
Van Acker, A. 1984. Transversal load distribution of linear loadings in hollow core floors. FIP
Conference, Calgary, Alberta, Canada, pp. 27–33.
Chapter 5
Beams are the main horizontal load-carrying members in skeletal structures. They are, by
definition, relatively small prismatic sections of large flexural (typically, - ) and
shear ( - ) capacity. In the European product standards, they form part of ‘linear
structural elements’ according to BS EN 13225 (2013) but note that this standard is for the
beam itself and does not extend to the design of beams within the framework. In a precast
concrete structure, they must at some point in time support the self-weight of the floor slabs
alone and should therefore be capable of resisting all of the possible load combinations
that precast construction brings – for example torsion will be present if, in the temporary
construction stage, the floor units are all positioned on one side of the beam. This must be
allowed for both in the design of the beam and at the end connections to the column.
Beams fall into two distinct categories: internal and external. Internal beams are usually
symmetrically loaded, that is floor slabs are on both sides of the beam, and therefore the
beam is symmetrical in cross section as shown in Figure 5.1a. The limiting design criterion is
often minimum depth in order to maximise headroom and reduce the drop beam, or ‘down-
stand’ depth defined in this figure. For this reason, internal beams are often pretensioned to
maximise their structural performance. To minimise the downstand, part of the beam may
be recessed within the depth of the floor slab, giving rise to the so-called inverted-tee beam.
Internal beams may be designed compositely, with the floor slab acting as a compression
flange – see Section 5.3.
External (edge) beams are, by nature, asymmetrically loaded. Take for example the deep
spandrel beam shown in Figure 5.2, torsion will result when floor slabs sit on the bearing
nib because the line of action of the load is not coincident with the shear centre of the beam.
Torsion must therefore be considered in design.
The lateral instability of slender beams, for example the upstand part of deep edge beams
of depth h and breadth , is limited according to EN 1992-1-1, clause 5.9. It is taken
into account for beams without sufficient lateral bracing in the finished structure. A lateral
deflection of should be assumed as a geometric imperfection in the verification of
beams under unbraced conditions, where l = total length of the beam. In finished structures
(persistent), bracing from connected members may be taken into account, such as tied in
floor slabs. Second-order effects in connection with lateral instability may be ignored in
code Exp. 5.40, if the distance between torsion restraints is
= £ (5.1)
215
= £ (5.2)
Typically, and , then before the beam is restrained by floor slabs

= ´ ´ - = , which can be a limiting criterion in some beams.
Cross section may be rectangular, but to avoid having to place formwork at an exter-
nal edge, the cross section tends to be L-shaped, as shown in Figure 5.1b. Beams with tall
upstands are known as ‘spandrel’ beams, as shown in Figure 5.3, where the connections to
resist overturning are provided by projecting rebars from the column. They are frequently
used around the perimeter of buildings such as in car parks where they form part of the
impact barrier. Spandrels are often used to form a dry envelope around the perimeter of
the building by making a temporary weather shield between successive storey heights.
Edge beams are not pretensioned – their non-symmetrical shape is the main reason, but as
beam depth is not a limiting factor there is no reason to do so. Edge beams may be designed
compositely with the floor slab, but for the same reasons as mentioned earlier, there is often
little need to do so.
Upstand breath
250–400 mm, typically in
50 increments
Floor slab level

with top of upstand
Floor slab thus
Upstand depth
150, 200, 250
boot depth from 150
Depth from 250 to to about 600 mm, typically
about 1000 mm, in 50 or 75 increments
typically in 50 or 75
increments
Breadth from 500 to

about 750 mm, typically
Breadth 250–500 mm in 75 or 100 increments
Floor slab thus

Difference in
floor level
(a)
Types of beam. (a) Internal rectangular and inverted-tee. ( )
Precast concrete beams 217
Upstand
breadth
250400flmm
Floorflslab
Downstand
possible
Column
position
Upstandflbreadth
140200flmmflminimum
possible
Upstandfldepthfl150flto
Upstandflbreadthfl75 aboutfl400flmm,flusuallyflto Upstandfldepthflupflto
toflb-125,flinflsame matchfldepthfloffloor aboutfl1500fladflhoc.flNo
incrementsflasflb slab standardflincrements
generally
Optionalflbearing
nibflforflcladding
Bootfldepthfl150fltoflabout
800flmm,fltypicallyflinfl75 Possiblefldownstand
b = Breadthflfromfl250flto orfl100flincrements
aboutfl400flmm,fltypically
(b) infl50florfl75flincrements
Types of beam. (b) Edge L- and spandrel beams.
L-shaped edge beams support non-symmetrical floor loads. The part of the beam sup-
porting the floor is called the “boot”, and the main web is the “upstand”. There are two
types of edge beam, shown in Figure 5.4:
Type I, where a wide upstand is part of the structural section (Figure 5.4b).
Type II, where a narrow upstand provides a permanent formwork to the floor slab and
is considered monolithic with the in situ concrete infill at the ends of the floor slab
(Figure 5.4a).
In type I beams, the minimum width of the upstand should be approximately

= - . The ledge width is the sum of the nominal slab bearing length ( ),
Deep spandrel edge beam to support double-tee slabs.
Edge spandrel (deep L-beams) supporting hollow core floor units. (a) Site construction dowelled
onto a column head with corbel. ( )
(b)
(c)
( ) Edge spandrel (deep L-beams) supporting hollow core floor units. (b) End detail
showing dowel holes; (c) Site construction using a narrow plate connector (see
Section 10.2.4) to face of columns.
a fixing tolerance ( ) and the clear space for in situ infill ( ), giving a total dimension
of . Thus, the minimum breadth of a type I beam is about . The precast
upstand width in type II beams is - , and the minimum breadth is about .
Minimum depth is often determined by the size of the connector in the end of the beam.
The minimum depth would be equal to the depth of the floor slab plus the minimum boot
depth of .
The design of beams is based on ordinary reinforced or prestressed concrete principles
for specified loads and support conditions. Support conditions may be simple or continuous.
Site placed tie bars Projecting

tie bars
Projecting stirrups
In situ concrete infill
‘Upstand’
Bearing ledge
Shear stirrups
‘Boot’
Part of floor slab Boot links
ignored in design
(a) (b)
L-section edge beams: (a) composite (left) and (b) non-composite.
Continuous and semi-rigid beam end connections. (Tested by Elliott and Ferriera at Nottingham
University, Nottingham, UK.)
Recent research (Elliott et al. 2005) has shown that semi-rigid supports are almost
equally as effective as fully rigid supports provided the fixity factor g is around
(pinned g = and rigid g = ) as shown in Figure 5.5. The ultimate failure bending
moment in these tests was M u = −177 and −184 kNm, yielding a mean bending ratio
, about of the balanced moment of
resistance using in a position of limited end rotation according to EN 1992-1-1,
clause 5.6.3(2). In fact, the test exceeded the predicted based on the as-measured yield
strength of 3 no. H16 top rebars by , showing that semi-rigid beam
supports may be conservatively designed.
Unlike monolithic r.c. design, where the cross section and reinforcement are designed to
satisfy project requirements, the reverse happens in precast concrete design. A predetermined
set of standardised beam sections is selected by a manufacturer according to the requirements
of most building structures. Flexural and shear reinforcements are computed for the optimum
reinforcement quantities appropriate to each size of the beam. Standardised designs are pre-
pared in advance for beams that may vary only in depth, breadth and quantity of reinforcement.
Simple computer programs or spreadsheets are used to do this. The following sections will illus-
trate these methods.
Although a designer is able to specify any grade of concrete, in practice, a manufacturer
would want to restrict this to two grades, one for r.c. work and the other for prestressed
work. For practical reasons of demoulding and detensioning, grade C32/40 is used for r.c
beams and C40/50 to C50/60 for prestressed beams. Similarly, one type of rebar is used –
high tensile deformed bar of . Rebars with is available in
some countries, as shown in Figure 5.6, note the deep ribbing to improve bond by wedging.
Although mild steel is perfectly suitable, its differential cost (compared with HT bar) and
smooth surface (difficult to make stable cages) make it less attractive. Mild steel is used for
projecting dowels which must be hand bent into position on site. One type of pretensioning
bar is used – a seven-wire helical strand between f pk = 1750 and 1860 N/mm 2 . Pretensioning
wire would not be used because of the large force demand in beams.
Cover to reinforcement must satisfy fire and durability requirements. The usual approach
is to fix the cover distance for the various faces of the beam and to quote these properties
as part of the beam specification. External surfaces having exposure XC3 usually have
a cover of 30 or 40 mm, whilst internal (protected having exposure XC1) surfaces have
a 25 mm cover. The clear distances between bars in tension should satisfy EN 1992-1-1,
clause 7.3.3 and code Table 7.3. For example if the steel stress under the quasi-permanent
combination load is (i.e. ) and crack width is limited to , bar
spacing £ . Minimum quantities of reinforcement according to clause 9.2.1.1(1) are
as follows:
(5.3)
where bt = mean width of the tension zone. For T-beams with the flange in compression,
the width of the web is taken into account for calculating the value of . For example
if , and = ´ = > .
High tensile rebar .
Sections containing less reinforcement than should be considered as unreinforced (see

