ACI STRUCTURAL JOURNAL
Title No. 115-S53
Shear Strength of Deep Hollow-Core Slabs
by Raed Tawadrous and George Morcous
Hollow-core (HC) slabs are the most commonly used and
economical precast/prestressed concrete flooring system. HC
slabs produced in the U.S. market typically have untopped depth
ranging from 6 to 12 in. (152 to 305 mm). Recently, deeper HC
slabs (for example, 16 in. [406 mm]) have been produced to satisfy
the growing need for longer spans and/or heavier loads. ACI
318-14, Section 7.6.3.1, requires a minimum shear reinforcement
to be provided where ultimate shear force is greater than 50% of
the factored concrete web shear strength (ϕVcw) for untopped HC
slabs deeper than 12.5 in. (318 mm). The 50% modification factor
was introduced in 2008 based on a limited testing conducted by HC
suppliers on deep HC slabs. This paper briefly summarizes param-
eters that affect concrete web shear strength of deep HC slabs and
presents the results of shear and flexure testing of 16 in. (406 mm)
HC slabs. Ten different shear strength provisions adopted by ACI
318-14, AASHTO LRFD 2014, CSA A23.3-04, JSCE 2007, fib MC
2010, AS 3600-2009, EN 1168, and Yang’s method were compared
using a database of 51 web shear tests (12 conducted by the authors
and 39 obtained from the literature) of deep HC slabs. Comparison
results indicated that the 10 shear strength provisions vary signifi-
cantly with respect to the accuracy and consistency of their predic-
tions and, therefore, different modification factors need to be used.
Keywords: code comparison; hollow-core slab; shear testing; web shear
strength.
INTRODUCTION
Hollow-core (HC) slabs date back to 1930s when the
German Schaefer and Kuen started producing compo-
nents similar to what is called a “hollow-core slab” today
(ASSAP 2002). In the late 1940s, production of HC slabs
started in Europe with an insulated structural slab made of
a HC pumice concrete enclosed within two layers of normal
reinforced concrete. Prestressed concrete HC slabs were
developed in the United States in the 1950s and became the
mainstream precast/prestressed concrete flooring system
due to technology advancement in the industry and ease of
the production process (Hawkins and Ghosh 2006). Two
methods are the most common in HC production. First, the
extrusion method, which forces a very-low-slump concrete
into an extrusion machine that forms the cores with augers
or tubes. Second, the slip-form method, using higher-slump
concrete and side forms attached to the casting machine
where cores are typically made by pneumatic tubes.
The American Concrete Institute (ACI) shear provisions
for prestressed concrete were developed by testing primarily
prestressed concrete beam sections (that is, deep sections).
The applicability of these provisions to prestressed HC slabs
without shear reinforcement was investigated by Anderson
(1978) and Becker and Buettner (1985) to address concerns
about shear strength of very-low-slump concrete used in
extruded HC slabs (that is, shallow sections). Anderson
ACI Structural Journal/May 2018
TECHNICAL PAPER
(1978) performed experimental testing on HC slabs with
untopped depth ranging from 6 to 14.5 in. (152 to 368 mm)
and shear span-depth ratio (a/d) ranging from 2.67 to 4.5,
while Becker and Buettner (1985) performed experimental
testing on 8 and 10 in. (203 and 254 mm) deep HC slabs with
an a/d ranging from 3.75 to 5.6. Both researchers concluded
that the actual shear strength of these slabs was in excess
of the strength predicted by ACI shear provisions. However,
shear design provisions for HC slabs remained the same in
the ACI design code.
As of the 1980s, deeper HC slabs started to be produced
based on the market needs in Europe, which led the European
researchers to investigate behavior of HC slabs that are 14
to 20 in. (356 to 508 mm) deep without shear reinforcement.
The European researchers concluded that the traditional
design method in Eurocode 2 (EC 2) overestimates the web
shear strength of HC slabs with depths greater than 12 in.
(305 mm) (Pajari 2005). New requirements for shear design
were adopted by ACI 318-08 for HC slab depth greater than
12.5 in. (318 mm) based on published work by Hawkins and
Ghosh (2006). They reported that few of the tested HC slabs
produced and tested by one supplier failed in web shear at
53% of the load predicted by ACI 318-05. The lower shear
strength of deep HC slabs is in agreement with the findings
of the European researchers (Yang 1994; Pajari 2005). More
information about these tests is presented in Appendix A.*
Therefore, since 2008, ACI 318 required the concrete web
shear strength (ϕVcw) to be multiplied by a 50% modification
factor for untopped HC slabs deeper than 12.5 in. (318 mm)
when minimum shear reinforcement is eliminated, which is
the common practice due to the difficulty of placing shear
reinforcement during HC slab production. Web shear provi-
sions of ACI 318-14, Section 22.5.8.3.2 will be referred to
throughout this paper as it is the current design code of HC
slabs in the United States.
RESEARCH SIGNIFICANCE
Precast/prestressed concrete HC slabs are the most
popular precast product for floorings due to their efficient
production, light weight, high stiffness, speed of construc-
*
The Appendix is available at www.concrete.org/publications in PDF format,
appended to the online version of the published paper. It is also available in hard copy
from ACI headquarters for a fee equal to the cost of reproduction plus handling at the
time of the request.
ACI Structural Journal, V. 115, No. 3, May 2018.
MS No. S-2017-035.R1, doi: 10.14359/51701298, was received July 15, 2017, and
reviewed under Institute publication policies. Copyright © 2018, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
699
tion, and economy. Deep HC slabs have gained a significant
portion of the flooring market in Europe and Middle East
and have attracted the attention of designers and producers
in the United States. However, the presence of modification
factor that reduces web shear strength of deep HC slabs
significantly impact their use in longer spans and heavier
loads. Therefore, this paper summarizes the different param-
eters that affect web shear strength of deep HC slabs and
present the experimental investigation conducted to evaluate
the shear and flexural capacity of 16 in. (406 mm) HC slabs.
Also, 10 different shear strength provisions adopted by ACI
318-14, AASHTO LRFD 2014, CSA A23.3-04, JSCE 2007,
fib MC 2010, AS 3600-2009, EN 1168, and Yang’s method
(Yang 1994) were reviewed and evaluated to determine the
accuracy and consistency of their predictions using a data-
base of 51 tests conducted by the authors and others on
deep HC slabs. The outcome of this review and comparison
is expected to help building designers and HC producers
to better predict the shear strength of deep HC slabs, and
consequently, promote their use in building construction.
PARAMETERS AFFECTING WEB SHEAR
STRENGTH OF HC SLABS
Many parameters affect concrete web shear strength of HC
slabs. Most of these parameters were addressed by several
researchers (Kani 1967; Walraven and Mercx 1983; Nilson
1987; Jonnson 1988; Bažant and Kazemi 1991; Shahawy et
al. 1992; Yang 1994; MacGregor 1997; Collins and Kuchma
1999; Angelakos et al. 2001; Pajari 2005; Bentz 2005;
Palmer and Schultz 2010, 2011). These parameters include:
• Load and support configuration;
• Shear span-depth ratio (a/d);
• Prestressing level;
• Concrete compressive strength;
• Geometry of cross section; and
• Overall unit depth.
Load and support configuration
Jonnson (1988) investigated the effect of load/support
configuration on concrete web shear capacity of HC slabs.
Jonnson concluded that having partial bearing width
supporting system for HC slabs can reduce its shear capacity
by up to 50% of the predicted capacity. This agrees with the
results of testing Specimen #1A conducted by the authors,
which will be presented in the “Experimental investigation”
section of this paper.
Shear span-depth ratio (a/d)
Concrete web shear strength is affected by a/d, where a
is the shear span and d is the member’s depth. This effect is
attributed to the arching action that takes place when the load
is closer to the support (Jonnson 1988; MacGregor 1997;
and Palmer and Schultz 2010). Jonnson (1988) studied the
effect of a/d ranging from 1.0 to 7.0 for 10.5 in. (267 mm)
HC slabs and concluded that the shear strength in case of
a/d = 1.0 can be twice the shear strength in case of a/d =
7.0. Other studies showed that a/d greater than 2.0 would
mitigate the arching effect (Hawkins et al. 2005; Palmer
700
and Schultz 2010), which was adopted in the experimental
investigation presented in the next section.
Prestressing level
Axial compressive stresses due to prestressing are recog-
nized by most shear design provisions in various codes.
Prestressing enhances the web shear-carrying capacity
because it tends to reduce the principal tensile stress, causing
an increase in the web shear strength as the web shear failure
occurs when the principle tensile stress reaches its limiting
concrete tensile strength. However, the critical section for
design is traditionally taken at h/2 from the face of the
support, which means that the section is more likely to be
within the transfer length of the prestressing strands espe-
cially for shallow members such as HC slabs. A generally
accepted model for the variation of the stress in prestressing
strand is a straight line from zero stress at the point where
