Transfer and Development

Lengths of Concentrically
Prestressed Concrete

J. R. Martí-Vargas, Ph.D.
Civil Engineer
Department of Construction
Engineering and Civil Engineering
Projects
Polytechnic University of Valencia
Valencia, Spain
M. A. Fernández-Prada, Ph.D.
Civil Engineer
Department of Construction
Engineering and Civil Engineering
Projects
Polytechnic University of Valencia
C. A. Arbeláez, Ph.D. Valencia, Spain
Civil Engineer
Department of Construction
Engineering and Civil Engineering
Projects
Polytechnic University of Valencia
Valencia, Spain

P. Serna-Ros, Ph.D.
Civil Engineer
Department of Construction
Engineering and Civil Engineering
Projects
Polytechnic University of Valencia
Valencia, Spain
Polytechnic University of Valencia
P. F. Miguel-Sosa, Ph.D.
Civil Engineer
Department of Construction
Engineering and Civil Engineering
Projects
Valencia, Spain

74 PCI JOURNAL
 At nominal strength of member
fps
Stress in the prestressing strand
fse
Prestress
only
Transfer length
Development length
Fig. 1. Idealized relationship between strand prestressing stress and development length.
An experimental investigation to develop a test to
characterize the bond between prestressing strand
and concrete has been carried out at the Polytechnic
University of Valencia in Spain. The proposed test
determines the transfer and development lengths by
measuring the force in the strand in a series of test
specimens with different embedment lengths. The
test sequentially reproduces the transfer of prestress
in a specimen and simulates service load through
a pull-out test on the same specimen. With this
technique, the Hoyer Effect is produced prior to the
pull-out phase of testing. An experimental program
determined the applicability of the test method and
the feasibility of a standardized test for quality con-
trol of prestressing strand bonding. Transfer- and
development-length tests were conducted on seven-
wire prestressing strand using 12 different concretes.
Results show that the test is valid and practical. Test
data indicate that ACI 318 equations are conserva-
tive. Details of testing equipment and procedures
are also presented in this paper.
September–October 2006 75
Flexural bond length
T
he behavior of pretensioned concrete members depends
on the bonding mechanism between the strand and the
surrounding concrete; strand prestressing forces are main-
tained solely by this mechanism. Two fundamental requirements
must be met to ensure bond integrity of prestressed concrete
members: the transfer of prestressing forces at strand release and
the ability of the strand to develop adequate stress forces when
subjected to overload conditions.
Section 12.9 of ACI 318-05 defines the following lengths
related to strand bond (Fig. 1):1
• Transfer length: the distance over which the strand
should be bonded to the concrete to develop the effec-
tive prestress in the prestressing steel fse.
• Flexural bond length: the additional length over which the
strand should be bonded so that a stress in the prestressing
steel at nominal strength of the member fps may develop.
• Development length: the sum of the transfer length
and the flexural bond length.
The development length is represented by the following
expression:1
 f  f −f 
ps se 
ld =  se  db +  d
  b (1)
 3000   1000 
where:
ld = development length in tension of
pretensioned strand
db = nominal diameter of prestress-
ing strand
fse = effective stress in prestressing
steel (after allowance for all prestress
losses)
fps = stress in prestressing steel at nom-
inal flexural strength of the member
The first term in Eq. (1) represents
the transfer length of the strand, and
the second term, the flexural bond
length. This expression is based on
test results for transfer length and flex-
ural bond length.2,3 The ACI equation
for ld first appeared in ACI 318-63 and
has remained unchanged through ACI
318-05.1,4
The mechanical bond of steel pre-
stressing strand to concrete is a function
of numerous factors listed by Comité
Euro-International du Béton (CEB).5
Analysis of the influence of these fac-
tors was the objective of many past and
recent studies.6–8 The impact of high-
strength concrete in the fabrication and
construction of prestressed concrete
and the development of new (higher
tensile strength, lower relaxation, and
larger diameter) steel strands, and
new materials for prestressing strands
(epoxy-coated steel strands and fiber-
reinforced polymers) have been the
subjects of many research programs.
Significant experimental results were
obtained by Cousins et al.,9 who ob-
served transfer and development lengths
of uncoated 0.5-in.-diameter (13 mm)
seven-wire strand much greater than
those calculated by ACI 318 equations.
In response to these test results, the Fed-
eral Highway Administration (FHWA)
issued a memorandum that required the
ACI development length expression
to be multiplied by a factor of 1.6 and
disallowed the use of 0.6-in.-diameter
(15 mm) strand in pretensioned concrete
for highways bridge applications.10
The FHWA memorandum generated
subsequent research to study bond of
pretensioned strand and to establish
design recommendations for transfer
and development lengths. Results from
these research projects indicated wide
variations in the measured transfer and
development lengths; a review of these
studies is presented by Buckner.11
76 PCI JOURNAL
Strand-concrete bond performance is
essential for pretensioned concrete ap-
plications. ACI 318-05 states that for
bond applications, quality assurance
procedures should be used to confirm
that the strand is capable of adequate
bond.1 Neither ASTM A416 nor ACI
318-05,1,12 however, include minimum
requirements for bond performance
of prestressing strands, and neither do
other standards.13 In spite of the large
number of experimental studies con-
ducted to date, there is no consensus on
a standard test method for bond qual-
ity.14 According to Rose and Russell,15
a standard test for the strand bond in
pretensioned concrete would benefit
all parties involved in the precast/pre-
stressed concrete industry: manufactur-
ers, producers, designers, builders, and
owners.
Objectives and Scope
of the Research
The purpose of this study is to de-
velop a test method for measuring the
transfer and development lengths of
prestressing strand. The most impor-
tant condition for the test was that the
strand release and the pull-out opera-
tions be performed sequentially on the
same specimen. In this way, the devel-
opment length is determined for speci-
mens that have been subject to prestress
transfer with the same procedures used
in the construction of precast, pre-
stressed concrete in practice. In deter-
mining the transfer and development
lengths of seven-wire strand, the study
included 12 different concretes in the
experimental program to validate the
test method.
Experimental
Determination
of Transfer and
Development Length
Various researchers use the follow-
ing two methods to experimentally de-
termine the transfer and development
lengths of prestressing strands:16
1. Measurement of the longitudinal
concrete strains in the surface of
a specimen.
2. Measurement of the strand-end
slip of a prestressed member.
These two test methods are the basis
of different standards to determine
the transfer length of prestressing
strands.17,18 These standards, however,
do not include the following procedures
to determine development length:
3. Breaking beams with concen-
trated loads at various distances
from the supports.
4. Pull-out tests.
These four test methods are dis-
cussed in this paper.
Measurement of Longitudinal
Concrete Strain
Electrical or mechanical strain
gauges are attached along the sides of
a pretensioned concrete member prior
to release of strand tension. When the
prestressing force is transferred, the
compressive strains in the concrete are
registered at various distances from the
ends of the member. Transfer length
is determined directly from the data
curve (concrete strains versus distance
from the end of a member) by the
slope-intercept method or the 95% av-
erage maximum strain (AMS) method
(Fig. 2).16,19,20 Some researchers apply
a smooth fit to the data curve,21 while
others use 100% of the average maxi-
mum strain.22
Measurement of Strand-End Slip
Based on Guyon’s theory, this meth-
od uses the amount of the strand-end
slip δ during the release of tension on
the casting bed.23 Guyon proposed an
expression where the transfer length
lt is linearly proportional to δ and in-
versely proportional to the initial strand
strain εsi.
δ
lt = α (2)
ε si
The coefficient α represents the
shape of the bond stress distribution
along the transfer length and is typi-
cally a value between 2 and 3. Several