EN 1992-1-1, Section 12). The maximum area of tension or compression reinforcement is
based on the gross section of concrete as £ outside lap locations.
Non-composite construction utilises the properties of the basic beam alone. For specified
cross section and flexural and shear reinforcement patterns, the following may be calculated:
1. Ultimate moment of resistance

2. Ultimate shear resistance
3. Torsional resistance
4. Bearing ledge resistance
5. Flexural stiffness ( deflection limit)
The reinforcement quantities are curtailed according to the distribution of design moments
and shear forces. Figure 5.7a and b show a typical and actual reinforcement pattern for
an L-beam. This beam will be used to demonstrate the design procedures in the following
sections.
A B C
4. 7.
7. 6.
6.
9. 11. Links continue
8.
10. 3.
1.
2. 5. 3. 1,2.
12.
A B C
7. 7. 7. 7. 7. 7.
9. 11. 11.
6. 6. 6. 6. 6. 6.
8.
5. 5. 5. 10. 10.
1. 1. 1. 2. 1. 3. 3.
1. 2. 1.
(a) A-A B-B C-C
(a) Typical reinforcement details in non-composite L-beams. ( )
200 345 170
07 06 07 06
17 02 17 02
07 06 07 06
585
07 06 07 06
01
01
12 12 12 12
1285
07 06 06
Mould face
40 07 40 07
Filling face
40 07
10 40
11 11
11
500
06 11
07 07 06
12 12 12 12
12 H20 16–500
03 13 Tie bar 13 16
06 13 03 06
13 15 15 15
13 13 15 15 15
200
15 15
06 06
14 05 05 05 05 14 05 05 05 05
(b) 715
( ) (b) Real cross-sectional detailing for spandrel beam. (Courtesy of Creagh Concrete,
N. Ireland.)
5.2.1.1 Type I beam

These beams may be designed using all types of floor units, that is hollow core, double-tee
and plank floors. Reinforcement at the top of the boot is ignored. Referring to Figure 5.8,
assume that the depth to the neutral axis (NA) > , then
(5.4)
bw
εcu = 0.0035
C1
hs
x
C2
N.A.
d
z1
z2
T
εs
b
Total area
bars As
Design method for ultimate moment of resistance of L-beam.
(5.5)
= - (5.6)
Fs - Fc = Fc
Hence, check X and < for under/singly reinforced sections. The limit 0.6d is not
obvious or apparent in EN 1992-1-1 (and often mistakenly taken in some text books as
0.45d where the section under consideration is not limited by rotation). According to clause
5.5(4), if the degree of moment redistribution = d = depth of the neutral axis at the
ultimate limit state after redistribution, then
d= + £ (5.7)
where from the NAD , for = + e = , then for no moment redis-

tribution d = and therefore £ . Note that for concrete > ,
the term (for the depth of an equivalent rectangular stress block) is replaced by
- - .
Returning to the section analysis:
= - = - - (5.8)
= + (5.9)
If < , then < , the section is rectangular and
= { - } (5.10)
(5.11)
Note that the limit is not given in the Eurocodes but is used in the same way as in BS
8110 to ensure that the depth compression is not less than - = – roughly
the size of a piece of large aggregate.
If > , the beam should be doubly reinforced, or the value of reduced to remain
singly reinforced.
(5.12)
= ´ - (5.13)
= - = - (5.14)
Fs¢ = f yk As - Fc1 - Fc2

(5.15)
¢
Fs = f yk As - f ck d - b - bw hs
Then
¢= ¢ ( - ¢) (5.16)
= + + ¢ - ¢ (5.17)
However, adding compression reinforcement of any reasonable area, for example 2 no.
H16, to the top of the upstand of the beam is often not practical because of the limited space
between the bars. It is better to increase the compressive resistance of the concrete by utilising
cast in situ infill concrete at the ends of the slabs. This gives rise to Type II beams.
5.2.1.2 Type II beam

These beams may be designed using only those floor types that permit the placement of fully
confined in situ infill, that is hollow core and plank floors. Double-tee units are not permit-
ted. In order to be economic and obtain maximum , the strength of the infill concrete
should be similar to that used in the beam; however, a lower strength, say
may be suitable. Referring to Figure 5.8, the effective breadth of the compression zone is
equal to the beam breadth minus the slab bearing length, usually taken as 75 mm but may
be greater. The calculation is as mentioned earlier except that in Equation 5.5, is replaced
with and is the strength of the in situ infill concrete (not the precast beam).
As As M Ed
This is the reverse calculation of that given earlier and is sometimes used in bespoke pre-
cast for the design of adhoc beams. The following procedure is not explicitly given in EN
1992-1-1 but implied from clause 3.1.7(3) and code Fig. 3.5 for £ .
Letting K = M Ed /fckbd2 < 0.206* for singly reinforced sections (5.18)
= +Ö - (5.19)
= - (5.20)
If < , the section is rectangular and
(5.21)
Otherwise the beam is L-shaped or inverted T-shaped such that the virtual moment of resis-
tance obtained from the area of the recess - is added to as follows, modifying
Equation 5.18 to
= + ( - ) - (5.22)
* when = = - ´ = .
Followed by Equations 5.19 and 5.21.

If > , is limited to - ´ = .
Compression steel is added to cater for the additional moment capacity, and the section
is doubly reinforced:
-
¢= (5.23)
- ¢
= + ¢ (5.24)
The design for shear follows the normal procedures for r.c. sections – that is the ultimate
shear capacity is the sum of the concrete resistance ( aggregate interlock + dowel action) plus
the ultimate ( yield stress) capacity of the shear stirrups. The big difference to the approach
in EN 1992-1-1 compared to BS 8110 is that these three variables are combined in a single
function known as the variable strut inclination (VSI) method. In BS 8110, the inclined
shear plane was taken at q = to the horizontal irrespective of the strength of concrete
and the breadth of the web, both of which affect the relationship between bending and shear
stresses and hence the inclination of the shear crack, EN 1992-1-1 allows q = ,
or q= . This is shown in the truss model in EN 1992-1-1 Fig. 6.5. Note the
assumption that the lever arm between the top and bottom truss is , which is a
reasonable assumption where bending moment is small in the presence of shear, but not, for
example in cantilevers where shear and bending are both large.
In L-beams, the effective breadth of the web used in shear calculations depends on
whether the NA of the section occurs in the upstand or in the boot of the beam as shown in
Figure 5.9. If the NA lies in the upstand (Figure 5.10a), then upstand width.* However,
if the NA lies in the boot the critical section may lie either in the upstand or in the boot, and
both cases must be considered as shown in Figure 5.10b and c as follows.
Cast in situ infill

into milled slots
In situ infill Interface shear in floor slab
concrete reinforcement
Floor slab
Interface shear
reinforcement
Bearing length Compression in

top of boot
Interface shear rebars or loops between L-beam and floor slab.
* Note that the upstand width is used in these shear calculations, whereas is used for the same parameter in
flexural calculations.
Shear stress
distribution
parabolic
bv
Maximum
x
hs
N.A.
Rectangular
(a)
bv
b
Maximum
Maximum
(b) (c)
Principles of shear stress in reinforced L-beam. (a) where < , (b) where > , (c) where
just > but the maximum shear stress is at the bottom of the upstand.
Elastic shear stress distribution function t = is used to determine the shear stress
at the two critical sections. Referring to Figure 5.10, the critical sections are at a level at
(1) the top of the boot, and (2) the NA. In this calculation, S is the first moment of area above
the critical section, I is the second moment of area of the whole beam ( using transformed
section), b is the effective breadth at the critical section and V is the shear force. We are not
interested in the actual value of τ, only where it is a maximum. Because and I are con-
stant, we require the maximum value of the term .
At the level at the top of the boot,
(5.25)
At the NA,
æ ö
- - ç - ÷
= è ø (5.26)
But let in the limit, then Equation 5.24 reads