bonding commences to the effective prestress at the end
of the transfer length. Numerous research has been done
on determining transfer length of prestressing strands as it
affects shear strength of prestressed members (Russell and
Burns 1993; Mitchell et al. 1993). Current design codes
propose different formulas for estimating transfer length (lt)
of seven-wire strands as a function of strand diameter (db):
ACI 318-14, Section 22.5.9 proposes a transfer length of
50db, while AASHTO LRFD (2014), Section 5.11.4 proposes
a transfer length of 60db. However, many other factors affect
the transfer length in pretensioned member, such as strand
surface condition, concrete compressive strength at release,
type of concrete, center-to-center spacing between strands,
concrete cover (Lim et al. 2013), section shape, strand
distribution, and de-tensioning method. Therefore, all these
factors can have indirect effect on the concrete web shear
strength of HC slabs.
Concrete compressive strength
Concrete shear strength is directly related to its tensile
strength as diagonal shear cracks start to take place when
principle tensile stresses reach the tensile strength. Specified
28-day concrete compressive strength fc′ is generally used in
evaluating the tensile strength due to the difficulty associated
with conducting direct tension tests and disparity of the test
results comparing to compression test results. Therefore, in
most design codes, the shear strength of a concrete member
is directly proportioned to (fc′)x, where the power x differs
from code to code to reflect the concrete tensile strength and
it ranges from 1/3 to 1/2.
Geometry of cross section
The Australian standard (AS 3600 2009) does not specify
a certain equation for shear design of prestressed members.
Instead, it requires the web shear strength to be calculated
based on the principal tensile strength at either the centroidal
axis, in the case of HC with circular voids, or the intersec-
tion of bottom flange and web, in case of HC with non-cir-
cular voids (NPCAA 2003). Pajari (2005) reported that
when using Yang’s method for calculating shear strength of
HC slabs, the location of the critical point is the centroidal
axis for slabs with circular voids. However, for slabs with
ACI Structural Journal/May 2018
non-circular voids, the critical section is located at the
junction of the web and the flange, which confirms that the
geometry of the cross section plays a role in determining the
shear strength of HC slabs.
Overall unit depth
Palmer and Schultz (2010) studied the effect of HC slab
depth on its shear strength using test data obtained from five
different experimental programs for a total of 198 slabs with
30 slabs whose depth was greater than 12.5 in. (318 mm).
This study could not clearly indicate the depth effect on
shear strength due to the disparity in the test data. Shioya et
al. (1989) tested concrete members with depth ranging from
4 to 18 in. (102 to 457 mm) without transverse reinforce-
ment and subjected to a uniformly distributed load. Shioya
et al. (1989) concluded that shear strength decreases as the
depth of the member increases.
EXPERIMENTAL INVESTIGATION
An experimental program was conducted to investigate
the web-shear and flexure-shear strength of 16 in. (406 mm)
HC slabs. A total of eight full-scale HC slab specimens were
fabricated in Lincoln, NE: four slabs, 16 ft (4.877 m) long
Fig. 1—Cross-section dimensions and properties of 16 in.
(406 mm) HC slabs. (Note: 1 in. = 25.4 mm.)
Fig. 2—Test setup and instrumentation. (Note: 1 ft = 3.28 m; 1 in. = 25.4 mm.)
ACI Structural Journal/May 2018
each, were tested in web-shear strength at both ends (the
A end was tested first and the B end was tested second for
a total of eight tests) using three-point loading setup; and
four slabs, 24 ft (7.315 m) long each, were tested for flex-
ure-shear strength using a four-point loading setup. Figure 1
shows the cross-section dimensions and geometric proper-
ties of the tested HC slabs. The specified concrete compres-
sive strength of the HC slabs was 9 ksi (62 MPa); however,
the average compressive strength at the time of testing was
found to be approximately 10 ksi (69 MPa). All slabs were
reinforced using seven 0.5 in. (12.7 mm) diameter Grade
270 (1860 MPa) low-relaxation seven-wire strands pre-ten-
sioned to 70% of the ultimate strength.
Test setup and instrumentation
The web-shear testing was conducted using one concen-
trated load applied at a shear span of 3 ft 8 in. (1.118 m)
(a/d = 2.75), as shown in Fig. 2(a). The flexure-shear testing
was conducted using two concentrated loads applied at 3 ft
(0.914 m) from the midspan point (a/d = 6.5), as shown in
Fig. 2(b). All loads were applied across the width of the spec-
imen using steel and wood beams and 400 kip (1780  kN)
hydraulic jack supported on a steel frame anchored to the
strong floor. Loads were measured using 450 kip (2000 kN)
calibrated load cell and deflections were measured using
string potentiometer with 0.001 in. (0.025 mm) accuracy.
Strand slippage was measured using two linear variable
differential transformers (LVDTs) during web-shear testing
for two interior strands at the location shown in Fig. 2(a).
Concrete strain gauges were used to measure the strain at
the top and bottom fibers of the slab during flexure-shear
testing. Two steel rollers that are 39 in. (990 mm) long were
initially used as bearing for the first web-shear test (#1A),
which did not support the exterior webs of the hollow core
slabs resulting in premature failure. Two 4 x 4 in. (102 x
701
102 mm) dimension lumber that are 4 ft (1.192 m) long were
used as bearings afterwards to support the full width of the
specimen, as shown in Fig. 3. All supports were located at
4 in. (102 mm) from the ends of tested HC slab based on
the recommendations of the precast producer as it is their
common practice in HC construction.
Web-shear test results
Figure 4 shows the shear-deflection relationship of the
eight shear tests (two tests: A and B, for each end of the four
HC specimens) conducted using three-point loading. These
relationships indicate that all specimens had almost linear
Fig. 3—Web-shear test specimen supported by 4 ft (1.219 m)
long bearing.
Fig. 4—Shear-deflection relationships. (Note: 1 in. =
25.4 mm; 1 lb = 4.448 N.)
Fig. 5—Local failure due to inadequate bearing length of Specimen #1A.
702
behavior with similar slope up to approximately 0.2  in.
(5.1  mm) deflection followed by nonlinear behavior with
much lower slope up to failure. Exceptions are: Specimen
#1A that failed prematurely due to using 39 in. (991  mm)
long bearings, causing horizontal cracks between the exte-
rior webs and the top flange, as shown in Fig. 5; and Speci-
mens #2B and #4B that experienced significant strand slip-
page due to the excessive cracking occurred while testing
the other ends (#2A and #4A, respectively), as shown in
Fig. 6. All specimens except #1A, #2B, and #4B experi-
enced typical diagonal web-shear cracks in areas adjacent
to the support, as shown in Fig. 7. The failure in these spec-
imens was sudden without visible cracking prior to failure.
Table 1 lists the ultimate load, ultimate shear, corresponding
deflection at load location, and corresponding strand slip-
page at failure. Also, the ratio of the measured shear force to
predicted web-shear force at the critical section (using provi-
sions of ACI-318-14, Section 22.5.8.3.2 without either resis-
tance reduction factor or modification factor) is presented.
The mean, standard deviation (STD) and coefficient of vari-
ation (COV) values were calculated with and without Spec-
imens #1A, #2B, and #4B, which did not experience typical
web-shear failure. These values indicate that consistency of
web-shear strength of deep HC specimens (COV = 5.5%)
when Specimens #1A, #2B, and #4B are excluded. It also
shows that measured-to-predicted web shear strength ratio is
0.76, which is significantly higher than the 0.5 modification
factor currently adopted by ACI 318.
Flexure-shear test results
Figure 8 shows the load-deflection relationships of the
four 24 ft (7.315 m) long HC specimens tested in flexure
using four-point loading. These relationships indicate that
the four specimens behaved linearly with approximately
the same slope up to the cracking load (averaged 54.5 kip
[242.4 kN] and 0.5 in. [12.7 mm] deflection), then nonlin-
early up to the ultimate load (averaged 82.3 kip [366 kN])
and 9.5 in. [241 mm] deflection). Figure 8 shows that the
measured cracking load of all HC specimens was higher
than predicted using ACI 318-14, Section 24.2.3.9 and
assuming a modulus of rupture of 7.5√fc and total prestress
loss of 15%. It also shows that the measured ultimate load
of all HC specimens was higher than predicted using strain
ACI Structural Journal/May 2018
compatibility and assuming a resistance reduction factor of
1.0. Table 2 lists the results of testing the four HC speci-
mens as well as the ratios of measured to predicted cracking
moment, ultimate moment, and flexure-shear strength. All
the ratios are slightly higher than 1.0, indicating the satisfac-
tory performance of all HC specimens based on ACI 318-14,
Section 22.2 predictions. Moreover, the low values of STD
and COV indicate the consistency of test results. All speci-
mens had vertical flexural cracks starting from the bottom
fibers at the loading points, as shown in Fig. 9, after reaching
the cracking load. As the load increased, these cracks prop-
agated upward to form inclined flexural-shear cracks, as
shown in Fig. 10.