Transfer Transfer
length length
95% Average sioned prestressing strand samples with
maximum strain
Specimen length
Fig. 2. Idealized relationship between concrete strain and specimen length.
researchers have proposed different the beam will exhibit bond failure with sig-
values for α based on experimental re- nificant strand slip. The development length
sults and theoretical studies.24 of the prestressing strand is determined as
the embedment length at which flexural
Beam Breakage by Concentrated crushing without bond failure occurs; this
Loading method has been used in other research.21,25
After strand release, a series of preten- Pull-Out Tests
sioned beams are tested in a three-point
or four-point flexural test. The variable Results of pull-out tests on specimens
in this beam series is the length from the with different embedment lengths can be
end of the beam to the point of maximum used to measure the tensile force reached
moment. If the embedment length tested by the prestressing strand in order to de-
is greater than the required development termine the development length. Mah-
length, the beam will reach the flexural moud et al. provide a concrete specimen
crushing load without exhibiting bond configuration for this method:22 a prism
failure. If the embedment length tested is specimen comprising two concrete parts
less than the required development length, with a concentric strand, each connected
by a hollow jack and a hollow load cell.
After release, force is gradually applied
with an internal hollow jack to separate
A the two concrete parts.
Strand
Test Methods to Study
Concrete
Basic Bond Properties
C Following are two testing techniques
often used to study the basic bond prop-
erties of prestressing strand:14
• Pull-out tests.
B • Push-in tests.
Pull-Out Tests
Fig. 3. The setup of plates A, B, and C
shows a push-in test based on Keuning Several experimental studies of un-
et al.31 tensioned prestressing strand have
September–October 2006 77
been conducted with this method.26,27
The influence of the embedment length
on the maximum and average values of
the bond stress are documented.14
The Moustafa pull-out test uses a
concrete block and a number of unten-
long embedment lengths.28 In this test,
the strands are individually pulled out.
This test has been used by researchers
to correlate the results of the pull-out
force of prestressing strand with the
results of transfer and development
lengths obtained in beams.6,15,25 Pull-
out tests with tensioned prestressing
strands were conducted to simulate an-
chorage bond behavior,29 and with the
exception of work done by Mahmoud
et al.,22 these tests do not reproduce the
transverse expansion (Hoyer Effect30)
of the prestressing strand.
Push-In Tests
The influence of prestressing strand
expansion can be analyzed by means of
a push-in test as follows (Fig. 3):
• A strand is pretensioned be-
tween two plates (A and B).
• A concrete specimen is cast
around the strand near the third
plate (C).
• The prestressing force is re-
leased at plate A, originating
movement of the strand toward
plate B and creating bond
stresses because the concrete is
fixed at plate C.
The push-in technique has been used
by researchers to simulate bond behav-
ior along the transfer length.29,32 For
greater control during the strand stress,
a testing frame with longer free strand
lengths was patterned by Rose and
Russell.15 A simulation of the transfer
length and the flexural bond length is
simultaneously performed in the test
proposed by Cousin et al.,33 wherein
a hydraulic actuator at the reaction
plate C moves the concrete specimen
toward plate A (Fig. 3). Strand stress
on the top side (toward plate A) of
the concrete specimen will decrease
(push in) and the strand stress on the
bottom side (at plate C) will increase
(pull out). Sophisticated measurement
procedures that do not disturb the bond
phenomenon (photoelasticity or the ul-