At the NA,
S æ b ö (5.27)
= d - ç - v ÷ hs d- hs
b è b ø
Values of S /bd 2 at the top of the boot

hs /d 0.1 hs /d 0.2 hs /d 0.3 hs /d 0.4 hs /d 0.5 hs /d 0.6
0.005 19 pt 0.045 0.080
Values of S /bd 2 at the neutral axis

bv /b hs /d 0.1 hs /d 0.2 hs / d 0.3 hs /d 0.4 hs /d 0.5 hs /d 0.6
0.1 0.036 0.023 0.018
0.2 0.052 0.040 0.036
0.3 0.068 0.058 0.054
0.4 0.075 0.072
0.5 0.093 0.090
0.6 0.110 0.108
Given values of , the maximum value of may be found. Equations 5.25 and
5.27 may be further simplified by considering the ratios . Tables 5.1 and 5.2
gives the values of for the two cases for typical values of . Critical values
are given in bold. From these data, it is possible to determine where the critical shear section
should be taken. Generally it is found that the critical shear section lies at the top of the boot
when > .
Note that for a rectangular section with . Therefore,
for the most common type of L-beam with , repre-
senting a reduction in the maximum shear stress.
It would be very unusual not to use shear reinforcement in beams, although some minor
elements where shear failure is neither likely nor critical may rely on the dowel action of the
main bars fully anchored at the support, and some wet case floor slabs do not have shear
reinforcement. However, the section may be checked for as follows.
6.2.2.(1). Fig. 6.3. Calculated where bars extend + beyond the section considered.
If bars extend the bond length beyond face of support, the shear is calculated at d from
face of support.
6.2.2.(1) = × (5.28)
but ³ Ö (5.29)
where = + < (5.30)
If > , designed shear reinforcement must be placed in the appropriate section. Shear
stirrups will be placed in the upstand of the beam as shown in Figure 5.7. Links must also
be provided in the boot of the beam to carry the bearing ledge forces, as explained in
Section 5.2.4. In the latter, shear stirrups must be provided in the upstand or in the boot
depending on where the effective breadth is taken. Designed shear stirrups must be addi-
tional to those required for torsion of bearing ledge reactions.
VRd, s
z = 0.9d
S
ASW per pair of legs
VEd
Variable strut inclination θ
(a) (of depth z) determined from (b)
Compression zone
VRd,max z = 0.9d
σc
θ θ
z tan θ
(b) VEd
Shear design in beams: (a) shear stirrups capacity VRd ,s. (From BS EN 1992-1-1, Design of con-
crete structures – Part 1-1: General rules and rules for buildings, BSI, London, Exp 6.8.) and
(b) compressive strut capacity VRd,max (From BS EN 1992-1-1, Design of concrete structures –
Part 1-1: General rules and rules for buildings, BSI, London, Exp 6.9.)
Referring to Figure 5.11 and to EN 1992-1-1, clause 6.2.3, the number of stirrups crossing
the shear plane . The ultimate force in each stirrup = area × stress ,
then the ultimate resistance of the stirrups is
= q( ) (5.31)
Note there is no in this analysis, and is not added to (as was in BS 8110).
Therefore, the area of designed shear reinforcement ( run along the beam)
is given by
Exp. 6.8 ³ (5.32)

q
The area of shear reinforcement must be appropriate to the critical shear section; in this
case, it is assumed this is at the top of the boot. Clearly, the VSI q has a big influence on
; it can alter it by 250%! The expression for q derives from the compressive strut,
shown in Figure 5.11 and code Fig. 6.5, equal to the compressive stress s acting perpendicular
to the shear plane on the breadth b × projected length z q. The vertical component of this
is q q. The compressive stress is given as s =
where a = for non-prestressed sections
Exp. 6.6 shear stress parameter = - and
= q q (5.33)
but q q= q q= q q= q + q = q+ q
and q q= q
a
Exp. 6.9 = = q (5.34)
q+ q
Rearranging and replacing with calculated at a distance d from edge of the support
- æ VEd ö
q= ç ÷ \ q (5.35)
è b z v1 f cd ø
but £ q£
For example let , , , and
-
, then q = ´ ´ ´ ´ = .
< = ´ ( ´ ´ ´ )
Use H10 at spacing (785)
The minimum area of links is some function of concrete tensile strength Ö and b (or
in a web or upstand) as follows:
Exp. 9.5N ³ (5.36)
The maximum area of links is given when the concrete stress reaches , as
a
Exp. 6.12 £ (5.37)
The minimum spacing of links is not specified in codes, but practically is about 50 mm for
small links (8–10 mm dia.) or about 70 mm (12 mm dia.). The maximum spacing of links is
as always 0.75d to avoid a ° shear plane passing between two links as follows:
9.2.2.(6) £ (5.38)
Finding Vrd , s knowing Asw and s.

Rearranging Exp. 6.8 and 6.9
q= - (5.39)
Exp. 6.8 = q (5.40)
Example 5.1
Calculate and of the deep wide L-beam as shown in Figure 5.4b.
The upstand is deep wide. Main steel at mid-span is 3 no. H25 bars
reducing to two H25 at the supports. Shear stirrups are H10 bars at 100 mm spacing. Use
and . Cover to the stirrups is 40 mm. Check the
contribution to shear of the concrete and dowel action .
Solution
Flexure
3 no. H25 bars are present at midspan
= - - - =
> >
= ´ ´ ´( - )= - ´
Then - ´ = ´ - ´ = ´
\ =
= <
Lever arm = - = = - ´ - =
= ´ + ´ =
(It is interesting to note that in the first ed. of this book using and
to BS 8110, )
Shear
At the support, 3 no. H16 longitudinal bars ( ) and H10 stirrups at 100 mm
spacing are present, .
= - - - =
From Tables 5.1 and 5.2 and Equations 5.5 and 5.27, using , and
, it is found that the critical shear section lies at the NA. However,
at the support, X will be small and the critical shear will be at the top of the boot.
, , = - = ,
.
Using Equations 5.39 and 5.40
´
q= - = < = °
´ ´
-
= ´ ´ ´ ´ ´ ´ =
From Equation 5.28 through 5.30

-
= ´ ´ ´ ´ ´ ´ =
-
= ´ Ö ´ ´ ´ =
where and = + = <
(It is interesting to note that in the first ed. of this book to BS 8110, with the
concrete contributing , both is close agreement)
EN 1992-1-1, Section 7.3 offers two alternatives to crack control by limiting the size
and spacing of rebars, either (1) by direct calculation in clause 7.3.2 for the minimum
allowable tensile force in the reinforcement s or (2) using tables without direct
calculation.
7.1 s = (5.41)
where
s = = maximum stress permitted in the reinforcement immediately after formation
of the crack, or less if s is limited by the stress in code Table 7.3 (see later)
without axial tension
for webs with £ or flange widths < , to for webs with
³ or flange widths > , intermediate values interpolated
, if cracking expected < days
= - = area of concrete within the tensile zone just before formation of the first
crack (may be taken as in most cases)
In Example 5.1 the reinforcement is fully stressed at , then

´ = . This is less than from Equation 5.3
(showing that from code Exp. 7.1 is probably only critical where
the section is reinforced too much (such that s < ).
Using tables. The design method given in clause 7.3.4 for the calculation of crack width
may be replaced by using Tables 7.2 and 7.3 (Table 5.3 in this book) by restricting the bar
diameter or spacing as a simplification. Where the minimum reinforcement given by 7.3.2 is
provided, crack widths are unlikely to be excessive if
Maximum bar spacing for crack control
N/mm 2 wk 0.3 mm wk 0.2 mm