Fig. 6—Flexure failure of Specimens #2B and #4B.
ANALYTICAL INVESTIGATION
Appendix B lists 10 shear design methods/code provisions
adopted by seven different international design codes: ACI
318, AASHTO LRFD, CSA A23.3, JSCE, fib Model Code,
AS 3600, and European Standard EN 1168. The equations
used in each method/provision, as well as a short description
of the governing parameters, are presented in Appendix B.
A database of 51 deep HC slab shear tests conducted by the
authors and others (Pajari 2005; Hawkins and Ghosh 2006;
Palmer and Schultz 2011) were obtained from the literature
and used to compare prediction accuracy and consistency
of the 10 shear design methods/provisions. The database
includes 43 HC slabs that are 16 in. (406 mm) and eight HC
slabs that are 20 in. (508 mm). These tests were conducted

Fig. 7—Web-shear failure cracks.

Table 1—Web-shear test results of 16 ft (4.877 m) long HC specimens

Specimen Ultimate load, lb Ultimate shear, kip Vtest/Vcw Corresponding deflection, in. Corresponding slippage, in.
#1A* 70.0 48.2 0.58 0.142 0.0035
#1B 86.0 58.8 0.71 0.310 0.0124
#2A 97.7 66.6 0.80 0.529 0.0893
#2B *
64.1 44.2 0.53 0.164 N/A
#3A 96.2 65.7 0.79 0.425 0.0565
#3B 96.5 65.8 0.79 0.563 0.0900
#4A 89.1 60.9 0.73 0.282 0.0070
#4B *
71.7 49.3 0.59 0.698 0.2186
Mean (all) 83.9 57.4 0.69
STD (all) 13.4 9.0 0.11
COV (all) 16.0% 15.6% 15.6%
Mean (excl. ) *
93.1 63.6 0.76
STD (excl. ) *
5.2 3.5 0.04
COV (excl.*) 5.6% 5.5% 5.5%
*
These specimens did not have web-shear cracking failure at h/2.
Notes: 1 lb = 4.448 N; 1 kip = 4.448 kN; 1 in. = 25.4 mm.

ACI Structural Journal/May 2018 703

Table 2—Flexure test results of 24 ft (7.315 m) long HC specimens
Specimen Cracking Cracking Cracking Ultimate Ultimate Ultimate Corresponding Corresponding compres-
No. load, kip moment, kip.ft Mtest/Mcr load, kip moment, kip.ft Mtest/Mn Vtest/Vci deflection, in. sive strain, × 10–6
#1 56.0 265.0 1.14 82.1 378.2 1.13 1.09 9.86 2288
#2 53.0 252.0 1.08 79.4 366.2 1.09 1.06 8.67 1954
#3 55.0 260.6 1.12 82.7 380.8 1.14 1.10 10.02 1327
#4 54.0 256.3 1.10 85.0 390.9 1.17 1.13 9.56 1869
Mean 54.5 258.5 1.11 82.3 379.0 1.13 1.1
STD 1.3 5.6 0.02 2.3 10.1 0.03 0.03
COV 2.4% 2.2% 2.2% 2.8% 2.7% 2.7% 2.6%

Notes: 1 kip = 4.448 kN; 1 kip.ft = 1.3557 kN.m; 1 in. = 25.4 mm.

Fig. 9—Flexure cracks under cracking load.

Fig. 8—Load-deflection relationships of 24 ft (7.315 m) long

HC specimens. (Note: 1 in. = 25.4 mm; 1 lb = 4.448 N.)
by researchers at the University of Nebraska–Lincoln
(UNL) (Tawadrous and Morcous 2017); researchers at Tech-
nical Research Center of Finland (VTT) (Pajari 2005), two
U.S. suppliers (A and B) (Hawkins and Ghosh 2006), and
researchers at the University of Minnesota (UMN) (Palmer
and Schultz 2011) for HC produced by the same U.S.
suppliers (A and B).
The HC slabs tested by VTT, UMN, and US Supplier B
were manufactured using the extrusion method, while the
HC slabs tested by UNL and US Supplier A were manu-
factured using the slip-form method. Appendix A shows
detailed cross-section dimensions of the tested HC slabs and
lists their geometric properties. Cross-section dimensions
and geometric properties of the HC slabs tested by UNL
were shown earlier in Fig. 1. In all the 51 shear tests, diag-
onal web-shear failure was observed. Flexure-shear strength
of these HC slabs was not critical as it did not control the
design of tested HC slabs. In addition, the flexure-shear test
results presented in Table 2 were above the values predicted
by ACI 318-14, Section 22.5.8.3.1.
Table 3 lists the measured web-shear strength Vtest of the
51 tests, source of test data, HC depth, average concrete
Fig. 10—Flexure-shear cracks under ultimate load.
compressive strength, and measured-to-predicted ratio
(Vtest/Vpred) using the 10 shear prediction methods/provi-
sions listed in Appendix B. Also, the mean, STD, and COV
values of Vtest/Vpred are listed for each method/provision to
evaluate their relative accuracy and consistency. Figure 11
presents these results using a box-whiskers plot that shows
the 25th and 75th percentiles (lower and upper box borders
respectively), median (cross line), mean (x), below 25th
and above 75th percentiles (whiskers), and outliers (dots)
for each prediction method/provision. This plot uses all the
51 data points, regardless of the data source and HC slab
depth. However, Fig. 12(a) to (j) present the ratio Vtest/Vpred
versus Vtest for each of the 10 shear design methods/provi-
sions grouped by data source and HC slab depth. The values/

