See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/372888164

Development and splice length of reinforcement in NZS 3101:2006 New

Zealand Concrete Standard – Model calibration and sensitivity analysis

Technical Report · August 2023

DOI: 10.13140/RG.2.2.23080.34565

CITATIONS READS

0 431

1 author:

Dorian Borosnyoi-Crawley

118 PUBLICATIONS 1,047 CITATIONS

SEE PROFILE

All content following this page was uploaded by Dorian Borosnyoi-Crawley on 04 August 2023.

The user has requested enhancement of the downloaded file.

Development and splice length of reinforcement in NZS 3101:2006
New Zealand Concrete Standard – Model calibration and sensitivity
analysis

Compiled by Dorian Borosnyoi-Crawley, CMEngNZ, PhD

August 2023
Disclaimer
Material contained in this document is intended as guidance only. All readers should satisfy
themselves as to the applicability of the recommendations made and should not act on the basis
of any matter contained in this document without considering, and if necessary, taking
appropriate professional advice from a Chartered Professional Engineer, on their own particular
circumstances. BCR Consulting Ltd and all contributors to this Report, either listed or not listed,
expressly disclaim all and any liability to any person in respect of anything done or omitted to
be done by any such person in reliance upon the whole or any part of the contents of this
document.

1
Executive summary
NZS 3101:2006 follows the principles of Limit State Design similarly to the Eurocodes, and
the development length and splice length formulas in NZS 3101 can be considered to be in the
same category of conservativism than that of the formulas recommended by Eurocode 2 (EN
1992-1-1).

There are multiple differences between the code formulas for calculating the development
length and splice length. For example, the concept of force transfer by bond, or the term bond
strength is not introduced in NZS 3101.2006.

After a detailed parameter analysis and a calibration based on the extension of international
databases it can be concluded that the development length and splice length formulas for cast-
in rebars in NZS 3101:2006 are not robust and further analysis of the topic is addressed to future
research.

2
Contents
1. Status quo in development and splicing of reinforcement in NZS 3101 4

1.1 Provisions for non-contact splices in NZS 3101 5

1.2 The bond strength in codes 8

1.3 Influencing parameters in codes 11

1.4 Reliability verification format in codes 15

2. Calibration of the NZS 3101 development and splice length models 28

2.1 Existing database 28

2.2 Database filtering in tension 29

2.3 Calibration of splices and anchorages in tension 29

2.4 Database filtering in compression 35

2.5 Calibration of splices and anchorages in compression 36

References 37

Appendix A – The tensile strength of concrete 41

Appendix B – NZS 3101:2006 Chapter 8: Stress development, detailing and

splicing of reinforcement and tendons 47

3
1. Status quo in development and splicing of reinforcement in NZS 3101
NZS 3101 was first published in 1982 and as a basis – with the exception of the provisions for
seismic loading –, ACI 318-77 has been used with minor modification. After its first revision,
NZS 3101 was re-issued in 1995 and minor changes have also been made to facilitate a planned
future harmonization with the Australian Concrete Structures Code. In particular, new sections
covering the design for durability and fire have been based on the corresponding sections of AS
3600, modified as appropriate for New Zealand conditions, materials and regulations. The other
non-seismic sections of NZS 3101 were still based largely on the provisions of the building
code of the American Concrete Institute, with some of the new provisions of ACI 318-89 being
incorporated in NZS 3101:1995. After a second revision, NZS 3101 was re-issued again in 2006
and is in force at the date of publishing of this Report. During the second revision of NZS 3101
various technical advancements and improvements have been incorporated that have been
developed since 1995. The non-seismic sections of NZS 3101 are largely based upon ACI 318-
02.

The current formulae for development length in tension and compression (Section 8 in NZS
3101:2006 – see Appendix) have been added to the standard in 1995. The general formulae are
as follows:

Basic development length in tension:

(0.5𝛼𝛼𝑎𝑎 𝑓𝑓𝑦𝑦 )
𝐿𝐿𝑑𝑑𝑑𝑑 = 𝑑𝑑𝑏𝑏
�𝑓𝑓𝑐𝑐′

Refined development length in tension:

𝛼𝛼𝑏𝑏
𝐿𝐿𝑑𝑑 = 𝐿𝐿 ≥ 300 𝑚𝑚𝑚𝑚
𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 𝑑𝑑𝑑𝑑

Basic development length in compression:

𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑𝑑𝑑 = 0.22
�𝑓𝑓′𝑐𝑐

Refined development length in compression:

𝐿𝐿𝑑𝑑 = 𝛼𝛼𝑏𝑏 𝛼𝛼𝑒𝑒 𝐿𝐿𝑑𝑑𝑑𝑑 ≥ 0.040𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏 ≥ 200 𝑚𝑚𝑚𝑚

Commentary Clause C8 of NZS 3101 reveals that the development length formulae for tension
shown above are based on research conducted before the 1980’s, namely the research of
Orangun, Jirsa, and Breen (1975, 1977), as the basis of the ACI 318 development and splice
length of straight reinforcement in tension, in combination with the research of Jirsa, Lutz and
Gergely (1979). The parameters accounting for the beneficial effects of transverse
reinforcement were checked in the original research based on the experiments of Untrauer and
Warren (1977).

It is noted that in ACI 318 the strength reduction factor ϕ is not used in the development length
and lap splice length equations since an allowance for strength reduction is already included in
the expressions for development and splice lengths. Similarly, Clause 2.3.2.2 of NZS 3101 sets
the strength reduction factor for anchorage and strength development of reinforcement as ϕ =
1.0.

4
The NZS 3101 provisions on development length in compression are similar to the provisions
given in ACI 318-89. According to ACI 318, the weaking effect of flexural tension cracks is
not present for rebars in compression, and usually end bearing of the rebars on the concrete is
also beneficial, therefore, shorter development lengths are specified for compression than for
tension. ACI 318 also adds that bond behaviour of compression bars is not complicated by the
problem of transverse tension cracking and thus compression splices do not require provisions
as strict as those specified for tension splices. It is noted that the minimum lengths specified for
column splices contained originally in the 1956 ACI Building Code and have been carried
forward in the later code editions and extended to compression bars in beams and to higher
strength steels later. No changes, however, have been made in the provisions for compression
splices since the 1971 ACI code edition.

The NZS 3101 provisions on lap splices in tension follow the recommendations of ACI
Committee 408 – also adopted by ACI 318 – that splice and development lengths for deformed
bars and wires are the same.

1.1 Provisions for non-contact splices in NZS 3101

For non-contact splices, NZS 3101 adopted the concept of the effective lap splice length
proposed first by Robinson et al (1974). In accordance with Clause 8.7.2.5 of NZS 3101, bars
spliced by non-contact lap splices in flexural members spaced transversely farther apart than
3db shall have splice length, Lds, given by Lds ≥ Ld + 1.5sL; where sL is the clear distance between
bars of a non-contact lap splice, mm (Figure 1.1). The angle of the diagonal compression struts
of the concrete is assumed to be appr. θ = 33°.

Figure 1.1 Lap length (Lds) of non-contact rebars

(reproduction of Fig. C8.14 of NZS 3101:2006)

Robinson et al (1974) hypothesised that a diagonal compression field forms between the spliced
rebars and a truss model has been proposed with 45 degrees angle for the inclination of the
compression struts (Figure 1.2). Accordingly, the effective lap splice length was predicted as
the total splice length reduced with the spacing of the spliced rebars, or in reverse, the splice
length is equal to the effective splice length increased with the spacing of the spliced rebars:

𝐿𝐿𝑠𝑠 = 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 + 𝑠𝑠𝐿𝐿

5
 𝑠𝑠𝐿𝐿

𝑠𝑠𝐿𝐿 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒

𝐿𝐿𝑠𝑠 = 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 + 𝑠𝑠𝐿𝐿

Figure 1.2 Concept of the effective lap splice length proposed by Robinson et al (1974)

Goto and Otsuka (1980) experimentally demonstrated the existence of diagonal internal cracks
between rebars in non-contact splices (Figure 1.3), and they reported that the splice length and
the rebar spacing both influences the inclination of these cracks. It was observed that in the lap
splices, relatively large primary transverse cracks generally occur at both ends of the splice.
Although not visible, if the lap length is long, primary horizontal cracks occur further inside the
splice. If the splice is not long, the diagonal internal cracks between the two rebars are formed
across the entire length of the splice. A large number of these diagonal cracks are generated
almost parallel with each other. Although it depends on the lap length, in general, the inclination
of these cracks was found to be about 30 to 45 degrees when the rebar spacing is small.
Increasing the rebar spacing increased the crack inclination angle. In the central part, the
inclination angle tended to be larger than in the vicinity of the splice ends, and this tendency
was particularly pronounced when the lap length was long (Goto and Otsuka, 1980).

Figure 1.3 Pattern of internal cracks around lapped splice after Goto and Otsuka (1980)
Note: rebar diameter 16 mm, clear spacing 8 mm, concrete cover 42 mm, total overlap length
250 mm

Figure 1.4 Fig. 1.3 redrawn by Sagan et al (1988)

6
Sagan et al (1988) studied non-contact splices under both monotonic and cyclic loading. The
specimens had bar spacings ranging from direct contact to 8 bar diameters clear spacing for
both 19 mm and 25 mm diameter bars, with concrete compressive strengths ranging from 21.4
to 42.1 MPa, and splice lengths of 30 and 40 bar diameters. It was observed that the bar spacing
affects the number of inelastic load cycles that can be attained before failure, but it does not
affect the ultimate strength of a splice. The ultimate load that can be carried by a splice was
found to be independent of the bar spacing. Sagan et al (1988) suggested that non-contact lap
splice design should consider the effects of the added confinement provided by the additional
concrete between the spaced bars, the reduction in the tensile strength of the concrete due to the
compression force transfer stresses in the concrete between the bars, and the reduction in the
effective lap splice length. The observed diagonal surface cracks over the spliced rebars were
found in the range of 20 to 70 degrees, but were predominantly 45 degrees. Since the concrete
cover was small (1.5db) the authors concluded that the surface cracking should be a reflection
of the internal cracking between the spliced bars where the actual load transfer is occurring. As
a result of the development of diagonal cracks, the overall lap splice length was suggested to
be determined by adding to the effective lap splice length the clear spacing multiplied by 0.75:

𝐿𝐿𝑠𝑠 = 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 + 0.75𝑠𝑠𝐿𝐿

Sagan et al (1988) also hypothesised a diagonal compression field between the spliced rebars,
however, they commented that the surface cracking in the concrete surrounding the splice
cannot be used to define a precise angle for the inclination of the internal diagonal compressive
struts. It was assumed that the surface cracking, nevertheless caused primarily by the stress
state, is also affected by the cover thickness, the existing transverse and longitudinal cracks, the
location and distribution of the transverse reinforcement and the coarse aggregate of the
concrete. Based on compression field theory and biaxial concrete strength theory Sagan et al
(1988) proposed an angle of 50 degrees for the inclination of the compression struts. In the
formula above, 0.75 represents the sine of 50 degrees. In a later paper, Sagan et al (1991) based
on the same research and the same proposed angle of 50 degrees for the inclination of the
compression struts suggested a different formula for the overall lap splice length to be
determined by adding to the effective lap splice length the clear spacing multiplied by 1.2
(representing the tangent of 50 degrees), that is:

𝐿𝐿𝑠𝑠 = 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 + 1.2𝑠𝑠𝐿𝐿

𝑠𝑠𝐿𝐿

𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 ∙ 𝑠𝑠𝐿𝐿 𝐿𝐿𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒

𝐿𝐿𝑠𝑠

Figure 1.5 Concept of the effective lap splice length by Sagan et al (1991)

7
Diagonal surface cracking over non-contact lap splices were observed by many other
researchers too. McLean and Smith (1997) observed an angle of 45 degrees. Hong and Chun
(2007) proposed that the effective lap splice length depends on the transverse reinforcement
ratio and the spacing of the non-contact splice. The influence of the rebar spacing is more
pronounced when the transverse reinforcement ratio is low. The slope of the strut developed
between the rebars becomes steeper as the transverse reinforcement ratio decreases. Masud
(2019) observed angles between 12 to 47 degrees, and the angle increased as the rebar spacing
increased. Hwang et al (2022) confirmed the significance of the influence of transverse
reinforcement and added that when a sufficient effective lap splice length and transverse
reinforcement is provided, the bond strength of noncontact lap splices is greater than that of
contact lap splices. However, when the spliced bars are too widely separated or transverse
reinforcement is not used, the bond strength of noncontact lap splices is less than that of contact
lap splices. Similar observations could be found in the literature as early as Chamberlin (1952,
1956, 1958) who observed in pullout tests that tied (i.e., contact) splices developed better
average bond stress than that of spaced splices, but in beam tests the difference was found to be
insignificant. Hamad and Mansour (1996) observed that the bond strength of non-contact
splices was greater than that of the contact splices when the spacing of spliced bars was
increased from zero to 5db, however, when the spacing was further increased beyond 7db, the
bond strength of the non-contact splices was less than that of the contact splices. Kilpatrick and
Gilbert (2012) observed that non-contact splices with small spacing showed similar bond
strength to that of contact splices, however, when the spacing was increased, the bond strength
of the non-contact splices was found to be inferior to that of contact splices.

It can be concluded that there is no consensus or generally accepted model in the literature to
describe the mechanisms of non-contact lap splices or to predict the lap length needed for non-
contact splices, and the reported experimental observations sometimes contradict to each other.
The topic is addressed to future research. It can be also concluded that the NZS 3101 model
does not take the influence of transverse reinforcement directly into account (apart from the
parameter αd for tension splices in certain cases) and provides a more conservative estimate for
the overall non-contact lap splice length than the models proposed by Robinson et al (1974),
Sagan et al (1988) and Sagan et al (1991) based on the concept of the effective lap splice length.

1.2 The bond strength in codes

The concept of force transfer by bond, or the term bond strength is not introduced in NZS 3101.
The design of lap splices and single rebar anchorages in accordance with Eurocode 2 (EN 1992-
1-1) introduces the bond strength, therefore, the use of bond strength in different codes is
summarized in the followings, with a focus on Eurocode 2.

The design bond strength (fbd) is a material property in Eurocode 2 (EN 1992-1-1) related to the
tensile strength of concrete (see the Appendix of this Report). Using the relationships between
the tensile and compressive strength of concrete, however, the design bond strength can be
expressed as a function of the characteristic compressive strength that reads (Borosnyoi and
Genesio, 2021):

𝛼𝛼𝑐𝑐𝑐𝑐 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 𝛼𝛼𝑐𝑐𝑐𝑐 [0.7𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 ]

𝑓𝑓𝑏𝑏𝑏𝑏 = 2.25𝜂𝜂1 𝜂𝜂2 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 2.25𝜂𝜂1 𝜂𝜂2 � � = 2.25𝜂𝜂1 𝜂𝜂2 � �=
𝛾𝛾𝑐𝑐 𝛾𝛾𝑐𝑐
2/3
= 2.25𝜂𝜂1 𝜂𝜂2 �𝛼𝛼𝑐𝑐𝑐𝑐 �0.7�0.3𝑓𝑓𝑐𝑐𝑐𝑐 ��/𝛾𝛾𝑐𝑐 �

8
Where:
fctd is the design value of concrete tensile strength
η1 is a coefficient related to the quality of the bond condition and the position of the bar
during concreting; the ‘top-bar effect’ parameter
η2 is a coefficient related to the bar diameter

Introducing η1 = η1 = αct =1.0 for simplicity and considering γc = 1.5, the expression of fbd
simplifies to:

2/3
𝑓𝑓𝑏𝑏𝑏𝑏 = 0.315𝑓𝑓𝑐𝑐𝑐𝑐 = 0.315(𝑓𝑓′𝑐𝑐 )2/3

Presuming that the same assumptions apply (i.e., the force transfer from the rebar to the concrete
develops by bond under a partially confined condition determined by the geometric constraints)
for the calculation of the development length in NZS 3101 and the basic required anchorage
length in Eurocode 2 (EN 1992-1-1), respectively, one can formally write (with introducing a
generalized parameter, K):

𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏 𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏

𝐿𝐿𝑑𝑑𝑑𝑑 = 𝐾𝐾 ~ 𝑙𝑙𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 =
�𝑓𝑓′𝑐𝑐 4𝑓𝑓𝑏𝑏𝑏𝑏

From which an equivalent term for the design bond strength can be expressed as:

�𝑓𝑓′𝑐𝑐
𝑓𝑓𝑏𝑏𝑏𝑏 ~
4𝐾𝐾

Which can be shown for tension loading in NZS 3101 to be:

𝑓𝑓𝑏𝑏𝑏𝑏 ~ 0.5�𝑓𝑓′𝑐𝑐

And the same for compression loading would be:

𝑓𝑓𝑏𝑏𝑏𝑏 ~ 1.14�𝑓𝑓′𝑐𝑐

From the above two expressions the one for tension loading together with the design bond
strength of Eurocode 2 (EN 1992-1-1) is represented as a function of specified compressive
strength of concrete in Figure 1.6. It can be observed that the two functions can result in
different values for the same concrete strength, especially in the range of higher strengths. This
highlights the differences in the basic assumptions of these standards and calls the attention that
one cannot make a fair judgement based on the comparison of such parameters since the design
approaches give different weights to the concrete strength in the calculation of development
length or lap length. This is especially true if one compares tension and compression loading.

As it was indicated before, the differences between the NZS 3101 development length formulae
for tension loads and compression loads are rooted in ACI 318, and were based on experimental
observations made before the end of the 1970’s. Clause C8.6.5 of NZS 3101 explains (with the
words of ACI 318) these experimental observations as follows: “The weakening effect of
flexural tension cracks is not present for bars in compression and usually end-bearing of the
bars on the concrete is beneficial. Therefore, shorter development lengths have been specified
for compression than for tension. The development length may be reduced by up to 25% when

9
the compression reinforcement is enclosed within a column by spiral or rectangular ties, hoops
or supplementary ties, or an individual spiral around each bar or group of bars is used.” In
Eurocode 2 (EN 1992-1-1), or in CEB-FIP Model Code 1990, or in fib Model Code 2010 no
such differentiation is made for the calculation of anchorage length in tension or in
compression, and the same design bond strength (fbd) is the basis of both calculations.

The expression 1.14�𝑓𝑓′𝑐𝑐 is not represented in Figure 1.6 since Clause 6.1.3.3 of fib Model
Code 2010 sets an upper limit for the design bond strength as 𝑓𝑓𝑏𝑏𝑏𝑏 < �𝑓𝑓′𝑐𝑐 .

𝑓𝑓𝑏𝑏𝑏𝑏 = 0.315(𝑓𝑓′𝑐𝑐 )2/3

Design bond strength fbd

6
and 0.5 �𝑓𝑓′𝑐𝑐 ,MPa

0.57�𝑓𝑓′𝑐𝑐
2
0.5�𝑓𝑓′𝑐𝑐

0
20 30 40 50 60 70 80 90 100

Specified concrete strength f’c,MPa

Figure 1.6 Design bond strength in Eurocode 2 (EN 1992-1-1) vs. the expression 0.5�𝑓𝑓′𝑐𝑐
represented as a function of specified compressive strength of concrete (note: see Appendix
for curve 0.57�𝑓𝑓′𝑐𝑐 )

5
Bond stress, MPa

0
(unloaded end) lb = 200 mm (loaded end)

Figure 1.7 Bond stress distribution over the embedment length during pullout testing
(Xu, 1992)

10
1.3 Influencing parameters in codes

As it has been introduced before, there is experimental evidence that bond strength (more
precisely: uniform bond strength, as it is understood in the definition of fbd, i.e., assumed to
have the same value over the embedment length; that is a simplification of the real bond stress
distribution, see Figure 1.7) strongly depends on the rebar diameter, the confinement conditions
(which is usually expressed as an influence of concrete cover and rebar spacing), the position
of the rebar in the cross section (e.g., related to the direction of concreting), the embedment
length, the steel stress level and the type of loading. These influences can be taken directly into
account within the bond strength parameter (e.g., the approach of fib Model Code 2010), or
with separate parameters as it is the case in Eurocode 2 (EN 1992-1-1) and NZS 3101.

For example, the concrete cover + rebar spacing parameters to take the indirect confinement
effects into account are quite different in NZS 3101 and in Eurocode 2 (EN 1992-1-1). In the
formula of refined development length in tension, NZS 3101 considers this with parameter αc,
as:

𝛼𝛼𝑏𝑏 𝛼𝛼𝑎𝑎 𝛼𝛼𝑏𝑏 𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏

𝐿𝐿𝑑𝑑 = 𝐿𝐿𝑑𝑑𝑑𝑑 = 0.5 ≥ 300𝑚𝑚𝑚𝑚
𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓′𝑐𝑐

Where αc, when cover to bars in excess of 1.5db or clear distance between adjacent bars in
excess of 1.5db is provided:

𝑐𝑐𝑚𝑚
1.0 ≤ 𝛼𝛼𝑐𝑐 = 1 + 0.5 � − 1.5� ≤ 1.5
𝑑𝑑𝑏𝑏

Where cm = the lesser of the concrete cover or the clear distance between rebars.

On the other hand, Eurocode 2 (EN 1992-1-1) takes the concrete cover + rebar spacing
parameter into account in the formula of design anchorage length with parameter α2, as:

𝑙𝑙𝑏𝑏𝑏𝑏 = 𝛼𝛼1 𝛼𝛼2 𝛼𝛼3 𝛼𝛼4 𝛼𝛼5 𝑙𝑙𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 ≥ 𝑙𝑙𝑏𝑏,𝑚𝑚𝑚𝑚𝑚𝑚 𝑤𝑤𝑤𝑤𝑤𝑤ℎ 𝛼𝛼2 𝛼𝛼3 𝛼𝛼5 ≥ 0.7

Where α2 is expressed as:

𝑐𝑐𝑑𝑑 − 𝑑𝑑𝑏𝑏
0.7 ≤ 𝛼𝛼2 = 1 − 0.15 ≤ 1.0
𝑑𝑑𝑏𝑏

In which cd = a dimension (in millimetres), as shown in Figure 1.8.

Figure 1.8 Values of cd for beams and slabs in Eurocode 2 (EN 1992-1-1)

11
The relationship between αc of NZS 3101 and α2 of Eurocode 2 (EN 1992-1-1) can be seen in
Figure 1.9.

Eurocode 2 (EN 1992-1-1) accounts for the beneficial effect of non-straight end rebars in the
formula of design anchorage length with parameter α1, that allows a reduction of α1 = 0.7 for
standard bends, hooks and loops in case of adequate concrete cover and rebar spacing (cd >
3db).

Both NZS 3101 and Eurocode 2 (EN 1992-1-1), however, takes into account the ‘top bar effect’;
parameter αa accounts for it in NZS 3101 and parameter η1 in Eurocode 2 (EN 1992-1-1).

In NZS 3101, there is no parameter that might directly take into account in the calculation of
the development length that increasing rebar diameter is resulted in decreasing bond strength,
while in Eurocode 2 (EN 1992-1-1) parameter η2 is expressed as (for db > 32 mm):

132 − 𝑑𝑑𝑏𝑏
𝜂𝜂2 =
100

1
α2 a)
0.95 0.925
0.9

0.85

0.8 0.775

0.75
αc
0.7
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6

1.6

1.4
b)
1.2
αc

1
α2
0.8

0.6
1 2 3 4 5
𝑐𝑐𝑚𝑚 𝑐𝑐𝑑𝑑
𝑜𝑜𝑜𝑜 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟
𝑑𝑑𝑏𝑏 𝑑𝑑𝑏𝑏

Figure 1.9 The relationship between αc of NZS 3101 and α2 of Eurocode 2 (EN 1992-1-1)

The beneficial effect of transverse reinforcement that may control the opening of splitting
cracks, as shown in Figure 1.10, is usually expressed in code formulae by the cross-sectional
area of transverse reinforcement. One possible way of considering this effect (as adopted in
NZS 3101) is to take as an average for the bars, using the total area of transverse bars crossing

12
the splitting plane, divided by the number of longitudinal bars in the layer, as it is expressed in
the formula, αd, below.