Strand
Adjustable strand
anchorage
Hydraulic
Frame
actuator
D A
Fig. 4. Design of anchorage-measurement-access (AMA) test equipment setup.
trasonic waves34,35) have not yet been
sufficiently developed.
Proposed Test
A test to measure the transfer and de-
velopment lengths of prestressing strand
has been developed.36 Figure 4 shows
the design of this test (called ECADA,
the Spanish acronym for “Ensayo para
Caracterizar la Adherencia mediante
Destesado y Arrancamiento,” which
translates to “Test to Characterize the
Bond by Release and Pull Out” in English).
The test equipment has been de-
signed to introduce some variations
from the Keuning test (Fig. 3):31
Fig. 5. Adjustable prestressing strand anchorage.
78 PCI JOURNAL
Embedment length
Concrete Sleeve
specimen
• The length between frame plates
A and C allows measurement
of the transfer and development
lengths of prestressing strand
in addition to providing data on
basic strand bond properties.
• Plates C and B, the two separa-
tors between these plates, and
the sleeve on the bottom end of
the strand specimen are de-
signed to simulate the sectional
stiffness of the specimen during
the transfer of the prestress-
ing force when the specimen
embedment length is equal to or
greater than the transfer length.
This test assembly is called the
anchorage measurement access
(AMA) system and is designed
AMA
system
Anchorage
Hollow load cell
Support separator
C B
to prevent the confining influ-
ence of plate C, measure the
force in the strand by means of
a hollow-load cell, and increase
the strand stress at plate B with a
pull-out operation.
Test Procedure
The test procedure for a specimen is
as follows:
A. Preparation stage
1. Strand is placed through the
holes between plates D and B.
2. Anchorage devices are put into
place, and a hydraulic actuator is
placed at plate A.
3. The strand is tensioned between
plates D and B using the hydrau-
lic actuator to pull the anchorage
devices separating plate D from
plate A.
4. Adjustable strand anchor-
age is engaged until it makes
contact with the plate D. The
initial force in the strand PI is
anchored. Then, the hydraulic
actuator is unloaded and stress is
relieved (Fig. 5).
5. Concrete is mixed and placed
into the prepared form in the
frame near plate C and consoli-
dated. After concrete placement,
the specimen is cured to achieve
the desired concrete properties
for testing.
B. Testing stage
B1. Prestress transfer phase
1. The hydraulic actuator is placed
at the frame plate A and is
Fig. 6. Pull-out phase of testing.
loaded to recover the force P0
(force in the strand just before
release) at the adjustable strand
anchorage at plate A.
2. The prestressing force P0 is
released at plate D by means of
the hydraulic actuator. At this
moment, the transfer of the pre-
stressing force to the concrete is
complete. The concrete speci-
men is supported at the reaction
frame plate C.
3. Stabilization period.
B2. Pull-out phase
1. The hydraulic actuator is posi-
tioned at the frame plate C.
2. Strand stress is increased using
the hydraulic actuator to pull
the anchorage device to separate
plate B from plate C (Fig. 6).
Using the same specimen, the
ECADA test sequentially reproduces
the transfer of the prestress (push-in)
and anchorage (pull-out) forces in the
strand in the same manner as in-service
pretensioned concrete members (re-
lease and overloading). In this way, the
Hoyer Effect is reproduced prior to the
September–October 2006 79
pull-out operation.
The force in the strand is recorded
during all test stages (tensioning, re-
lease, and pull-out) by a hollow-load
cell placed between plate B and the an-
chorage device at plate B. A pressure
sensor controls the force exerted by the
hydraulic actuator. The instrumentation
used in the test is simple and reusable.
To prevent interference with the bond
phenomenon, an internal measuring
device was not used in the specimens.
If the AMA system stiffness and the
sectional stiffness of the specimen are
equal, the value of the measured force
in the strand at end B (after release) is
a measure of the strand force inside
the concrete specimen. System com-
ponents are manipulated so the AMA
system stiffness matches the specimen
sectional stiffness. From the charac-
teristics of the AMA system and those
of the specimens, one may obtain the
theoretical estimation of loss of force
ΔPest registered in the strand during
the transfer of prestress for specimen
embedment lengths equal to or greater
than the transfer length.36
Interpretation of Specimen Test
Figure 7 shows the force in the strand
and the force exerted by the hydraulic
actuator versus time. The represented
force in the strand P0 corresponds to
the variation registered at the onset of
the testing. Two possible test results
are presented depending on the speci-
men embedment length:
• If the specimen embedment
length is equal to or greater than
the transfer length, after the
stabilization period, the force in
the strand at the AMA system
shows a drop ΔP equal to ΔPest
(Fig. 7). This drop is due to
compatibility of strains between
concrete and strand. The force in
the strand after the stabilization
period PT (prestressing force
transferred [PT = P0 - ΔP] to the
concrete) will be the effective
prestressing force after elastic
shortening loss (PE = P0 - ΔPest).
• If the specimen embedment length
is less than the transfer length, the
force in the strand at the AMA
 ƒP