704 ACI Structural Journal/May 2018

Table 3—Comparisons of web-shear strength predictions using different provisions/methods
Vtest/VPred
Specimen h, ACI- ACI- AASHTO- AASHTO- AS EN
Source/Supplier No. fc, ksi in. Vtest, kip Detailed Approximate General Simplified CSA JSCE fib 3600 1168 Yang
1B 10.00 16 58.8 0.69 0.59 1.01 1.18 0.77 0.98 0.69 0.96 0.59 0.96
2A 10.00 16 66.6 0.78 0.67 1.15 1.33 0.87 1.10 0.78 1.09 0.67 1.08
Tawadrous and
3A 10.00 16 65.7 0.77 0.66 1.13 1.31 0.86 1.09 0.77 1.08 0.66 1.07
Morcous 2017
3B 10.00 16 65.8 0.77 0.67 1.13 1.32 0.86 1.09 0.77 1.08 0.66 1.07
4A 10.00 16 60.9 0.72 0.62 1.05 1.22 0.79 1.01 0.72 1.00 0.62 0.99
416-A1 9.75 16 88.8 1.22 1.11 1.38 1.98 1.41 1.46 1.27 1.11 0.92 1.08
416-A2 9.75 16 99.8 1.37 1.25 1.55 2.22 1.58 1.65 1.43 1.24 1.04 1.21
416-B1 9.30 16 77.6 1.11 0.98 1.64 1.81 1.22 1.27 1.10 1.01 0.87 1.00
416-B2 9.30 16 88.0 1.26 1.11 1.86 2.05 1.38 1.44 1.25 1.15 0.99 1.14
414-A 10.30 16 65.5 0.92 0.82 1.02 1.48 1.07 1.01 0.97 0.85 0.71 0.69
413 8.00 16 65.8 1.09 0.92 1.19 1.84 1.14 1.24 1.03 1.01 0.86 0.88
Pajari 2005
422-A1 10.26 20 81.7 0.79 0.72 0.95 1.26 1.00 0.99 0.90 0.69 0.59 0.68
422-A2 10.26 20 93.3 0.90 0.82 1.09 1.44 1.14 1.13 1.03 0.79 0.68 0.77
422-B 11.00 20 105.9 0.98 0.88 1.18 1.57 1.28 1.24 1.16 0.87 0.75 0.82
422-C1 10.80 20 125.3 1.08 1.05 1.40 1.64 1.51 1.34 1.36 0.94 0.82 0.80
422-C2 10.80 20 115.7 1.00 0.97 1.29 1.52 1.39 1.24 1.26 0.86 0.76 0.74
422-C3 10.80 20 110.5 0.96 0.92 1.23 1.45 1.33 1.18 1.20 0.82 0.72 0.70
1A 6.00 16 49.0 0.65 0.29 1.76 1.12 1.44 0.91 0.59 0.75 0.57 0.67
1B 6.00 16 43.6 0.58 0.26 1.56 1.00 1.28 0.81 0.53 0.67 0.51 0.60
Hawkins and 2A 6.00 16 47.3 0.63 0.28 1.70 1.08 1.39 0.88 0.57 0.72 0.55 0.65
Ghosh 2006
(Supplier A) 2B 6.00 16 46.4 0.61 0.28 1.66 1.06 1.36 0.86 0.56 0.71 0.54 0.64
3A 6.00 16 42.6 0.56 0.26 1.53 0.97 1.25 0.79 0.52 0.65 0.50 0.59
3B 6.00 16 52.5 0.70 0.31 1.88 1.20 1.54 0.98 0.64 0.80 0.61 0.72
1 10.83 16 61.7 0.81 0.54 0.91 1.30 0.95 0.97 0.86 0.70 0.62 0.74
2 9.31 16 68.8 0.95 0.61 1.09 1.52 1.09 1.14 0.99 0.82 0.72 0.86
3 8.76 16 56.8 0.79 0.50 0.90 1.25 0.90 0.94 0.81 0.67 0.60 0.71
Hawkins and 4 8.93 16 65.7 0.91 0.58 1.04 1.45 1.04 1.09 0.94 0.78 0.69 0.82
Ghosh 2006
(Supplier B) 5 10.30 16 79.5 1.04 0.70 1.17 1.67 1.23 1.25 1.11 0.91 0.79 0.95
6 10.77 16 84.0 1.10 0.74 1.23 1.77 1.30 1.32 1.17 0.96 0.84 1.01
7 9.73 16 70.2 0.92 0.62 1.03 1.48 1.09 1.10 0.98 0.80 0.70 0.84
8 9.99 16 64.2 0.84 0.56 0.94 1.35 0.99 1.01 0.90 0.73 0.64 0.77

ranges of the governing parameters in these data groups can • The measured values of concrete compressive strength
be described as follows: are used in prediction;
• Average concrete compressive strength fc ranges from 6 • Corresponding code provisions are used for calculating
to 11 ksi (42 to 76 MPa); different parameters, such as material limits, location
• HC slab depth h ranges from 16 to 20 in. (406 to of critical section, and transfer length of prestressing
508 mm); strands;
• Prestressing strand area to cross-sectional area ratio ρps • The 50% modification factor for Vcw considered in ACI
ranges from 0.4% to 0.72%; 318-14 calculations is 1.0; and
• a/d ranges from 2.4 to 3.2; and • Total prestressing loss is 15%.
• Jacking stress-to-ultimate strength ratio fpi/fpu ranges
from 0.6 to 0.7. DISCUSSION
The predicted shear strength value, Vpred, was calculated Comparing the plots presents in Fig. 12(a) to (j) indicates
using the following assumptions: that AASHTO LRFD simplified and general procedures
• Strength reduction factor is 1.0; underestimated Vcw of deep HC slabs (Vtest/Vpred > 2.0 in