Figure 1.10 Splitting failure modes and transverse reinforcement (fib, 2014)

In the formula of refined development length in tension, NZS 3101 considers αd (when
transverse reinforcement with at least 3 bars, spaced less than 8db, transverse to the bar being
developed, and outside it, are provided within Ld) as:

𝐴𝐴𝑡𝑡𝑡𝑡 𝑓𝑓𝑦𝑦𝑦𝑦
1.0 ≤ 𝛼𝛼𝑑𝑑 = 1 + �� � � � ≤ 1.5
𝑠𝑠 80𝑛𝑛𝑛𝑛𝑏𝑏
Where:
Atr is the smaller of area of transverse reinforcement within a spacing s crossing plane of
splitting normal to concrete surface containing extreme tension fibres, or total area of
transverse reinforcement normal to the layer of bars within a spacing, s, divided by n,
mm2. Transverse reinforcement used for shear, flexure or temperature may be included
in Atr. If longitudinal bars are enclosed within spiral or circular hoop reinforcement,
Atr = At when n ≤ 6, where At is the area of a bar formed into a spiral or circular or
rectangular hoop reinforcement, mm2
s is the maximum spacing of transverse reinforcement within Ld, or spacing of stirrups
or ties or spacing of successive turns of a spiral, all measured centre-to-centre, mm
fyt is the lower characteristic yield strength of transverse reinforcement, MPa
n is the number of longitudinal bars in the layer through which a potential plane of
splitting would pass, or the number of bars uniformly spaced around circular sections
db is the nominal diameter of rebar, mm

Eurocode 2 (EN 1992-1-1) takes the effect of transverse reinforcement into account in the
formula of design anchorage length with parameter α3, as:

Σ𝐴𝐴𝑠𝑠𝑠𝑠 − Σ𝐴𝐴𝑠𝑠𝑠𝑠,𝑚𝑚𝑚𝑚𝑚𝑚
0.7 ≤ 𝛼𝛼3 = 1 − 𝐾𝐾 ≤ 1.0
𝐴𝐴𝑠𝑠
Where:
ΣAst is the cross-sectional area of the transverse reinforcement along the design anchorage
length lbd, mm2
ΣAst,min is the cross-sectional area of the minimum transverse reinforcement, mm2 (= 0.25As
for beams and 0 for slabs), mm2
As is the area of a single anchored bar with maximum bar diameter, mm2
K is shown in Figure 1.11

13
 As Ast As Ast As Ast

K = 0.1 K = 0.05 K=0

Figure 1.11 Values of K for beams and slabs in Eurocode 2 (EN 1992-1-1)

The type of loading (e.g., cyclic loading during seismic actions or sustained loading etc.), or
the bond degradation due to e.g., corrosion, crack accumulation by repeated loading or fire
exposure is not directly taken into account in either NZS 3101 or Eurocode 2 (EN 1992-1-1) in
the calculation of development lengths or anchorage lengths. A summary of the influencing
parameters covered by the two codes is given in Table 5.1.

Table 5.1 Summary of influencing parameters in different codes

NZS 3101 Eurocode 2

‘top-bar effect’ parameter 𝛼𝛼𝑎𝑎 𝜂𝜂1
excess reinforcement parameter 𝛼𝛼𝑏𝑏 –
bar size parameter – 𝜂𝜂2
bar shape parameter – 𝛼𝛼1
concrete cover parameter 𝛼𝛼𝑐𝑐 𝛼𝛼2
transverse reinforcement parameter 𝛼𝛼𝑑𝑑 𝛼𝛼3
welded transverse reinforcement parameter – 𝛼𝛼4
confinement by transverse pressure – 𝛼𝛼5

As it was shown, Eurocode 2 (EN 1992-1-1) sets a limit for the allowable contribution of
confinement provided by the concrete cover and transverse reinforcement as:

𝛼𝛼2 𝛼𝛼3 𝛼𝛼5 ≥ 0.7

NZS 3101 sets these limits as:

1.0 ≤ 𝛼𝛼𝑐𝑐 ≤ 1.5

1.0 ≤ 𝛼𝛼𝑑𝑑 ≤ 1.5

Figure 1.12 provides a comparison for this potential contribution of confinement in accordance
with NZS 3101 and Eurocode 2 (EN 1992-1-1). The diagrams indicate the development lengths
needed to achieve 500 MPa steel stress by the basic anchorage length and basic development
length formulas, respectively, together with the maximum allowable reduction due to
confinement, as:

𝜎𝜎𝑠𝑠 𝑑𝑑𝑏𝑏
𝑙𝑙𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 = → 𝑙𝑙𝑏𝑏𝑏𝑏 = 0.7 ∙ 𝑙𝑙𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟
4𝑓𝑓𝑏𝑏𝑏𝑏
14
and:
𝜎𝜎𝑠𝑠 𝑑𝑑𝑏𝑏 𝐿𝐿𝑑𝑑𝑑𝑑
𝐿𝐿𝑑𝑑𝑑𝑑 = 0.5 → 𝐿𝐿𝑑𝑑 =
�𝑓𝑓′𝑐𝑐 2.25

It can be seen in Figure 1.12 that NZS 3101 apparently provides a wider margin to the designers
for taking the contribution of confinement provided by the concrete cover and transverse
reinforcement into account.

C20 concrete C30 concrete

600 600

500 500
Steel stress, MPa

Steel stress, MPa

400 400

300 300

200 200

EC2: 54db → 38db EC2: 41db → 29db

100 100
NZS 3101: 56db → 25db NZS 3101: 46db → 20db
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Ld/db Ld/db

C40 concrete C50 concrete

600 600

500 500
Steel stress, MPa

Steel stress, MPa

400 400

300 300

200 200
EC2: 34db → 24db EC2: 29db → 20db
100 100
NZS 3101: 40db → 18db NZS 3101: 35db → 16db
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Ld/db Ld/db

Figure 1.12 Allowable maximum contribution of confinement provided by the concrete cover
and transverse reinforcement in accordance with NZS 3101 and Eurocode 2 (EN 1992-1-1)

1.4 Reliability verification format in codes

As it was indicated earlier, the strength reduction factor ϕ is not used in the development length
and lap splice length equations (see Clause 2.3.2.2 of NZS 3101 with ϕ = 1.0 for anchorage and
strength development of reinforcement) since ACI 318 claims that an allowance for strength
reduction is already included in the expressions for development and splice lengths. It is not
explained, however, what is the exact magnitude of this strength reduction. To be able to
precisely compare the Eurocode 2 (EN 1992-1-1) anchorage length and lap splice length

15
formulae with the NZS 3101 development length and lap splice length formulae at the same
level of reliability, this assumption would be needed to be known.

Both the Eurocodes and the AS/NZS 1170 + NZS 3101 (as well as ACI 318) follow the
principles of Limit State Design with clear definition of limit states (e.g., SLS, ULS etc.) and
factoring both the loads (actions) and the capacities (resistances). In the US nomenclature, Limit
State Design is called as Load and Resistance Factor Design (LRFD). A structure designed by
Limit State Design is detailed to withstand all actions likely to occur during its specified
intended life (design life), and to remain fit for use, with an appropriate level of reliability in
every limit state. The appropriate levels of reliability are either prescribed explicitly or defined
implicitly in building codes. In the followings a brief comparison is given for the basics of the
Limit State Design approach interpreted in Eurocode 2 (EN 1992-1-1) and in more general in
Eurocodes as well as in AS/NZS 1170 + NZS 3101, related especially to the stress development,
detailing and splicing of reinforcement provisions.

In the Eurocodes (EN 1990 to EN 1999), the design values of the basic variables, Xd (design
value of a material or product property) and Fd (design value of an action), are usually not
introduced directly into the partial factor design equations (i.e., for example Rd ≥ Ed). They are
introduced by their representative values Xrep and Frep, which may be:
a) characteristic values, i.e., values with a prescribed or intended probability of being
exceeded, e.g., for actions, material properties and geometrical properties, or
b) nominal values, which are treated as characteristic values for material properties, and as
design values for geometrical properties.

The representative values Xrep and Frep, should be divided and/or multiplied, respectively, by
the appropriate partial factors to obtain the design values Xd and Fd.

The design value of a material or product property, Xd, can be expressed in general terms as:

a) where a lower value for design resistance is used:

𝑋𝑋𝑘𝑘,𝑖𝑖𝑖𝑖𝑖𝑖
𝑋𝑋𝑑𝑑 = 𝜂𝜂
𝛾𝛾𝑚𝑚

b) where an upper value for design resistance is used:

𝑋𝑋𝑑𝑑 = 𝜂𝜂𝛾𝛾𝑚𝑚 𝑋𝑋𝑘𝑘,𝑠𝑠𝑠𝑠𝑠𝑠

Where η is the mean value of a conversion factor taking into account volume and scale effects,
effects of moisture and temperature, and any other relevant parameters, and γm is the partial
factor for the material or product property to take into account the possibility of an unfavourable
deviation of a material or product property from its characteristic value, and the random part of
the conversion factor η. In accordance with EN 1990, the above expressions can be used for
capacity design.

The design value of geometrical data such as dimensions of members that are used to assess
action effects and/or resistances, ad, may be represented by nominal values:
𝑎𝑎𝑑𝑑 = 𝑎𝑎𝑛𝑛𝑛𝑛𝑛𝑛

The design value of an action, Fd, can be expressed in general terms as:
𝐹𝐹𝑑𝑑 = 𝛾𝛾𝑓𝑓 𝐹𝐹𝑟𝑟𝑟𝑟𝑟𝑟 = � 𝛾𝛾𝑓𝑓,𝑖𝑖 𝜓𝜓𝑖𝑖 𝐹𝐹𝑘𝑘,𝑖𝑖
𝑖𝑖

16
Where γf is a partial factor for the action which takes account of the possibility of unfavourable
deviations of the action values from the representative values, and ψ is the load combination
factor to take into account the reductions in the design values of variable actions for
simultaneously occurring, accompanying variable actions.

In the Eurocodes (EN 1990 to EN 1999), the terms action, F, and effect of action, E, are
differentiated as:
• Action, F, is either the set of forces (loads) applied to the structure (direct action); or
the set of imposed deformations or accelerations caused for example, by temperature
changes, moisture variation, uneven settlement, or earthquakes (indirect action),
• Effect of action, E, is a local or global response of either a structural member (e.g.,
internal force, moment, stress, strain), or the whole structure (e.g., deflection, rotation).

Design values for model uncertainties may be incorporated into the partial factor design
equations (i.e., for example Rd ≥ Ed) through the partial factor γSd (partial factor taking into
account uncertainties in modelling the effects of actions, and in some cases, in modelling the
actions) and partial factor γRd (partial factor covering uncertainty in the resistance model, plus
geometric deviations if these are not modelled explicitly), such that:

𝐸𝐸𝑑𝑑 = 𝛾𝛾𝑆𝑆𝑆𝑆 𝐸𝐸𝑘𝑘 ; 𝑅𝑅𝑑𝑑 = 𝑅𝑅𝑘𝑘 /𝛾𝛾𝑅𝑅𝑅𝑅

It should be noted that in a more general case the effects of actions depend on material
properties.

The design value of the effects of actions for a specific load case, Ed, can be expressed in general
terms as:
𝐸𝐸𝑑𝑑 = 𝛾𝛾𝑆𝑆𝑆𝑆 𝐸𝐸�𝛾𝛾𝑓𝑓,𝑖𝑖 𝐹𝐹𝑟𝑟𝑟𝑟𝑟𝑟,𝑖𝑖 ; 𝑎𝑎𝑑𝑑 � 𝑖𝑖 ≥ 1

In most cases, the following simplification can be made:

𝐸𝐸𝑑𝑑 = 𝐸𝐸�𝛾𝛾𝐹𝐹,𝑖𝑖 𝐹𝐹𝑟𝑟𝑟𝑟𝑟𝑟,𝑖𝑖 ; 𝑎𝑎𝑑𝑑 � 𝑖𝑖 ≥ 1
With:
𝛾𝛾𝐹𝐹,𝑖𝑖 = 𝛾𝛾𝑆𝑆𝑆𝑆 𝛾𝛾𝑓𝑓,𝑖𝑖

The design resistance, Rd, can be expressed in general terms as:

1 𝑋𝑋𝑘𝑘,𝑖𝑖
𝑅𝑅𝑑𝑑 = 𝑅𝑅 �𝜂𝜂𝑖𝑖 ; 𝑎𝑎 � 𝑖𝑖 ≥ 1
𝛾𝛾𝑅𝑅𝑅𝑅 𝛾𝛾𝑚𝑚,𝑖𝑖 𝑑𝑑

In most cases, the following simplification can be made:

𝑋𝑋𝑘𝑘,𝑖𝑖
𝑅𝑅𝑑𝑑 = 𝑅𝑅 �𝜂𝜂𝑖𝑖 ; 𝑎𝑎 � 𝑖𝑖 ≥ 1
𝛾𝛾𝑀𝑀,𝑖𝑖 𝑑𝑑
With:
𝛾𝛾𝑀𝑀,𝑖𝑖 = 𝛾𝛾𝑅𝑅𝑅𝑅 𝛾𝛾𝑚𝑚,𝑖𝑖

When considering a limit state of rupture or excessive deformation of a section, member or

connection in ultimate limit state (ULS), associated with collapse or with other similar forms
of structural failure, it shall be verified that:

𝑅𝑅𝑑𝑑 ≥ 𝐸𝐸𝑑𝑑

17
The relation between individual partial factors in Eurocodes introduced above is schematically
shown in Figure 1.13 (reproduced from Fig. C3 of EN 1990):

Figure 1.13 Relation between individual partial factors in Eurocodes (EN 1990)

In accordance with Clause 2.3.2 of NZS 3101, structures and structural members shall be
designed for strength as follows:
a) The loads and forces giving rise to the ultimate limit state design action, S*, shall be
determined from the governing ultimate limit state combinations specified in AS/NZS
1170 or other referenced loading standard,
b) The design strength of a member or cross section at the ultimate limit state shall be taken
as the nominal strength, Sn, for the relevant action calculated in accordance with the
requirements and assumptions of NZS 3101, multiplied by the applicable strength
reduction factor, ϕ, specified in Clause 2.3.2.2 of NZS 3101,
c) Each member shall be proportioned so that the design strength is equal to or greater than
the design action, in accordance with the following relationship:
S* ≤ ϕSn
where S is replaced by the actions of moment, axial force, shear or torsion as appropriate.

For anchorage and strength development of reinforcement the strength reduction factor is ϕ =
1.0.

In accordance with ISO 22111 “Bases for design of structures - General requirements” a single
resistance factor ϕ (= 1/γM) can also be used instead of the partial factor for resistance γM, which
can consist of a combination of the partial factors for material properties γm and model
uncertainty γRd, respectively, in the same manner as it was explained above in the context of the
Eurocodes.

Based on the findings above it can be concluded that in ultimate limit state (ULS), associated
with collapse or with other similar forms of structural failure, the design equations of the
Eurocodes and that of NZS 3101 can be expressed and considered to be equivalent in general
terms as:

𝑅𝑅𝑑𝑑 ≥ 𝐸𝐸𝑑𝑑

𝜙𝜙𝜙𝜙𝑛𝑛 ≥ 𝑆𝑆 ∗

It can be also observed that due to the strength reduction factor ϕ = 1.0 for anchorage and
strength development of reinforcement in NZS 3101, there is a difference between the steel
stress taken into account for the calculation of the design anchorage length in accordance with

18
Eurocode 2 (EN 1992-1-1) for yield condition (i.e., design yield strength, σs = fyd = fyk/γs), and
the steel stress taken into account for the calculation of the refined development length in
accordance with NZS 3101 for yield condition (i.e., nominal strength = lower characteristic
yield strength, σs = fy).

Capacity safety factors in codes (e.g., ϕ in NZS 3101) may be calibrated by structural reliability
theory methods in accordance with ISO 2394 “General principles on reliability for structures”.
The reason for the reliability-based calibration is to provide an acceptable uniform benchmark
of safety for both new structures being designed, and existing structures being assessed. Code
calibration is a decision optimization problem where the expected societal net benefits are
maximized by optimizing the parameters in the design code. In accordance with ISO 2349,
within this formulation all relevant benefits and consequences associated to the structures
should be considered; i.e. benefits in general for the society, construction costs, maintenance
costs during the design life, and failure costs.

Reliability index, β is often found as a suitable measure for expressing the probability of failure,
pf, and thus level of safety of a structure or a structural member. Reliability index, β is defined
as:

𝛽𝛽 = −𝛷𝛷−1 (𝑝𝑝𝑓𝑓 )

Where Ф-1 is the inverse standard normal probability distribution function. Values of the
reliability index, β are shown in Figure 1.14.

7
6
Reliability index, β

5
4
3
2
1
0
10-10 10-8 10-6 10-4 10-2 100
Probability of failure, pf

Figure 1.14 Reliability index (β) shown as a function of the probability of failure (pf).

For example, ASCE/SEI 7-16 presents theoretical relationships for load factors, γQ (i.e., an
increasing function of the bias in the estimation of the nominal load, the variability in the load,
and the target reliability index), and resistance factors, ϕ (i.e., a decreasing function of the bias
in the estimation of the nominal resistance, the variability in the resistance, and the target
reliability index), that reads:

𝛾𝛾𝑄𝑄 = �𝜇𝜇𝑄𝑄 ⁄𝑄𝑄𝑛𝑛 ��1 + 𝛼𝛼𝑄𝑄 𝛽𝛽𝑉𝑉𝑄𝑄 �

𝜙𝜙 = (𝜇𝜇𝑅𝑅 ⁄𝑅𝑅𝑛𝑛 )𝑒𝑒𝑒𝑒𝑒𝑒[−𝛼𝛼𝑅𝑅 𝛽𝛽𝑉𝑉𝑅𝑅 ]

Where:
μQ is the mean load (for a normal distribution random variable)
Qn is the nominal load
αQ is a sensitivity coefficient
VQ is the coefficient of variation for the load
19
μR is the mean strength (for a log-normal distribution random variable)
Rn is the code-specified strength
αR is a sensitivity coefficient
VR is the coefficient of variation for the strength
β is the reliability index

In ASCE/SEI 7-16 the engineer is cautioned that load and resistance criteria necessary to
achieve a reliability-based performance objective are coupled through the reliability index, β,
and adjustments to the load factors, γQ without corresponding adjustments to the resistance
factors, ϕ would lead to an unpredictable change in structural performance and reliability.

ISO 2394 specifies that the design of structures shall be supported by risk-based robustness
assessments and/or by consideration of robustness provisions in dependence of the exposures
acting on the structure and the consequences of failure. For such purpose, a five-tier system of
Consequence Classes are defined in ISO 2394, which, in general terms, explain and categorize
structures and consequences of failure in a similar way than that is given in Table 3.2 of
AS/NZS 1170.0, for the Importance levels IL1 to IL5. Supplement to AS/NZS 1170.0
highlights that much of the philosophy and some of the text in AS/NZS 1170.0 has been drawn
from ISO Standards, including ISO 2394, ISO 3898, ISO 4356 and ISO 8930. For the level of
safety of structures, ISO 2394 specifies tentative minimum target reliability indices, βT based
on two different considerations:
• By limiting the value of the lifesaving costs based on the Life Quality Index (LQI), or
• By monetary optimization based on the ratio of failure costs to construction costs
(CF/CC)

Table 1.2 summarizes the tentative annual (i.e., related to one year reference period) minimum
target reliability indices, βT,1 based on ISO 2394.

Table 1.2 Tentative annual minimum target reliability index (βT,1) and probability of failure
(pf) based on ISO 2394
Tentative annual minimum target reliability index (βT,1) and probability of failure (pf)
Relative costs of
Based on LQI Based on monetary optimization / Consequence Classes (CC)
safety measures
acceptance criterion CC2 (CF/CC ≈ 2) CC3 (CF/CC = 2-5) CC4 (CF/CC = 5-10)
Large βT,1 = 3.1 βT,1 = 3.1 βT,1 = 3.3 βT,1 = 3.7
pf = 1.0×10-3 pf = 1.0×10-3 pf = 4.8×10-4 pf = 1.1×10-4
Medium βT,1 = 3.7 βT,1 = 3.7 βT,1 = 4.2 βT,1 = 4.4
pf = 1.1×10-4 pf = 1.1×10-4 pf = 1.3×10-5 pf = 5.4×10-6
Small βT,1 = 4.2 βT,1 = 4.2 βT,1 = 4.4 βT,1 = 4.7
pf = 1.3×10-5 pf = 1.3×10-5 pf = 5.4×10-6 pf = 1.3×10-6

Table 1.3 summarizes the recommended minimum values for target reliability index, βT for
ultimate limit states (ULS) in Eurocode 0 (EN 1990). It is emphasized that design using EN
1990 with the partial factors given in Eurocodes (EN 1991 to EN 1999) is considered generally
to lead to a structure with a β value greater than 3.8 for a 50 year reference period. Eurocode 2
(EN 1992-1-1), in particular, claims that design using the partial factors given in it and the
partial factors given in Eurocode 0 (EN 1990) is considered to lead to a structure associated
with Reliability Class RC2. It can be observed that the Eurocodes set a higher safety level than
the tentative minimum levels set by ISO 2394.

20
Table 1.3 Recommended minimum values for target reliability index (βT) in in Table B2 of
Eurocode 0 (EN 1990)
Minimum values for βT and pf
Consequence Class Reliability Class 1 year reference 50 years reference
period period
βT,1 = 5.2, pf = βT,50 = 4.3, pf =
CC3 RC3
1.0×10-7 8.5×10-6
βT,1 = 4.7, pf = βT,50 = 3.8, pf =
CC2 RC2
1.3×10-6 1.1×10-4
βT,1 = 4.2, pf = βT,50 = 3.3, pf =
CC1 RC1 -5
1.3×10 4.8×10-4

Target reliability indices, βT vary from country to country and can be defined for individual
structural members or components. The acceptable value of target reliability indices, βT depends
on the type of failure, the cost of failure, the cost of enhanced safety and the existing safety
level. The type of failure has a major role in defining target reliability indices; if a structural
member or component can fail suddenly without any warning then it shall be designed for a
higher Consequence Class. This principle is reflected in e.g., ASCE/SEI 7-16 for target
reliability indices for 50 years reference period in case of load cases that do not include
earthquake, tsunami, or extraordinary events as: βT,50 = 3.0 for structural elements that have
failure modes that are not sudden and do not lead to widespread progression of damage - for
example, beams resisting tension-controlled flexure with ϕ = 0.90; βT,50 = 3.5 for structural
elements that either have failure modes that are sudden or lead to widespread progression of
damage - for example, beams resisting shear with ϕ = 0.75; and βT,50 = 4.0 for structural
elements that have failure modes that are sudden and lead to widespread progression of damage
- for example, beams resisting compression-controlled flexure with ϕ = 0.65. Note: the strength
reduction factors, ϕ used as examples above are those from ACI 318.

fE
R
fR
E σR
σE
Resistance
μE μR Effect of
fZ Z=R-E β·σZ ti
pf σZ
μZ Safety margin
β·σZ

Figure 1.15 Derivation of the reliability index (β) with denoting Z = R – E safety margin

AS/NZS 1170.0 does not define annual target reliability indices, βT, however, introduces annual
probability of exceedance values for different limit states, load types, importance levels and
21
design working lives. AS/NZS 1170.0 explains that the annual probability of exceedance
depends on the chosen notional importance level and on the design working life of the structure.
Once these two parameters are determined for a structure, the annual probability of exceedance
of an action can be obtained from AS/NZS 1170.0. Then, for each probability, design values
for actions can be found in different parts of the AS/NZS 1170 series. Since the reliability index,
β is a complex measure of the safety margin (i.e., Z = Rd – Ed) and may be interpreted as a
measure of the distance from the most likely realization point to failure (see the graphical
representation on Figure 1.15), the annual probability of exceedance given in AS/NZS 1170.0
cannot be directly translated to reliability index.

Table 1.4 Description of consequences classes in Table B1 of Eurocode 0 (EN 1990) and
consequences of failure for importance levels in Table 3.1 of AS/NZS 1170.0
Table B1 of Eurocode 0 (EN 1990)
Consequences Description
Class
CC1 Low consequence for loss of
human life, and economic,
social or environmental
consequences small or
negligible
CC2 Medium consequence for
loss of human life,
economic, social or
environmental consequences
considerable
CC3 High consequence for loss
of human life, or economic,
social or environmental
consequences very great
- -
Table 3.1 of AS/NZS 1170.0
Consequences Description Importance
of failure level
Low Low consequence for loss of IL1
human life, or small or
moderate economic, social or
environmental consequences
Ordinary Medium consequence for loss IL2
of human life, or considerable
economic, social or
environmental consequences
High High consequence for loss of IL3 or IL4
human life, or very great
economic, social or
environmental consequences
Exceptional Circumstances where reliability IL5
must be set on a case by case
basis

A comparison between Table 3.1 of AS/NZS 1170.0 and Table B1 of EN 1990 is given in Table
1.4. It can be observed that the AS/NZS 1170.0 Importance Levels, IL represent very similar
general consequences to those covered by the Eurocode Consequences Classes, CC.