160 300 200
200 250 150
240 200 100
280 150 50
320 100 —
360 50 —
• For cracking caused dominantly by restraint, the bar sizes given in Table 7.2 are not
exceeded where the steel stress is the value obtained immediately after cracking (i.e. s
in code Expression 7.1).
• For cracks caused mainly by loading, either the provisions of Tables 7.2 or 7.3 are
complied with. The steel stress s should be calculated on the basis of a cracked
section under the relevant combination of actions, that is dead load plus quasi-
permanent live load.
The steel stress is calculated using the partially cracked flexural stiffness according to EN
1992-1-1, clause 7.4.3(3). Members which are not expected to be loaded above the level
which would cause the tensile strength of the concrete to be exceeded anywhere within
the member should be considered to be uncracked. Members which are expected to crack
but may not be fully cracked will behave in a manner intermediate between the uncracked
and fully cracked conditions and, for members subjected mainly to flexure, an adequate
prediction of behaviour is given by code Expression 7.18:
7.18 a = za + -z a (5.42)
where
α is the deformation ‘parameter’ considered, which may be, for example a strain, a cur-
vature, or a rotation
a and a are the values of the ‘parameter’ for the uncracked and fully cracked condi-
tions, respectively
where ζ is a distribution coefficient (allowing for tensioning stiffening at a section), where

z = - b s s , where s is the stress at first cracking and s is the applied service stress.
For situations of bending stresses and cracking, Equation 5.42 can be more easily and
alternatively expressed in terms of the flexural rigidity I and the cracking and
applied moments as follows:
Effective rigidity
= + - b £ (5.43)
where
Flexural ridigity, uncracked
Flexural ridigity, cracked
Ecm
Ec ef =
+ j(t,to )
where
j is the creep coefficient
b= for long-term loads when the section is flexurally cracked, or b = for short-
term loads at installation
is the cracking moment of resistance
is the elastic section modulus at tension face = -
is the applied service bending moment
Therefore, is the cracked rigidity plus a proportion of the difference - depend-

ing on the ratio by which the applied exceeds the cracking . The power 2 term sug-
gests that the moment versus curvature relationship - is parabolic after cracking, and
of course linear before cracking where . For I beams with narrow webs, the power
term can be as large as 5 or 6 as the cracks develop rapidly up narrow webs and -
is much flatter.
Flexurally Uncracked Section

Transformed area of steel = - at d from the top, where the modular ratio
Second moment of area of concrete alone
Depth to the centroidal axis
Ac h + As m - d
xu = (5.44)
Ac + As m -
\ + - + - - (5.45)
Flexurally Cracked Section

Transformed area of steel =
Depth to the centroidal axis
b xc + mAs d
xc = (5.46)
bxc + mAs
\ + - (5.47)
(if top steel is present, the aforementioned terms will have added expressions
and - ¢ )
Generally, it is not necessary to calculate the deflections as simple rules, for example limits
to effective span/depth l/d ratio are given in EN 1992-1-1, clause 7.4.2(2), generally referred
to as “cases where calculations may be omitted.” l/d expressions only hold true for fairly reg-
ular rectangular slabs or T-shaped beams, otherwise moment versus curvature as mentioned
earlier would be used. Limiting values for l/d are given in code Exp’s 7.16a and 7.16b, and
multiplying this by correction factors allows for the type of reinforcement used and other
variables, for example provided/ required, up to a limit of 1.5. The maximum l/d
under any circumstance is 40. For spans greater than 7 m which support partitions liable
to be damaged by excessive deflection, l/d is multiplied by the ratio 7/l ( effective span in
metres). For flanged beams where the flange/web breadth > , l/d is multiplied by 0.8.
As usual, l/d, a function of beam curvature, depends on the required reinforcement ratio
r= , a concrete parameter r = Ö , and a factor K (see code Table 7.4) to take
account of the type of span, for example simply supported = 1, cantilever = 0.4, etc. (do not
confuse this with ). If r £ r (often the case for slabs or wide flanged beams), l/d
will result as follows:
7.16a = é
ë + Ö r r + Ö [r r - ] ù
û r£r (5.48)
If r £ r and the section is designed for compression reinforcement (not nominal) with
r¢ = ¢
7.16b = éë + Ö [r r - r¢ ] + Ö Ö r¢ r ùû r>r (5.49)
Example 5.2
Calculate the limiting bar spacing for the L-beam in Example 5.1. Use ,
and long-term creep coefficient j = . The beam
also contains 2 H10 bars at the top of a 50 mm cover. The service moment may be
taken as (based on from Example 5.1). The exposure
class is XC1. Determine the maximum allowable effective span using l/d ratio as the
controlling parameter.
Solution
= + =
Uncracked
- = ´ ´ = = from the top
- ¢= ´ ´ = ¢=
Then
= ´ - = ´
Then
b = ´ =
Cracked
from the top
Then (note at ultimate , only 4 mm difference)
= + éë - ´ ùû = ´
s = - = ´ ´ ´ - ´ =
Table 5.3 for gives spacing £ . OK

Also EN 1992-1-1, Table 7.2 allows 25 mm dia. bars. OK
l/d ratio using and
-
r =Ö ´ = r= = ´ = r r= <
Equation 5.48. = + ´Ö ´ =
£ ´ =
It is interesting to check the deflection due to using .

Assuming uniformly distributed loading (UDL):
d= = ´ ´ ´ ´ ´ ´ =
Limiting d = = = , in good agreement.
The boot of the beam must be reinforced using links around the full perimeter of the boot.
If the depth of the boot is less than 300 mm, it should be designed as a short cantilever in
bending. Otherwise the behaviour is nearer to the strut-and-tie action. The bending method
gives a slightly greater area of tie back steel (about 5%–10%).
As with all projecting nibs (a nib is a short bearing ledge), it is first necessary to preclude
a shear failure at the root of the nib. The enhanced shear stress given in EN 1992-1-1, clause
6.2.2(6) usually takes care of any vertical shear problems. If not, then the depth of the boot
should be increased in preference to providing shear reinforcement. A ‘shallow’ boot is
where the lever arm a in Figure 5.12 is greater than 0.6 d″ (i.e. d″/a ≤ 1.67), where d″ is the
effective depth to the steel at the top of the boot from the bottom of the beam. Otherwise,
the nib is classed as ‘deep’. This information is not explicitly mentioned in EN 1992-1-1
but can be found in the Published Document PD 6678, clause B.4.1 (PD 6678 2010) under
corbels, where the angle of the compressive strut is more relaxed and should be less than
tan θ < 2.5, or d″/a ≤ 2.5. These differences, that is 2.5 versus 1.67, will make little difference
to the design of the reinforcement in the boot.
Referring to Figure 5.12, the compressive strut in the boot is inclined at θ to the vertical,
-
where q = , where = ¢¢ - is the centre-to-centre distance of the boot link, and
c the edge distance to the centroid of the steel bar at the top of the boot.
If the floor slab is placed in direct contact with a shallow bearing nib, a horizontal force
resulting from possible contractions or other movement (e.g. thermal effects) of the floor
slab relative to the beam will develop at the interface. Referring to Figure 5.12, if the floor
bv lb
a
V Shrinkage contration
c V μV or other movement
μV
c
H
μV
T θ
d˝
Boot reinforcement design in L-beams.
reaction is V per unit length of the beam, the horizontal force is μV, where μ is the coefficient
of friction between two concrete surfaces taken as 0.7. Thus, the horizontal tie force is
= q+[ + ]m (5.50)
The horizontal bars placed at the top of the boot must satisfy
(5.51)
The bars are formed into links but do not contribute to vertical shear strength of the beam
unless the boot is sufficiently deep that > (see Tables 5.1 and 5.2). The upstand width
is fairly small, typically 150 mm, such that the bars at the top of the boot must extend a full
anchorage length in the rear face of the beam. This means that the bars are stressed beyond
a point which is more than four diameters from the corner of the bar, and the bend radius
must be checked so that the bursting stresses caused by small bend radii are not a problem.
The usual practice is to provide H8 or H10 links at a spacing ¢¢ + < < to
EN 1992-1-1, clauses 9.3.1.1(3) and 9.3.2(4).
The compressive strut force is
= (5.52)
q
which must be resisted by a compressive strut in the uncracked part of the nib. The uncracked
zone may extend to a point, based on the maximum depth to the NA, at 0.6d″ from the
bottom of the beam. The limiting compressive strength of the concrete, which is subjected
to transverse tension at the bottom of the beam according to clause 6.5.2(2), is
6.56 (5.53)
Thus, the strut capacity per unit length of the beam based on ´ = is
as follows:
= - ¢¢ q (5.54)
Combining Equations 5.52 and 5.54, an upper limit for the vertical reaction V is
= - ¢¢ q q (5.55)
The vertical force T in the stirrup in the main body of the beam is given by
= + (5.56)
where a is the centre distance from the stirrup to the line of action of the slab reaction. If the
floor slab is fully tied to the beam using reinforced in situ strips capable of generating the
frictional force μV, then this force may be ignored in the aforementioned design, in which
case . The area of one leg of vertical stirrups is
(5.57)
This steel must be in addition to any design shear requirement.