ƒP

ƒPest

Fig. 7. Specimen test results show strand force and hydraulic actuator force versus time: a prestressed concrete specimen where
embedment length > transfer length and a specimen with embedment length < transfer length.

80 PCI JOURNAL
 system shows a drop ΔP greater
than ΔPest after the stabilization
period (Fig. 7). The prestressing
force transferred to the concrete
PT will be less than PE.
After the stabilization period, if the
transfer of prestress force is complete
(PT ≈ PE), the force in the strand is in-
creased by the pull-out operation. The
maximum force reached in the strand
PR will depend on the specimen em-
bedment length. Related to the nominal
development force PD, the following
conditions are noted:
• If the specimen embedment
length is less than the devel-
opment length, PD cannot be
reached in the strand (PR < PD).
• If the specimen embedment
length is equal to or greater than
the development length, PD is
reached in the strand during the
pull-out operation.
In addition to the definition of the
development length in ACI,1 Buck-
ner indicates that the stress fps must be
developed without strand-end slip.11
Thus, the development length will cor-
respond to the lesser embedment length
for which PD is reached without strand
slip at the free end of the specimen. In
light of this, a displacement transducer
is used as additional instrumentation at
the free end of the specimen during the
pull-out process.
Interpretation of a Complete Test
With the ECADA method, the trans-
fer and development lengths are deter-
mined by testing a series of specimens
with different embedment lengths. The
transfer length will be the lesser em-
bedment length of the test specimens
where the effective prestressing force
PE in the strand is reached after the
stabilization period. The development
length will correspond to the lesser em-
bedment length of the test specimens in
which the nominal development force
PD in the strand is reached in the pull-
out operation and in the absence of
strand slip at the free end of the speci-
men. The resulting determination of the
transfer and development lengths will
depend on the sequences of lengths of
the specimens tested.
Experimental Program
An experimental program to verify
the feasibility of the ECADA method
for determining the transfer and de-
velopment lengths of prestressing
strand was conducted. Twelve differ-
ent concretes were tested with a range
of water-cement ratios (w/c) of 0.3 to
0.5, cement quantities C of 590 lb/yd3
to 843 lb/yd3 (350 kg/m3 to 500 kg/m3),
and compressive strengths at time of
testing f ʹci ranging from 3481 psi to
7977 psi (24 MPa to 55 MPa).
The concrete components are CEM
I 52.5 R type cement,37 crushed lime-
stone aggregate, washed rolled lime-
stone sand, and ISOCRON FM-211
high-range water-reducing admixture.
Table 1 shows the different concretes
tested. The prestressing strand is a

Table 1. Test Program

Tested Embedment Lengths, in.
f ciʹ
Cement C (lb/yd3) w/c
(psi)
Series A Series B