Clause C2.3.2.2 of NZS 3101 explains that the design strength of a member, as used in NZS
3101, is the nominal strength calculated in accordance with the provisions and assumptions
stipulated in NZS 3101 multiplied by a strength reduction factor, ϕ as detailed in Clause 2.3.2.2
of NZS 3101 for the ultimate limit state (ULS) and 2.6.3.2 for serviceability limit state (SLS)
load combinations involving seismic forces. The rules for computing the nominal strength of a
member are based on chosen limits of stress, strain, cracking or crushing, and conform to
research data for each type of structural action and to established structural engineering practice.
Clause C2.3.2.2 of NZS 3101 further explains that the basis for the selected values of strength
reduction factors are detailed in the study by MacGregor (1983), which ascertained that for the
values of ϕ similar to those in Clause C2.3.2.2 of NZS 3101 and load factors corresponding to
AS/NZS 1170, target values of the reliability index, β (that is called as safety index in NZS
3101) of 3.0 for dead and live load, 2.5 for dead and live and wind forces and 2.0 for dead and
live and earthquake forces applied. These values for the safety index are within the range
implicit in AS/NZS 1170.

22
The above target reliability indices set by NZS 3101 would mean pf = 1.35×10-3 (for βT = 3.0),
pf = 6.21×10-3 (for βT = 2.5), and pf = 2.28×10-2 (for βT = 2.0) probabilities of failure. The length
of reference period corresponding to the given target reliability indices is not detailed in NZS
3101, however, it can be noted that the reliability indices in those earlier studies (e.g., the paper
by MacGregor (1983) above, based on the NBS Special Publication 577, June 1980) were
determined for structural members based on a service period of 50 years.

As it was demonstrated earlier, structural concrete design in accordance with Eurocode 2 (EN
1992-1-1) and the partial factors given in Eurocode 0 (EN 1990) is considered to lead to a RC2
structure associated with an annual target reliability index of βT,1 = 4.7 (i.e., pf = 1.3×10-6
probability of failure) and a target reliability index for 50 years reference period of βT,50 = 3.8
(i.e., pf = 1.1×10-4 probability of failure).

Based on the findings above it can be assumed that the safety level of the design of stress
development, detailing and splicing in accordance with NZS 3101 and AS/NZS 1170 is
different from that of the design in accordance with Eurocodes. A complete, reliability-based
comparison was not aimed within the framework of this Report and the question is addressed
to future research.

It should be noted that design standards, including NZS 3101 and Eurocode 2 (EN 1992-1-1)
set their target reliability indices for new structures. In their paper, Steenbergen et al (2018)
suggest that: “Currently, EN 1990 does not have the notion from ISO 2394 and the JCSS
Probabilistic Model Code, that the target reliability level depends on the marginal safety costs.
Especially for existing structures this is important as it takes more effort to increase the
reliability level compared to a new structure. Consequently, for an existing structure one should
use lower reliability levels. This is also in agreement with the recommendations of the new fib
Model Code 2010.” The authors also add: “In general ISO 2394 and JCSS Probabilistic Model
Code, which are both based on the work of Rackwitz (2000) seem to provide a more appropriate
reliability differentiation for existing structures than EN 1990 and ISO 13822 since costs of
safety measures are taken into account.” In their paper, Steenbergen et al (2018) clarify the link
between the design working life and the reliability index based on the principles developed by
Rackwitz (2000), and guidance is provided for specification of the target reliability level of
existing structures for a given design working life.

As it was indicated in the beginning of this Chapter, the strength reduction factor ϕ is not used
in the development length and lap splice length equations (see Clause 2.3.2.2 of NZS 3101 with
ϕ = 1.0 for anchorage and strength development of reinforcement) since an allowance for
strength reduction is already included in the expressions for development and splice lengths,
however, the exact magnitude of this strength reduction is unknown. In the context of ACI 318-
19, Mahrenholtz et al (2020) assumed ϕc = 0.65 for concrete and ϕy = 0.9 for steel in expressing
an equivalent term for the design bond strength, ϕc·fb′ (not included in ACI 318-19) that would
be needed for the transfer of the design yield capacity ϕy·Ty of the rebar to the concrete by bond
as follows:

𝜙𝜙𝑐𝑐 ∙ 𝑓𝑓 ′ 𝑏𝑏 ∙ 𝜋𝜋 ∙ 𝑑𝑑𝑏𝑏 ∙ 𝐿𝐿𝑑𝑑 = 𝜙𝜙𝑦𝑦 ∙ 𝑇𝑇𝑦𝑦

𝜋𝜋
𝑇𝑇𝑦𝑦 𝑓𝑓𝑦𝑦 ∙ 4 𝑑𝑑𝑏𝑏2 𝑓𝑓𝑦𝑦 ∙ 𝑑𝑑𝑏𝑏
𝜙𝜙𝑐𝑐 ∙ 𝑓𝑓 ′ 𝑏𝑏 = 𝜙𝜙𝑦𝑦 = 𝜙𝜙𝑦𝑦 = 𝜙𝜙𝑦𝑦
𝜋𝜋 ∙ 𝑑𝑑𝑏𝑏 ∙ 𝐿𝐿𝑑𝑑 𝜋𝜋 ∙ 𝑑𝑑𝑏𝑏 ∙ 𝐿𝐿𝑑𝑑 4 ∙ 𝐿𝐿𝑑𝑑

23
 𝑓𝑓𝑦𝑦 ∙ 𝑑𝑑𝑏𝑏
0.65 ∙ 𝑓𝑓 ′ 𝑏𝑏 = 0.9 ∙
4 ∙ 𝐿𝐿𝑑𝑑

Where the development length Ld in accordance with ACI 318-19 can be written as (with
introducing a generalized parameter, K):
𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑 = 𝐾𝐾
�𝑓𝑓′𝑐𝑐

Therefore, the term nominal bond strength, fb′ (also not included in ACI 318-19) can be
expressed as:

𝜙𝜙𝑦𝑦 �𝑓𝑓′𝑐𝑐 0.9 �𝑓𝑓′𝑐𝑐

𝑓𝑓 ′ 𝑏𝑏 = ∙ = ∙
𝜙𝜙𝑐𝑐 4𝐾𝐾 0.65 4𝐾𝐾

Mahrenholtz et al (2020) applied the same assumptions for the derivation of the mean bond
strength, ub (also not included in ACI 318-19) as it was explained by Genesio et al (2017). The
mean bond strength decreases with increasing development length. It is assumed that in case of
pullout failure, the mean bond strength for a bond length Ld is approximately 10% lower
compared to pullout tests with a short bond length (Spieth, 2002) which can be taken into
account with a devisor 0.9 (Mahrenholtz et al, 2020). To calculate the mean bond strength, the
nominal bond strength is divided by the ratio of nominal bond strength and mean bond strength.
A ratio of 0.75 is assumed which is commonly used in the context of concrete failure modes of
adhesive anchors (Genesio et al, 2017). These assumptions would lead to:

𝑓𝑓 ′ 𝑏𝑏 1 0.9 �𝑓𝑓 ′ 𝑐𝑐
𝑢𝑢𝑏𝑏 = = ∙ ∙
0.9 ∙ 0.75 0.9 ∙ 0.75 0.65 4𝐾𝐾

Mahrenholtz et al (2020) assumed db ≤ 19 mm and the maximum permissible influence of

confinement in accordance with ACI 318-19 that resulted in:

𝑢𝑢𝑏𝑏 = 1.76 �𝑓𝑓′𝑐𝑐

Based on the logic applied both in Genesio et al (2017) and in Mahrenholtz et al (2020), one
can formally write:

𝑓𝑓𝑏𝑏𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 ~ 𝑢𝑢𝑏𝑏

Where:
𝛾𝛾𝑐𝑐 ∙ 𝑓𝑓𝑏𝑏𝑏𝑏 40𝑑𝑑𝑏𝑏 0.55 1.5 ∙ 0.315(𝑓𝑓′𝑐𝑐 )2/3 0.55
𝑓𝑓𝑏𝑏𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 = � � = 4 = 1.35(𝑓𝑓′𝑐𝑐 )2/3
0.75 10𝑑𝑑𝑏𝑏 0.75

The required minimum mean bond strength fbm,req corresponds to a bond length equal to 10db.
The rationale of the values for fbm,req given in Table 1.5 is explained in Genesio et al (2017) as
follows. Starting from the design value of bond strength in accordance with EN 1992-1-1, fbd,
the average bond resistance fbm can be derived by multiplying fbd with the material partial factor
γc = 1.5 and the ratio 1/0.75 to calculate the mean value from a given characteristic value
assuming a coefficient of variation of 15% at 90% confidence level for a high number of
samples, as well as including a multiplicator that accounts for the assumption that the decrease

24
of the bond strength with increasing embedment length follows a power law with an exponent
of 0.55 (see fib Model Code 2010, Eq. (6.1-19)). Considering that the bond tests are conducted
with a bond length equal to 10db, in C20/25 concrete, and that for an anchorage far away from
the concrete edge (see Eq. (8.10) of EN 1992-1-1) leads to an anchorage length of 40db to reach
steel yielding, the derivation of the required minimum mean bond strength reads (Genesio et al,
2017):

𝛾𝛾𝑐𝑐 ∙ 𝑓𝑓𝑏𝑏𝑏𝑏 40𝑑𝑑𝑏𝑏 0.55

𝑓𝑓𝑏𝑏𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 = � �
0.75 10𝑑𝑑𝑏𝑏

Where (Borosnyoi and Genesio, 2021):

𝛼𝛼𝑐𝑐𝑐𝑐 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 𝛼𝛼𝑐𝑐𝑐𝑐 [0.7𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 ]

𝑓𝑓𝑏𝑏𝑏𝑏 = 2.25𝜂𝜂1 𝜂𝜂2 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 2.25𝜂𝜂1 𝜂𝜂2 � � = 2.25𝜂𝜂1 𝜂𝜂2 � �=
𝛾𝛾𝑐𝑐 𝛾𝛾𝑐𝑐
2/3
= 2.25𝜂𝜂1 𝜂𝜂2 �𝛼𝛼𝑐𝑐𝑐𝑐 �0.7�0.3𝑓𝑓𝑐𝑐𝑐𝑐 ��/𝛾𝛾𝑐𝑐 �

If η1 = η2 = αct = 1.0 are taken as a simplification, and considering γc = 1.5 then the above
expression for the design bond strength simplifies to a direct function of the characteristic
concrete compressive strength:

2/3
𝑓𝑓𝑏𝑏𝑏𝑏 = 0.315𝑓𝑓𝑐𝑐𝑐𝑐 = 0.315(𝑓𝑓′𝑐𝑐 )2/3

The values of fbm,req and fbd are given in Table 1.5. According to an unpublished database of
confined pullout tests in C20/25 non-cracked concrete with cast-in rebars that were carried out
at the University of Stuttgart, the mean bond strength is fbm ≈ 10 N/mm2 (Genesio et al, 2017).
It can be seen that the above derivation is in good agreement with the experimental observations
with fbm,req = 10.0 N/mm2 for C20/25 concrete.

Table 1.5 Design bond strength, fbd, according to EN 1992-1-1 for good bond conditions and
required minimum mean bond strength, fbm,req, for a bond length of 10db
Design bond Required minimum
strength, fbd, mean bond strength,
Concrete strength according to EN fbm,req, for a bond
class 1992-1-1 for good length of 10db,
bond conditions, N/mm2
N/mm2
C12/15 1.6 7.1
C16/20 2.0 8.6
C20/25 2.3 10.0
C25/30 2.7 11.6
C30/37 3.0 13.1
C35/45 3.4 14.5
C40/50 3.7 15.9
C45/55 4.0 17.2
C50/60 4.3 18.4

25
Figure 1.16 indicates fbm,req and ub over the specified compressive strength of concrete, f’c, in
the range of f’c = 20 MPa (i.e., lower boundary limit for durability in NZS 3101 for a specified
intended life of 50 years) and f’c = 50 MPa (i.e., upper boundary limit for fbm,req).

30

25

Mean bond strength, MPa

𝑓𝑓𝑏𝑏𝑏𝑏,𝑟𝑟𝑟𝑟𝑟𝑟 = 1.35(𝑓𝑓′𝑐𝑐 )2/3
20

15

10
𝑢𝑢𝑏𝑏 = 1.76 �𝑓𝑓′𝑐𝑐
5

0
20 40 60 80 100

Specified concrete strength f’c,MPa

Figure 1.16 Required minimum mean bond strength, fbm,req, and mean bond strength, ub,
represented as a function of the specified compressive strength of concrete, f’c

With a similar logic, one may formally write:

𝑓𝑓𝑏𝑏𝑏𝑏 ~ 𝜙𝜙𝑐𝑐 ∙ 𝑓𝑓 ′ 𝑏𝑏

Based on the assumptions made by Mahrenholtz et al (2020) this would lead to:

𝑓𝑓𝑏𝑏𝑏𝑏 = 0.315(𝑓𝑓′𝑐𝑐 )2/3 ~ 𝜙𝜙𝑐𝑐 ∙ 𝑓𝑓 ′ 𝑏𝑏 = 0.65 ∙ 0.9 ∙ 0.75 ∙ 𝑢𝑢𝑏𝑏 = 0.768�𝑓𝑓′𝑐𝑐

Figure 1.17 indicates fbd (Eurocode 2; EN 1992-1-1) and ϕc·fb′ over the specified compressive
strength of concrete, f’c, in the range of f’c = 20 to 50 MPa (together with the expressions that
were illustrated earlier in Figure 1.6).

6
𝜙𝜙𝑐𝑐 ∙ 𝑓𝑓 ′ 𝑏𝑏 = 0.768�𝑓𝑓′𝑐𝑐
Design bond strength, MPa

5
𝑓𝑓𝑏𝑏𝑏𝑏 = 0.315(𝑓𝑓′𝑐𝑐 )2/3
4

3
0.5�𝑓𝑓′𝑐𝑐 0.57�𝑓𝑓′𝑐𝑐
2

0
20 30 40 50
Specified concrete strength f’c,MPa

Figure 1.17 Design bond strength (Eurocode 2; EN 1992-1-1), fbm,req, and design bond
strength (Mahrenholtz et al, 2020), ϕc·fb′, represented as a function of the specified
compressive strength of concrete, f’c

26
Multiple observations can be made:
• Figure 1.16 reveals that the expression 𝑢𝑢𝑏𝑏 = 1.76 �𝑓𝑓′𝑐𝑐 cannot capture the mean bond
strength of cast-in rebars, fbm,req, as it is derived by Eurocode 2 (EN 1992-1-1) and
Genesio et al (2017). Since both fbm,req and ub target to express the bond strength at a
50% probability level, the two curves indicated in Figure 1.16 would be expected to
coincide, or at least run very closely to each other if both models are acceptable
estimations of the physical phenomena to be described.
• As it was indicated in this Report, Genesio et al (2017) highlighted that according to an
unpublished database of confined pullout tests in C20/25 non-cracked concrete with
cast-in rebars that were carried out at the University of Stuttgart, the mean bond strength
is fbm ≈ 10 N/mm2. The proposed model for fbm,req can reproduce this observation quite
precisely (see Figure 1.16), therefore, the accuracy of the fbm,req model can be
hypothesised. The expression 𝑢𝑢𝑏𝑏 = 1.76 �𝑓𝑓′𝑐𝑐 , however, gives a very conservative
estimate for the fbm ≈ 10 N/mm2 value (the actual value is ub = 7.87 N/mm2).
• Comparing the findings of Figures 1.16 and 1.17 one can observe that the ratios of
fbm,req/fbd and ub/(ϕc·fb′) are different (fbm,req/fbd = 4.29 and ub/(ϕc·fb′) = 2.28). As it was
mentioned above, both fbm,req and ub target to express the bond strength at a 50%
probability level, therefore, this observation allows the assumption that fbd and ϕc·fb′
express the design bond strength at different levels of probability.
• It can be stipulated that the selection of a strength reduction factor ϕc = 0.65 for concrete
in the context of ACI 318-19 is inappropriate if the target is the construction of a mean
bond strength expression, ub, to capture the experimental observations, fbm, for cast-in
rebars (Genesio et al, 2017) at a 50% probability level, as it is derived by Eurocode 2
(EN 1992-1-1), for a bond length equal to 10db.
• It is also noted, however, that in accordance with the North American acceptance
criteria, ICC-ES AC308 (2019) Clause 10.25.2, the minimum required tension bond
strength for C20/25 non-cracked concrete with cast-in rebars is τmin,req = 7.5 N/mm2
rather than fbm,req = 10 N/mm2.

The above observations confirm the need for a complete, reliability-based comparison as future
research. Such comparison was not aimed within the framework of this Report.

27
2. Calibration of the NZS 3101 development and splice length models
The robustness of code formulae can be tested by comparisons made to experimental databases.
This process is referred to as calibration in this Report. Through calibration, the accuracy and
the precision of the code formulae can be observed, together with the sensitivity of the
parameters included in the formulae. It is noted here that the terms accuracy and precision can
be interpreted in multiple ways, therefore, the understanding of these terms in the context of
this Report is given in the followings. A schematic visual interpretation is shown in Figure 2.1.
True
l Average
Probability density

Accuracy l

Value

Precision
Figure 2.1 Interpretation of accuracy and precision

In the context of calibration in this Report, true value represents the recorded observations
available in the experimental database. In the literature, depending on the actual discipline, true
value can be also referred to as reference value. The term value indicated in Figure 2.1 is
understood in the followings as the predicted value for the true value, calculated by the code
formulae. Accuracy means, in general, the closeness (or proximity) of the predicted value (or
the average of the predicted values) to the true value. In the literature accuracy is also referred
to as trueness or bias, and in metrology it is attributed to systematic errors. Precision means, in
general, the closeness of agreement among a set of the predicted values. In the literature a
variety of different terms can be found with the meaning of precision, such as variability,
dispersion, scatter or spread, and in metrology precision is attributed to random errors, related
to repeatability and reproducibility of repeated measurements. One common quantity that is
used for the representation of precision is the standard deviation.

2.1 Existing database

For the calibration of the models proposed in NZS 3101 two separate databases were used. The
database for splices and anchorages in tension is an extension of the database compiled by
Schoening (2018), based on 1230 anchorage and lap tests that were conducted between the
1950s and 2018. The extended database used for this Report comprises 1394 test data, of which
905 are splice specimens and 489 are anchorage specimens. The database does not cover
anchorages with bends and hooks or lapped bars with different bar diameters The database for
splices and anchorages in compression is that of ACI 408 Compression Lap Splice Database
04-2021. In this chapter the terms anchorage and development are used interchangeably. The
term anchorage or anchorage length is used when referring to the length over which the rebar
was developed in an experiment by straight end anchorage; and the term development length is
used when referring to the prediction of the development length according to NZS 3101 or
another model.

28
2.2 Database filtering in tension

For the calibration of the NZS 3101 development and splice length models in tension the
following filter was used to narrow the database for the limits that fit to the NZS 3101
provisions:
• Concrete strength – 20 to 85 MPa, in accordance with Clause 8.6.3.2 of NZS 3101 (max.
70 MPa) with an allowance of 15 MPa overstrength in accordance with Clause 2.6.5.5
of NZS 3101
• Rebar diameter – between 8 mm and 40 mm
• Rebar stress – below 675 MPa, in accordance with Clause 5.3.3 of NZS 3101 (max. 500
MPa) with an allowance of ϕo,fy =1.35 overstrength in accordance with Clause 2.6.5.5
of NZS 3101
• Rebar spacing – larger than 25 mm, in accordance with Clause 8.3.1 of NZS 3101
• Splice or anchorage length – larger than 10db in general, with a cut-off for ≥ 300 mm,
in accordance with Clause 8.6.3.1 of NZS 3101

The above filtering resulted in a reduction of the 1394 original test data to 507 splice specimens
and 78 anchorage specimens. The database collects the following experimental data:
• Ld,exp the experimental splice length or anchorage length
• σs,exp the experimental maximum rebar stress at failure
• fcm,exp the mean compressive strength of the concrete in the specimen
• fy the yield strength of the longitudinal rebars
• fR the relative projected rib area (when available)
• db the diameter of the longitudinal rebars
• nb the number of the longitudinal rebars spliced or anchored
• sb the spacing of the longitudinal rebars
• cx, cy concrete cover at the face and the side of the beam
• dt the diameter of the transverse reinforcement
• nt the number of the transverse reinforcement
• st the spacing of the transverse reinforcement

2.3 Calibration of splices and anchorages in tension

The NZS 3101 development and splice length models give an estimate on the characteristic
level since both the concrete and the steel material properties are taken into account as their
characteristic strength, i.e., for the concrete as the specified strength and for the steel as the
lower characteristic yield strength. To be able to calibrate the model and compare predicted
values with experimental values collected in the database, a transformation to the mean level
of estimation is needed.

NZS 3101 does not provide a relationship between the mean and characteristic strength of
concrete or steel, however, technical literature can offer such information. For this Report, the
provisions of the fib Model Code 2010 are adopted for concrete. The relationship between the
mean (fcm) and characteristic (f’c) compressive strength of concrete can be taken as:

𝑓𝑓𝑐𝑐𝑐𝑐 = 𝑓𝑓′𝑐𝑐 + 8 𝑀𝑀𝑀𝑀𝑀𝑀

29
For the yield strength of steel Wisniewski et al (2012) is used, who collected worldwide data on
the mechanical properties of concrete, reinforcing steel and pre-stressing steel, and reported a
coefficient of variation for the yield strength of steel in the range of Vsy = 2.3 to 10.7 percent.
For adopting the same format of:

𝑓𝑓𝑦𝑦𝑦𝑦 = 𝑓𝑓𝑦𝑦 + ∆

for the mean value of yield strength, the possible values of Δ are shown below:

Vsy = 2.3 % Vsy = 5.0 % Vsy = 10.7 %

Grade 300 11.8 MPa 26.9 MPa 64.1 MPa
Grade 400 15.7 MPa 35.8 MPa 85.4 MPa
Grade 500 19.7 MPa 44.8 MPa 106.8 MPa

For the model calibration in this Report a value of Δ = 50 MPa has been selected.
Based on these assumptions, the NZS 3101 development and splice length model can be
transformed to the mean level of estimation as:

(0.5𝛼𝛼𝑎𝑎 𝛼𝛼𝑏𝑏 𝑓𝑓𝑦𝑦 ) 0.5𝛼𝛼𝑎𝑎 𝛼𝛼𝑏𝑏 (𝑓𝑓𝑦𝑦 + 50 𝑀𝑀𝑀𝑀𝑀𝑀)

𝐿𝐿𝑑𝑑𝑑𝑑 = 𝑑𝑑𝑏𝑏 → 𝐿𝐿𝑑𝑑𝑑𝑑,𝑚𝑚 = 𝑑𝑑𝑏𝑏
𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐′ 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐′ + 8 𝑀𝑀𝑀𝑀𝑀𝑀

As the first step, the ratio Ld,calc/Ld,exp of the predicted mean development (or splice) lengths,
Ld,calc to the experimental anchorage (or splice) lengths, Ld,exp was generated for the filtered 507
splice specimens and 78 anchorage specimens as:

𝐿𝐿𝑑𝑑,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 �0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏 ���𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 �

=
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒

The visual representation of the Ld,calc/Ld,exp ratio over the normalized values of the experimental
anchorage (or splice) lengths (Ld,exp/db) is shown separately for splices in Figure 2.2, anchorages
in Figure 2.3, and combined for splices and anchorages in Figure 2.4. It can be observed that
the NZS 3101 development (or splice) length model is not robust. It is accurate only in the small
vicinity of 30db splice or anchorage lengths, otherwise overestimates the experimental data for
shorter lengths and underestimates the experimental data for longer lengths.