In a deep boot, the floor slab reactions would be carried directly into the web of the beam
by diagonal strut action assuming that q = °. If the level of the bearing surface is above the
NA, the only steel required would be the horizontal steel .
In fact, the design of all the earlier reinforcement should be carried in two stages, before
and after in situ concrete has been added to the ends of the slab. This is because the in situ
concrete increases the bearing length to the full ledge width and hence reduces the lever
arm a. Before the in situ concrete is added, the lever arm is = + - - , and the
slab reaction is due to the self-weight of the slab plus the in situ concrete infill. Afterwards,
= + - , and the slab reaction is due to superimposed dead and live loading.
Example 5.3
Calculate the minimum bearing ledge capacity of the boot of the L-beam in Example
5.1 and shown in Figure 5.7(a). Assume that the line of action of the force is at the mid-
point of the bearing ledge, that is at from the edge. Use boot links
. Cover to the inside face of the upstand = .
Solution
Boot links are at 150 mm spacing
Minimum upstand links are H10 at 300 mm spacing
¢¢ = - - =
= - =
= + + =
-
q= = °
Horizontal tie steel capacity

-
= ´ ´ ´ ´ =
Equation 5.50, changing the subject,
VRd = =
°+ ´ +
Equation 5.54, changing the subject, = ´ ´ ´ ´ ´ °

°´ - = (clearly not critical)
Vertical force in the stirrups in the upstand
-
= ´ ´ ´ ´ =
Equation 5.56, changing the subject,
= =
+ ´
Thus, the minimum bearing ledge capacity is 93.6 kN/m run.
Precast reinforced beams may act compositely with certain types of floor slabs, such as hollow
core and plank units, by the introduction of appropriate interface shear mechanisms and cast
in situ concrete infill as shown in Figures 5.13 and 5.14. The projecting stirrups in Figure 5.13
are not closed over the top of the beam but are hooked over to comply with EN 1992-1-1,
Fig. 8.5a where the straight part of the hook should be ³ F (in this beam, they were 7Φ). This
helps to handle the site reinforcement into position not only to form a composite beam, but
also to provide negative continuity (as was shown in Figure 5.5). The projecting stirrups in
Figure 5.14 are closed and anchored into the precast beam, but the site-placed bars are posi-
tioned inside the cage before erection and manhandled through the sleeves in the column.
Typical details are shown in Figure 5.15. It is usual, but not obligatory, for only internal
beams to be designed compositely as there is rarely a need to enhance the strength of edge
beams in this way.
Flexural and shear reinforcement in prestressed rectangular.
Composite beam design with negative moment continuity through sleeves in columns.
Site placed straight

or hooked bars placed Hollow core
in milled slots floor units
Projecting dowels
or loops act as
shear connectors
(a) Precast beam
Interface shear reinforcement in composite beams using (a) single dowel and slotted holes in
hollow core unit (hcu). ( )
Intermittent
milled slots
Chamfered end to
hollow core unit
(b)
Continuous mesh in top of

cast in situ concrete topping
Projecting loops to
enclose top steel
Precast soffit
or plank unit
(c)
( ) Interface shear reinforcement in composite beams using (b) projecting loop into slot-
ted holes in chamfered hcu, (c) projecting loop into the topping of half slab.
The characteristic of the in situ infill concrete is either 20 or . The main

benefit in using a composite beam is to increase the flexural strength and stiffness ( reduce
deflections). These must be carefully considered against the additional cost and design
responsibility, particularly if the floor slab is designed by other parties. For these reasons,
composite r.c. beams are not often used – unlike composite prestressed beams which have
greater benefits (see Section 5.5).
It is necessary to reinforce the cast in situ concrete such that it will develop full design
strength . In the case of hollow core units as shown in Figures 5.14 and 5.15a, the
milled slots over the tops of all the cores are removed for a distance of approximately 500 mm
to receive site-placed tie steel bars, usually of 10 or 12 mm diameter with .
The tie steel may be loose bars or bent bars projecting from the beam. The tie steel also
serves several other functions, including diaphragm action (Section 8.4) and stability tie
steel (Section 11.4), but in this context, a steel area of 0.2% of the transverse area has been
found by experimentation to be adequate (Lam et al. 2000, Lam 2002, Fu and Lam 2006),
for example H12 bars at 300 mm spacing for floor depths upto 200 mm. Flooring manu-
facturers should be consulted over the practicalities of opening cores as the end of the floor
unit may become unstable where many cores are opened.
The ends of the hollow core unit may be chamfered with sloping ends, as shown in
Figure 5.15b, to benefit the placement of in situ infill. The chamfer is usually around
deep – manufacturers will provide exact details as shown in
Figure 5.16. In this case, not all of the hollow cores will be opened as slots. Experimental
results show that opened cores at 300 mm centres are sufficient.
Full interaction between the in situ concrete in the cores and the precast hollow core unit is
assumed. The full depth of the slab is used in design. The effective breadth of the flange is taken
as equal to the actual length of the filled cores, and this is equal to one bond length for the
straight transverse steel that is placed in the core, that is 40.4 diameters for
bar in infill, according to EN 1992-1-1, clause 8.4.3 and 8.4.4. If the length
Sloping ends to hollow core units enable placement of infill around shear connectors to
beams. (From Bison Manufacturing Ltd., Swadlincote, United Kingdom.)
of the slot becomes excessive, say greater than , the ends of the transverse bar may
be hooked. The bars should be placed at the mid depth of the slot; however, due to lazy
site practice, they tend to rest at the bottom of the core at less than 50 mm from the bottom
(the effect of this is not known).
In the case of solid plank flooring, Figure 5.15c, transverse tie steel will automatically
be present as part of the topping/floor design, but a minimum area ratio of 0.2% is recom-
mended (this ratio is not in the Eurocode) as mentioned before. The positions of the trans-
verse bars are more accurate than in the case of hollow core units, with the top cover being
about 50 mm. Full interaction between the precast plank and the in situ topping is assumed.
The full depth of the slab and an effective flange breadth of 1/10 of the simply supported
span of the beam, , are used in design.
Composite beams are not designed for vertical shear (reinforcement is designed for the
noncomposite section). However, interface shear calculations for shear forces due to imposed
loads should be made according to Section 4.4.4. except that where the bars are placed hori-
zontally between the upstand of the beam and the floor slab (Figure 5.31), the resistances
due to cohesion and friction m s are considered not to be effective in a state of flexural
compression. The entire shear resistance is carried by the bars such that the area of interface
shear bars is m . This ensures a reasonably high value for μ and the beam con-
tact surface is either heavily roughened (scabbled after demoulding) or provided with shear
keys (indentations conforming with BS EN 1992-1-1, Fig. 6.9); then m = . Note that the
contact breadth may be small, typically 150 mm in 300 mm wide beams, resulting in large
interface shear stress. Interface reinforcement is always used (irrespective of the value of )
in the form of loops or dowels, as shown in Figures 5.13 through 5.15.
Flexural design is carried out in two stages and the resultant effects are superimposed elas-
tically, even though the ultimate design is plastic. At stage 1, the self-weight of the beam,
precast floor units and the cast in situ infill/topping are carried by the precast beam alone.
The area of reinforcement required for this is called . At stage 2, after the in situ has
developed full strength, the imposed dead ( services, partitions, ceilings, etc.) and live loads
are carried by the composite beam. The area of steel here is . Then the total steel area
= + . See Figure 5.17.
There is a fundamental difficulty in this approach as that part of the concrete at the top of
the precast beam may be called on to resist compressive stress at both stage 1 and stage 2.
To solve this in a rigorous manner, the strain history of the beam should be followed and the
resulting stresses found. In this case, an idealised ‘rectangular stress block’ approach is not
appropriate. However, it is found that in most cases the depth of the compressive stress block
at stage 2, given in Figure 5.17b as , is less than the depth of the slab.
Referring to Figure 5.17a, let the ultimate design moment due to self-weight of the beam
and slab before the in situ infill has hardened be . Then if the strength of the precast
concrete is :
(5.58)
from which the lever arm and neutral axis depth are found according to the EN 1992-
1-1 rectangular stress block approach. Check £ and £ . The area of steel is
(5.59)
Cast in situ strength fcki