0.30 7977 2.0, 5.9, 9.8, 13.8, 17.7, 21.7 3.9, 7.9, 11.8, 15.7, 19.7, 23.6
843 0.35 6817 13.8, 15.7, 17.7, 19.7, 21.7, 23.6
0.40 4496 13.8, 17.7, 21.7, 23.6, 25.6, 29.5 19.7, 23.6, 27.6, 31.5, 35.4, 39.4
0.35 6817 13.8, 17.7, 21.7, 25.6, 29.5, 33.5 15.7, 19.7, 23.6, 27.6, 31.5, 35.4
758
0.40 5221 13.8, 17.7, 21.7, 25.6, 29.5, 33.5 15.7, 19.7, 23.6, 27.6, 31.5, 35.4
0.35 6526 13.8, 17.7, 21.7, 25.6, 29.5, 33.5 15.7, 19.7, 23.6, 27.6, 31.5, 35.4
0.40 5511 9.8, 15.7, 21.7, 27.6, 33.5, 53.1 13.8, 15.7, 17.7, 19.7, 21.7, 23.6
674
0.45 4061 17.7, 19.7, 21.7, 23.6, 25.6, 27.6
0.50 3481 19.7, 23.6, 27.6, 31.5, 35.4, 39.4 21.7, 25.6, 29.5, 33.5, 37.4, 41.3
0.40 6672 17.7, 19.7, 21.7, 23.6, 25.6, 27.6
590 0.45 5366 17.7, 19.7, 21.7, 23.6, 25.6, 27.6
0.50 4061 13.8, 17.7, 21.7, 25.6, 29.5, 33.5 15.7, 19.7, 23.6, 27.6, 31.5, 35.4
Additional series
C: 13.8, 31.5, 37.4, 53.1
843 0.30 7977 D: 11.8, 13.8, 15.7, 17.7, 19.7, 21.7
E: 15.7, 15.7, 17.7, 17.7
F: 17.7, 17.7, 19.7, 39.4

Notes: 1 lb/yd3 = 0.5932 kg/m3; 1 psi = 6.895 kPa; 1 in. = 25.4 mm.

September–October 2006 81
low-relaxation, seven-wire strand as
specified in UNE 36094:97 Y 1860 S7
13.0,13 with diameter of 0.5 in. (13 mm)
and a guaranteed ultimate strength of
270 ksi (1860 MPa).
The concrete specimens are 4 in. × 4
in. (100 mm × 100 mm) in cross section
and are prestressed with a concentri-
cally located single strand at a prestress
level of 75% of the ultimate tensile
strength. All specimens were subjected
to the same concrete consolidation and
curing conditions. The release and pull-
out processes were gradual.
For each concrete type, the transfer
and development lengths are deter-
mined from a series of 6 or 12 speci-
mens. Table 1 lists the embedment
lengths that were necessary to deter-
mine the transfer and development
lengths in each concrete type. Addi-
tional tests on the C 843 (w/c = 0.3)
concrete had a double objective: to ver-
ify the operation of the designed AMA
system and to verify the applicability
and effectiveness of the test method.
For each specimen, the release and the effective prestressing force after
pull-out operations of the strand with elastic shortening loss PE. In contrast,
the ECADA test method were sequen- all the test specimens with embedment
tially performed. Strand release oc- length less than 15.7 in. (400 mm) pre-
curred 24 hours after concrete place- sented values of PT less than PE. The
ment, and the pull-out operation took test specimens with embedment lengths
place subsequent to the stabilization equal to 15.7 in. (400 mm) present data
period, typically two hours after re- that border on values for complete
lease. The nominal force to attain PD in transfer of prestress. Therefore, for the
the strand was established at 35.6 kip embedment lengths tested, it may be
(158 kN), corresponding to 0.1% of the concluded that the transfer length de-
nominal yield strength of strand.13 termined by the ECADA test method is
17.7 in. (450 mm).
Figure 9 shows the maximum force
reached in the strand without strand
Test Results and slip at the free end of the specimen dur-
Analysis ing the pull-out operation. Data in Fig.
9 show that all test specimens that at-
tained the effective prestressing force
Figure 8 shows the prestressing PE present values of PR greater than
force transferred to the concrete PT ver- PE. Specimens of embedment length
sus the embedment length for the C 843 equal to or greater than 23.6 in. (600
(w/c = 0.3) concrete. Test specimens mm) present values of force PR greater
with embedment lengths equal to or than the nominal force to develop PD
greater than 17.7 in. (450 mm) present (35.6 kip [158 kN]). Breakage occurred
identical values of PT, with PT equal to in the anchorage device without strand

45 200 45 200
Development length Development length
40 180 40 180

35 160 35 160

140 140
30 30
Force (kip)

Force (kip)

120 120
25 25
(kN)

(kN)
100 100
20 Transfer length 20
80 80
Transfer length
15 15
60 60
10 10
40 40
Cement 590 lb/yd 3 Cement 674 lb/yd3
5 w/c 0,45 20 5 w/c 0,35 20

0 0 0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
Embedment length (in.) Embedment length (in.)