2.5

Mean: 1.203
2
SD: 0.383
Ld,calc/Ld,e

1.5

0.5

0
0 10 20 30 40 50 60 70 80 90
Ld,exp/db
Figure 2.2 Ratio Ld,calc/Ld,exp predicted by NZS 3101 in tension for the filtered 507 splice
specimens
30
 2.5

Mean: 1.165
2
SD: 0.463

Ld,calc/Ld,e 1.5

0.5

0
0 10 20 30 40 50 60 70 80 90

Ld,exp/db
Figure 2.3 Ratio Ld,calc/Ld,exp predicted by NZS 3101 in tension for the filtered 78 anchorage
specimens

2.5

2
Mean: 1.198
SD: 0.394
Ld,calc/Ld,e

1.5

0.5

0
0 10 20 30 40 50 60 70 80 90

Ld,exp/db

Figure 2.4 Ratio Ld,calc/Ld,exp predicted by NZS 3101 in tension for all filtered specimens

To understand which parameter of the NZS 3101 development (or splice) length model has
correlation with the experimental data, correlation charts have been created for the following
rearrangements of the NZS 3101 model:

𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 ~
0.5𝑑𝑑𝑏𝑏

𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

𝑑𝑑𝑏𝑏 ~
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 ~
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
𝛼𝛼𝑐𝑐 𝑜𝑜𝑜𝑜 𝛼𝛼𝑑𝑑 ~
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

31
The correlation charts are shown in Figure 2.5. The following observations can be made for the
NZS 3101 model:
• there is strong positive correlation with the stress developed in the rebars in the
experiments, σs,exp
• there is strong positive correlation with the concrete strength in the experiments,
�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒
• there is strong positive correlation with the rebar diameter, db
• there is weak positive correlation with the transverse reinforcement parameter, αd
• there is weak or negligible correlation with the concrete cover parameter, αc

It is reported in the literature that concrete cover and rebar spacing do have an influence on the
development length, therefore, the possible correlation of such an influence was further
analysed. Additional correlation charts have been prepared for the minimum, cmin = min(cx ; cy)
and maximum, cmax = max(cx ; cy) concrete covers; for the ratio of the maximum and minimum
concrete covers, cmax/cmin; for parameter cm defined in NZS 3101; and for parameter cd defined
in Eurocode 2 (EN 1992-1-1). The correlation charts are shown in Figure 2.6. The following
observations can be made for the NZS 3101 model:
• there is weak or negligible correlation with the parameter cm defined in NZS 3101
• there is weak or negligible correlation with cmin, cmax or cmax/cmin
• there is weak positive correlation with the parameter cd defined in Eurocode 2 (EN 1992-
1-1)

The biggest shortcoming of the NZS 3101 development (or splice) length model is its
robustness. The parameter analysis demonstrated that the current composition (i.e., linear
combination) of the main, strongly correlating parameters (i.e., stress in the rebars, σs,exp, rebar
diameter, db, and concrete strength, fcm,exp) cannot provide an even accuracy for both short and
long development (or splice) lengths (see Figures 2.2 to 2.4). One way of addressing this non-
linearity can be the introduction of different exponents for the main influencing parameters.
One indicative example is shown below for the 507 splice specimens and 78 anchorage
specimens filtered for the NZS 3101 boundary conditions (emphasizing that this model is not
intended to be a general proposal, it serves only as an illustration):
1.88
𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒
𝐿𝐿𝑑𝑑𝑑𝑑 = 0.0075𝑑𝑑𝑏𝑏 � �
�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

For simplicity, the weakly correlating influencing factors, such as the concrete cover (or any
combination of concrete cover and rebar spacing) and the transverse reinforcement is not
considered in the formula above. As a comparison, the ratio Ldm/Ld,exp for the predicted mean
development (or splice) lengths, Ldm, to the experimental anchorage (or splice) lengths, Ld,exp,
was generated, and it is illustrated in Figure 2.7 over the normalized values of the experimental
anchorage (or splice) lengths (Ld,exp/db). Comparing Figures 2.4 and 2.7 it can be observed that
the accuracy of the prediction is significantly improved, and the illustrated formula with a non-
linear combination of the main influencing parameters can provide a more even accuracy over
the complete range of Ld,exp/db for the 507 splice specimens and 78 anchorage specimens filtered
for the NZS 3101 boundary conditions. It can also be observed that the precision of the
prediction is not improved that might confirm the potential benefit of including the parameters
that account for confinement (concrete cover, transverse reinforcement area etc.).

32
Correlation charts for the parameters in the NZS 3101 development (or splice) length model

(a) (b)
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑝𝑝 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑
0.5𝑑𝑑𝑏𝑏

100 200 300 400 500 600 700 4 5 6 7 8 9 10

𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

(c)
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒

0 5 10 15 20 25 30 35 40 45
𝑑𝑑𝑏𝑏

(d) (e)
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

1 1.1 1.2 1.3 1.4 1.5 1 1.1 1.2 1.3 1.4 1.5
𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑

Figure 2.5 Correlation charts for the parameters in the NZS 3101 development (or splice)
length model

𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

• Chart a): stress developed in the rebars in the experiments, 𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 ~ 0.5𝑑𝑑𝑏𝑏

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart b): concrete strength in the experiments, �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑

𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝛼𝛼𝑐𝑐 𝛼𝛼𝑑𝑑 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

• Chart c): rebar diameter, 𝑑𝑑𝑏𝑏 ~ 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart d): NZS 3101 concrete cover parameter, 𝛼𝛼𝑐𝑐 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart e): NZS 3101 transverse reinforcement parameter, 𝛼𝛼𝑑𝑑 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

33
 Correlation charts for concrete cover related parameters

(a) (b)
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0 20 40 60 80 100 120 140 160 0 10 20 30 40 50 60 70 80 90

𝑐𝑐𝑚𝑚 𝑐𝑐𝑑𝑑

(c) (d)
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑥𝑥𝑥𝑥 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒 𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0 10 20 30 40 50 60 70 80 90 0 20 40 60 80 100 120 140 160 180

𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚

(e)
0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0 20 40 60 80 100 120 140 160 180

𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 ⁄𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚

Figure 2.6 Correlation charts for concrete cover related parameters

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart a): parameter cm defined in NZS 3101, 𝑐𝑐𝑚𝑚 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart b): parameter cd defined in Eurocode 2, 𝑐𝑐𝑑𝑑 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart c): minimum concrete cover, 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏
• Chart d): maximum concrete cover, 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 ~ 𝐿𝐿
𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 0.5𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏

• Chart e): ratio of max. and min. concrete covers, ~ 𝐿𝐿
𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 �𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

34
 1.88
𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒
𝐿𝐿𝑑𝑑𝑑𝑑 = 0.0075𝑑𝑑𝑏𝑏 � �
�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

2.5

2
Mean: 1.063
SD: 0.410
Ldm/Ld,exp

1.5

0.5

0
0 10 20 30 40 50 60 70 80 90
Ld,exp/db
Figure 2.7 Ratio Ldm/Ld,exp predicted for the splice and anchorage specimens filtered for NZS 3101

The current state-of-the-art (see e.g., fib Bulletin 72 (fib, 2014) and fib Model Code 2010) for
the prediction of the bond-splitting capacity of lap splices or single rebar anchorages utilizes a
non-linear combination of selected influencing parameters in the following form for the mean
stress developed, fstm:
𝑓𝑓𝑐𝑐𝑐𝑐 0.25 𝑙𝑙𝑏𝑏 0.55 25 0.20 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 0.25 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚 0.10
𝑓𝑓𝑠𝑠𝑠𝑠𝑠𝑠 = 54 ∙ � � � � � � �� � � � + 𝑘𝑘𝑚𝑚 𝐾𝐾𝑡𝑡𝑡𝑡 �
25 𝑑𝑑𝑏𝑏 𝑑𝑑𝑏𝑏 𝑑𝑑𝑏𝑏 𝑐𝑐𝑚𝑚𝑚𝑚𝑚𝑚

Schoening (2018) proposed a rearranged and extended formula of the above for the direct
prediction of the mean value of splice length or development length as it is shown below, where
the term [αc + αtr + αtp] accounts for the confinement provided by concrete cover, transverse
reinforcement and transverse pressure:

𝜎𝜎𝑠𝑠 1.82 25 0.455 𝑑𝑑𝑏𝑏 0.36 −1.82

𝐿𝐿𝑑𝑑𝑑𝑑 = 45𝑑𝑑𝑏𝑏 � � � � � � �𝛼𝛼𝑐𝑐 + 𝛼𝛼𝑡𝑡𝑡𝑡 + 𝛼𝛼𝑡𝑡𝑡𝑡 �
435 𝑓𝑓𝑐𝑐𝑐𝑐 25

This model is not analysed further within the framework of this Report, however, can serve as
a basis for future advances of the development (or splice) length models in NZS 3101; readers
can find details about the robustness and accuracy of the above prediction in the thesis of
Schoening (2018).

2.4 Database filtering in compression

For the calibration of the NZS 3101 development and splice length models in compression, the
filtering methodology explained in Chapter 2.2 of this Report was used with one difference:
• Splice or anchorage length – larger than 10db in general, with a cut-off for ≥ 0.040fydb
≥ 200 mm, in accordance with Clause 8.6.5.1 of NZS 3101

The database for splices and anchorages in compression was that of ACI 408 Compression Lap
Splice Database 04-2021 that complies experimental data for 176 specimens from compression

35
tests conducted on columns with lap spliced rebars from the literature. The above filtering
resulted in a reduction of database to 125 specimens.

2.5 Calibration of splices and anchorages in compression

As it was introduced earlier, the refined development (or splice) length in compression proposed
in NZS 3101 reads:

𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏
𝐿𝐿𝑑𝑑 = 𝛼𝛼𝑏𝑏 𝛼𝛼𝑒𝑒 𝐿𝐿𝑑𝑑𝑑𝑑 = 0.22𝛼𝛼𝑏𝑏 𝛼𝛼𝑒𝑒 ≥ 0.040𝑓𝑓𝑦𝑦 𝑑𝑑𝑏𝑏 ≥ 200 𝑚𝑚𝑚𝑚
�𝑓𝑓′𝑐𝑐

The ratio Ld,calc/Ld,exp of the predicted mean splice lengths, Ld,calc to the experimental splice
lengths, Ld,exp was generated for the filtered 125 specimens (during which the parameters αb and
αe were both set to unity):

𝐿𝐿𝑑𝑑,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 0.22𝜎𝜎𝑠𝑠,𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑏𝑏 ⁄�𝑓𝑓𝑐𝑐𝑐𝑐,𝑒𝑒𝑒𝑒𝑒𝑒

=
𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒 𝐿𝐿𝑑𝑑,𝑒𝑒𝑒𝑒𝑒𝑒

The visual representation of the Ld,calc/Ld,exp ratio over the normalized values of the experimental
splice lengths (Ld,exp/db) is shown in Figure 2.8.a for the filtered 125 specimens and in Figure
2.8.b for a sub-group of the filtered data, not showing the results corresponding to Ld,exp/db ≈
10. It can be observed that the NZS 3101 model is not robust and in most of the cases
underestimates the experimental data. Regarding the very limited size of the database, the splice
lengths in compression are not analysed any further in this Report.
1.75

1.5
a)
1.25
Ld,calc/Ld,exp

0.75

0.5

0.25

0
0 5 10 15 20 25 30 35 40
Ld,exp/db

1.75

1.5
b)
1.25
Ld,calc/Ld,exp

0.75

0.5

0.25

0
0 5 10 15 20 25 30 35 40
Ld,exp/db

Figure 2.8 Ratio Ld,calc/Ld,exp predicted by NZS 3101 in compression

36
References
ACI 318 Building Code Requirements for Structural Concrete
Angst, U. M., Geiker, M. R., Michel, A., Gehlen, C., Wong, H., Isgor, O. B., Elsener, B.,
Hansson, C. M., Francois, R., Hornbostel, K., Polder, R., Alonso, M. C., Sanchez, M.,
Correia, M. J., Criado, M., Sagues, A., Buenfeld, N. (2017) The steel–concrete interface,
Materials and Structures (2017) Vol. 50, No. 143, pp. 1-24,
https://doi.org/10.1617/s11527-017-1010-1
AS 3600:2018 Concrete structures
AS/NZS 1170.0 Structural design actions. Part 0: General principles
ASCE/SEI 7-16 Minimum Design Loads and Associated Criteria for Buildings and Other
Structures
Borosnyoi, A., Genesio, G. (2021) Post-installed rebars for seismic applications in concrete-
to-concrete connections, SESOC Journal, Vol. 34, No.2, September 2021, pp. 52-65
Carino, N. J., Guthrie, W. F., Lagergren, E. S. (1994) Effects of Testing Variables on the
Measured Compressive Strength of High-Strength (90 MPa) Concrete, NISTIR, Report No.
5405, National Institute of Standards and Technology, Gaithersburg
CEB Bulletin 228 (1995) State of the Art Report, High Performance Concrete, Recommended
Extensions to the Model Code 90, Research Needs, CEB Bulletin d'Information, No. 228,
CEB – Comité Euro-international du Béton, Lausanne, 1995, p. 60
CEB-FIP Model Code 1990 (1993) CEB Bulletin d'Information, No. 213/214, CEB – Comité
Euro-international du Béton, Lausanne, 1993, p. 460
Chamberlin, S. J. (1952) Spacing of Spliced Bars in Tension Pull-Out Specimens, ACI
Journal, December 1952, Vol. 49, No. 12, pp. 261-274, https://doi.org/10.14359/11817
Chamberlin, S. J. (1956) Spacing of Reinforcement in Beams, ACI Journal, July 1956, Vol.
53, No. 7, pp. 113-134. https://doi.org/10.14359/11505
Chamberlin, S. J. (1958) Spacing of Spliced Bars in Beams, ACI Journal, February 1958,
Vol. 54, No. 2, pp. 689-697, https://doi.org/10.14359/11463
Concrete NZ (2021) Concrete production guide for New Zealand, July 2021, p. 60
Cook, J. E. (1989) 10,000 psi Concrete, Concrete International, Vol. 11, No. 10, October
1989, pp. 67-75.
Cramer, J., Javidmehr, S., Empelmann, M. (2021) Simulation of Crack Propagation in
Reinforced Concrete Elements, Applied Sciences, 2021, Vol. 11, 785, pp. 1-19,
https://doi.org/10.3390/app11020785
Day, R. L. (1994) Strength Measurement of Concrete Using Different Cylinder Sizes: A
Statistical Analysis, Cement, Concrete, and Aggregates, CCAGPD, Vol. 16, No. 1, June
1994, pp. 21-30.
de Rooij, M. R., Bijen, J. M. J. M., Frens, G. (1998) Introduction of syneresis in cement paste,
Proceedings, 2nd Int. RILEM Conf. on The Interfacial Transition Zone in Cementitious
Composites, Haifa, Israel, 1998, E & FN Spon, pp. 59-66
Domone, P., Illston, J. (2010) Construction Materials: Their nature and behaviour, 4th Edition,
Spon Press, 584 p.
EN 1990 Eurocode - Basis of structural design
EN 1992-1-1 Eurocode 2: Design of concrete structures - Part 1-1 : General rules and rules for
buildings
EN 206 Concrete - Specification, performance, production and conformity
fib Bulletin 42 (2008) Constitutive modelling of high strength / high performance concrete,
State-of-art report prepared by Task Group 8.2, fédération internationale du béton (fib),
Lausanne, January 2008, p. 134

37
fib Bulletin 72 (2014) Bond and anchorage of embedded reinforcement: Background to the fib
Model Code for Concrete Structures 2010. Technical report prepared by Task Group 4.5,
fédération internationale du béton (fib), Lausanne, May 2014, p. 170.
fib Model Code 2010 (2013) Model Code for Concrete Structures 2010, prepared by Special
Activity Group 5, fédération internationale du béton (fib), Lausanne, 2013, Ernst & Sohn,
p.436
Genesio, G., Piccinin, R., Silva, J. (2017) Qualification of a system for post-installed
reinforcing bars under the rules established by EOTA and ICC-ES, Proceedings ConSC-
2017, 3rd International Symposium on Connections between Steel and Concrete, Stuttgart,
Germany, 2017, pp. 186-197
Gonnerman, H. F. (1925) Effect of Size and Shape of Test Specimen on Compressive
Strength of Concrete, ASTM Proceedings, Vol. 25, pp. 237-255.
Goto, Y., Otsuka, K. (1980) Experimental studies on cracks formed in concrete around
deformed tension bar, Proceedings of the Japan Society of Civil Engineers, No. 294,
February 1980, pp. 85-100 (in Japanese)
Hamad, B. S., Mansour, M. Y. (1996) Bond Strength of Noncontact Tension Lap Splices, ACI
Structural Journal, May-June 1996, Vol. 93, No. 3, pp. 316-326.
https://doi.org/10.14359/9691
Hong, S-G., Chun, S-C. (2007) Splice Strengths of Noncontact Lap Splices Using Strut-and-
Tie Model, Journal of the Korea concrete Institute, April 2007, Vol. 19, No. 2, pp. 199-
207, (in Korean) https://doi.org/10.4334/JKCI.2007.19.2.199
Hwang, H-J., Yang, F., Ma, G. (2022) Effect of Noncontact Lap Splices in Reinforced
Concrete Beams, ACI Structural Journal, March 2022, Vol. 119, No. 2, pp. 3-18,
https://doi.org/10.14359/51734374
ICC-ES AC308 (2019) Acceptance criteria for post-installed adhesive anchors in concrete
elements
ISO 13822:2010 Bases for design of structures - Assessment of existing structures
ISO 22111:2019 Bases for design of structures - General requirements
ISO 2394:2015 General principles on reliability for structures
ISO 3898:2013 Bases for design of structures - Names and symbols of physical quantities and
generic quantities
ISO 4356:1977 Bases for the design of structures - Deformations of buildings at the
serviceability limit states
ISO 8930:2021 General principles on reliability for structures - Vocabulary
JCSS Probabilistic Model Code (2002) Joint Committee for Structural Safety (JCSS),
https://www.jcss-lc.org/jcss-probabilistic-model-code
Jirsa, J. O., Lutz, L. A., Gergely, P. (1979) Rationale for Suggested Development, Splice, and
Standard Hook Provisions for Deformed Bars in Tension, Concrete International, Vol.1,
No. 7, pp. 47-61
Juhász, K. P. (2019) The effect of the synthetic fibre reinforcement on the fracture energy of
the concrete, Proceedings IC-CMTP5, 5th International Conference on Competitive
Materials and Technology Processes, Miskolc-Lillafüred, Hungary, IOP Conf. Series:
Materials Science and Engineering, Vol. 613 (2019) 012037, pp. 1-9,
https://doi.org/10.1088/1757-899X/613/1/012037
Kilpatrick, A. E., Gilbert, R. I. (2012) A Preliminary Investigation of the Strength and
Ductility of Lapped Splices of Reinforcing Bars in Tension, Proceedings, 22nd
Australasian Conference on the Mechanics of Structures and Materials (ACMSM 2012),
Sydney, Australia, 2012, pp. 305-311
MacGregor, J. G. (1983) Load and Resistance Factors for Concrete Design, ACI Journal, Vol.
80 (1983) No. 4, pp. 279-287, https://doi.org/10.14359/10845

38
Mahrenholtz, C., Eligehausen, R., Reinhardt, H-W. (2020) Qualification and Design of Cast-
in-Place and Post-Installed Reinforcing Bar Anchorages, ACI Structural Journal, March
2020, Vol. 117, No. 2, pp. 3-16, https://doi.org/10.14359/51720192
Masud, M. (2019) Non-contact lap splices in geometrically dissimilar bridge column to
drilled shaft connections, PhD thesis, Faculty of the Department of Civil and
Environmental Engineering, University of Houston, 2019, p. 241, https://uh-
ir.tdl.org/handle/10657/5788
McLean, D. I., Smith, C. L. (1997) Noncontact Lap Splices in Bridge Column-Shaft
Connections, Final Report WA-RD 417.1, Washington State Department of Transportation,
July 1997, p. 55, https://www.wsdot.wa.gov/research/reports/fullreports/417.1.pdf
Mehta, P. K., Monteiro, P. J. M. (2006) Concrete: Microstructure, Properties, and Materials,
3rd Edition, McGraw-Hill Education, 2006, 659 p.
NBS Special Publication 577 (1980) Development of a Probability Based load Criterion for
American National Standard A58, Eds.: Ellingwood, B., Galambos, T. V., MacGregor, J.
G. and Cornell, C. A., p. 236,
https://nehrpsearch.nist.gov/static/files/NIST/PB80196512.pdf (accessed July 2022)
Nilson, A. H., Darwin, D. and Dolan, C. W. (2004) Design of Concrete Structures, 13th
Edition, McGraw-Hill, p.779.
NZS 3101:2006 Concrete structures standard
Orangun, C. O., Jirsa, J. O., Breen, J. E. (1975) The Strength of Anchored Bars: A
Reevaulation of Test Data on Development Length and Splices, Research Report No. 154-
3F, Center for Highway Research, The University of Texas at Austin, Jan. 1975
Orangun, C. O., Jirsa, J. O., Breen, J. E. (1977) A Reevaulation of Test Data on Development
Length and Splices, ACI Journal, March 1977, Vol. 74, No. 3, pp. 114-122,
https://doi.org/10.14359/10993
Price, W. H. (1951) Factors Influencing Concrete Strength, ACI Journal, Vol. 22, No. 6,
February 1951, pp. 417-432. https://doi.org/10.14359/12003
Rackwitz, R. (2000) Optimization — the basis of code-making and reliability verification,
Structural Safety, Vol. 22 (2000) No. 1, pp. 27-60. https://doi.org/10.1016/S0167-
4730(99)00037-5
Riedel, P., Leutbecher, T., Piotrowski, S., Heese, C. (2019) Ratios of Compressive Strengths
of Ultra-High-Performance Concrete Specimens of Different Shapes and Sizes, ACI
Materials Journal, Vol. 116, No. 6, November 2019, pp. 139-149,
https://doi.org/10.14359/51716983
Robinson, J. R., Zsutty, T. C., Guiorgadze, G., Lima, L. J., Hoang, L. H., Villatoux, J. P.
(1974) La Couture des Jonctions par Adherence (Bar overlap adhesion connections),
Annales de l'Institut Technique du Batiment et des Travaux Publics, Serie Beton, No. 141,
CEB Bulletin d’Information, No. 99, March 1974, Comité Européen du Béton (CEB), p.
145 (in French)
Sagan, V. E., Gergely, P., White, R. N. (1988) The Behavior and Design of Noncontact Lap
Splices Subjected to Repeated Inelastic Tensile Loading, Technical Report NCEER-88-
0033, National Center for Earthquake Engineering Research, Buffalo, New York,
December 8, 1988, p. 164
Sagan, V. E., Gergely, P., White, R. N. (1991) Behavior and Design of Noncontact Lap
Splices Subjected to Repeated Inelastic Tensile Loading, ACI Structural Journal, July-
August 1991, Vol. 88, No. 4, pp. 420-431, https://doi.org/10.14359/3037
Schoening, J. C. (2018) Anchorages and Laps in Reinforced Concrete Members under
Monotonic Loading, PhD thesis, Fakultät für Bauingenieurwesen der Rheinisch-
Westfälischen Technischen Hochschule Aachen, 2018, p. 394, https://d-
nb.info/1195446640/34

39
Steenbergen, R. D. J. M., Rózsás, Á. and Vrouwenvelder, A. C. W. M. (2018) Target
reliability of new and existing structures - A general framework for code making, HERON,
Vol. 63 (2018) No. 3, pp. 219-242. https://heronjournal.nl/63-3/2.pdf (accessed July 2022)
Untrauer, R. E., Warren, G. E. (1977) Stress Development of Tension Steel in Beams, ACI
Journal, August 1977, Vol. 74, No. 8, pp. 368-372, https://doi.org/10.14359/11018
Wisniewski, D. F., Cruz, P. J. S., Henriques, A. A. R., Simoes, R. A. D. (2012) Probabilistic
models for mechanical properties of concrete, reinforcing steel and pre-stressing steel,
Structure and Infrastructure Engineering, Vol. 8, No. 2, February 2012, pp. 111–123
https://doi.org/10.1080/15732470903363164
Xu You-lin (1992) Experimental Study of Bond-Anchorage Properties for Deformed Bars in
Concrete. Proceedings CEB - International Conference: Bond in Concrete, Riga 1992, Part
I. pp. 9-17
Yi, S-T., Yang, E-I., Choi, J-C. (2006) Effect of specimen sizes, specimen shapes, and
placement directions on compressive strength of concrete, Nuclear Engineering and
Design, Vol. 236, pp.115–127, https://doi.org/10.1016/j.nucengdes.2005.08.004

40
Appendix A – The tensile strength of concrete
The bond strength of deformed cast-in rebars is related to the tensile strength of concrete in
design codes (e.g., Eurocode 2 (EN 1992-1-1), CEB-FIP Model Code 1990 or fib Model Code
2010), as an approximation. Based on the presumed connection between the tensile and
compressive strength of concrete, a direct relationship can be derived between the design bond
strength and the specified compressive strength of concrete (i.e. the characteristic compressive
strength of concrete) based on e.g., the provisions given in Eurocode 2 (EN 1992-1-1), that
reads (Borosnyoi and Genesio, 2021):

𝑓𝑓𝑏𝑏𝑏𝑏 = 2.25𝜂𝜂1 𝜂𝜂2 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 0.315(𝑓𝑓′𝑐𝑐 )2/3

Tensile and compressive strength of concrete, however, are not proportional to each other.
Particularly for higher strength concretes an increase in the compressive strength leads only to
a small increase in the tensile strength (fib, 2008), that is usually taken into account in the
simplified relationships by assuming a nonlinear function between tensile and compressive
strength. One of the reasons for this phenomenon is the increasing brittleness of the cement
paste with increasing strength. Due to the finer and denser pore structure of high strength
concrete (compared to lower strength concrete) the crack propagation in the high strength
cement paste is less hindered by air voids or even by the aggregates. Crack propagation in
concrete is further described in the followings.

a)
b) c) d)

Figure A.1 a) Direct tensile test of ∅150/300 mm concrete cylinder (Cramer et al, 2021), and
schematic illustration of the fracture process zone and the crack propagation of concrete under
uniaxial tension (after Juhász, 2019 and Cramer et al, 2021), where b) micro-crack formation,
c) micro-crack connecting, d) tensile fracture failure

Several indirect and direct testing methods exist for the testing of the tensile strength of
concrete. Uniaxial concrete tensile strength is determined most effectively using a direct tensile
test. Figure A.1 illustrates a ∅150/300 mm concrete cylinder after tensile fracture failure in
direct tensile testing and the schematic representation of the fracture process zone and the crack

41
propagation of concrete under uniaxial tension. As it is shown, the micro-cracks in the cement
paste accumulate at the weakest section of the test specimen and connect into a macroscopic
crack at failure. The location of this failure plane cannot be predicted before testing (and is
rarely located at the mid-section of the cylindrical specimen) and the minimum tensile strength
fct,min of a specimen corresponds to the actual fracture tensile strength of the weakest section
(Juhász, 2019 and Cramer et al, 2021). In lower strength concretes both the micro-cracks and
the macroscopic crack formed at failure tend to follow the boundaries of the aggregate particles
(see Figure A.2) due to the high porosity and low strength of the Interfacial Transition Zone
(ITZ) in the cement paste at the aggregate interfaces. The cement paste close to the aggregate
surface is remarkably different from the bulk paste, and significantly more porous and weaker
than the rest of the paste. By applying load on the concrete, micro-cracking starts in the ITZ
and propagate into the cement paste until crack paths are formed through the concrete, as it is
schematically shown in Figure A.1. In high strength concretes the influence of the ITZ is less
pronounced.