ACI Structural Journal/May 2018 705

Table 3 (cont.)—Comparisons of web-shear strength predictions using different provisions/methods
Vtest/VPred
Specimen h, ACI- ACI- AASHTO- AASHTO- AS EN
Source/Supplier No. fc, ksi in. Vtest, kip Detailed Approximate General Simplified CSA JSCE fib 3600 1168 Yang
1 9.97 16 79.0 1.06 0.67 1.74 1.76 1.21 1.30 1.06 0.95 0.83 1.14
2 9.97 16 104.6 1.41 0.88 3.79 2.35 3.14 1.72 1.41 1.26 1.10 1.53
3 9.97 16 121.0 1.29 1.03 1.75 2.01 1.81 1.14 1.64 1.12 1.13 1.76
4 9.97 16 98.0 1.26 0.82 2.73 2.09 2.02 1.60 1.30 1.11 0.98 1.43
5 9.97 16 134.4 1.31 0.86 2.71 2.16 1.99 1.65 1.36 0.97 1.00 1.44
Palmer and 6 9.97 16 137.0 1.28 0.84 2.76 2.11 2.05 1.66 1.33 1.02 0.99 1.41
Schultz 2011
(Supplier B) 7 9.97 16 68.8 0.89 0.61 1.04 1.41 1.08 1.07 0.97 0.70 0.67 0.87
8 9.97 16 72.8 0.90 0.63 1.07 1.42 1.11 1.11 1.00 0.77 0.68 0.91
9 9.97 20 100.2 0.80 0.53 0.97 1.29 1.01 1.14 0.91 0.56 0.63 0.89
10 9.97 20 113.7 0.92 0.60 1.11 1.47 1.15 1.30 1.04 0.63 0.72 1.01
11 9.97 16 72.9 0.89 0.62 1.05 1.40 1.09 1.07 0.98 0.75 0.67 0.92
12 9.97 16 91.6 1.14 0.80 1.36 1.79 1.41 1.36 1.28 0.96 0.86 1.17
13 7.82 16 62.2 0.74 0.38 1.40 1.29 0.96 1.07 0.67 0.79 0.59 0.84
14 7.82 16 77.3 0.87 0.45 2.44 1.52 1.92 1.28 0.79 0.97 0.70 0.97
15 7.82 16 69.7 0.82 0.43 1.94 1.44 1.44 1.20 0.76 0.92 0.67 0.91
Palmer and 16 7.82 16 74.5 0.88 0.46 2.33 1.54 1.80 1.28 0.81 0.94 0.71 1.01
Schultz 2011
(Supplier A) 17 7.82 16 75.4 0.83 0.49 1.19 1.33 0.95 1.20 0.86 0.77 0.65 0.88
18 7.82 16 84.2 0.93 0.54 1.75 1.49 1.07 1.34 0.95 0.86 0.68 0.99
19 7.82 16 128.3 1.06 0.84 1.63 1.56 1.62 1.23 1.47 1.15 0.87 1.26
20 7.82 16 82.6 0.92 0.54 1.67 1.47 1.05 1.34 0.95 0.86 0.67 0.98
Mean 0.94 0.67 1.49 1.52 1.30 1.19 0.99 0.89 0.74 0.95
STD 0.21 0.24 0.58 0.33 0.42 0.22 0.27 0.17 0.15 0.25
COV 0.23 0.36 0.39 0.22 0.32 0.19 0.27 0.19 0.21 0.26

Notes: fc is average concrete compressive strength; h is total HC slab depth; Vtest is experimentally observed failure shear at critical point; Vpred is predicted web-shear strength; STD
is standard deviation; COV is coefficient of variation; ACI is American Concrete Institute; AASHTO is American Association of State Highway and Transportation Officials; CSA
is Canadian Standards Association; JSCE is Japan Society of Civil Engineers; fib is Fédération Internationale du Béton; AS is Australian Standard; EN 1168 is European Standard;
Yang is method developed by Yang; 1 ksi = 6.89 MPa; 1 in. = 25.4 mm; 1 kip = 4.448 kN.

codes provided significantly different predictions as the mean

Fig. 11—Box-whiskers plot of ratio Vtest/Vpred of different
shear provisions.
some data points), while ACI approximate procedure over-
estimated Vcw of deep HC slabs (Vtest/Vpred < 0.3 in some data
points). Though AASHTO LRFD 2014 general procedure,
CSA A23.3-04, and fib MC 2010 provisions are based on the
Modified Compression Field Theory (MCFT), these three
value of Vtest/Vpred was 1.49, 1.3, and 0.99, respectively. This
is primarily due to differences in material strength limits and
location of the longitudinal strain specified in each provi-
sion. CSA A23.3-04 and fib MC 2010 methods are simplified
non-iterative methods; however, AASHTO LRFD general
procedure is an iterative method. The fib MC 2010 method
provided the most accurate results when it was compared to
other code provisions with a mean Vtest/Vpred, STD, and COV
value of 0.99, 0.27, and 0.27, respectively. However, these
values for AASHTO LRFD and CSA are 1.49, 0.58, 0.39
and 1.3, 0.41, 0.32, respectively.
Even though, AASHTO LRFD 2014 simplified procedure
is similar to ACI 318-14 detailed web-shear provisions, both
codes provided very different predictions. The mean Vtest/
Vpred value using AASHTO LRFD and ACI was 1.52 and
0.67, respectively. This difference is believed to be due to
the concrete tensile strength used in each provision, where
AASHTO LRFD uses 1.9√fc, however, the ACI uses 3.5√fc.
JSCE 2007 yields the least scattered predictions (COV =
0.19) while being slightly conservative (mean = 1.19). One

706 ACI Structural Journal/May 2018

Fig. 12—Comparison of HC web-shear strength prediction methods/provisions. (Note: 1 kip = 4.448 kN.)