45 200 45 200
Development length Development length
40 180 40 180

35 160 35 160

140 140
30 30
Force (kip)

Force (kip)

120 120
25 25
(kN)
(kN)

100 100
20 20
80 80
15 Transfer length 15 Transfer length
60 60
10 10
40 40
Cement 758 lb/yd3 Cement 843 lb/yd3
5 w/c 0,40 20 5 w/c 0,35 20

0 0 0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
Embedment length (in.) Embedment length (in.)

P0 PT PR

Fig.10. Test results of prestressing force versus embedment length for various concretes. Note: 1 in. = 25.4 mm.

82 PCI JOURNAL
slip at the free end of the specimen. 50 220
With the embedment lengths tested, it 45 200
may be concluded that the development

Force in the strand (kip)

40 180
length determined by the ECADA test 160
35
method is 23.6 in. (600 mm). 140
30
The recorded prestressing force 120
25

(kN)
transferred and the force developed 100
for the different embedment lengths 20
80
(Fig. 8 and 9) indicate a similar rela- 15
60
tionship to that of the idealized dia- 10 40
gram represented in Fig. 1 but with a 5 20
slight discontinuity at the point of slope 0 0
change. This result agrees with data re- 0 5 10 15 20 25 30 35 40 45 50 55
ported by Mahmoud et al.22 Embedment length (in.)
From the results obtained in the se- Series A
Series E
Series B
Series F
Series C
Average P0
Series D
Calculated value PE
ries of six specimens made with C 843
(w/c = 0.3), it is concluded that the test Fig. 8. Results of prestressing force transferred versus embedment length for concrete
is systematic and reliable. For the other C 843 (w/c = 0.3). Note: 1 in. = 25.4 mm.
concretes, a series of 6 or 12 specimens
with different embedment lengths were
tested under the same conditions as
concrete C 843 (w/c = 0.3). Results are
shown in Fig. 10.
Figures 11 and 12 show results for
the transfer and development lengths,
respectively, for all tested concretes
versus the compressive strength of
concrete at the time of testing. A de-
crease in the transfer and development
lengths with an increase of compres-
sive strength of concrete can be ob-
served. Results for lengths are less than
those of ACI 318-051 (where fse is ap-
plied to the corresponding strand stress
for the average force PE for all tested
concretes and where fps corresponds to
strand force PD). For transfer lengths, Fig. 9. Results of maximum prestressing strand force reached without free-end slip
test results are between 50% and 80% versus embedment length for concrete C 843 (w/c = 0.3). Note: 1 in. = 25.4 mm.
of those calculated by ACI 318-05
equations.1 For development length,
test results are between 45% and 65%
of those calculated by ACI 318-05
equations.1
For the materials used and the estab-
lished test parameters, good correlation
to experimental results is shown in the
following expressions:

lt = 29.85 - 0.0016 f ʹci (3)

ld = 35.08- 0.0014 f ʹci (4)

where:
lt = transfer length (in.)
ld = development length (in.)
f ʹci = compressive strength of
concrete at time of testing (psi)
The application feasibility of the Fig. 11. Transfer length versus concrete compressive strength. Note: 1 psi = 6.895 kPa.