Figure A.2 Schematic cracking pattern in normal-strength concrete (Domone and Illston,
2010) and failure surface of a concrete specimen indicating the interface cracks (unpublished,
courtesy of Borosnyoi-Crawley)

There is a general consensus that the Interfacial Transition Zone (ITZ) is about 30 to 50 microns
wide (Mehta and Monteiro, 2006) and its structure as a schematical impression is shown in
Figure A.3. In the ITZ special forms of hydration products are developed. One often finds large,
hexagonal Ca(OH)2 crystals, sometimes tens of microns long but only one or two microns thick,
and clusters of ettringite needles that are mainly rod-shaped and adhere to the contact layer on
the aggregate interface (de Rooij et al, 1998). The formation of the ITZ is not well explained in
the literature but attributed mostly to two mechanisms (Domone and Illston, 2010): 1) the ‘wall
effect’, whereby the cement grains cannot pack efficiently next to the aggregate (Mehta and
Monteiro, 2006), and 2) localised bleeding (mix water separation) at the interface, leading to a
higher local water-to-cement ratio (de Rooij et al, 1998). The porosity of the ITZ is substantially
higher than that of the bulk cement paste (Figure A.3).

42
 Figure A.3 Schematical impression of the ITZ structure (Domone and Illston, 2010) and
porosity in the ITZ between cement paste and aggregate or steel rebar (Angst et al, 2017)

To describe the distribution of the tensile strength of concrete, a probability function is required.
In the JCSS Probabilistic Model Code (JCSS, 2002) a logarithmic normal distribution is
assumed, but Eurocode 2 (EN 1992-1-1) and fib Model Code 2010 assume a normal
distribution. For the tensile strength of concrete the coefficient of variation is considered a
constant quantity, in contrast to the compressive strength of concrete where the standard
deviation is considered a constant (Cramer et al, 2021). With regard to the scatter of the tensile
strength Cramer et al (2021) indicates a coefficient of variation V < 10%. In Eurocode 2 (EN
1992-1-1) and fib Model Code 2010, the coefficient of variation for the tensile strength is V =
18%. Cramer et al (2021) proposes that for one concrete batch, a coefficient of variation V <
10% is realistic and for several batches in a concrete member (or in the absence of tensile tests),
V = 18% may be reasonable. Corresponding to the V = 18% coefficient of variation for the
tensile strength of concrete, Eurocode 2 (EN 1992-1-1) and fib Model Code 2010 prescribe the
lower and upper characteristic values related to the mean value as:

𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 = 0.7𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐

𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.95 = 1.3𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐

In view of the above, the design bond strength given in Eurocode 2 (EN 1992-1-1) can be
rearranged and expressed as a function of the lower characteristic value of the tensile strength
of concrete (for good bond conditions and db < 32 mm) as:

𝛼𝛼𝑐𝑐𝑐𝑐 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 1.0𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05

𝑓𝑓𝑏𝑏𝑏𝑏 = 2.25𝜂𝜂1 𝜂𝜂2 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 2.25𝜂𝜂1 𝜂𝜂2 � � = 2.25 ∙ 1.0 ∙ 1.0 � �=
𝛾𝛾𝑐𝑐 1.5
= 1.5𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05

NZS 3101 defines the lower characteristic value of the tensile strength of concrete as:

𝑓𝑓𝑡𝑡 = 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 = 0.38𝜆𝜆 �𝑓𝑓′𝑐𝑐

Where:

43
 0.6𝜌𝜌
𝜆𝜆 = 0.4 + ≤ 1.0
2200

Commentary clause C5.2.4 of NZS 3101 explains that the values of tensile strength given by
the above formula have been calibrated to closely match the values specified in the fib Model
Code 2010 while being defined by a simpler formula. Similarly, the λ factor accounting for
lightweight concrete has been adopted from the fib Model Code 2010. If an estimate is
required of the average direct tensile strength of concrete, this may be obtained by multiplying
the lower characteristic value of the tensile strength by a factor of 1.43. Similarly, an estimate
of the upper characteristic direct tensile strength may be obtained by multiplying the lower
characteristic value by a factor of 1.86. These values are derived from to the V = 18%
coefficient of variation for the tensile strength of concrete, in accordance with Eurocode 2
(EN 1992-1-1) and fib Model Code 2010 as:

𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.43𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 =
0.7

Based on Figure A.4 it is assumed in this Report that λ = 1.0 for normal weight concretes; for
New Zealand conditions, Figure A.4 indicates the relationship assumed between the hardened
density of concrete with compressive strength (reproduced from Fig. 21 of Concrete NZ -
Concrete production guide for New Zealand, July 2021).

Figure A.4 Hardened density of concrete and its relationship with compressive strength
(Concrete NZ, 2021)

Design codes usually propose relationships between the mean tensile strength and the mean
compressive strength of concrete, together with relationships between the characteristic
compressive strength and the mean compressive strength of concrete. Such relationships are
missing from NZS 3101. It is possible, however, to generate an expression for the mean tensile
strength of concrete based on the NZS 3101 lower characteristic value of the tensile strength
by applying the ratio proposed in Eurocode 2 (EN 1992-1-1), with accepting the assumption of
V = 18% constant coefficient of variation for the tensile strength and normal probability
distribution. The mean tensile strength of concrete for NZS 3101 would read:

44
 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 0.38�𝑓𝑓′𝑐𝑐
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.43𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 = = = 0.54�𝑓𝑓′𝑐𝑐
0.7 0.7

With this, a comparison is possible to be made to other code formulas. In Table A.1 a collection
of different proposals for the relationship between the mean tensile strength and the
characteristic compressive strength of concrete is shown. Figure A.5 indicates the values of the
calculated mean tensile strength of concrete, fctm, in accordance with the formulae introduced
in Table A.1, over the specified compressive strength of concrete, f’c, in the range of f’c = 20 to
50 MPa.

4.5

Eurocode 2
Mean tensile strength fctm, MPa

4
NZS 3101

3.5

2.5

1.5
20 25 30 35 40 45 50
Specified compressive strength f’c, MPa

Figure A.5 Mean tensile strength of concrete, fctm, and its relationship with specified
compressive strength, f’c

Table A.1 Code proposals for the relationship between the mean tensile strength of concrete
and the characteristic compressive strength of concrete
Eurocode 2 (EN 1992-1-1) (for ≤C50/60) 2
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 0.3(𝑓𝑓′𝑐𝑐 )3
fib Model Code 2010 (for ≤C50/60)
2
CEB-FIP Model Code 1990 𝑓𝑓′𝑐𝑐 3
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.4 � �
10
0.6
CEB-FIP Working Group on HSC/HPC 𝑓𝑓′𝑐𝑐 + 8
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.8 � �
(1995) 18
𝑓𝑓 ′ + 8
CEB Bulletin 228 (1995) 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 2.64 ∙ �𝑙𝑙𝑙𝑙 � 𝑐𝑐 � − 0.1�
10
0.56
𝑓𝑓′𝑐𝑐 + 8
fib Bulletin 42 (2008) 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.61 � �
10
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 5 �𝑓𝑓′𝑐𝑐 (𝑝𝑝𝑝𝑝𝑝𝑝)
Nilson et al. (2004)
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 0.415�𝑓𝑓′𝑐𝑐 (𝑀𝑀𝑀𝑀𝑀𝑀)
AS 3600:2018 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 1.4𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 = 0.504�𝑓𝑓′𝑐𝑐

45
It can be seen in Figure A.5 that the difference between the Eurocode 2 (EN 1992-1-1) proposal
and the NZS 3101 proposal is not considerable in this strength range. Therefore, it may be
acceptable to adopt for NZS 3101 the relationship between the lower characteristic value of the
tensile strength of concrete and the design bond strength given in Eurocode 2 (EN 1992-1-1)
for good bond conditions and db < 32 mm as:

𝑓𝑓𝑏𝑏𝑏𝑏 = 1.5𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05 = 1.5 ∙ 0.38 �𝑓𝑓′𝑐𝑐 = 0.57 �𝑓𝑓′𝑐𝑐

It is noted that in the formulas referred in this Appendix, the characteristic compressive strength
of concrete corresponds to ∅150/300 mm standard cylinder specimens. In accordance with
NZS 3101, however, the specified compressive strength of concrete corresponds to ∅100/200
mm standard cylinder specimens. It is exhaustively published in the technical literature for
almost 100 years, see e.g., Gonnerman (1925), Price (1951), Carino et al (1994), Day (1994),
Cook (1989), and accepted for normal strength concrete, that for a given concrete composition
and defined testing procedure the compressive strength measured on specimens with a constant
slenderness (at a fixed height-to-diameter ratio of e.g., 2:1) decrease with increasing specimen
size. This tendency holds true also for high strength concrete (fib, 2008; Riedel et al, 2019). An
approx. 4 % lower compressive strength can be measured on ∅150/300 mm standard cylinder
specimens, in comparison to the tests on ∅100/200 mm standard cylinder specimens. This
difference is not huge, but far not negligible. In a sensitive situation, and especially under low
level of quality control, it might result in a different concrete strength class for ∅100/200 mm
size cylinders than it could be obtained for ∅150/300 mm size cylinders. The different specified
compressive strength of concrete may be a safety concern.

Fracture mechanics provides strength-dependent size effect relationships. Based on Bazant’s

and others’ theoretical studies Yi et al (2006) established an empirical size effect relationship
(Figure A.6):
0.49𝑓𝑓 ′ 𝑐𝑐
𝑓𝑓𝑐𝑐𝑐𝑐 (𝑑𝑑) = + 0.81𝑓𝑓 ′ 𝑐𝑐
⁄
�1 + 𝑑𝑑 2.6

Where the compressive strength of a general cylinder, fcy(d) and the compressive strength of
standard ∅150/300 cylinder, f’c are in MPa and the cylinder diameter d is in cm (Yi et al, 2006).

Figure A.6 Size effect on compressive strength of concrete measured on cylindrical specimens
(Yi et al, 2006)

46
Appendix B – NZS 3101:2006 Chapter 8: Stress development, detailing and
splicing of reinforcement and tendons

47
 NZS 3101:Part 1:2006

8 STRESS DEVELOPMENT, DETAILING AND SPLICING OF REINFORCEMENT

AND TENDONS

8.1 Notation
Ab area of an individual bar, mm2

nd
Asp area of flexural reinforcement provided, mm2
Asr area of flexural reinforcement required, mm2
At area of a bar formed into a spiral or circular or rectangular hoop reinforcement, mm2 A3

ala
Atr smaller of area of transverse reinforcement within a spacing s crossing plane of splitting normal
to concrete surface containing extreme tension fibres, or total area of transverse reinforcement
normal to the layer of bars within a spacing, s, divided by n, mm2. If longitudinal bars are

Ze
enclosed within spiral or circular hoop reinforcement, Atr = At when n ≤ 6.
Av area of shear reinforcement within a distance s, mm2
Aw area of an individual wire to be developed or spliced, mm2
bw web width, or diameter of circular section, mm

ew
cb neutral axis depth corresponding to balanced conditions, mm
cm the smaller of the concrete cover or the clear distance between bars, mm
d distance from extreme compression fibre to centroid of tension reinforcement, mm

sN
db nominal diameter of bar, wire or prestressing strand, or in a bundle, the diameter of a bar of
equivalent area, mm
di diameter of bend measured to the inside of the bar, mm
f ć specified compressive strength of concrete, MPa
rd
fps calculated stress in prestressing steel at design load, MPa
fs stress in reinforcing bar, MPa
da

fse effective stress in prestressing steel after losses, MPa

fy lower characteristic yield strength of non-prestressed reinforcement, MPa
fyt lower characteristic yield strength of transverse reinforcement, MPa
an

Lb distance from critical section to start of bend, mm

Ld development length, mm
Ldb basic development length of a straight bar, mm
St

Ldh development length of hooked bars, equal to straight embedment between critical section and
point of tangency of hook, plus bend radius, plus one bar diameter, mm. (Refer to Figure 8.1)
Lds splice length of bars in non-contact lap splices in flexural members, mm
ht

Mn nominal flexural strength of section, N mm

n number of bars uniformly spaced around circular sections, or the number of longitudinal bars in
ig

the layer through which a potential plane of splitting would pass

s maximum spacing of transverse reinforcement within Ld, or spacing of stirrups or ties or spacing
of successive turns of a spiral, all measured centre-to-centre, mm
yr

sb for a particular bar or group of bars in contact, the centre-to-centre distance or, measured
perpendicular to the plane of the bend, to the adjacent bar or group of bars or, for a bar or group
p

of bars adjacent to the face of the member, the cover plus one half of db, mm
clear distance between bars of a non-contact lap splice, mm
Co

sL
sw spacing of wires to be developed or spliced, mm
u4, u8 residual elongation after 4 and 8 cycles respectively A3
V* design shear force at section at the ultimate limit state, N
Vs nominal shear strength provided by the shear reinforcement, N A3
α1, α2 parameters used in determining development lengths for standard hooks
αa , αb , αc , αd , αe parameters used in determining development lengths for straight reinforcing bars
βb ratio of area of reinforcement to be cut off to total area of tension reinforcement at the section,
including those bars which are to be cut off

8-1
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006
8.2 Scope
This section presents general provisions that shall apply to detailing of reinforcement and tendons,
including spacing and design of anchorage, development and splices.

Provisions specific to particular elements are presented within the sections specific to those elements.

nd
8.3 Spacing of reinforcement
8.3.1 Clear distance between parallel bars

ala
The clear distance between parallel reinforcing bars in a layer shall be equal to or greater than the largest
of the nominal diameter of the bars, or 25 mm, except that bars in slabs may be placed in two bar bundles.
8.3.2 Nominal maximum size of aggregate

Ze
The nominal maximum size of the aggregate shall be equal to or less than three-quarters of the minimum
clear spacing between individual reinforcing bars or bundles or pretensioning tendons or post-tensioning
ducts.

ew
8.3.3 Placement of parallel bars in layers
Where parallel reinforcement is placed in two or more layers in beams, the bars in the upper layers shall
be placed directly above those in the bottom layer with the clear distance between layers shall be the
larger of the nominal diameter of the bars or 25 mm.
8.3.4 Bundled bars
sN
Except in slabs, groups of parallel reinforcing bars bundled in contact and assumed to act as a unit shall
only be used when the bundles are within the perimeter stirrups or ties. Bundles shall not contain any
rd
more than four bars. Bars larger than 32 mm shall not be bundled in beams or girders. Individual bars in
a bundle cut off within the span of flexural members shall terminate at different points with at least 40 bar
diameters stagger. Where spacing limitations and minimum clear cover are based on bar size, a unit of
da

bundled bars shall be treated as a single bar of a diameter derived from the equivalent total area.
8.3.5 Spacing of principal reinforcement in walls and slabs
an

A3 The requirements for the spacing of reinforcement in walls is given by 11.3.11.3.8, 11.3.11.3.9, 11.3.12.2
and 11.4.5.5. The requirements for the spacing of reinforcement in slabs is given by 9.3.8.3, 12.5.6.3, and
12.8.2.3.
St

8.3.6 Spacing of outer bars in bridge decks or abutment walls

In bridge decks or abutment walls, the maximum spacing between adjacent bars in the outermost layer
shall be 300 mm.
ht

8.3.7 Spacing between longitudinal bars in compression members

In spirally reinforced and tied compression members, the clear distance between longitudinal bars shall be
ig

equal to or greater than 1.5db, or 40 mm.

8.3.8 Spacing between splices
yr

The limit on clear distance between bars shall also apply to the clear distance between a contact lap
splice and adjacent splices or bars.
p

8.3.9 Spacing between pretensioning reinforcement

Co

Except for hollow-core floor systems as provided for in 8.3.9, the clear distances between pretensioning
reinforcement at each end of the member shall be equal to or greater than 4 db for individual wires or 3db
for strands. Closer vertical spacing and bundling of strands is permitted in the middle portion of the spans,
but the requirements of 8.3.3 shall be satisfied. In hollow-core floor systems the clear distance between
prestressing strands shall be equal to or greater than 2db.
8.3.10 Bundles of ducts for post-tensioned steel
Ducts for post-tensioning steel may be bundled if it can be shown that the concrete can be satisfactorily
placed and provision is made to prevent the steel, when tensioned, from breaking through the duct.

8-2
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006
8.4 Bending of reinforcement
8.4.1 Compliance with NZS 3109
Bending and re-bending of reinforcing bars shall comply with the provisions of NZS 3109 including its
amendments.
8.4.2 Bending of steel bar reinforcement

nd
8.4.2.1 Minimum bend diameter for main bars
The diameter of bend, measured to the inside of the bar, shall be equal to or greater than the greater of

ala
the appropriate value given in Table 8.1 for steel reinforcement manufactured to AS/NZS 4671 or the
value given by Equation 8–1, except that Equation 8–1 need not apply in the case where two transverse
bars of db greater than or equal to the bar being bent are placed in contact with the inside of the bend or
where the stress at the start of the bend is less than fy/2. Where transverse bars are required they shall

Ze
extend for a minimum distance of 3db beyond the plane of the last bent bar.
Table 8.1 – Minimum diameters of bend

ew
fy Bar diameter, db Minimum diameter
(MPa) (mm) of bend, di
(mm)
300 or 500 6 – 20 5 db
24 – 40
sN
The diameter of bend measured to the inside of the bar shall be equal to or greater than:
6 db
rd
 d f d
d i ≥ 0.92 0.5 + b  s ' b ............................................................................................................ (Eq. 8–1)
da

 sb  fc

where fs is the stress in the bar at the start of the bend. This may be taken as fy, or a lower value if this
an

justified by a rational analysis which allows for the influence of diagonal cracking on stress.

8.4.2.2 Minimum bend diameter in fatigue situations

St

In members subjected to frequently repetitive loading situations, the minimum diameter of bends in A2
flexural reinforcement shall comply with 2.5.2.2.
ht

8.4.2.3 Stirrup and tie bends

The inside diameter of bends of stirrups shall be greater than or equal to the diameter of the largest
enclosed bar, and greater than or equal to the values given in Table 8.2.
ig

Table 8.2 – Minimum diameters of bends for stirrups and ties

yr

fy Stirrup or tie Minimum diameter

(MPa) diameter db of bend, di
p

(mm) (mm)
Plain bars Deformed bars
Co

300 or 500 6 – 20 2 db 4 db
24 – 40 3 db 6 db

8.4.2.4 Bends in galvanised deformed bars

Where deformed bars are galvanised before bending, the minimum bend diameter shall be:
(a) 5db for bar diameters of 16 mm or less;
(b) 8db for bar diameters of 20 mm or greater.
8.4.3 Bending of welded wire fabric
The inside diameter of bends in welded wire fabric, plain or deformed, shall be equal to or greater than
four wire diameters for deformed wire larger than 7 mm and two wire diameters for all other wires. Bends

8-3
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

with an inside diameter of less than eight wire diameters shall be equal to or greater than four wire
diameters from the nearest welded intersection.

8.5 Welding of reinforcement

8.5.1 Compliance with AS/NZS 1554:Part 3

d
Except as provided herein, all welding shall conform to AS/NZS 1554:Part 3. In the design and execution

lan
of welding of reinforcing bar, appropriate account shall be taken of the process of manufacture.
8.5.2 In-line quenched and tempered steel bars
Welding, including tack welding, and hot bending of bars that have been manufactured by the in-line
quenched and tempered process shall not be permitted.

a
Ze
8.5.3 Welds in proximity to bends
Welds in reinforcing bars shall be at least 3db away from the commencement of bends or that part of a bar
which has been bent and re-straightened in accordance with NZS 3109.

ew
8.6 Development of reinforcement
8.6.1 Development of reinforcement – General
Calculated tension or compression in reinforcement at each section of a reinforced concrete member shall

thereof. Hooks may be used in developing bars in tension.

8.6.2 Development of shear and torsion reinforcement
sN
be developed on each side of that section by embedment length or end anchorage or a combination
rd
The development of shear and torsion reinforcement shall comply with the relevant requirements of 7.5.7
and 7.6.3 respectively.
da

8.6.3 Development length of deformed bars and deformed wire in tension

8.6.3.1 Development length in tension

The development length, Ld, of deformed bars and wire in tension shall be calculated from either 8.6.3.2 or
an

8.6.3.3, but Ld shall be equal to or greater than 300 mm.

8.6.3.2 Basic development length in tension

St

Unless a more detailed determination of Ld is made in accordance with 8.6.3.3, the development length,
Ldb shall be calculated from:
ht

Ldb =
(0.5α a f y ) d ....................................................................................................................... (Eq. 8–2)
b
rig

'
fc

where αa = 1.3 for top reinforcement where more than 300 mm of fresh concrete is cast in the member
py

below the bar, or 1.0 for all other cases.

The value of f ć used in Equation 8–2 shall not exceed 70 MPa.

Co

8.6.3.3 Refined development length in tension

The development length, Ld, in tension may be determined from:

αb
Ld = L db ≥ 300 mm ........................................................................................................... (Eq. 8–3)
α cα d

with αb, αc and αd being defined as follows:

(a) Reinforcement provided in a flexural member (not subjected to seismic forces nor required for
temperature or shrinkage in restrained members) in excess of that required:

8-4
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

αb = Asr/Asp ......................................................................................................................... (Eq. 8–4)

(b) When cover to bars in excess of 1.5db or clear distance between adjacent bars in excess of 1.5 db is
provided:

d
 cm 
α c = 1 + 0.5 − 1.5  ........................................................................................................ (Eq. 8–5)

lan
 db 

with the limitation of 1.0 ≤ αc ≤ 1.5

a
where cm = the lesser of the concrete cover or the clear distance between bars.

Ze
(c) When transverse reinforcement with at least 3 bars, spaced less than 8db, transverse to the bar being
developed, and outside it, are provided within Ld:

ew
 Atr  f yt 
α d = 1 +    ..................................................................................................... (Eq. 8–6)

 s  80 nd b 

sN
with the limitation of 1.0 ≤ αd ≤ 1.5. nd b refers to the terminating reinforcement. A3

Transverse reinforcement used for shear, flexure or temperature may be included in Atr.
8.6.4 Development length of plain bars and plain wire in tension
rd
The development of plain bars and wire in tension shall rely on hooks. The development length shall be
twice the value for Ldh calculated from Equation 8–12.
da

8.6.5 Development length of deformed bars and deformed wire in compression

8.6.5.1 Development length in compression

an

Development length Ld of deformed bars in compression shall be computed from either 8.6.5.2 or 8.6.5.3,
but Ld must be greater than 200 mm.
St

8.6.5.2 Basic development length in compression

Unless a more detailed determination of Ld is made in accordance with 8.6.5.3 the development length in
compression, Ldb, shall be calculated from:
ht

0.22f y
Ldb = d b ........................................................................................................................... (Eq. 8–7)
rig

'
fc

with limitations of
py

Ldb ≥ 0.040fydb ≥ 200 mm ............................................................................................................ (Eq. 8–8)

Co

The value of f ć used in Equation 8–7 shall not exceed 70 MPa.