Precast strength fck
b1 b2
X2
X1
d2
d1
As1 resulting from As2 resulting from

(a) stage 1 loading (b) stage 2 loading
Ultimate limit state design in composite beams (a) stage 1 based on precast alone when in situ
topping is wet, (b) stage 2 compositely when topping has hardened.
Let the ultimate design moment due to imposed dead and live loads be . Then if the
depth of the slab is and the strength of the infill/topping concrete is :
(5.60)
where = + , and is the breadth of the in situ topping (exclude the breadth of the
beam in stage 1). The lever arm and neutral axis depth are found according to the
EN 1992-1-1 rectangular stress block approach. Check < < and £ .
The stage 2 area of steel is
(5.61)
Then = +
The remainder of the precast beam will be designed and reinforced according to
Section 5.2. Shear forces are resisted by the precast section where discrete precast floor
elements are used to guard against a plane of weakness in shear in the joints (this does not
happen in bending where the joints transmit horizontal compression). The design procedure
is as given in Section 5.2.3.
In everyday design, deflections are controlled by the procedure of limiting the span-to-
effective depth ratio l/d according to EN 1992-1-1, clause 7.4.2, a procedure which is
entirely satisfactory for most loading conditions in singular sections. The method involves
equating beam curvature and strain distributions with a limiting deflection of span/250
according to EN 1992-1-1, clause 7.4.1(4). This cannot be adopted in a composite beam
because there are deflections due to stage 1 loads that (obviously) respond to a completely
different flexural stiffness based on together with the modular ratio for
rebars, than those due to stage 2 loads based on with and a =
. Moreover, the Young’s modulus of concrete changes with time due to creep such that
stage 1 deflections take place as the floor slabs are being positioned between 7 and 28 days
after casting typically, whilst stage 2 deflections take place over many years, 500,000 h
(57 years) being the design ‘period’. It is often forgotten that stage 1 loads continue to act
over the long term due to the viscoelastic deformation and must be considered in stage 2
calculations.
The effects of the relative shrinkage of the precast beam to the in situ concrete must
also be considered in deflection calculations, particularly in the case of plank floors where
the volume of wet concrete is large. A value of relative shrinkage strain of = ´ -
is adopted. This is not necessary when using hollow core units as the shrinkage of these
units is small and undisturbed by the in situ infill placed into individual opened cores.
The design method adopts the area–moment and partially cracked section method.
Referring to Figure 5.18a and b, stage 1 section properties for the precast beam alone
and (with subscript 1) are given in Section 5.2.6 with long-term modular ratio using
of the beam. Prior to the in situ infill hardening, the beam attains equilibrium based on
the stage 1 partially cracked section , where
= + - b (5.62)
Floor slab and

wet in situ infill
As΄ d1΄ Xc1 As΄ d1΄

Xu
d1 d1
h
As As
b1 b1
(a) (b)
Floor slab and

in situ infill
b2,ef b2,ef
As̋ at d˝
As̋
hS Xu2 Xc2
As΄
Xpc1 d2΄
d2 As΄ d2΄ d2
As
(c) (d) As
Section properties at stages of deflection in reinforced concrete composite beams. (a) Stage 1:
unracked 1, (b) Stage 1: cracked 1, (c) Stage 2: uncracked 2 (partially cracked from stage 1),
(d) Stage 2: cracked 2 . ( )
Mk1 + Mk2
Kcr2
Kef, 2
Mcr2
Ku2
Mk1
Insitu
infill added
Kcr1
Kef, 1
Mcr1
Ku1
δ1 δ2
(e)
) Section properties at stages of deflection in reinforced concrete composite beams.

(e) Moment deflection (for 1 > 1 and 2 > 2.
and the corresponding depth to the partially cracked centroidal axis is as follows:
= + - b (5.63)
where b = for long-term deflections.

Referring to Figure 5.18c, part of the beam is pre-cracked and the stage 2 uncracked section
properties and to a partially cracked depth of + are
Stage 2 concrete area = +
Stage 2 concrete centroid = + +
Transformed rebar area = + - + -
Rebar centroid = + - + -
=
[ + ] (5.64)
[ + ]
usually lies inside the slab but note that unlike non-composite beams, it may not lie close
to the stage 2 NA in the ultimate limit state.
= + + - + -
+ - - + - - (5.65)
where long-term modular ratio based on an assumption as the mean of the in situ
infill and precast concrete floor slab (approximately equal share in terms of stiffness). Note
m for top steel and is - because the bars replace their own areas of concrete,
whereas the bottom steel, being in the cracked portion, does not. are tie bars at the top
of the in situ infill provided for diaphragm action and robustness (See Chapters 8 and 11)
and may be ignored without loss of accuracy, particularly if a small-diameter helical strand
is used.
Referring to Figure 5.18d, the stage 2 cracked section is
+
= (5.66)
+
If £ (the most likely condition unless the beam is very deep and the slab shallow, in
which case composite action is probably not worth the extra effort)
= + + - + - - (5.67)
Note that the cracked section can be much closer to the quasi-uncracked section
(because it is already cracked from stage1), unlike in fresh one-stage sections where < .
For UDL acting on a beam of effective length l, referring to Figure 5.19, the mid-span long-
term deflections are as follows:
Stage 1. d = (5.68)
b
X
d
h–X
Definition of terms for deflection check in composite beams.
Stage 2. d = - (5.69)
d +d <
where
is stage 1 dead load
- is the stage 2 quasi-permanent combination of load
The value of is appropriate to the loading stage, that is at stage 1, it is , and, at

stage 2, it is , where = +j , where j is the creep coefficient for the
beam and composite section, respectively.
Example 5.4
The composite beam shown in Figure 5.20 is simply supported over a span of 6.0 m.
It carries 200 mm deep hollow core slabs that have a span of 6.0 m and a self-weight
(including in situ infill) of . The imposed floor loading for offices is
live and dead. Indoor exposure class is XC1. Calculate the area of reinforce-
ment required to satisfy the ultimate bending moment at mid-span, the shear stirrups at
the supports and the mid-span deflection at mid-span for the short-term deflection due to
self-weight and the long-term deflection due to full service load.
Use and for the precast beam, and
and for the in situ infill, and . Bearing length
at the bottom of the beam (no recess or pocket).
The creep coefficient for the precast beam from a maturity of days equivalent
after heat curing to installation at days j , and long-term j for the
composite section shall be calculated according to Section 4.3.4).
Solution
Using EN 1990, Exp. 6.10(b), where g = and g = (shown to be critical over
Exp. 6.10(a), where g = and g = ´ = ).
200 mm deep hollow core unit

with 450 mm long milled slots
450 100 450

170
200
400
40 cover to H10 links
300
( ) gf ( )
Stage 1 loading
Self-weight of floor slab 3 kN/m2 6 m 18.00
Self-weight of beam 0.4 0.3 25 3.00
125
.
Stage 2 loading
Live 5 kN / m2 6 m 30.00 150
. 45.00
Dead 1 kN/m2 6 m 6.00 125
. 7.50
Exp. 6.10(a) gives ´ + ´ = <
Quick estimate for bar size . Average z for a 500 mm

deep beam . Then, . Try 2 no.
H32 + 1 no. H25 (2099).
Stage 1 flexure
Try + bars with mean
= - - - = , and
= +Ö - = <
= = - = <
Then
Stage 2 flexure
= + = = + + =
= >
= < <
Then
Then total
= + =
Use 2 no. H32 ± 1 no. H25 bars ( )

(Note that using and g = and 1.6 to BS 8110,
but requiring bars, 15% more)
Shear reinforcement
Shear span = - - ´ =
at d from edge of support
Using
= ´ - ´ =
-
q= ´ ´ ´ ´ ´ = ° °= >
run per leg
= ´ ´Ö ´ ´ = per leg
Use H10 at 250 mm spacing (314)