ACI Structural Journal/May 2018 707


Fig. 13—Measured web-shear stress versus concrete
compressive strength. (Note: 1 ksi = 6.9 MPa.)
Fig. 14—Histogram of measured-to-predicted shear stress
using ACI 318-14 Detailed Method for 51 tested deep HC
specimens. (Note: 1 ksi = 6.9 MPa.)
the other hand, AS 3600-09 provides the same consistency
(COV = 0.19) while overestimating the web-shear strength
by 11%. Although the AS 3600 web-shear provision uses the
same concept as the ACI code, there is a tangible difference
between the predicted values. This is attributed to the loca-
tion of the critical section at which the web-shear strength to
be calculated. The method implemented in the AS 3600 code
requires calculating web shear strength by equating the prin-
cipal tensile stress developed at the critical section with the
concrete tensile strength, which is more complicated than
other methods especially for preliminary design purpose.
EC 2 provisions overestimate the web-shear strength by
an average of 26% when compared to the experimental
results, while Yang’s method is considered the second most
accurate method (mean Vtest/Vpred = 0.95), after fib MC 2010
(mean Vtest/Vpred = 0.99). This method requires more calcula-
tions at different locations than other methods, as shown in
Appendix B.
Figure 13 plots the web-shear stress values of the 51 spec-
imens calculated by dividing the ultimate shear force by the
web width times section depth versus the concrete compres-
sive strength. This plot indicates that there is a positive
correlation between the web shear strength of deep HC slabs
and concrete compressive strength with coefficient of deter-
mination greater than 0.5 (R2 = 0.59), which confirms the
708
dependence of shear strength on the concrete compressive
strength. Figure 14 shows the frequency distribution (that is,
histogram) of the ratio of measured-to-predicted shear stress
using the ACI 318-14 detailed method without any modi-
fication factors. This plot indicates that there is need for a
modification factor as the ratio is less than 1.0 for more than
50% of the data. However, it also indicates that a modifica-
tion factor of 0.5 is conservative. Using the 95%-fractile, a
modification factor of 62% is recommended to ensure the
adequacy of the prediction method for most cases (only the
gray bar in Fig. 14 will be overestimated).
CONCLUSIONS
Based on experimental and analytical evaluation presented
in the paper on the web-shear strength and flexure strength
of 16 and 20 in. (406 and 508 mm) HC slabs, the following
conclusions can be made:
1. fib MC 2010 provisions provide the closest ratio of
Vtest/Vpred to 1.0 for web-shear strength of deep HC slabs,
while the AASHTO LRFD simplified and general proce-
dures provide the highest ratios of Vtest/Vpred (1.52 and 1.49,
respectively), which means the most conservative predic-
tions. On the other hand, the ACI approximate procedure and
EC2 provide the least conservative predictions of web-shear
strength of deep HC slabs (mean Vtest/Vpred = 0.67 and 0.74,
respectively) when modification and resistance reduction
factors are taken as 1.0.
2. Although Yang’s method provides accurate predictions
of web-shear strength of HC slabs (mean Vtest/Vpred = 0.95),
it involves intensive calculations when compared to other
methods.
3. JSCE and AS 3600 web-shear provisions provide
consistent predictions of web-shear strength of deep HC
slabs as they resulted in the least COV for Vtest/Vpred (19%).
JSCE predictions are more conservative than AS 3600
predictions as their mean Vtest/Vpred is 1.19, compared to 0.89
for AS 3600. On the other hand, AASHO LRFD general
procedure, ACI approximate procedure, and CSA provide
the most scatter predictions as their COV for Vtest/Vpred were
39%, 36%, and 32%, respectively.
4. The length of bearing supports of HC slabs is a crucial
factor to its shear strength. A continuous support across the
entire width of HC slabs is necessary to achieve the full
strength. Although the shear strength of Specimen #1A was
adversely affected by using a short bearing, test results do
give engineers some guidance on shear strength at real situa-
tions with partially missing plank bearing, such as HC plank
notching around columns on steel structures.
5. The measured flexural strength of the 16 in. (406 mm)
HC slabs is 13% higher than that predicted using strain
compatibility (with strength reduction factor of 1.0) and the
measured cracking moment is 11% higher than that predicted
using gross section properties, prestress losses of 15%, and
modulus of rupture of 7.5√fc′.
6. Concrete compressive strength has a significant
effect on the web-shear strength of deep HC slabs due to
the absence of shear reinforcement. However, it is not the
only governing parameter as a/d, level of prestressing, and
cross-section geometry also affect the web-shear strength.
ACI Structural Journal/May 2018
 7. The current ACI 318-14 web shear modification factor
adopted in 2008 was based on a small set of test data (14).
Using a database of 51 web shear test specimens indicated
that the 50% modification factor can be increased to 62%
using the 95% fractile rule.
AUTHOR BIOS
ACI member Raed Tawadrous is a PhD Candidate at the University of
Nebraska–Lincoln, Lincoln, NE. He received his BS in civil engineering
from Minia University, El-Minia, Egypt, in 2011, and his MS in structural
engineering from Cairo University, Cairo, Egypt, in 2014. His research
interests include analysis and design of precast/prestressed concrete
structures.
ACI member George Morcous is a Professor at the University of Nebraska–
Lincoln since 2005. His research interests include design and testing of
reinforced and prestressed concrete buildings and bridges.
ACKNOWLEDGMENTS
The authors would like to thank Concrete Industries, Inc., for the finan-
cial support of this research. Special thanks to Mark Lafferty for his valu-
able input to this work.
REFERENCES
AASHTO, 2014, “LRFD Bridge Design Specifications,” seventh
edition, American Association of State Highway and Transportation Offi-
cials, Washington, DC, 2014.
ACI Committee 318, 2005, “Building Code Requirements for Structural
Concrete (ACI 318R-05) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 430 pp.
ACI Committee 318, 2008, “Building Code Requirements for Structural
Concrete (ACI 318R-08) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 473 pp.
ACI Committee 318, 2014, “Building Code Requirements for Structural
Concrete (ACI 318R-14) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 520 pp.
Anderson, A. R., 1978, “Shear Strength of Hollow-Core Members,”
Technical Bulletin 78-81, Concrete Technology Associates, Tacoma, WA,
33 pp.
Angelakos, D.; Bentz, E. C.; and Collins, M. P., 2001, “Effect of Concrete
Strength and Minimum Stirrups on Shear Strength of Large Members,” ACI
Structural Journal, V. 98, No. 3, May-June, pp. 290-300.
AS 3600, 2009, “Concrete Structures,” Standards Australia International
Ltd., Sydney, Australia, 198 pp.
ASSAP, 2000, “The Hollow Core Floor Design and Applications,” Associa-
tion of Manufacturers of Prestressed Hollow Core Floors, Verona, Italy, 236 pp.
Bažant, Z. P., and Kazemi, M. T., 1991, “Size Effect on Diagonal Shear
Failure of Beams without Stirrups,” ACI Structural Journal, V. 88, No. 3,
May-June, pp. 268-276.
Becker, R. J., and Buettner, D. R., 1985, “Shear Tests of Extruded
Hollow-Core Units,” PCI Journal, V. 30, No. 2, Mar.-Apr., pp. 40-54. doi:
10.15554/pcij.03011985.40.54
Bentz, E. C., 2005, “Empirical Modeling of Reinforced Concrete
Shear Strength Size Effect for Members without Stirrups,” ACI Structural
Journal, V. 102, No. 2, Mar.-Apr., pp. 232-241.
Collins, M. P., and Kuchma, D., 1999, “How Safe Are Our Large, Lightly
Reinforced Concrete Beams, Slabs, and Footings?” ACI Structural Journal,
V. 96, No. 4, July-Aug., pp. 482-490.
CSA A23.3-04, 2004, “Design of Concrete Structures,” Canadian Stan-
dards Association, Mississauga, ON, Canada, 258 pp.
ACI Structural Journal/May 2018
EN 1168, 2005, “Precast Concrete Products. Hollow Core Slabs,” Euro-
pean Committee for Standardization, Brussels, Belgium, 84 pp. fib, 2010,
“Model Code 2010—Final Draft,” Fédération Internationale du Béton,
Lausanne, Switzerland.
EN 1992-1-1, 2004, “Design of Concrete Structures – Part 1-1: General
Rules and Rules for Buildings (Eurocode 2),” European Committee for
Standardization, Brussels, Belgium, 227 pp.
Hawkins, N. M., and Ghosh, S. K., 2006, “Shear Strength of Hollow-
Core Slabs,” PCI Journal, V. 51, No. 1, Jan.-Feb., pp. 110-115.
Hawkins, N. M.; Kuchma, D. A.; Mast, R. F.; Marsh, M. L.; and Reineck,
K.-H., 2005, “Simplified Shear Design of Structural Concrete Members,”
National Cooperative Highway Research Program (NCHRP) Report 549,
Transportation Research Board, Washington, DC, 338 pp.
Jonnson, E., 1988, “Shear Capacity of Prestressed Extruded Hollow-
Core Slabs,” Nordic Concrete Research, V. 7, pp. 167-187.
JSCE, 2010, “Standard Specifications for Concrete Structures – 2007:
Design,” Japan Society of Civil Engineers, Tokyo, Japan, 469 pp.
Kani, G. N. J., 1967, “How Safe Are Our Large Concrete Beams?” ACI
Journal Proceedings, V. 64, No. 3, Mar., pp. 128-141.
Lim, S. N.; Choi, Y. C.; Oh, B. H.; Kim, J. S.; Shin, S.; and Lee, M. K.,
2013, “Bond Characteristics and Transfer Length of Prestressing Strand
in Pretensioned Concrete Structures,” 8th International Conference on
Fracture Mechanics of Concrete and Concrete Structures, Toledo, Spain,
Mar. 10-14, 8 pp.
MacGregor, J. G., 1997, Reinforced Concrete, Mechanics and Design,
third edition, Prentice Hall, Upper Saddle River, NJ, 939 pp.
Mitchell, D.; Cook, W. D.; Khan, A. A.; and Tham, T., 1993, “Influ-
ence of High Strength Concrete on Transfer and Development Length
of Pre-Tensioning Strand,” PCI Journal, V. 38, No. 3, pp. 52-66. doi:
10.15554/pcij.05011993.52.66
Nilson, A. H., 1987, Design of Prestressed Concrete, second edition,
John Wiley and Sons, New York, NY, 608 pp.
NPCAA, 2003, “Hollow-Core Flooring Technical Manual,” National
Precast Concrete Association Australia, Glenelg, Australia, 35 pp.
Pajari, M., 2005, “Resistance of Prestressed Hollow-Core Slabs against
Web Shear Failure,” Research Notes 2292, VTT Building and Transport,
Kemistintie, Finland, 62 pp.
Palmer, K. D., and Schultz, A. E., 2010, “Factors Affecting Web-Shear
Capacity of Deep Hollow-Core Units,” PCI Journal, V. 55, No. 2,
pp. 123-146. doi: 10.15554/pcij.03012010.123.146
Palmer, K. D., and Schultz, A. E., 2011, “Experimental Investigation of
the Web-Shear Strength of Deep Hollow-Core Units,” PCI Journal, V. 56,
No. 4, pp. 83-104. doi: 10.15554/pcij.09012011.83.104
Russell, B. W., and Burns, N. H., 1993, “Design Guidelines for Transfer,
Development and Debonding of Large Diameter Seven Wire Strands in
Pretensioned Concrete Girders,” Rep. No. FHWA/TX-93+1210-5F, Texas
Department of Transportation, Austin, TX, 300 pp.
Shahawy, M. A.; Issa, M.; and Batchelor, B. V., 1992, “Strand Transfer
Lengths in Full Scale AASHTO Prestressed Concrete Girders,” PCI Journal,
V. 37, No. 3, May-June, pp. 84-96. doi: 10.15554/pcij.05011992.84.96
Shioya, T.; Iguro, M.; Nojiri, Y.; Akiyama, H.; and Okada, T., 1989,
“Shear Strength of Large Reinforced Concrete Beams,” Fracture
Mechanics: Application to Concrete, SP-118, American Concrete Institute,
Farmington Hills, MI, pp. 259-280.
Tawadrous, R., and Morcous, G., 2017, “Shear Strength of 16 in. Thick
Hollow-Core Slabs,” PCI Convention and National Bridge Conference,
Cleveland, OH, 13 pp.
Walraven, J. C., and Mercx, W. P., 1983, “The Bearing Capacity of
Prestressed Hollow-Core Slabs,” Heron, V. 28, No. 3, pp. 1-46.
Yang, L., 1994, “Design of Prestressed Hollow-Core Slabs with Reference
to Web Shear Failure,” Journal of Structural Engineering, ASCE, V. 120,
No. 9, pp. 2675-2696. doi: 10.1061/(ASCE)0733-9445(1994)120:9(2675)
709
ARE YOU A RESEARCHER?
SIGN UP FOR TODAY!
ORCID provides a persistent digital identifer that distinguishes you
from every other researcher and, through integration in key research
workflows such as manuscript and grant submission, supports
automated linkages between you and your professional activities,
ensuring that your work is recognized.
Individuals may use ORCID services freely and it’s as easy as 1-2-3:

1 REGISTER

2 ADD YOUR INFO

3 USE YOUR ORCID ID

For more information and to register, visit:

WWW.ORCID.ORG
1 APPENDIX A

2 CROSS SECTION DIMENSIONS AND GEOEMTRIC PROPERTIES

3 OF TESTED HC SLABS

5 Fig. A-1– Cross-section dimensions of 16 in. (400 mm) HC slabs type 3 (Pajari, 2005).

6 (Note: 1 in. = 25.4 mm)

8 Fig. A-2– Cross-section dimensions of 20 in. (500 mm) HC slabs type 5 (Pajari, 2005).

9 (Note: 1 in. = 25.4 mm)

1
1

2 Fig. A-3– Cross-section dimensions of 16 in. (400 mm) HC slabs Supplier “A” (Hawkins

3 and Ghosh, 2006; and Palmer and Schultz, 2011). (Note: 1 in. = 25.4 mm)

6 Fig. A-4– Cross-section dimensions of 16 in. (400 mm) HC slabs Supplier “B” (Hawkins

7 and Ghosh, 2006; and Palmer and Schultz, 2011). (Note: 1 in. = 25.4 mm)

2
1

2 Fig. A-5– Cross-section dimensions of 16 in. (400 mm) Heavy-HC slabs Supplier “B”

3 (Palmer and Schultz, 2011). (Note: 1 in. = 25.4 mm)

6 Fig. A-6– Cross-section dimensions of 20 in. (500 mm) HC slabs Supplier “B” (Palmer

7 and Schultz, 2011). (Note: 1 in. = 25.4 mm)

3
1 Table A-1: Geometric properties of HC specimens
Specimen  bw.dp  APS  ACP 
Source  h (in.)  dp (in.)  bw (in.)  A (in.2)  Ybot (in.) Q (in3.)  Ig (in.4)  SCP (in.3) ZCP (in.) 
No.  (in.2)  (in.2)  (in.2) 

1B  16  14.25  13.88  197.72  2.13  349  7.96  907  10941  1782  224  4.17 
2A  16  14.25  13.88  197.72  2.13  349  7.96  907  10941  1782  224  4.17 
Tawadrous and 
3A  16  14.25  13.88  197.72  2.13  349  7.96  907  10941  1782  224  4.17 
Morcous, 2017 
3B  16  14.25  13.88  197.72  2.13  349  7.96  907  10941  1782  224  4.17 
4A  16  14.25  13.88  197.72  2.13  349  7.96  907  10941  1782  224  4.17 
416‐A1  16  14.13  11.46  161.93  1.99  340  7.87  882  10803  1623  214  3.82 
416‐A2  16  14.13  11.46  161.93  1.99  340  7.87  882  10803  1623  214  3.82 
416‐B1  16  14.33  11.46  164.22  1.68  340  7.87  882  10803  1623  214  3.82 
Paraj, 2005 
416‐B2  16  14.33  11.46  164.22  1.68  340  7.87  882  10803  1623  214  3.82 
414‐A  16  14.10  11.10  156.51  1.99  340  7.99  903  10803  1294  197  2.22 
413  16  14.20  11.22  159.32  1.38  340  7.99  903  10803  1294  197  1.99 
422‐A1  20  17.54  12.80  224.51  2.45  448  10.12  1420  21466  2839  298  5.38 
422‐A2  20  17.54  12.80  224.51  2.45  448  10.12  1420  21466  2839  298  5.38 
422‐B  20  17.84  12.80  228.35  2.45  448  10.12  1420  21466  2839  298  5.38 
Paraj, 2005 
422‐C1  20  18.00  12.80  230.40  3.21  448  10.12  1420  21466  2839  298  5.38 
422‐C2  20  18.00  12.80  230.40  3.21  448  10.12  1420  21466  2839  298  5.38 
422‐C3  20  18.00  12.80  230.40  3.21  448  10.12  1420  21466  2839  298  5.38 
1A  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 
1B  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 
Hawkins and Ghosh,  2A  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 
2006 (Supplier A)   2B  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 
3A  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 
3B  16  15.00  15.90  238.50  0.92  397  8.19  1010  11767  1662  247  3.38 

4
1 Table A-1 (con.): Geometric properties of HC specimens
Specimen  bw.dp  APS  ACP 
Source  h (in.)  dp (in.)  bw (in.)  A (in.2)  Ybot (in.) Q (in3.)  Ig (in.4)  SCP (in.3) ZCP (in.) 
No.  (in.2)  (in.2)  (in.2) 

1  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
2  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
3  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
Hawkins and  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
4 
Ghosh, 2006 
(Supplier B)   5  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
6  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
7  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
8  16  13.90  11.72  162.91  1.79  334  8.34  900  11055  1625  213  3.15 
1  16  14.00  12.11  169.54  1.48  343  8.14  926  11270  1837  230  4.22 
2  16  14.00  12.09  169.26  1.48  348  8.12  941  11452  1882  234  4.29 
3  16  14.00  12.03  168.42  1.48  343  8.13  926  11275  1844  230  4.27 
4  16  14.00  12.23  171.22  1.48  348  8.22  937  11386  1913  238  4.42 
5*  16  13.44  16.62  223.37  2.13  470  8.25  1208  14258  2297  308  3.80 
palmer and Schultz,  6*  16  13.44  17.28  232.24  2.13  463  8.25  1161  13517  2168  299  3.69 
2011 (Supplier B)    7  16  13.57  11.79  159.99  2.13  359  7.81  976  11826  1625  214  3.14 
8  16  13.57  12.08  163.93  2.13  341  7.62  920  11150  1510  200  3.10 
9  20  17.67  15.24  269.29  2.13  493  9.94  1538  23143  3075  329  5.32 
10  20  17.67  15.19  268.41  2.13  494  9.91  1542  23223  3040  327  5.22 
11  16  13.70  12.25  167.83  2.13  347  8.20  932  11329  1848  231  4.27 
12  16  13.70  11.88  162.76  2.13  336  8.17  907  11052  1782  224  4.17 