8.6.5.3 Refined development length in compression

The development length in compression, Ld, may be determined from:

Ld = αbαeLdb ................................................................................................................................. (Eq. 8–9)

with αb as defined in 8.6.3.3(a) and αe as follows:

8-5
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

When transverse reinforcement with at least three bars, transverse to the bar being developed and
A A
outside it, are provided within Ldb, and tr ≥ b
s 600

αe = 0.75, or
= 1.0 for all other cases.

nd
8.6.6 Development length of plain bars and plain wires in compression
The development length for plain bars and wires in compression shall be twice the calculated value Ld or

ala
Ldb for a deformed bar or wire.
8.6.7 Development of bundled bars
Development length of individual bars within a bundle, in tension or compression, shall be that for the

Ze
individual bar, increased by 20 % for a three-bar bundle, and 33 % for a four-bar bundle.
8.6.8 Development of welded plain and deformed wire fabric in tension

8.6.8.1 Development length of wire fabric

ew
Development length, Ld, of welded plain and deformed wire fabric measured from the point of critical
section to the end of the wire shall be computed from either 8.6.8.2 or 8.6.8.3.

sN
8.6.8.2 Development length of welded wire fabric – cross wires considered
The yield strength of plain and deformed wires of welded wire fabric shall be considered developed by
embedding at least two cross wires, with the first one equal to or greater than 50 mm from the critical
section. However, development length Ld measured from the critical section to the outermost cross wire
rd
shall be equal to or greater than 100 mm:

3.25α b Aw f y
da

Ld ≥ .................................................................................................................... (Eq. 8–10)

'
s w fc
an

where αb is given by 8.6.3.3(a), but Ld shall be equal to or greater than 150 mm for plain wire fabric or
greater than 100 mm for deformed wire fabric.
St

8.6.8.3 Development length of welded wire fabric – cross wires not considered
The development length of welded deformed and plain wire fabric, with no cross wires or when the cross
wires within the development length as required by 8.6.8.2 are ignored, shall be determined by 8.6.3 or
ht

8.6.4 as appropriate and shall be equal to or greater than 200 mm.

8.6.9 Development of prestressing strand
ig

8.6.9.1 Development length of pretensioning strand

Three or seven-wire pretensioning strand shall be bonded beyond the critical section for a development
yr

length given by:

p

 2 d
Ld ≥  f ps − f se  b .................................................................................................................. (Eq. 8–11)
 3  7
Co

8.6.9.2 Development of pre-stressing strand

Where bonding of a strand does not extend to the end of a member, the bonded development length
specified in 8.6.9.1 shall be doubled.

A3 8.6.9.3 Prestressing strand transfer length

The transfer length of the strand, i.e. the distance over which the strand must be bonded to the concrete to
develop the prestress fse in the strand, shall be taken as fsedb/21 provided the concrete strength at the time
of transfer is equal to or greater than 21 MPa. For shear strength calculations in the development length
see 19.3.11.2.3.

8-6
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

8.6.10 Standard hooks

8.6.10.1 Standard hooks – definition

The term “standard hook” as used herein shall mean either:
(a) A semi-circular turn plus an extension of at least four bar diameters but equal to or greater than
65 mm at the free end of the bar; or

d
(b) A 90° turn plus an extension of at least 12 bar diameters at the free end of the bar for a deformed bar

lan
and 16 bar diameters for a plain bar; or A2
(c) A stirrup hook, which is defined as a 135° turn around a longitudinal bar plus an extension of at least
eight stirrup bar diameters for plain bars and six stirrup bar diameters for deformed bars at the free
end of the bar embedded in the core concrete of the member.

a
The standard hooks defined in this clause are illustrated in Figure 8.1.

Ze
A2

ew
sN
rd
da
an
St
ht
rig

Figure 8.1 – Standard hooks

8.6.10.2 Bars > 32 mm in diameter

py

Bars with diameter greater than 32 mm shall not be developed in tension by the use of standard hooks.

8.6.10.3 Development length of standard hooks in tension

Co

8.6.10.3.1 Calculation of development length for hooked bars

For the following two situations described in (a) and (b), the development length, Ldh, for hooks in tension
shall be determined from Equation 8–12:
(a) Where the bar is anchored by a standard hook inside a volume of concrete confined by closed ties,
spirals or stirrups perpendicular to the plane of the hook;
(b) Where the bar, is anchored by a standard hook inside a volume of concrete that is not confined by
reinforcement perpendicular to the plane of the hook, but
(i) The spacing between that bar and any adjacent bar or fixing loaded in a similar direction is
greater than or equal to three times db over the development length (Ldh) for that bar; and A3
(ii) The distance normal to the axis of the bar to the side or edge of the element is greater than or
equal to two times db over the development length (Ldh) for that bar. A3
8-7
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

fy db
Ldh = 0.24α bα 1α 2 ≥ 8d b ................................................................................................... (Eq. 8–12)
'
fc
where
f ć shall not be taken greater than 70 MPa
αb is given by 8.6.3.3 (a)

nd
α1 = 0.7 for 32 mm bars or smaller with side cover normal to the plane of the hook ≥ 60 mm, and
cover on the tail extension of 90° hooks equal to or greater than 40 mm
= 1.0 for all other cases

ala
α2 = 0.8 where confined by closed stirrups or hoops spaced at 6db or less and which satisfy the
A A
relationship tr ≥ b
s 1000

Ze
= 1.0 for all other cases

8.6.10.3.2 Determination of development length where not covered by 8.6.10.3.1

For situations other than as described by 8.6.10.3.1(a) and (b), the development length of a hook shall be

ew
determined from a rational analysis or suitable testing that takes into account the effects of the proximity of
the anchored bar to edges of elements and to other loaded embedded items.

sN
8.6.10.3.3 Development length of standard hooks anchoring around transverse bars
The development length Ldh of a deformed bar terminating in a standard hook as determined from 8.6.10.3
may be reduced by 20 %, provided that two transverse bars having a diameter equal to or larger than that
of the bent bar are placed in contact with the inside of the bend and extend for a distance equal to or
rd
greater than 3db beyond the centreline of the bent bar.

8.6.10.4 Hooks in compression

da

Hooks shall not be considered effective in developing reinforcement in compression.

8.6.11 Mechanical anchorage
an

8.6.11.1 General
A3 For reinforcement complying with AS/NZS 4671, any mechanical device used alone as an anchorage, or
used in combination with an embedment length beyond the point of maximum stress in the bar, shall be
St

capable of developing the upper bound breaking strength of the reinforcing bar without damage to the
concrete or overall deformation of the anchorage.
ht

In addition, when tested with a bar complying with AS/NZS 4671, the mode of failure of the anchored bar
shall be by ductile yielding of the bar, with the bar developing its ultimate tensile strength at a location
outside the mechanical anchorage and away from any zone of the bar affected by working (e.g. by cold
ig

forging).
yr

8.6.11.2 Upper bound breaking strength of the reinforcing bar – definition

A3
The upper bound breaking strength of the reinforcing bar may be derived from 1.25 times the upper
characteristic yield strength specified by AS/NZS 4671, or otherwise shall be determined from an
p

appropriate testing programme.

Co

8.6.11.3 Adequacy of mechanical devices

Mechanical anchorage systems relying on interconnecting threads or mechanical interlock with the bar
deformations for attachment of the anchorage to the bar shall meet both the permanent extension and
fatigue strength criteria of 8.7.5.2. Where the mechanical anchor and ends of the bars are threaded as the
A3
means of achieving the connection between components, there shall be no thread stripping or evidence of
significant distortion of the threads at the failure load of the bar.

8.6.11.4 Brittle fracture resistance

Mechanical anchors for the anchorage of reinforcing steel shall be proven by an appropriate test method
to possess resistance to brittle fracture at the service temperatures at which they are intended for use.
Where the mechanical anchors and ends of the bars are threaded as the means of achieving the
8-8
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

connection between components, and/or the end of the bar is enlarged by cold forging prior to threading, A3
appropriate testing of the processed bar end shall be applied to ensure that the potential for brittle fracture
is avoided. Anchors manufactured from cast iron shall not be used.
8.6.12 Development of flexural reinforcement

nd
8.6.12.1 Bending across the web
Tension reinforcement may be developed by bending across the web to be anchored or made continuous
with reinforcement on the opposite face of member.

ala
8.6.12.2 Critical sections
Critical sections for development of reinforcement in flexural members are at points of maximum stress
and at points within the span where adjacent reinforcement terminates, or is bent.

Ze
8.6.12.3 Extension of tension reinforcement
Except at supports of simply supported spans and at the free end of cantilevers, tension reinforcement
shall extend beyond the point at which, according to the bending moment envelope and standard flexural

ew
theory, it is:
(a) Required at maximum stress for a distance equal to the development length, Ld, plus the effective
depth of the member, and

sN
(b) No longer required to resist flexure for a distance of 1.3 times the effective depth of the member.

8.6.12.4 Termination in a tension zone

Flexural reinforcement shall not be terminated in a tension zone unless one of the following conditions is
satisfied:
rd
(a) Shear at the cut-off point is less than two-thirds of the shear strength provided by the concrete; or
(b) The shear strength provided by the web reinforcement , Vs, measured for a distance of 1.3d along the
da

terminating bar from the cutoff point is equal to or greater than:

an

'
fc
Vs = 1.2 b w d ..................................................................................................................... (Eq. 8–13)
16
St

d
and the spacing, s, of stirrups or ties is equal to or less than the smaller of d/2 or .
8β b
ht

8.6.12.5 End anchorage in flexural members

Adequate end anchorage shall be provided for tension reinforcement in flexural members where
ig

reinforcement stress is not directly proportional to moment, such as: sloped, stepped, or tapered footings;
brackets, deep flexural members; or members in which tension reinforcement is not parallel to the
yr

compression face.

8.6.12.6 Anchorage of flexural reinforcement in external beam-column joints A3

p

Longitudinal reinforcement in a beam terminating at an external beam-column joint shall be anchored by a

Co

90o hook in which the leg of the hook is bent vertically into the joint zone. The distance from where the
beam reinforcement terminates shall be:
(a) As close as possible to the vertical column reinforcement on the opposite face of the column from
where the beam reinforcement enters the joint zone; and
(b) Equal to or greater than the larger of the development length, Ldh, or three quarters of the column
depth, 0.75hc.
Joint zone ties shall satisfy 15.3.6.2(b).

8-9
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006
8.6.13 Development of positive moment reinforcement in tension

8.6.13.1 Limitation in area of bars

At least one-third the maximum positive moment reinforcement in simply supported members and one
quarter the maximum positive moment reinforcement in continuous members shall extend along the same
face of member into the support. In beams, such reinforcement shall extend into the support at least

nd
150 mm unless a lesser distance is demonstrated by test to be adequate and to provide the structural
robustness required by AS/NZS 1170.0.

ala
8.6.13.2 Critical sections
Where a flexural member is part of a primary horizontal force-resisting system, positive moment
reinforcement required to be extended into the support by 8.6.13.1 shall be anchored to develop the lower
characteristic yield strength, fy, in tension at the face of support.

Ze
8.6.13.3 Limitation in diameter of bars at simple supports
A3 The positive tension reinforcement at simple supports shall be limited in diameter to enable the bars
extending to the free end of the member to be fully developed from a point at a distance Mn/V * from the

ew
centre of the support. Mn is the nominal bending moment capacity provided by the reinforcement at the
centre of the support, calculated as the area of the positive tension reinforcement at the support multiplied
by fy and by the internal lever arm. V * is the shear at the face of the support. Where the support induces

sN
compression in the anchorage zone of the reinforcement the development length may be reduced by
25 %.

8.6.13.4 Limitation in diameter of bars at points of inflection

rd
The positive (and negative) tension reinforcement at points of inflection shall be limited in diameter to
enable the bars, from a point at a distance Mn/V * from the point of inflection, to be fully developed
satisfying the requirements that:
da

Mn
Ld ≤ + 12d b ........................................................................................................................ (Eq. 8−14)
an

V*

and
St

Mn
Ld ≤ + d ............................................................................................................................. (Eq. 8−15)
V*
ht

A3 For both the positive and the negative tension reinforcement, the value of Mn /V * shall be calculated at the
point of inflection, where Mn equals the area of the positive or negative tension reinforcement at the point
ig

of inflection multiplied by fy and by the internal lever arm.

8.6.14 Development of negative moment reinforcement in tension
yr

8.6.14.1 Anchorage of bars

Negative moment reinforcement in a continuous, restrained or cantilever member, or in any member of a
p

rigid jointed frame, shall be anchored in or through the supporting member by embedment length, hooks
Co

or mechanical anchorage.

8.6.14.2 Embedment length adjacent to supports

Negative moment reinforcement shall have an embedment length into the span as required by 8.6.1 and
8.6.12.3.

8.6.14.3 Embedment length beyond the point of inflection

At least one-third the total tension reinforcement provided for negative moment at a support shall have an
embedment length beyond the point of inflection, according to the appropriate bending moment envelope,
for a distance equal to or greater than 1.3 times the effective depth of the member.

8 - 10
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006
8.6.14.4 Limitation in diameter of bars
The requirements of 8.6.13.4 at points of inflection for negative reinforcement shall be satisfied.

8.7 Splices in reinforcement

8.7.1 General

nd
Splices in reinforcement shall be shown on the design drawings or specified in the specifications.
8.7.2 Lap splices of bars and wire in tension

ala
8.7.2.1 Bar sizes of lap splices
Lap splices shall not be used for bars larger than 40 mm in diameter.

Ze
8.7.2.2 Lap splices of bundled bars
Lap splices of bundled bars shall be based on the lap splice length required for individual bars of the same
size as the bars spliced, and such individual splices within the bundle shall not overlap each other. The
length of lap, as prescribed in 8.7.2.3 or 8.7.3, shall be increased by 20 % for a three-bar bundle and 33 %

ew
for a four-bar bundle.

8.7.2.3 Length of lap splices of deformed bars or wire

The minimum length for lap splices of deformed bars and deformed wire in tension shall be equal to or

lapping unless using hooks or other anchorages.

8.7.2.4 Length of lap splices of hooked plain bars or wire

sN
greater than the development length, Ld, in 8.6.3. Plain straight bars or wires shall not be spliced by
rd
The length of lap splices for hooked plain bars or wire with a standard hook shall be equal to or greater A3
than the development length required by 8.6.4. For bars with 50 mm of cover concrete or less, hooks
da

shall be in a plane at a right angle to the adjacent concrete surface. Such splices shall not be used in
potential plastic hinge regions of members.

8.7.2.5 Length of non-contact lap splices

an

Bars spliced by non-contact lap splices in flexural members spaced transversely farther apart than 3db
shall have splice length, Lds, given by Lds ≥ Ld + 1.5 sL.
St

8.7.2.6 Strength developed at sections

In computing the strength developed at each section, spliced bars shall be rated at the specified splice
strength.
ht

8.7.2.7 Strength of bars where cut off

Bars cut off near the section under consideration taking the requirements of 8.6.12.3 into account shall be
ig

rated only at a fraction of fy, defined by the ratio of the embedded length past this section to the required
development length.
yr

8.7.2.8 Lap splices of stirrups, ties and hoops

Stirrups, ties and rectangular hoops in beams, columns, piers, beam-column joints or walls may be spliced
p

A2
by lapping provided that the requirements in either (a) or (b) and (c) are satisfied: A3
Co

(a) Lapping bars shall be terminated with standard hooks in accordance with 8.4.2.1 and 8.6.10.1, and
the splice length shall be:
(i) For plain bars, equal to or greater than the development length required by 8.6.4;
(ii) For deformed bars, equal to or greater than Ldh in 8.6.10.3.

When the lapped splice is located in cover concrete, the hooks shall be placed in a plane at right
angles to the surface of the concrete. When located in a plastic region, the hooks shall be anchored
around longitudinal reinforcement of at least equal or greater diameter.
(b) Straight lapped splices with deformed bars may be used where not specifically excluded in (i) to (iii)
below:

8 - 11
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

A3 (i) Straight lapped splices shall not be used in nominally ductile, limited ductile or ductile plastic
regions, or in beam-column joints;
(ii) Straight lapped splices shall not be used in cover concrete where the reinforcement is required to
A2
provide confinement to the concrete;
(iii) Straight lapped splices shall not be used in cover concrete where the shear stress due to shear

nd
and torsion exceeds 0.5 fc' ;

A3 (c) 90° hooked lap splices may be used in beam-column joints. The distance from where each bar enters
the beam-column joint to the vertical leg of the hook is equal to or greater than three-quarters of the

ala
column depth provided the vertical legs of the hooks are located as close as possible to the vertical
reinforcement on the far side of the joint from which each bar entered the joint. The lap distance is
equal to or greater than that required for development (8.6.10) and the distance from where each bar
enters the beam-column joint is equal to or greater than three-quarters of the column depth;

Ze
Where stirrup or tie legs are formed by overlapping deformed reinforcement bars with straight laps, as
permitted by (b), the proportion of lapped splices in the cover concrete at any cross section shall be equal
to or less than 50 % of the stirrup or tie reinforcement. In this context, cover concrete is defined as the

ew
concrete lying on the outside of a line connecting the centres of longitudinal bars located closest to the
perimeter of the member.
8.7.3 Lap splices of bars and wires in compression

8.7.3.1 General
sN
The minimum length of a lap splice in compression shall be the development length in compression Ld, in
accordance with 8.6.5 and 8.6.6, but equal to or greater than 0.069fydb for fy of 430 MPa or less, or
(0.12 fy - 22) db for fy greater than 430 MPa, or 300 mm.
rd
8.7.3.2 Lap splices in compression with stirrups and ties
da

In compression members with stirrups and ties where at least three sets of ties are present over the length
of the lap, and
an

Atr Ab
≥ .............................................................................................................................. (Eq. 8–16)
s 1000
St

or where transverse reinforcement as required by either 10.4.7.4.3 or 10.4.7.4.5 has been provided, a lap
length of 0.8 times that specified in 8.7.3.1 may be used but the lap length shall be equal to or greater
than 300 mm.
ht

8.7.3.3 Lap splices in compression with spiral reinforcement

In spirally reinforced compression members, if at least three turns of spiral are present over the length of
ig

the lap, and

yr

Atr nAb
≥ .............................................................................................................................. (Eq. 8–17)
s 6000
p

a lap length of 0.8 times that specified in 8.7.3.1 may be used, but the lap length shall be equal to or
Co

greater than 300 mm.

8.7.4 Welded splices

8.7.4.1 Classification of welded splices

Welded splices shall be classified as follows:
(a) A “full strength” welded splice is one in which the bars are butt welded to develop in tension the
breaking strength of the bar;
(b) A “high strength” welded splice is one in which the bars are lap welded or butt welded to develop the
lower characteristic yield strength of the bar or better.

8 - 12
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006
8.7.4.2 Limitations on the classification of welded splices for grade > 450 MPa reinforcement
Butt welded splices in reinforcement with a lower characteristic yield stress of more than 450 MPa shall
not be classified as “full strength” unless either:
(a) Yielding of the reinforcement will not occur; or
(b) Proof testing using a portion of the actual bar to be welded and the selected welding procedure,
demonstrates that failure of the bar occurs away from the weld.

nd
8.7.4.3 Exceptions for welded splices
The requirements 8.7.4.1(b) may be waived when the conditions of 8.7.5.4 are satisfied.

ala
8.7.5 Mechanical connections

8.7.5.1 Definition of mechanical connection

Ze
A mechanical connection is defined as a connection that relies on interlocking threads or mechanical
interlock with the bar deformations to develop the connection capacity.

8.7.5.2 Performance requirements for mechanical connections

ew
Mechanical connections shall:
(a) satisfy the requirements of 8.6.11 for mechanical anchors; A3
(b) when tested in tension or compression, as appropriate, to the application, exhibit a change in length

sN
at a stress of 0.7fy in the bar, measured over the length of the coupler, of less than twice that of an
equal length of unspliced bar;
(c) satisfy the requirements of 2.5.2.2 when used in situations where fatigue may develop.
rd
8.7.5.3 Use of welded splices and mechanical connections
Welded splices in tension or compression shall meet the requirements of 8.7.4.1 (a) or (b).
da

Mechanical connections in tension or compression shall meet the requirements of 8.7.5.2.

8.7.5.4 Use of welded splices and mechanical connections – an exception

an

The requirements of 8.7.4.1(b) and 8.7.5.2, as appropriate, may be waived when splices:
(a) Are staggered at least 600 mm; and
(b) Can develop at least twice the calculated force in the bars to be spliced at the section; and
St

(c) Can develop equal to or greater than 0.7 fy based on the total area of effective bars across the
section; and
(d) Where the level of any resulting premature cracking is not likely to affect the performance of the
ht

structure, then the change of length shall be not more than six times that of an equal length of
unspliced bar.
ig

8.7.5.5 Identification and marking

Each coupler or coupling sleeve shall be legibly and durably marked with the identification of the
yr

manufacturer and the nominal bar size for which it is intended. Each coupler or coupling sleeve shall be
traceable back to its production data and production batch.
p

8.7.5.6 Installation
Co

The method of installation of mechanical connection systems shall be specified for all conditions that arise
on a job site. This may be by reference to manufacturers’ written instructions. Connection systems that
rely on a minimum length of engagement between the coupler or coupling sleeve and the bar for the
development of the connection strength shall incorporate a system for positively locating the coupler or
coupling sleeve and defining when adequate engagement has been achieved.
8.7.6 Splices of welded plain or deformed wire fabric
Lap splices shall be detailed by satisfying one of the following conditions:
(a) The overlap measurement between outermost cross wires of each fabric sheet is equal to or greater
than the spacing of cross wires plus 50 mm, nor less than 1.5 Ld or 150 mm whichever is greater,
where Ld is the development length for fy as given in 8.6.8.2; or

8 - 13
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

(b) When cross wires are ignored or no cross wires are present within the lapped length and the lap is a
contact or near contact lap splice, the splice length shall be equal to or greater than Ld, where Ld is
the development length given by 8.6.8.3.

8.8 Shrinkage and temperature reinforcement

nd
8.8.1 Floor and roof slab reinforcement
Reinforcement for shrinkage and temperature stresses normal to the principal reinforcement shall be
provided in structural floor and roof slabs where the principal reinforcement extends in one direction only.

ala
At all sections where it is required, such reinforcement shall be developed for its lower characteristic yield
strength in conformance with 8.6.1 or 8.7.2. Such reinforcement shall provide at least the ratio of
reinforcement area to gross concrete area of 0.7/fy, but equal to or greater than 0.0014.

Ze
8.8.2 Large members
In a large member whose size is not governed by stress considerations, or where exact analysis is
impractical, minimum reinforcement on all surfaces should be the greater of 1000 mm2 per metre width in
each direction, with bars not further apart than 300 mm, or, where appropriate, as required by 2.4.4.8.

ew
8.9 Additional design requirements for structures designed for earthquake effects

sN
8.9.1 Splices in reinforcement

8.9.1.1 Placement of splices

Full strength welded splices meeting the requirements of 8.7.4.1(a) may be used in any location. For all
other splices the following restrictions apply:
rd
A3 (a) With the exception noted in (b) below, no portion of any splice shall be located within the beam-
column joint region, or within one effective depth of the member from the critical section of a limited
da

ductile or ductile plastic region in a beam where stress reversals in spliced bars could occur;
(b) Hooked lap splices which comply with 8.7.2.8 may be used in beam-column joints;
(c) In a column framing top and bottom into beams or other moment-resisting elements, the centre of the
an

splice must be within the middle quarter of the storey height of the column unless it can be shown that
a high level of protection is provided against the formation of plastic regions, as defined in
Appendix D.
St

8.9.1.2 Lap splices in region of reversing stresses

Reinforcement in beams and columns shall not be spliced by lapping in a region where reversing stresses
ht

at the ultimate limit state may exceed 0.6fy in tension or compression unless each spliced bar is confined
by stirrup-ties so that:
ig

Atr db fy
≥ ............................................................................................................................... (Eq.8−18)
yr

s 48f yt

except that where there is no alternative load path in a structure for the forces being carried by an element
p

in the event of failure of the element, lap splicing shall not be permitted at all.
Co

8.9.1.3 Requirements for welded splices or mechanical connections

A3 For welded splices or mechanical connections to be used in members that are subjected to seismic
forces, such splices shall comply with 8.7.4.1 or 8.7.5.2. In addition to the requirements of 8.7.5.2,
mechanical splices and anchorages shall satisfy the cyclic load performance requirements specified by
ISO 15835-1 and ISO 15835-2 as follows:
(a) When tested in accordance with 5.6.2 of ISO 15835-2, the residual elongations after 4 cycles, u4,
shall be less than 0.3 mm, and after 8 cycles u8 shall be less than 0.6 mm;
(b) Where high cycle fatigue is a consideration, the mechanical connection shall satisfy the requirements
of 5.4 of ISO 15835-1. The testing shall comply with 5.5 of ISO 15835-2.