Reduce to 150 mm spacing in the region of end connectors for a distance of 450 mm.
Creep coefficients
Notional size : for precast beam and for composite beam
for the beam , and for the hollow core unit and infill mean of
25 and the hollow core unit (taken as
= = ´ + =
a = and (composite) a = and 0.949, a = and 0.877
Long-term RH for exposure class XC1 =50%, at installation
j long-term (beam) and (composite) 1.485. j at installation (beam)
b = and b =
b at a maturity of 5 days and b at 28 days
b long-term and 819 days, b at installation 502 and 845 days
b long-term (beam) and (composite) ,b at installation
Creep coefficients:
j at installation for the beam
j long-term for the beam
j for composite
Short-term deflection
Stage 1. Beam and hollow core unit
j = = = =
= = = = + + =
. Section cracked, use effective stiffness
= = ´ b= short-term
= ´ + - ´ ´ = ´
Deflection at stage 1.
d = ´ ´ ´ ´ =
Long-term deflection
This is due to stage 1 loads, which continue to act, and stage 2 loads.
Stage 1. Beam and hollow core unit
j = = = =
. Section cracked, use
= = ´ b=
b = ´ =
= + - ´ = ´
Then
= + - ´ =
d = ´ ´ ´ ´ = (increased by 4.4 mm)
Stage 2. Beam with composite hollow core unit
j = = = =
Concrete
a = =
= ´ = + = + =
+ = + ´ =
= ´ = <
Section cracked, use with ´ =
= ´ + - ´ = ´
d = ´ + ´ ´ ´ ´ =
Final long-term deflection d + d =
. OK.
For interest only, check the condition to validate the l/d ratio according to EN 1992-1-1,
clause 7.4.2(2) but use mm as using mm will be similar to checking a
slab with the result that l/d is around 25. This shows the limitation of the l/d ratio method
for some sections and geometry.
-
r =Ö ´ = r= = ´ = r r= <
Equation 5.49. l/d basic = + ´Ö ´ =
> = ,
fits well between .
The design of prestressed beams is less versatile than for reinforced beams because the ten-
dons’ positions are restricted to a predetermined pattern by an array of holes in the jack-
ing heads, which is usually a permanent fixture at precasting works. Figure 5.21 shows
the end template for an inverted-tee beam with quite a large eccentricity from the centre
on the tendons to the centroid of the section, resulting in an upward camber as shown
in Figure 5.22. A neoprene (or similar) strip is fixed at this particular factory, ready to
receive the floor slab on site, but most slabs such as hollow core units with up to about
8–10 m span (depending on depth) do not have a bearing material. The upward camber
is often controlled by adding two tendons (which need not be fully stressed) at the top of
Preparation of mould and rebars for prestressed concrete inverted-tee beam.
Prestressed concrete inverted-tee beam – note the upward curvature due to eccentric pretensioning.
the upstand. Top tendons also help to avoid cracking during transportation and handling.
Figure 5.23a shows a full array of possible tendon positions in an inverted-tee beam,
and an example of a typical tendon layout. Note the symmetry. The tendons are placed
at all the corners; a 40–50 mm centroidal cover distance is being used in most cases.
Rectangular section beams are more efficient in terms of capacity per cross-sectional area,
50
Small chamfer
say 10 × 10 mm
Clear zone
100
for services
50
B
50
A
40
50
to
50 50c/c
Chamfer 25 × 25 mm
to 50 × 50 mm
1 2 3 4
Strand grid
(a) reference
Essential for prevention

of top corner splitting
and hangers for stirrups
Shear stirrups at
end of beam
Boot links
Essential for
(b) stirrup hangers
(a) Array of pretensioning tendons (usually strands) in inverted-tee beams. (b) Typical strand
and link arrangement.
but inverted-tee beams hold the advantage of allowing the floor slab to be recessed in the
upper part of the beam, as shown in Figure 5.1a.
The minimum breadth of the beam is a function of the type of the floor slab to be used.
The breadth is equal to twice the recess width, plus the upstand width. The same rea-
soning as for the L-beam is used if floor ties are intended to be placed within the recess
and concealed in the depth of hollow core floor slabs. The minimum recess width for this
condition is 100–125 mm. If the ties are to be located elsewhere, the recess width may be
90–100 mm. The minimum upstand breadth is 250–300 mm. Beam depths depend on the
following four factors:
1. The flexural moment of resistance

2. Shear capacity
3. The size of the end connector
4. The depth of the boot required to carry the floor loads
The design procedure is identical to the design of prestressed floor units given in Section 4.3.1.
with the additional consideration of satisfying transfer, as well as working, stress conditions.
This is because strands can be either deflected or debonded. There is much more freedom in
selecting the strand pattern than in floor units as the design of the beam can be optimised
( economy of strands) by choosing a pattern that will simultaneously satisfy transfer at the
ends of the beam and working loads at the point of maximum imposed bending moment.
It is desirable that the permissible stress at transfer after the initial losses (due to
elastic shortening) is made equal (or as close as possible) to the working stress after
all losses. In most beam designs, the initial and final losses are about 8% and 25%, respec-
tively, meaning that the ratio should be at least 0.55. In fact, the transfer strength
for is often at about , and therefore transfer stresses will (nearly) always
govern parallel, unbonded tendons. To overcome this problem, it is desirable to debond a
small number of strands from the end of the beam to about one-fourth span, say four in a
typical situation. It is also wise to actually restrict the top fibre stress to something less than
, say a half to two-thirds this value, whilst accepting that there will be a small loss in
moment capacity. This factor of safety is based on the experience gained from transporting
and handling highly stressed prestressed beams, and the need to avoid flexural cracking
at the sides of the beams for the sake of durability. During transportation, beams are sup-
ported on bearers at about 500–750 mm from the ends, and the negative bending moment
due to self-weight of the cantilevered ends causes tension at the top in addition to the nega-
tive prestress.
The pretensioning force which takes place at the moment of detensioning includes the
immediate relaxation Ds pr of the tendons plus most of the elastic shortening loss Ds . The
total loss ratio is defined here as x = Ds + Ds s , calculated according to Section 4.3.2,
where s = h = . The loss ratio Ds s for Class 2 strand stressed to 70% and
detensioned after 20 h is fixed as ´ ´ ´ ´ ´ -
´ - = . The
force at transfer is therefore - x , and the bottom and top fibre stresses are as follows:
s = -x + <+ (5.70)
s = -x - >- (5.71)
Manipulation of the simultaneous Equations 5.70 and 5.71 will give optimum values for the
initial prestressing force and the eccentricity , as follows:
Ac é - a öù
Ppi =
- x êë
f ck t - fctm t + {( fck t + fctm t } + a ÷ø úû
(5.72)
where a = x = initial prestress loss (expressed as a decimal fraction)

The number of tendons required is
= h (5.73)
where h is the degree of prestress.

The eccentricity is given by
f ck t + f ctm t
zcp = (5.74)
- x Ppi b
æ ö
where b = ç + ÷ and is based on the value obtained from Equation 5.72.
Z
è b Z t ø
Given and , the actual bottom and top fibre stresses may be calculated from Equations
5.70 and 5.72 from which the actual (as opposed to the guessed at value in Equation 5.70)
value of ξ is found, and if this differs by more than 2%–3%, iteration should take place. The
elastic shortening loss ratio is given by
Ds s = s h (5.75)
where
s = +
(5.76)
=
at the centroid of the tendons

prestress after initial relaxation loss = s - Ds
Total losses may be determined as given in Section 4.3.2. The final bottom and top stresses
are calculated in the usual manner:
s = + £
s = - ³- (or zero if durability > XC1)

The service moment of resistance is given by the lesser of:
= s + (5.77)
or
= s + (5.78)
where and are defined in Section 4.3.4.