September–October 2006 83

Fig. 12. Development length versus concrete compressive strength.
ECADA test method for the determi-
nation of the transfer and development
length of prestressing strands has been
verified. Moreover, studies on basic
bond properties of prestressing strand
can be made using this test, and the
test can be applied to any type of pre-
stressing tendons and wires. Therefore,
the proposed test is valid as a basis for
standardization of bond quality in the
pretensioning industry.
Conclusions
Based on the results of this research,
the following conclusions can be made:
• A valid test method for measur-
ing the transfer and development
lengths of prestressing strands is
proposed.
• Sequential testing operations on
the same specimen are repre-
sentative of procedures used
in actual precast/prestressed
construction.
• The test is systematic and reliable.
• The instrumentation used in the
test is simple and reusable. Inter-
nal devices that might distort the
bond phenomenon are not used.
• An increase in concrete com-
pressive strength at the time of
testing results in a decrease of
strand transfer and development
lengths.
• For concretes with compres-
sive strengths from 3481 psi to
84 PCI JOURNAL
7977 psi (24 MPa to 55 MPa) at
the time of testing, the transfer
length results are between 50%
and 80% of those calculated by
ACI 318-05 equations, and the
test development lengths are
between 45% and 65% of those
calculated by ACI 318-05 equa-
tions.
• Expressions to predict the
transfer and development lengths
from concrete compressive
strength (at time of testing and
for test conditions) are presented.
Recommendations
The proposed test method can be
standardized for different purposes:
• For quality control of prestress-
ing strand: Under standard
conditions, the transfer and
development lengths can be
determined. Standard conditions
must include concrete materials
specifications, concrete dosage,
and methodology for making
specimens.
• To evaluate the bond quality of
a prestressing strand in construc-
tion site conditions: The mate-
rial characteristics (prestressing
strand and concrete) must be
taken into consideration.
• As an acceptance-rejection
test of a prestressing strand in
specific working conditions:
The test would be limited to
checking whether the effective
prestressing force and/or the
nominal force to develop in the
strand are achieved.
• To determine the effects of
manufacturing procedures on
bond features of a prestressing
strand.
Acknowledgments
The contents of this paper are part of
the research programs being carried out
by the Concrete Technology and Sci-
ence Institute of the Polytechnic Uni-
versity of Valencia. The authors grate-
fully acknowledge the collaboration of
the companies PREVALESA and ISO-
CRON. Financial support provided by
IMPIVA (Project IMTEIE/2000/85),
the Ministry of Education and Culture
(Project MAT2000-0346-P4-03), the
Ministry of Science and Technology,
and FEDER funds (Project MAT2003-
07157) made it possible to conduct this
research. The authors also express their
gratitude to the technical staff of the
Concrete Structures Laboratory at the
Polytechnic University of Valencia for
their assistance in the preparation and
testing of the specimens.
References
1. ACI Committee 318. 2005. Building
Code Requirements for Structural Con-
crete (ACI 318-05) and Commentary
(ACI 318R-05). Farmington Hills, MI:
ACI.
2. Kaar, P. H., R. W. LaFraugh, and
M. A. Mass. 1963. Influence of Con-
crete Strength on Strand Transfer
Length. PCI Journal, Vol. 8, No. 5
(September–October): pp. 47–67.
3. Hanson, N., and P. Kaar. 1959. Flexural
Bond Test of Pretensioned Prestressed
Beams. ACI Journal, Vol. 30, No. 7
(January): pp. 783–802.
4. ACI Committee 318. 1963. Building
Code Requirements for Reinforced
Concrete (ACI 318-63). Detroit, MI:
ACI.
5. Comité Euro-International du Béton
(CEB). 1987. Anchorage Zones of Pre-
 stressed Concrete Members. State-of-
the-Art Report. Bulletin No. 181. Swit-
zerland: CEB.
6. Shing, P. B., D. E. Cooke, D. M. Fran-
gopol, M. A. Leonard, M. L. McMul-
len, and W. Hutter. 2000. Strand De-
velopment and Transfer Length Tests
on High Performance Concrete Box
Girders. PCI Journal, Vol. 45, No. 5
(September–October): pp. 96–109.
7. Steinberg, E., J. T. Beier, and S. Sar-
gand. 2001. Effects of Sudden Pre-
stress Force Transfer in Pretensioned
Concrete Beams. PCI Journal, Vol. 46,
No. 1 (January–February): pp. 64–75.
8. Barnes, R. W., J. W. Grove, and N. H.
Burns. 2003. Experimental Assessment
of Factors Affecting Transfer Length.
ACI Structural Journal, Vol. 100, No. 6
(November–December): pp. 740–748.
9. Cousins, T. E., D. W. Johnston, and P.
Zia. 1990. Transfer and Development
Length of Epoxy-coated and Uncoated
Prestressing Strand. PCI Journal, Vol.
35, No. 4 (July–August): pp. 92–103.
10. Chief, Bridge Division, Federal High-
way Administration (FHWA). 1998.
Prestressing Strand for Pretension Ap-