8 - 14
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 1:2006

Splices not satisfying this stiffness requirement shall be used only if they are staggered so that no more A3
than two-thirds of the reinforcement area is spliced within any 900 mm length of the member.
8.9.2 Development length
For calculation of development length, the reduction provisions of 8.6.3.3(a), 8.6.8.2 and 8.6.10.3.1 by
αb (equal to Asr/Asp) shall not apply. A3

nd
ala
Ze
ew
sN
rd
da
an
St
ht
ig
p yr
Co

8 - 15
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

C8 STRESS DEVELOPMENT, DETAILING AND SPLICING OF REINFORCEMENT

AND TENDONS

C8.1 Notation

nd
The following symbols, which appear in this section of the Commentary, are additional to those used in
Section 8 of the Standard:

ala
As area of non-prestressed tension reinforcement, mm2
A3
cb minimum top cover to reinforcing bar, mm

Ze
cp minimum clear spacing between reinforcing bars in a horizontal layer, mm
cs minimum side cover to reinforcing bar, mm
hc column depth parallel to the longitudinal beam bars being considered, mm
beam depth, mm

ew
hb
Ls splice length, mm
Ns restraining force developed by a circular tie against one vertical column bar, N
R tensile strength of circular tie or spiral, N

C8.2 Scope sN
The provisions of Section 8 apply to detailing of reinforcement and tendons including the design of
rd
anchorage, development and splices. Provisions also cover minimum bend radii, minimum reinforcement
in walls and shrinkage and temperature reinforcement in slabs. Requirements are also given for detailing
da

spiral, circular and rectangular hoop reinforcement in columns and for stirrups and ties in flexural
members.
an

The provisions for development include deformed and plain bars and wire, bundled bars, welded and
smooth wire fabric and prestressing strand.

They also cover standard hooks in tension, mechanical anchorage and anchorage of transverse
St

reinforcement. A comprehensive set of requirements is given to govern development of flexural

reinforcement. Provisions for splices deal with lap splices, welded splices and mechanical connections.
ht

C8.3 Spacing of reinforcement

ig

C8.3.1 Clear distance between parallel bars

The spacing limits of this clause have been developed from successful practice over many years,
yr

remaining essentially unchanged through many codes. The minimum limits were established to permit
concrete to flow readily into spaces between bars and forms without developing honeycomb, and to
p

prevent the concentration of bars on a line that might result in shear or shrinkage cracking. The spacing
between two bar bundles in slabs must conform with 8.3.1
Co

C8.3.4 Bundled bars

Bond research 8.1 showed that bar cut-offs for beams and splices for columns should be staggered.
Bundled bars should be tied, wired or otherwise fastened together to ensure that they remain in position.

The limitation that bars larger than 32 mm be not bundled in beams or girders of buildings has been taken
from the ACI 318 Code which applies primarily to buildings. The 1974 American Association of State
Highway and Transportation Officials (AASHTO) design criteria 8.2 for reinforced concrete bridges
permitted two-bar bundles of bars up to 57 mm in bridge girders or columns, usually more massive than
those in buildings. Conformity with crack control requirements in the Standard will effectively preclude
bundling of bars larger than 35 mm as tensile reinforcement. The Standard phrasing “bundled in contact,
assumed to act as a unit”, is intended to preclude bundling more than two bars in the same plane. Typical
C8 - 1
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

bundle shapes are triangular, square, or L-shaped patterns for three or four-bar bundles. As a practical
provision, bundles more than one bar deep in the plane of bending may not be hooked or bent as a unit.
Where end hooks are required, it is preferable to stagger them. Bending and hooking of bundles must be
established in this manner, even at supports.

C8.3.5 Spacing of principal reinforcement in walls and slabs

d
These maximum spacing limits have remained essentially unchanged for many years. The spacing of

lan
reinforcement in topping slabs shall be such that effective anchorage required for diaphragm action, in
accordance with 13.3.5, is assured.

C8.3.6 Spacing of outer bars in bridge decks or abutment walls

a
Experience has shown that the spacing of reinforcement at greater than 300 mm centres in the exposed
surfaces of bridge decks or abutment walls is likely to result in long-term maintenance problems, due to

Ze
shrinkage effects of direct exposure to the weather and the fatigue effects of live loading. Two cases are
given which would permit the spacing to be increased to a maximum of 450 mm. An example of (a) would
be the soffits of cantilever slabs, while an example of (b) would be the earth face of abutment retaining

ew
walls.

C8.3.7 Spacing between longitudinal bars in compression members

These requirements for minimum bar spacing, like those in 8.3.1, were developed originally to provide

spacing permitted is an extension of the original provision to larger bars.

C8.3.8 Spacing between splices

sN
access for concrete placing in columns. Use of the bar diameter as a factor in establishing the minimum
rd
Commentary C8.3.1 and C8.3.7 are applicable here. See also 8.3.5.
da

C8.3.9 Spacing between pretensioning reinforcement

These requirements are provided to prevent weakened planes for splitting bond failure developing in the
cover concrete in the anchorage zones. Provision has been made for reducing the clear distance for
an

hollow-core flooring systems.

C8.3.10 Bundles of ducts for post-tensioned steel

St

When ducts for post-tensioning steel in a beam are arranged closely together vertically, provision must be
made to prevent the reinforcement, when tensioned, from breaking through the duct. Horizontal
disposition of ducts must allow proper placement of concrete. Generally a clear spacing of 11/3 times the
nominal maximum size of the coarse aggregate, but equal to or greater than 25 mm, proves satisfactory.
ht

Where concentration of reinforcement or ducts tends to create a weakened plane in the concrete cover,
reinforcement should be provided to control cracking.
rig

C8.4 Bending of reinforcement

py

C8.4.2.1 Minimum bend diameters for main bars

The minimum bend diameters given in Table 8.1 are generally twice the bend test diameters specified in
AS/NZS 4671.
Co

When large steel stresses need to be developed in the bend, radial bearing and splitting stresses in the
concrete may become excessive. Equation 8–1 controls the diameter of the bend when there is a
combination of high tensile stress at the bend, large bar diameter and low concrete strength.

The quantities shown in Table 8.1 are based on the assumption that sb = db + 40. The requirements for
the development length may be difficult to satisfy when beam bars are anchored in exterior columns in
accordance with 9.4.3.2.5. In such cases the means by which the development length of hooked bars can
be reduced, as permitted in 8.4.2.1, should be investigated, or alternatively the diameter of bend must be
increased.

C8 - 2
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

Clause 8.4.2.1 also permits the addition of transverse bars to allow the use of Table 8.1 values for di in
cases where Equation 8–1 in would call for larger values.

The arrangement of the transverse bars is shown in Figure C8.1, and is based on the fact that excessive
bearing stresses will not extend past the first 60° of bend that is closest to the critical section.

nd
The transverse bars should extend for a distance of at least 3 db beyond the centreline of the outermost
bars in each layer.

ala
Ze
ew
sN
Figure C8.1 – Arrangement of additional transverse bars to reduce bearing stress

C8.4.2.2 Minimum bend diameter in fatigue situations

rd
Bends in primary reinforcement should be avoided in regions of high stress range. The minimum diameter
of bend of slab reinforcing bars, for example, of cranked transverse reinforcement in bridge deck slabs, is A2
increased to the bar bend diameter equal to or greater than that specified in 2.5.2.2. High localised areas
da

of stress concentration, which occur in tight radius bends, can cause fatigue failure to propagate from
these locations.
an

C8.4.2.3 Stirrup and tie bends

It is not intended that a tie is to have different bend diameters should it pass round longitudinal bars of
different diameters.
St

C8.4.3 Bending of welded wire fabric

Welded wire fabric of plain or deformed wire can be used for ties and stirrups. The wire at welded
ht

intersections does not have the same uniform ductility and ease of bending as in areas which were not
heated. These effects of the welding temperature are usually dissipated over a length of approximately
ig

four wire diameters. Minimum bend diameters permitted are in most cases the same as those required in
the bend test for wire material. The requirements of 8.4.3 for welded wire fabric are shown in Figure C8.2.
p yr
Co

di ≥ 4 db if db > 7 mm
di  2 db if db ≤ 7 mm

Figure C8.2 – Bends in welded wire fabric for stirrups and ties

C8 - 3
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006
C8.5 Welding of reinforcement

C8.5.1 Compliance with AS/NZS 1554: Part 3

Reinforcing steels not conforming to AS/NZS 4671 will require different welding techniques and the
designer and fabricator must become familiar with these techniques before designing a weld or attempting
to weld the steel.

d
Due to the low carbon metallurgy of reinforcing steel manufactured to AS/NZS 4671, the steel is

lan
considered readily weldable. However, AS/NZS 4671 permits a range of manufacturing processes for the
production of steel reinforcement. Due care must be exercised for welding of such reinforcement because
the welding process can alter the metallurgy and microstructure of the as-rolled bars. In certain situations

a
this may result in lower yield strengths and lower ultimate tensile strengths in the heat affected zones of
the weld sites. This may lead to detrimental behaviour with loss of ductility in the bar and fracture of the

Ze
bar may occur. Refer to AS/NZS 1554:Part 3 and the reinforcement manufacturer's recommendations for
details of appropriate welding techniques.

In line quenched and tempered bars are subject to loss of strength when welded.

ew
Steel which has been heavily strained can become embrittled if heated, particularly into the critical range
of approximately 250 °C – 450 °C. This can be avoided by welding (or any other heating procedure) at

sN
some distance from bends or, provided the steel is not quenched and tempered, by a stress relieving heat
treatment of the bend zone.

C8.5.2 In-line quenched and tempered steel bars

rd
Welding of in-line quenched and tempered bars can have detrimental effects on the strength and ductility
of the bars and associated connection. AS 3600 requires designers to assume that the strength of such
reinforcement has a design strength of 250 MPa when raised to the temperatures associated with welding,
da

galvanising or hot bending. Such a requirement is considered inappropriate in a seismically active country
where concentration of yielding at a weld position would be undesirable and could result in brittle failure.
an

C8.6 Development of reinforcement

St

C8.6.1 Development of reinforcement – General

The development length concept for anchorage of reinforcement was first introduced in the 1971 ACI
Building Code, to replace the dual requirements for flexural bond and anchorage bond contained in earlier
ht

editions of the ACI Building Code. It is no longer necessary to consider the flexural bond concept which
placed emphasis on the computation of nominal peak bond stresses. Consideration of an average bond
resistance over a full development length of the reinforcement is considered more meaningful, partly
rig

because most bond tests consider average bond resistance over a length of embedment of the
reinforcement, and partly because unpredictable extreme variations in local bond stresses exist near
flexural cracks 8.1.
py

The term "development" used in this Standard implies the development of the required strength of the
reinforcement at a critical section. This may be the computed tensile stress, yield strength, fy, or breaking
Co

stress.

The development length concept is based on the attainable average bond stress over the length of
embedment of the reinforcement. In application the development length concept requires the specified
minimum lengths or extensions of reinforcement beyond all points of peak stress in the reinforcement. In
flexural members such peak stresses generally occur at the points specified in 8.6.12.2. This development
length or anchorage is necessary on both sides of such peak stress points.

Often the reinforcement continues for a considerable distance on one side of a critical stress point so that
calculations need involve only the shorter distance; for example, the negative moment reinforcement
continuing through a support to the middle of the next span. However, bars often need to be extended by

C8 - 4
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

a greater distance to satisfy special requirements applicable to tension zones of flexural members in
accordance with 8.6.12, 8.6.13 and 8.6.14.

C8.6.3 Development length of deformed bars and deformed wire in tension

8.3
A combination of recommendations from ACI 318 and other research have been used to form the
equations in 8.6.3.2 and 8.6.3.3.

nd
C8.6.3.2 Basic development length in tension
Concrete cover, clear distance between bars in a layer and bar diameter are the principal quantities which

ala
determine the basic development length of a bar. Transverse reinforcement, crossing splitting cracks, will,
to a certain extent, improve anchorage. Accordingly, empirical expressions have been derived to
determine the basic development length Ldb. Additional parameters, which may beneficially influence

Ze
development, are then taken into account separately in 8.6.3.3.

The basic development length, Ldb, in this clause is proportional to the lower characteristic yield stress of
the bar. Equation 8–2 includes a to recognise that for top reinforcement the reduction in the quality of

ew
bond because the excess water used in the mix for workability and air entrapped during the mixing and
placing operations may rise toward the top of the finished concrete before setting is complete. Entrapped
below bars, this water and air leaves the bar less bonded to the concrete on the underside. For horizontal
top bars in a structural member, bond resistance reflects this weakened underside restraint because the

sN
loss can be of the order of 50 % in extreme cases.

The symbol f ć is limited to 70 MPa because data on the development and bond of bars at concrete
strengths above 70 MPa is not readily available to date.
rd
To allow designers to reduce Ldb, if desired, a more rigorous determination of Ld may be undertaken using
8.6.3.3.
da

C8.6.3.3 Refined development length in tension

To reduce the development length specified in 8.6.3.2, three factors may be considered:
an

(a) Because the development length required is proportional to the tensile stress to be developed, the full
development length may be reduced proportionally when the stress is lower than the yield strength.
This is achieved by the modification factor Asr /Asp. It should be noted, however, that this reduction
St

must not be used at or near critical sections of members subjected to earthquake forces, nor should
the area of reinforcement required only to control shrinkage and temperature effects in restrained
members be omitted when computing Asr.
ht

(b) Concrete splitting is a common mode of failure when the strength of bars is developed, and when the
surrounding concrete cannot sustain the circumferential tensile stresses induced by the bond
stresses. As either cover or clear distance between bars is increased, improved resistance to
ig

concrete splitting is achieved.

yr

Equation 8–5 recognises that an increased cover or clear distance between bars will result in increased
bond strength. Equation 8–5 is based entirely on test results. Figure C8.3 indicates typical splitting cracks
along embedded bars. cm is the smallest of cb, cs and cp.
p
Co

Figure C8.3 – Definition and significance of distances cb, cs and cp

C8 - 5
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

The beneficial effect of transverse reinforcement 8.4, which may control the opening of splitting cracks as
shown in Figure C8.3, is expressed by the area of transverse reinforcement Atr. The effectiveness of ties,
stirrups, hoops or spirals in crossing a potential splitting crack is illustrated in Figure C8.4. For such
reinforcement to be effective, at least three bars must cross a potential crack over the development length.
However, transverse reinforcement used for any other purpose, such as shear resistance or to provide
stability for compression bars or confinement of concrete, may be included for this purpose.

nd
In the case of Figure C8.4 (a), Atr is effective only for the outer bars. The designer would have several
choices in this case. A different Ld could be calculated for the inner and outer longitudinal bars, or the

ala
effect of transverse reinforcement could be ignored, or Atr could be taken into account as an average for
the bars, using total area of transverse bars crossing the plane of splitting divided by the number of
longitudinal bars in the layer, n. The last approach was checked 8.3 using data reported by Untrauer and
Warren 8.5 and it was found to give a reasonable estimate of measured values. This approach is

Ze
incorporated in the definition of Atr in this section.

Where a number, n, of longitudinal bars are enclosed by a spiral, the value of Atr to control splitting, as
shown for a circular member in Figure C8.4 is given by:

ew
6 At
Atr = ≤ At .........................................................................................................................(Eq . C8–1)

sN
n

(a) For horizontal splitting through the layers of

rd
ΣAt
bars can put Atr =
n
2 At
da

=
4

(b) For splitting through the cover concrete

an

Atr = At
St

(c) For wide sections multiple stirrups are effective

ΣAt
ht

Atr =
n
(
2 At + At* )
ig

=
7
(d) Spiral
yr

6 At
Atr = ≤ At
n
p
Co

Figure C8.4 – Basis for calculation of Atr

The 300 mm minimum development length shall not be multiplied by the αa, αb, αc or αd factors.

Because the multiplier in 8.6.3.3(c), which allows for the beneficial effects of transverse reinforcement,
involves additional calculations, the designer may always assume that Atr is zero, so that the multiplier
becomes unity.

C8 - 6
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

C8.6.4 Development length of plain bars and plain wire in tension

As required by 5.3.1, plain bars other than those explicitly listed should only be used when special
verifiable reasons exist. The development of plain bars in tension must not rely on straight development
length.

C8.6.5 Development length of deformed bars and deformed wire in compression

d
These provisions are similar to those of the previous structural Concrete Standard, NZS 3101:1995.

lan
The weakening effect of flexural tension cracks is not present for bars in compression and usually end
bearing of the bars on the concrete is beneficial. Therefore, shorter development lengths have been
specified for compression than for tension. The development length may be reduced by up to 25 %

a
according to 8.6.5.3 when the compression reinforcement is enclosed within a column by spiral or

Ze
rectangular ties, hoops or supplementary ties, or an individual spiral around each bar or group of bars is
used. The interpretation of the effective transverse reinforcement area in the calculation of Atr for 8.6.5.3 is
defined in the notation of the Standard and illustrated in Figure C8.4.

ew
C8.6.7 Development of bundled bars
An increased development length for individual bars is required when three or four bars are bundled
together. The extra extension is needed because the grouping reduces the effective surface area over
which bond stresses to the surrounding concrete can be transferred.

sN
The designer should also note 8.3.4 relating to the cutoff points of individual bars within a bundle and
8.7.2.2 relating to splices of bundled bars.
rd
C8.6.8 Development of welded plain and deformed wire fabric in tension

C8.6.8.1 Development length of wire fabric

da

The requirements of either 8.6.8.2 or 8.6.8.3 may be used to calculate the development length, Ld,
required for plain or deformed wire fabric in tension.
an

C8.6.8.2 Development length of welded wire fabric – cross wires considered

Figure C8.5 shows the development requirements for wire fabric with the development primarily
dependent on the location of the cross wires rather than the bond characteristics of the plain or deformed
St

wire. An embedment of at least two cross wires 50 mm or more beyond the point of critical section is
adequate to develop the yield strength of anchored wires 8.6.
ht
rig
py
Co

Figure C8.5 – Development of welded wire fabric

C8.6.8.3 Development length of welded wire fabric – cross wires not considered
If no cross-wires are present or assumed to be present the development of the plain or deformed wire will
be dependent upon bond. Hence the requirements of 8.6.3 or 8.6.4 will govern, except that the minimum
development length is reduced to 200 mm.

C8.6.9 Development of prestressing strand

The development requirements for prestressing strand are intended to provide bond integrity for the
strength of the member. The provisions are based on tests performed on normal density concrete
C8 - 7
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

members with a minimum cover of 50 mm. These tests may not represent the behaviour of strand in low
water-cement ratio, no-slump concrete. Fabrication methods should ensure consolidation of concrete
around the strand with complete contact between the steel and concrete. Extra precautions should be
exercised when low water-cement ratio, no-slump concrete is used. In general, this clause will control only
for the design of cantilever and short-span members. The requirement of doubled development length for
strand not bonded through to the end of the member is also based on test data 8.7.

nd
The expression for development length Ld may be rewritten as:

ala
 f se 

 3 d b  f ps  f se d b  
 
Ld  ........................................................................................................ (Eq . C8–2)
7

Ze
where Ld and db are in mm, and fps, and fse are in MPa. The first term represents the transfer length of the
strand, that is, the distance over which the strand must be bonded to the concrete to develop the prestress
fse in the strand. The second term represents the additional length over which the strand must be bonded

ew
so that a stress fps may develop in the strand at nominal strength of the member.

The variation of strand stress along the development length of the strand is shown in Figure C8.6.

sN
rd
da
an
St
ht
ig
yr

Figure C8.6 – Variation of steel stress with distance from free end of strand
p

The expressions for transfer length and for the additional bonded length necessary to develop an increase
Co

in stress of (fps – fse) are based on tests of members prestressed with clean 8 mm, 9 mm and 12 mm
diameter strands for which the maximum value of fps was 1980 MPa 8.7, 8.8, 8.9.

The transfer length of strand is a function of the perimeter configuration area and surface condition of the
strand, the stress in the strand and the method used to transfer the strand force to the concrete. Strand
with a slightly rusted surface can have an appreciably shorter transfer length than clean strand. Gentle
release of the strand will permit a shorter transfer length than abruptly cutting the strands.

The provisions of 8.6.9 do not apply to plain wires nor to end anchored tendons. The length for smooth
wire could be expected to be considerably greater due to the absence of mechanical interlock. Flexural
bond failure would occur with plain wire when first slip occurred.

C8 - 8
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

C8.6.9.3 Prestressing strand transfer length

A3
The design transfer length is based on the stress in the strand after prestress loss. However, it depends
on the concrete strength at transfer. The expression for transfer length is valid provided f′c equals or
exceeds 21 MPa. An increase in concrete strength reduces the transfer length as does a slower rate of
transfer. It should be noted that the transfer and development lengths given in 8.6.9.1 and 8.6.9.3 allow for

nd
the potential increase in development length with time8.15.

C8.6.10 Standard hooks

8.3, 8.10

ala
Clause 8.6.10 is based on the recommendations of ACI Committee 408 . Refer to Figure 8.1 for a
diagrammatic explanation of Ldh.

One or more bars anchored by standard hooks with a development length according to Equation 8–12,

Ze
and in close proximity to each other, should develop the strength of the bars provided the bars are
included in a viable "strut and tie" mechanism. A viable mechanism consists of equilibrating internal
actions where the bond stresses along the hook and the bearing stresses in the bend of the hook are
balanced by (i) compression fields in the surrounding concrete and (ii) tension fields produced by

ew
reinforcement bounding and passing through the volume of concrete in which the bars are anchored.

Note that Reference 8.11 recommends that for the strut developed inside of the bend of the hook the
angle of the strut to the straight shaft of the hook (length Lb, in Figure 8.1) should not be greater than 55°.

sN
If the angle is greater than 55° then the pull-out of a cone of concrete, before the yield strength of the bar
is reached, is likely. This failure mode should be avoided. A typical situation where a concrete cone pull-
out can occur at the connection of a floor to a wall, is where starter bars are anchored with a standard
hook close to the adjacent face of the wall.
rd
Typical situations where the "strut and tie" mechanism exists include: beam-column joints, column and
beam stubs (for anchoring bars outside the beam-column joints) and longitudinal bars terminated by
da

standard hooks at (i) the end of cantilevers elements (slabs, beams and foundation pads) and (ii) where
curtailment of the longitudinal bars occurs within elements, where the traditional shear "truss" or "strut and
tie" mechanisms exist.
an

Where a "strut and tie" mechanism does not exist, the failure mode of the bar under tension may be the
pull-out of a cone of concrete, before the yield strength of the bar is reached. It is possible to prevent the
St

pull-out of a concrete cone and the bar embedded in it, by tying back the cone into the "strut and tie"
mechanism with appropriate tension reinforcement.
ht

Meshes or grillages of reinforcement in the plane of the concrete element, such as a wall panel, are
ineffective in preventing a cone type of failure 8.12. One method for determining adequate embedment or
development lengths terminated with standard hooks (not complying with 8.6.10.3) may be found in
ig

Reference 8.12.
yr

A study of the failures of hooked bars indicates that splitting of the cover parallel to the plane of the hook
is the primary cause of failure and that the splitting originates at the inside of the hook where stress
concentrations are very high. For this reason, Equation 8–12 is a function of db which governs the
p

magnitude of compressive stresses on the inside of the hook. Only standard ACI hooked bars were tested
Co

and the influence of a larger radius of bend was not evaluated. The test results indicate that as the straight
lead length increases, the lateral splitting force which develops in the side cover decreases; this is
reflected in an improvement in hook capacity.

The recommended provisions include adjustments to reflect the resistance to splitting provided by
enclosure in transverse reinforcement. If the side cover is large so that side splitting is effectively
eliminated, as in mass concrete, the product of the factors αb, α1 and α2 as given in 8.6.10.3 may be used.
Minimum values of Ldh are indicated to prevent failure by direct pullout in cases where the standard hook
may be located very near the critical section. No distinction is made between top bars and other bars.

In many cases where the value of Ldh given by Equation 8–12 is used, the value of di required will be
greater than that given in Table 8.1 as it will be governed by Equation 8–1. In such cases, if it is desired to

C8 - 9
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006
reduce the value of di to that given in Table 8.1, the value of Ldh will have to be increased above that given
to be used by Equation 8–12 in order to give an increased value for the lead length Lb as shown in
Figure 8.1 which will allow a reduced value of di from Equation 8–1.

C8.6.10.1 Standard hooks – definition

The standard hooks defined in this section are shown in Figure 8.1.

nd
C8.6.10.3 Development length of standard hooks in tension
The required development length Ldh for hooked bars in tension in accordance with 8.6.10, may be larger

ala
than what might be available in a column when the requirements of 9.4.3.2 shown in Figure C9.18 are to
be satisfied. In such situations it is better to improve the bearing conditions in the bend than to provide
extra straight anchorage length beyond the 90° bend. When transverse bars, as shown in Figure C8.1, are

Ze
provided, a 20 % reduction in the development length Ldh of Figure 8.1 may be made. When beam bars
are anchored within column bars in the core of a beam-column joint, the application of the multiplier α1 =
0.7 in 8.6.10.3(b) is appropriate.

ew
The bars placed in the bend help reduce the local bearing stresses and reduce the tendency for splitting
cracks to form in the plane of the bend. The extension of these bars by 3db beyond the plane of the bar
does not imply any limit on the spacing of adjacent bent bars.

sN
When the same bar is required to develop yield strength in compression, the bent portion of the
anchorage must be disregarded in satisfying the requirements of 8.6.5.1. However, when bars are
anchored in column cores, as described above, the confinement may be considered to be equivalent to
that implied in 8.6.5.3. The development of bars in compression will commence closer to the inner face of
rd
exterior columns.