Tables 5.4 and 5.5 give the minimum value of for a range of typical sizes for prestressed
inverted-tee beams using and , respectively. The different strengths are
used to demonstrate the potential increases in . Note that the transfer strength in both
cases remains . In compiling the tables, the tendons were arranged for
Service and ultimate moments of resistance for 500 mm wide grade C45/55 prestressed
inverted-tee beams
Ppi zcp MsR MRd

500 200 150 300 11 1267 58.9 155.0 260.1
500 300 150 300 15 1728 71.9 258.3 420.4
500 400 150 300 18 2074 89.4 383.1 623.1
500 500 150 300 22 2535 104.1 544.4 884.1
500 600 150 300 26 2996 122.0 743.6 1200.5
500 200 200 300 12 1383 66.7 198.4 340.7
500 300 200 300 16 1844 77.7 311.1 513.8
500 400 200 300 20 2305 94.2 457.8 733.9
500 500 200 300 24 2765 109.4 631.5 1001.2
500 600 200 300 27 3111 131.5 831.3 1323.2
500 200 250 300 14 1613 67.9 240.5 384.5
500 300 250 300 17 1959 91.7 379.3 654.3
500 400 250 300 21 2420 105.3 534.5 842.5
500 500 250 300 25 2881 118.5 716.1 1123.6
500 600 250 300 29 3342 135.0 932.8 1458.0
All sections use f ck 50 N/mm2 , f ck (t ) 30 N/mm2, f ctm 3.02 N/mm2 for exposure XC1, f pk 1770 N/mm2, Ap 93 mm2
per strand. Prestress at transfer sb (t ) = 14.5 –15.0 N/mm2 .
Service and ultimate moments of resistance for 600 mm wide grade C50/60 prestressed
inverted-tee beams
Ppi zcp MsR MRd

600 200 150 350 13 1498 57.1 185.8 325.6
600 300 150 350 18 2074 70.3 314.9 514.3
600 400 150 350 22 2535 85.7 469.6 771.9
600 500 150 350 27 3111 102.1 676.2 1110.8
600 600 150 350 31 3572 121.9 910.4 1495.0
600 200 200 350 15 1728 63.7 245.7 420.4
600 300 200 350 19 2189 80.5 382.5 646.7
600 400 200 350 24 2765 92.7 558.8 903.8
600 500 200 350 28 3226 112.6 768.2 1233.1
600 600 200 350 33 3802 127.2 1022.1 1637.5
600 200 250 350 16 1844 76.1 310.0 485.8
600 300 250 350 21 2420 85.2 464.5 785.9
600 400 250 350 26 2996 102.2 664.3 1063.7
600 500 250 350 30 3457 118.0 878.1 1385.2
600 600 250 350 34 3918 139.0 1140.4 1816.8
All sections use f ck 50 N/mm2, f ck (t ) 30 N/mm2 , f ctm 3.02 N/mm2 for exposure XC1, f pk 1770 N/mm2 , Ap 93 mm2
per strand. Prestress at transfer sb (t ) = 14.5 –15.0 N/mm2 .
200
Total ultimate dead + live load (kN/m) 175
150
125
100 400 500 600 700 mm deep
75
50
25
0
3 4 5 6 7 8 9 10 11 12 13 14 15
Clear span between column faces (m)
Load versus span charts for non-composite prestressed inverted-tee beams 600 mm wide at
soffit with 350 200 mm upstand. f ck 45 N/mm2, f ctm 3.02 N/mm2 for exposure XC1, pre-
stress at transfer σ ( ) = 14.0–14.5 N/mm . Bearing length 140 mm.
2
cases so that the prestress at transfer was in the range s = - , values that
included a reduction for the self-weight at transfer. Figure 5.24 gives a design graph for
the ultimate beam load versus depth for a range of 600 mm wide inverted-tee beams using
. The tendons were arranged so that , also includ-
ing the top and bottom fibre stresses due to self-weight at transfer.
Example 5.5
Calculate the initial prestressing requirements for an inverted-tee beam shown in Figure 5.25.
Use ,
strand dia. = = = h= = .
Assume x = initial losses, including an initial relaxation loss ratio of 0.004. Exposure
class is XC1.
Solution
Geometric data:
, ,
- -
. Then a = b= ´
-
Prestressing force per strand = ´ ´ ´ =
= = ´ + + =
Limiting transfer stresses are
s <+ ´ = s >- =-
125 350 125
50
200
50
400
50
spacing
12.5 dia. strands 50 × 50 chamfer

600
Detail to Examples 5.5 and 5.6.
Equation 5.72
´ é æ öù -
= ´ê +ç ´ ÷ú ´ =
´ - ë è øû
Equation 5.74
-
= ´ ´ ´ ´ =
Equation 5.73 rounded down to 26 to prevent the

possibility of overstress.
The actual prestressing force , and the first estimate

after initial losses . The maximum fibre stresses at transfer are
Equation 5.70
s = + =+ <+
Equation 5.71
s = - =- >-
At the level of the strands
Equation 5.76
s = ´ + =+
Initial relaxation loss given as 0.005

´
Equation 5.75 x = + = (original assumption of
0.07 is OK) ´ ´
Having established strand, , at , the strands are

arranged in the beam as shown in Figure 5.25 as follows:
yi
SNi y i
2 550 1100
2 350 700
4 250 1000
2 200 400
4 150 600
4 100 400
8 50 400
S = 26 S = 4600
Then,
The actual eccentricity = - = - = , closest possible to 92 mm
in the available array of tendon positions. The complete design is as follows:
Losses
Ds + Ds
´ - -
Ds = ´ ´ ´ ´ ´ =
-
= - ´ ´ =
Ds = ´ =
where
s = + ´ =
s = < ´ = OK
Transfer stress check
s = + ´ = <
s = - ´ =- >

Precast Concrete Structures Second Edition PDF Compressed 1 299

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Precast Concrete Structures Second Edition PDF Compressed 1 299

Uploaded by

Copyright:

Available Formats

www.EngineeringEBooksPdf.

Boca Raton London New York

CRC Press is an imprint of the

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-4987-2399-2 (Hardback)

Library of Congress Cataloging‑in‑Publication Data

Names: Elliott, Kim S.

Visit the Taylor & Francis Web site at

Preface to the First Edition xiii

1 What is precast concrete 1

2 Materials used in precast structures 21

2.3 Steel reinforcement 52

3 Precast frame analysis 65

4 Precast concrete floors 125

4.2.1 Definitions of precast floor units and slab fields 136

5 Precast concrete beams 215

5.3 Composite reinforced beams 239

6 Precast concrete columns 313

7 Shear walls 359

7.4.1 Precast concrete infill walls 376

8 Horizontal floor diaphragms 395

9 Joints and connections 429

10 Beam and column connections 513

11 Ties in precast concrete structures 599

12 Design exercise for 10-storey precast skeletal frame 629

12.3 Solution 637

12.8 Stability ties 676

1. British lecturers are holistic towards structural engineering.

1 FIP, Federation International de la Prefabrication; an international, but predominantly European, organisation

Council (United Kingdom), Betoni (Finland), Bevlon (Netherlands), C&CA (Australia),

Betonfertigteile GmbH (Germany), VarioPlus (Germany), Spancrete (United Staes), AB

North Galaxy Buildings, Belgium (c Ergon, Belgium).

Prestressed concrete beams 55 m long by VBI, the Netherlands.

fctd(t) Value of fctd at transfer or at time t

rm Column end moment ratio = M01/M02

Ac,co Compound value of Ac

Es Young’s modulus for steel or rebars

Le Effective diagonal length of infill wall

V Rds,b V Rd due to bottom steel stirrups in deep beam wall

εcd Drying shrinkage strain

ρ Reinforcement ratio A s /Ac

Abbreviations and Other Nomenclature

Architectural-structural precast concrete frame at Scottish Office, Leith, United Kingdom.

(a) Site cast using stack casting between sheets of polythene. ( )

Simple bearing nib and corbel.

A precast concrete structure is an assemblage of precast elements which, when suitably

* This book does not address seismic actions.

Precast concrete ‘skeletal’ sway frame, Europark, Rome.

Precast concrete ‘skeletal’ known as ‘semi-rigid’ frame, Recife University, Brazil.

Precast ‘skeletal’ and ‘wall-frames’, Barwa City, Qatar.

Rigid joint Mechanism

Precast concrete wall-frames for residential dwellings near to Helsinki.

Mixed precast and concrete with structural steelwork.

Mixed construction of precast concrete floors in a timber beam–column frame. (Courtesy of

BS EN 1990. 2002. Eurocode: Basis of structural design, BSI, London, UK.

Compaction problems in precast concrete columns.

The constitutive stress–strain relationship is idealised as a parabola–rectangle in

42.5N 15–20 ≥ 40–55 – 50–70

strength ηfcd is established by national codes to be as close as possible. BS EN 1992-1-1,

For example for fck = 90 N/mm 2 fcd loses 20%.

That is 0.8 reduces to 0.7 at 90 N/mm 2 .

fcm(t) = fck(t) + 8 N/mm 2 (2.4)

Self-compacting concrete, flowing and compacting without segregation, used to manufacture