plications—Development Length Re-
visited. Memorandum (October 26).
Washington, DC: FHWA.
11. Buckner, C. D. 1995. A Review of
Strand Development Length for Pre-
tensioned Concrete Members. PCI
Journal, Vol. 40, No. 2 (March–April):
pp. 84–105.
12. ASTM Subcommittee A01.05. Standard
Specification for Steel Strand, Uncoated
Seven-Wire for Prestressed Concrete. A
416/A 416M-99. West Conshohocken,
PA: ASTM International.
13. Asociación Española de Normalización
y Certificación (AENOR). 1997. Alam-
bres y Cordones de Acero para Ar-
maduras de Hormigón Pretensado. [In
Spanish.] UNE 36094. Madrid, Spain:
AENOR.
14. International Federation for Structural
Concrete (fib). 2000. Bond of Rein-
forcement in Concrete. State-of-Art Re-
port. Bulletin No. 10. Switzerland: fib.
15. Rose, D. R., and B. W. Russell. 1997.
Investigation of Standardized Tests to
Measure the Bond Performance of Pre-
stressing Strand. PCI Journal, Vol. 42,
No. 4 (July–August): pp. 56–80.
September–October 2006 85
16. Thorsen, N., 1956. Use of Large Ten-
dons in Pretensioned Concrete. ACI
Journal (February): pp. 649–659.
17. Réunion Internationale des Labo-
ratoires et Experts des Matériaux
(RILEM). 1979. Specification for the
Test to Determine the Bond Properties
of Prestressing Tendons. RPC6. Sys-
tèmes de Constructions et Ouvrages.
Bagneux, France: RILEM.
18. Fédération Internationale de la Précon-
trainte (FIP). 1982. Test for the Determi-
nation of Tendon Transmission Length
under Static Conditions. Report on Pre-
stressing Steel: 7. London, U.K.: FIP.
19. Deatherage, J. H., E. Burdette, and C.
K. Chew. 1994. Development Length
and Lateral Spacing Requirements of
Prestressing Strand for Prestressed
Concrete Bridge Girders. PCI Journal,
Vol. 39, No. 1 (January–February): pp.
70–83.
20. Russell, B. W., and N. H. Burns. 1996.
Measured Transfer Lengths of 0.5 and
0.6 in. Strands in Pretensioned Con-
crete. PCI Journal, Vol. 41, No. 5
(September–October): pp. 44–65.
21. Lu, Z., T. E. Boothby, C. E. Bakis, and
A. Nanni. 2000. Transfer and Devel-
opment Lengths of FRP Prestressing
Tendons. PCI Journal, Vol. 45, No. 2
(March–April): pp. 84–95.
22. Mahmoud, Z. I., S. H. Rizkalla, and E.
R. Zaghloul. 1999. Transfer and Devel-
opment Lengths of Carbon Fiber Rein-
forcement Polymers Prestressing Rein-
forcing. ACI Structural Journal, Vol.
96, No. 4 (July–August): pp. 594–602.
23. Guyon, Y. 1953. Béton Précontrainte.
Étude Théorique et Experiméntale. Ed.
Eyrolles. Paris, France: 711 pp.
24. Balázs, G. 1993. Transfer Length of
Prestressing Strand as a Function of
Draw-in and Initial Prestress. PCI
Journal, Vol. 38, No. 2 (March–April):
pp. 86–93.
25. Logan, D. R. 1997. Acceptance Cri-
teria for Bond Quality of Strand for
Pretensioned Prestressed Concrete Ap-
plications. PCI Journal, Vol. 42, No. 2
(March–April): pp. 52–90.
26. Salmons, J. R., and T. E. McCrate. 1977.
Bond Characteristics of Untensioned Pre-
stressing Strand. PCI Journal, Vol. 22,
No. 1 (January–February): pp. 52–65.
27. Brearley, L. M., and D. W. Johnston.
1990. Pull-out Bond Tests of Epoxy-
coated Prestressing Strand. Journal of
Structural Engineering, Vol. 116, No.
8 (August): pp. 2236–2252.
28. Moustafa, S. 1974. Pull-out Strength
of Strand and Lifting Loops. Technical
Bulletin 74-B5. Concrete Technology
Corp.
29. Abrishami, H. H., and D. Mitchell.
1993. Bond Characteristics of Preten-
sioned Strand. ACI Materials Journal,
Vol. 90, No. 3 (May–June): pp. 228–
235.
30. Hoyer, E., and E. Friedrich. 1939. Be-
itrag zur Frage der Haftspannung in
Eisenbetonbauteilen. Beton und Eisen,
Vol. 50, No. 9: pp. 717–736.
31. Keuning, R. W., M. A. Sozen, and C.
P. Siess. A Study of Anchorage Bond
in Prestressed Concrete. No. 251.
University of Illinois Structural Re-
search Series.
32. Vandewalle, L., and F. Mortelmans.
1994. Anchorage of Strands. Proceed-
ings of the Fédération Internationale
de la Précontrainte (FIP) XII Interna-
tional Congress. Washington, DC: pp.
J16–J22.
33. Cousins, T. E., M. Badeaux, and S.
Moustafa. 1992. Proposed Test for
Determining Bond Characteristics of
Prestressing Strand. PCI Journal, Vol.
37, No. 1 (January–February): pp. 66–
73.
34. Linger, D. A., and S. R. Bhonsle. 1963.
An Investigation of Transfer Length
in Pretensioned Concrete Using Pho-
toelasticity. PCI Journal, Vol. 8, No.
4 (July–August): pp. 13–30.
35. Chen, H., and K. Wissawapaisal.
2001. Measurement of Tensile Forces
in a Seven-Wire Prestressing Strand
Using Stress Waves. Journal of Engi-
neers Mechanics (June): pp. 599–606.
36. Martí, J. R. 2003. Experimental Study
on Bond of Prestressing Strand in
High-Strength Concrete. [In Span-
ish.] Doctoral thesis, Polytechnic Uni-
versity of Valencia. Microfilm UMI
3041710. Ann Arbor, MI: ProQuest
Information and Learning Co.
37. AENOR. 2000. Cemento. Parte 1:
Composición, especificaciones y crite-
rios de conformidad de los cementos
comunes. [In Spanish.] UNE EN 197-
1:2000. Madrid, Spain: AENOR.