C8.6.11 Mechanical anchorage

da

A3 Mechanical end anchorages and bar couplers should be made adequate for strength both for prestressing
tendons and for reinforcing bars.
an

Mechanical anchors and couplers need to be resistant to brittle fracture under their normal service
temperature conditions. Anchors and couplers manufactured by processes involving heat treatment of the
steel and bars to be coupled that have their ends enlarged for threading by cold forging will have their
St

mechanical properties including their brittle fracture resistance altered by these processes. Therefore their
manufacturing processes need to be subjected to appropriate testing of the finished products’ mechanical
properties and quality control procedures need to be employed to provide quality assurance of mechanical
ht

anchor and coupler systems.

Mechanical anchors and couplers manufactured from Grade 500/7 spheroidal graphite iron are not to be
ig

used. This material may contain casting defects and is likely to be prone to brittle failure under impact
loading at normal service temperatures8.14.
p yr
Co

C8 - 10
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

C8.6.12 Development of flexural reinforcement

C8.6.12.2 Critical sections

Critical sections for a typical continuous beam are indicated in Figure C8.7, together with the different
criteria which determine where bars may be cut off.

d
lan
8.6.12.3(a)

a
8.6.12.3(b)

Ze
ew
sN
rd
da
an
St
ht
rig
py
Co

8.6.13.4

8.6.13.2 and 8.6.13.4

Figure C8.7 – Development of flexural reinforcement in a typical continuous beam

C8.6.12.3 Extension of tension reinforcement

The moment diagrams customarily used in design are approximate; some shifting of the location of
maximum moments may occur due to changes in loading, settlement of supports, unaccounted lateral
C8 - 11
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

forces or other causes. A diagonal tension crack in a flexural member without stirrups may shift the
location of the calculated tensile stress approximately a distance d towards a point of zero moment. When
stirrups are provided, this effect is less severe, although still present to some extent.

To provide for shifts in the moment demand, the Standard requires the extension of reinforcement by a
distance 1.3d beyond the point at which it is theoretically no longer required to resist flexure, and by d

d
beyond the length Ld. Cut-off points of bars to meet this requirement are illustrated in Figure C8.7.

lan
When bars of different sizes are used, the extension should be in accordance with the diameter of bar
being terminated. A bar bent to the opposite face of a beam and continued to the point where the bar
crosses the mid-depth of the member may logically be considered effective in satisfying this clause.

a
The same principles apply to the curtailment of vertical reinforcement in walls, as implied by the design

Ze
moment envelope for cantilever walls shown in Appendix D, Figure CD.7.

C8.6.12.4 Termination in a tension zone

Evidence of reduced shear strength and consequent loss of ductility when bars are cut off in a tension

ew
zone, as in Figure C8.7, has been reported by several investigators 8.13. As a result, the Standard does
not permit flexural reinforcement to be terminated in a tension zone unless special conditions are satisfied.
Flexural cracks tend to open early wherever any reinforcement is terminated in a tension zone. If the steel

sN
stress in the continuing reinforcement and the shear strength are each near their limiting value, diagonal
tension cracking tends to develop prematurely from these flexural cracks. Diagonal cracks are less likely
to form where shear stress is low (see 8.6.12.4(a)). Diagonal cracks can be restrained by closely spaced
stirrups (see 8.6.12.4(b)). Tension bars bent into the web at an angle not exceeding 45° and terminating at
a distance of at least d/2 away from the tension face may be considered exempt from the requirements of
rd
this clause, because such bars do not terminate in a tension zone. These requirements are not intended
to apply to tension splices which are covered by 8.7.2 and the related 8.6.3.
da

C8.6.12.5 End anchorage in flexural members

Members such as brackets, members of variable depth and others where steel stress fs does not
an

decrease linearly in proportion to a decreasing moment require special consideration for proper
development of the flexural reinforcement. For the bracket shown in Figure C8.8 the stress in the
reinforcement at nominal strength is almost constant at approximately fy from the face of support to the
St

load point. In such a case, development of the flexural reinforcement depends largely on the anchorage
provided at the loaded end. A welded cross bar of equal diameter may be used as a means of providing
effective end anchorage. An end hook in the vertical plane, with the minimum diameter bend, is not totally
ht

effective because an unreinforced concrete corner may exist near loads applied close to the corner.
A3
rig
py
Co

Figure C8.8 – Consideration of the critical anchorage for a special member

C8 - 12
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

C8.6.12.6 Anchorage of flexural reinforcement in external beam-column joints A3

A major proportion of the joint zone shear force is resisted by a diagonal compression force which acts
from the compression corner of the tension zone corner of the joint zone. Where beam flexural tension
reinforcement is anchored by a 90° hook the diagonal compression force bears against the hook, inducing
a vertical tension force in the vertical leg of the hook. To enable the joint zone to reach its full strength the

d
tension force in the vertical leg must be transferred by bond to the adjacent vertical reinforcement in the
column. To enable this force transfer to occur, reinforcement that encloses both the vertical leg of the

lan
hooked bar and the longitudinal column reinforcement is required.

C8.6.13 Development of positive moment reinforcement in tension

a
C8.6.13.1 Limitation in area of bars

Ze
Certain proportions of the maximum positive moment reinforcement are required to be carried into the
support to provide for some shifting of the moment due to changes in loading, settlement of supports,
lateral forces and other causes.

ew
C8.6.13.2 Critical sections
When a flexural member is part of a primary lateral force-resisting system, forces greater than those
anticipated in design may cause reversal of moment at supports; therefore the required positive

sN
reinforcement should be well anchored into the support. This anchorage is to assure ductility of response
in the event of unexpected overstress, such as from blast or earthquake. It is not sufficient to use more
reinforcement at lower stresses. A3
rd
C8.6.13.3 Limitation in diameter of bars at simple supports
Flexural bond considerations require the anchorage length to be checked in regions of members where
da

the bending moment is zero, that is at simple supports and at points of contraflexure. In such regions the
area of tension reinforcement provided may be small and the shear force relatively large, resulting in high
flexural bond stresses. Clauses 8.6.13.3 and 8.6.13.4 ensure bond failure will not occur. In Figure C8.9,
an

Mn is the nominal moment supplied by the reinforcement at the support. Mn/V * can be increased by 30 %
if the reaction confines the end of the reinforcement.
St

Figure C8.9 and Figure C8.10 illustrate the use of the provisions of 8.6.13.3 and 8.6.13.4.
ht
rig
py
Co

C8 - 13
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

nd
ala
Ze
ew
sN
rd
Figure C8.9 – Procedure for determining maximum size bar at simple support

Figure C8.10 illustrates the use of the provisions of 8.6.13.4.

da
an
St
ht
ig
yr

Figure C8.10 – Procedure for determining the maximum size of bars “A” at a point of inflection for
positive reinforcing
p

In routine design it may often be found that Mn /V * > Ld and hence no further check need then be made.
Co

When a requirement of 8.6.13.3 or 8.6.13.4 is not satisfied, the designer should either reduce the diameter
of bars, whereby Ld is reduced, or increase the area of positive reinforcement at the section considered,
whereby Mn is increased, or undertake both of these steps.

C8 - 14
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

nd
ala
Figure C8.11 – Anchorage into exterior column

C8.6.14 Development of negative moment reinforcement in tension

Ze
Figure C8.11 and Figure C8.12 illustrate two methods of satisfying requirements for anchorage of tension
reinforcement beyond the face of support. For anchorage of reinforcement with hooks, see commentary
discussion C8.6.10.

ew
Clause 8.6.14.3 provides for possible shifting of the moment diagram at a point of inflection, as discussed
under C8.6.13.3. This requirement may exceed that of 8.6.13.3 and the more restrictive of the two
provisions governs.

The principles involved in 8.6.14.4 are the same as those considered in C8.6.13.3.
sN
rd
da
an
St

NOTE – Usually anchorage in the column becomes part of the adjacent beam reinforcement.
Figure C8.12 – Anchorage into adjacent beam
ht

C8.7 Splices in reinforcement

ig

For ductility of a member, lap splices should be adequate to develop more than the yield strength of the
reinforcement; otherwise a member may be subject to sudden splice failure when the yield strength of the
reinforcement is reached. The lap splice lengths specified in the Standard satisfy this ductility requirement
yr

for members.
p

Splices should, if possible, be located away from points of maximum tensile stress.
Co

The use of welded splices or mechanical connections with capacity less than the actual breaking strength
is permitted if the design criteria of 8.7.5.4 are met. Therefore, lap welds of reinforcing bars, either with or
without back-up material, welds to plate connections, and end-bearing splices are allowed under certain
conditions.

C8.7.1 General
The designer's written approval should be obtained for any welding as what seems to be an unimportant
weld to a site operative could affect a critical member.

C8 - 15
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006
C8.7.2 Lap splices of bars and wire in tension

C8.7.2.1 Bar sizes of lap splices

Research on lap splices with bars of diameter greater than 40 mm is limited. There is insufficient data to
establish lap lengths for either tensile or compressive lap splices for these bars.

d
C8.7.2.2 Lap splices of bundled bars

lan
The increased length of lap required for bars in bundles is based on the reduction in the exposed
perimeter of the bars. Where the factors in this clause are applied it is not intended that the factors in 8.6.7
should also be applied.

a
C8.7.2.3 Length of lap splices of deformed bars or wire
This clause follows the recommendations of ACI Committee 408 8.10, 8.3. The recommendation that splice

Ze
and development for deformed bars and wire are the same is also adopted by ACI 318. Statistical studies
have shown that no additional factors are necessary for splices . Straight plain bars shall not be spliced
except with hooks or mechanical anchorages.

ew
In determining the required splice length, Ls, the distance cp to be used is illustrated in Figure C8.13.
Where all bars are spliced at the same location, cp is the clear distance between bars. Where the splices
are staggered and the overlap is less than Ls, cp reflects this improvement. With staggered splices, the

sN
spacing between bars generally will not be as critical as is the cover to the centre of the bar.
rd
da
an
St
ht
rig

Figure C8.13 – Definition of cp for splices

py

C8.7.2.4 Length of lap splices of plain bars or wire

This clause will apply when, under 5.3.1, there are specific or special necessities to use plain bars for
other than ties, stirrups, spirals or hoops. The required standard hooks, as required by 8.6.10.1 and 8.4.2,
Co

such as shown in Figure 8.1, are not intended to pass around and engage other reinforcement. When the
hooks are located near the surface of a member, the hooks should be embedded in the core concrete of
the member. One application of lap splices in plain bars is detailed in 8.7.2.8.

C8.7.2.5 Length of non-contact lap splices

To ensure that the effective splice length, Ls, is maintained in splices with transverse spacings, sL, of bars
larger than 3db, Lds is introduced which assumes an approximately 33° diagonal compression field as
illustrated in Figure C8.14.

C8 - 16
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

nd
Figure C8.14 – The spacing of spliced bars

ala
C8.7.2.8 Lap splices of stirrups, ties and hoops
Deformed bars and plain bars used for stirrups, ties or rectangular hoops may be spliced when necessary,

Ze
provided standard hooks as required by 8.6.10.1 and 8.4.2 such as shown in Figure 8.1, are used. The
hooks must be embedded in the concrete core, that is, the plane of the hook must not be in the cover
concrete. Hooks anchored through and inside the core may be oriented to suit convenience in
construction. Hooks required for lap splicing of transverse reinforcement need not engage a longitudinal

ew
bar for anchorage.
Bond forces generate hoop tensile stresses in the concrete surrounding the reinforcement. With closely A3
spaced bars the hoop tensile stresses generate tensile stresses which under certain conditions may lead

sN
to spalling of the concrete. Such spalling may destroy the development of the lapped bar and the
effectiveness of the stirrup or tie leg.
rd
da

Lap in cover
concrete
an

Not more than 50 per cent

of stirrup or tie
reinforcement may be
St

lapped in cover concrete

Figure C8.15 – Laps in stirrups and ties

ht

Locating 50 % or more of the lapped stirrups in the core concrete reduces the tendency for bond stresses
ig

in the stirrup lap zone to result in spalling of the cover concrete.

C8.7.3 Lap splices of bars and wires in compression

yr

Recent bond research has been primarily related to bars in tension. Bond behaviour of compression bars
is not complicated by the problem of transverse tension cracking and thus compression splices do not
p

require provisions as strict as those specified for tension splices.

Co

C8.7.3.2 Lap splices in compression with stirrups and ties

Effective tie legs included in the evaluation of Atr are those which cross a potential splitting crack which
develops in the plane at which two spliced bars might be in contact with each other. An example is shown
in Figure C8.17. A3

C8.7.3.3 Lap splices in compression with spiral reinforcement

Compression lap lengths may be reduced when the lap splice is enclosed throughout its length by spirals
because there is increased splitting resistance. Spirals should meet requirements of 10.3.10.7, 10.3.10.8
and 10.3.10.5. Because spirals do not cross a potential splitting crack when spliced bars in contact are
aligned radially, they are less efficient in confining a splice. Therefore the area of the spiral is required to
be N/6 times longer than that of a tie crossing a crack at 90° assumed in 8.7.3.2. Potential radial splitting

C8 - 17
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

cracks, developing when all spliced bars touch a circular spiral may eventually link up with circumferential
cracks because the diameter of the strained spiral increases. The two mechanisms are illustrated in
A3 Figure C8.17.

C8.7.4 Welded splices and mechanical connections

Designers should avoid the need to weld reinforcing steel if possible as follows:

nd
(a) Where butt jointing is required there is a good range of coupling devices available. Lapping,
particularly of smaller bars, may also be an option;
(b) Tack welding of stirrups or ties to main bars may result in a reduction in capacity of the main bar,

ala
either through metallurgical changes, or the generation of notches due to undercut if the procedures
of AS/NZS 1554:Part 3 are not followed;
(c) Where welds are required to provide lightning protection, care should be taken to choose a route

Ze
through non-critical members.
Welds complying with 8.7.4.1(a) can withstand the most severe strain or stress cycles. Hence they are
acceptable in all locations, in particular, for splicing main longitudinal reinforcement in plastic hinge
regions and in beam-column joints. Weld quality should comply with the requirements of AS/NZS 1554:

ew
Part 3, Section 9 for “Direct Butt Splices”.

The categories of splices in 8.7.4.1(b) will be adequate for large bars in main members outside plastic
hinge regions and for welded splices in stirrups, ties, spirals or hoops. The limit of the breaking strength of

sN
the bar will ensure that the strength of the connection will be greater than the maximum design force in the
bar. Weld quality should comply with the requirements of AS/NZS 1554:Part 3, Section 9 for “Other
splices”.
rd
C8.7.4.2 Limitations on the classification of welded splices for grade > 450 MPa reinforcement
The current Standard for welding of reinforcement, AS/NZS 1554.3, allows the use of welding
da

consumables with a minimum strength of 550 MPa (E5518) or 620 MPa (E6218). It is considered unlikely
that the use of the lower strength electrode will provide an appropriately high probability that the full
strength of a Grade 500 bar will be achieved. Whether the full strength of the bar in the upper
an

characteristic range for Grade 500 reinforcing can be achieved with the higher strength electrode requires
verification.
St

Yielding of the weld is undesirable as the plastic deformation is limited to a short length. This can greatly
reduce the ductility of the bar and lead to a brittle type of failure of a member. For this reason welded
Grade 500 reinforcement should be approached with caution where plastic deformation may be required.
ht

C8.7.5.2 Performance requirements for classification as a “high-strength” mechanical connection

A stiffness criterion is imposed on mechanical splices of C8.7.5.2(b) to ensure that large premature cracks
ig

are not produced by excessive extensions in splicing devices. Accordingly the displacement of the spliced
bars relative to each other and measured in a test over the length of the connector, should not exceed
yr

twice the elongation of the same size of unspliced bar over the same measured distance when subjected
to 0.7 fy.
p

C8.7.5.3 Use of welded splices and mechanical connections

See commentary on 8.7.4.1(c). This clause describes the situation where welded splices or mechanical
Co

connections with capacity less than the actual breaking strength of the bar may be used. It provides a
relaxation in the splice requirements where the splices or connections are staggered and an excess
reinforcement area is available. The criterion of twice the computed tensile stress is used to cover
sections containing partial tensile splices with various percentages of the total reinforcement continuous.

C8.7.6 Splices of welded plain or deformed wire fabric

The strength of lap splices is dependent on either the anchorage obtained from the cross-wires, as shown
A3 in Figure C8.16 and as detailed in 8.6.8.2, or the development of the individual wires as detailed in 8.6.8.3.

C8 - 18
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

d
lan
A3
Figure C8.16 – Lap splice of welded fabric

C8.8 Shrinkage and temperature reinforcement

a
So-called shrinkage and temperature reinforcement is required at right angles to the principal

Ze
reinforcement to prevent excessive cracking and to tie the structure together to assure behaviour as
assumed in the design. The amount specified (0.7/fy) is empirical but follows closely values which have
been used satisfactorily for many years.
A3

ew
It should be kept in mind that the reinforcement ratios given in this clause are minimum values and apply
to the situation where restraint against shrinkage has been minimised. It is well known that if the slabs are

sN
fully restrained against shrinkage and temperature movement, much higher reinforcement ratios are
required to avoid severe cracking. In most cases it is possible to select structural form, construction joint
positions and pouring sequences to minimize restraint in suspended slabs, and this Standard has followed
the practice of most leading national codes in giving reinforcement ratios appropriate to this situation.
rd
For the condition of full restraint, first principles require that the yield strength of the reinforcement passing
da

through any potential crack position should be greater than the ultimate tensile strength of the
corresponding cross-sectional area of concrete during the period after initial setting. This would require,
for example, a reinforcement percentage of the order of 0.45 % for the case of a specified 28 day concrete
an

compressive strength f ć of 25 MPa and characteristic yield strength of reinforcement fy of 300 MPa.

Splices and end anchorages must be designed for the full specified yield strength.
St

C8.9 Additional design requirements for structures designed for earthquake effects
ht

C8.9.1 Splices in reinforcement

C8.9.1.1 Placement of splices

rig

Splices other than those described by 8.7.4.1(a) should not be used in potential plastic hinge regions or in
beam-column joints where anchorage conditions may be very critical. Therefore splices should be located
away from critical sections of potential plastic hinges by the distances specified. In a column, if plastic
py

hinges form, their location will be at the top and bottom ends of storey heights, adjacent to beams or
footings. When plastic hinge development is not expected because columns, designed in accordance with
Appendix D, Method A, have considerable reserve flexural strength, splices may be located also in the top
Co

and bottom ends of storey heights, the preferred position usually being immediately above a floor.

C8.9.1.2 Lap splices in regions of reversing stresses

Transverse reinforcement provided around splices with Grade 430 bars in accordance with Equation
818, was found to ensure that at least 85 % of the nominal strength of a column section with all bars
spliced could be sustained in at least 20 cycles of reversed loading without distress. Such splices were
found to sustain even a few limited excursions beyond yield. In determining the splice length from 8.6.3,
the beneficial effect of this transverse reinforcement may be utilised for Atr. Transverse ties must cross
potential sliding failure planes between two spliced bars as shown in Figure C8.17. A3

C8 - 19
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

d
a lan
Ze
ew
A3 Figure C8.17 – Bar force transmission by shear-friction at lapped splices

sN
At locations along beams and the mid-height of columns, where it can be shown that reversing stresses
do not exceed 0.6fy, in tension or compression, transverse reinforcement provided to satisfy other
requirements, such as 8.7.3.2 and 8.7.3.3, may be considered to ensure satisfactory splice performance.
rd
C8.9.1.3 Requirements for welded splices or mechanical connections
In members that are subjected to seismic forces, welded splices or mechanical connections are required
to develop the breaking strength of the bars. This is due to the consideration of the severe consequences
da

for the structure if failures of such classes of connections do occur. The requirement is analogous to lap
splices of bars being required to develop more than the yield strength of the bars and not being reduced in
length because more reinforcement is provided than that required at the lap location (8.6.3.3 and
an

9.4.3.2.3).

In determining the criteria for welded splices and mechanical connections, standard of workmanship,
St

difficulty of inspection, and the final reliability of the splice in service has been taken into account. Even so,
designers should be aware of the necessity for a site testing programme to ensure that these splices meet
A3 the requirements of 8.7.5.2. and 8.9.1.3.
ht
rig

Splices conforming to 8.7.4.1(a) and 8.7.4.1(b) may be located in the same plane or section.

REFERENCES
py

8.1 ACI Committee 408, “Bond Stress – The State of the Art”, ACI Journal, Proceedings Vol. 63, No.
Co

11, Nov. 1966, pp. 1161-1188. Also “ACI Manual of Concrete Practice”, American Concrete
Institute, Detroit, 1979, Part 2.
8.2 “Standard Specifications for Highway Bridges”. American Association of State Highway and
Transportation Officials, Washington, D.C., 11th edition, 1974, 284 pp. lnterim Specifications, 1973.
8.3 Jirsa, J.O., Lutz, L.A. and Gergely, P., ”Rationale for Suggested Development, Splice and Standard
Hook Provisions for Deformed Bars in Tension”, Concrete International, Vol. I, No.7, July 1979, pp.
47-61.
8.4 Paulay, T. and Priestley, M.J.N., “Seismic Design of Reinforced Concrete and Masonry Buildings”,
John Wiley, New York, 1992.
8.5 Untrauer, Raymond E. and Warren, George E., "Stress Development of Tension Steel in Beams",
ACI Journal, Proceedings Vol. 74, No. 8, Aug. 1977, pp. 368 - 372.

C8 - 20
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless your actions are covered by Part 3 of the Copyright Act 1994.
 NZS 3101:Part 2:2006

8.6 Anderson, A.R., “Bond Properties of Welded Wire Fabric”, ACI Journal, Proceedings Vol. 50, No. 4,
April 1952, pp. 681-692.
8.7 Kaar, P. and Magura, D., “Effect of Strand Blanketing on Performance of Pretensioned Girders”,
Journal, Prestressed Concrete Institute, Vol. 10, No. 6, Dec. 1965, 15 pp.

nd
8.8 Hanson, N.W. and Kaar, P.H., “Flexural Bond Tests Pretensioned Beams”, ACI Journal,
Proceedings Vol. 55, No. 7, Jan. 1959, pp. 783-802. Also, Development Department Bulletin D28,
Portland Cement Association, 1959, 20 pp.
8.9 Kaar, P.H., La Fraugh, R.W. and Mass, M.A., “Influence of Concrete Strength on Strand Transfer

ala
Length” Journal, Prestressed Concrete Institute, Vol. 8, No. 5, Oct. 1963, pp. 47-67. Also,
Development Department Bulletin D71, Portland Cement Association, Oct. 1963, 21 pp.
8.10 ACI Committee 408, “Suggested Development, Splice and Standard Hook Provisions for Deformed

Ze
Bars in Tension”, Concrete International, Vol. l, No. 7, July 1979, pp. 45-46.
8.11 Comite European du Beton, Commission “Strength of Reinforced and Prestressed Concrete Beams
CEB Approach”, ACI Symposium 1976, Philadelphia, U.S.A.
8.12 Restrepo, J.I., Crisafulli, F.J. and Park, R., "Earthquake Resistance of Structures: The Design and

ew
Construction of Tilt-up Reinforced Concrete Buildings", Research Report 96-11, Department of Civil
Engineering, University of Canterbury, New Zealand, 1996.
8.13 Ferguson, Philip M. and Matloob, Farid N., “Effect of Bar Cut-off on Bond and Shear Strength of

sN
Reinforced Concrete Beams”, ACI Journal, Proceedings Vol. 56, No. I, July 1959, pp. 5-24.
8.14 Kirkcaldie, D.K. and Lloyd, J.N., “Some Concrete Topics Being Addressed in New Zealand
A3
Transport Agency Bridge Manual Revisions – Shrinkage and Creep in Concrete: Mechanical
Coupling of Reinforcing Steel”, NZ Concrete Industry Conference, Queenstown, October 2013.
rd
8.15 Dang, C.N., Royce, W.F., Hale, W.M. and Mariti-Vargas, J.R., “Measured Transfer Length in
17.8mm Strand for Pretensioned Beams”, ACI Structural Journal, Vol. 113, No.1, January 2016.
da
an
St
ht
ig
p yr
Co

C8 - 21
© The Crown in right of New Zealand, administered by the New Zealand Standards Executive. Access to this standard has been sponsored by the Ministry of Business, Innovation and Employment
under copyright licence LN001319. You are not permitted to reproduce or distribute any part of this standard without prior written permission from Standards New Zealand, on behalf of New Zealand
Standards Executive, unless
View publication stats your actions are covered by Part 3 of the Copyright Act 1994.