5
1 Table A-1 (cont.): Geometric properties of HC specimens
Specimen  bw.dp  APS  ACP 
Source  h (in.)  dp (in.)  bw (in.)  A (in.2)  Ybot (in.) Q (in3.)  Ig (in.4)  SCP (in.3) ZCP (in.) 
No.  (in.2)  (in.2)  (in.2) 

13  16  14.31  16.13  230.82  0.92  398  7.78  981  11790  1727  259  3.07 
14  16  14.43  17.00  245.31  0.92  414  7.78  1014  12153  1731  259  3.06 
15  16  14.50  15.91  230.70  0.92  395  7.78  976  11738  1662  247  3.05 
Palmer and Schultz,  16  16  14.50  15.94  231.13  0.92  398  7.83  985  11841  1765  264  3.09 
2011 (Supplier A)   17  16  14.33  15.38  220.40  1.22  389  7.84  973  11643  1719  255  3.11 
18  16  14.33  15.44  221.26  1.22  392  7.78  918  11729  1745  259  3.06 
19  16  14.34  15.25  218.69  2.14  391  7.91  915  11751  1761  258  3.20 
20  16  14.34  15.19  217.82  1.22  386  7.86  900  11572  1740  257  3.15 
2 * Specimens with 16 in. Heavy HC slabs (refer to Figure 5 in App. I for cross section)
3
4 Note: h = HC slab depth; dp = distance from extreme compression fiber to centroid of prestressing steel; bw = total web width; Aps = area of prestressing strand; A = total area
5 of concrete; Ybot = distance from centroid to the bottom of slab; Q = first area moment of section about centroid; Ig = gross moment of inertia about section centroid; Scp =
6 first area moment of section about critical-point axis (Yang’s method); Acp = area of section above the critical point (Yang’s method); and zcp = vertical distance from
7 centroid to critical point (Yang’s method). 1 in. = 25.4 mm.

6
1 APPENDIX B

2 SHEAR DESIGN METHODS / CODE PROVISIONS

3 1. ACI 318-14

4 The shear design procedures in ACI 318-14 for prestressed members specify two methods

5 to calculate Vc for prestressed concrete. The two methods are often called: the approximate

6 method and the detailed method. The approximate method employs a simple empirical

7 equation (Cumming at al. 1998; and Eisuke et al. 2013). This method is applicable to

8 members with effective prestress of at least 40% of the tensile strength of flexural

9 reinforcement. Vc is given in ACI 318-14 Table 22.5.8.2

10
 V 
Vc   0.6 f c '  700 u d p bw d
 M u 11 (psi, in.)

2 f c 'bw d  Vc  5 f c 'bw d
12

13
Vu
d p  1 .0
Mu 14

15 The detailed method accounts for two different types of inclined cracks that may appear

16 on prestressed concrete members: web shear cracks (Vcw), and flexural shear cracks (Vci). On

17 the basis of this assumption, the detailed method takes Vc as the smaller of the two shear

18 strengths values at the formation of web shear cracks and flexural shear cracks. The shear

19 strength at the formation of the flexural shear crack is a sum of the shear that causes an initial

20 flexural crack and an additional increase of the shear required to turn this initial flexural

21 crack into a flexural shear crack. On the other hand, the web shear cracking equation is a

22 simplified formula of the principal stresses where the maximum principal tensile stress is

23 equal to the tensile strength of concrete 4√ (fc’). Details for Vcw derivation can be found in

24 ACI 318-63.

7
 Vi M cre
1 Vci  0.6 f c' bw d p  Vd  ACI 318-14 Eq. 22.5.8.3.1a
M max

2 Vcw  (3.5 f c'  0.3 f pc )bw d p  V p ACI 318-14 Eq. 22.5.8.3.2

3 Where,

4 dp = effective depth, defined as the distance from the extreme compression fiber to the

5 centroid of the prestressing steel ≥ 0.8 h

6 λ = lightweight concrete modification factor

7 f’c = concrete compressive strength

8 bw = web width

9 Vd = shear force at section due to unfactored dead load

10 Vp = vertical component of effective prestress force at section

11 Vi = factored shear force at section due to externally applied loads occurring

12 simultaneously with Mmax

13 Vu =total factored shear force

14 fpc = compressive stress in concrete at centroid of cross section resisting externally applied

15 loads after allowance for all prestress losses

16 Mcre = moment causing flexural cracking at section due to externally applied loads

17 Mmax = maximum factored moment at section due to externally applied loads

18 Mu = Total factored moment

19

20 2. AASHTO LRFD 2014

21 Two methods are presented in AASHTO LRFD 2014 for calculating shear strength of

22 prestressed members without transverse shear reinforcement: the general procedure and the

23 simplified procedure. The general procedure adopts design expressions based on the

24 Modified Compression Field Theory (MCFT), by Collins et al., 1996. In this method, Vc is a

8
1 function of a factor that indicate the ability of diagonally cracked concrete to transmit tension

2 and shear, β. The coefficient β is a function of the longitudinal strain in the flexure

3 reinforcement of the member, aggregate size, member depth, and distance between layers of

4 longitudinal crack control reinforcement. Hollow-core slabs are not explicitly mentioned in

5 AASHTO LRFD specifications, but an equivalent type of slabs is referred to as voided slabs

6 is used instead. The value of Vc is given by:

7 Vc  0.0316  f c' bv d v (ksi, in.) AASHTO Eq. 5.8.3.3-3

8 Where

4.8 51
9  AASHTO Eq. 5.8.3.4.2-2
(1  750 s ) (39  s xe )

1.38
10 s xe  s x AASHTO Eq. 5.8.3.4.2-5
a g  0.63

11 f’c = concrete compressive strength

12 bv = effective web width

13 dv ≥ 0.9de or 0.72h, whichever is greater

14 de ≥ effective depth from extreme compression fiber to the centroid of the tensile force in

15 the tensile reinforcement

16 h = height of the member

17 εs = the net longitudinal tensile strain in the section at the centroid of the tension

18 reinforcement

19 ag = maximum aggregate size

20 Sx = lesser of dv or maximum distance between layers of longitudinal crack control

21 reinforcement.

22

9
1 The simplified procedure is a similar approach to the ACI 318-14 detailed method. While

2 the simplified procedure incorporates practically the same Vci as the ACI 318-14 detailed

3 method, Vcw is modified to use a lower-bound estimate of the diagonal cracking load,

4 concrete contribution in shear, Kuchma et al. 2008. AASHTO LRFD uses concrete

5 contribution of 1.9√ (fc’) (psi), however, ACI 318-14 uses 3.5√ (fc’). The values of Vci and

6 Vcw are given by:

Vi M cre
7 Vci  0.02 f c' bv d v  Vd   0.06 f c' bv d v (ksi, in.) AASHTO Eq. 5.8.3.4.3-1
M max

8 Vcw  (0.06 f c'  0.3 f pc )bv d v  V p (ksi, in.) AASHTO Eq. 5.8.3.4.3-3

9 Where

10 bv and dv are as defined in the general shear procedure

11 fpc = compressive stress in concrete after all prestress losses have occurred either at

12 centroid of the cross section resisting live load or at the junction of the web and

13 flange when the centroid lies in the flange.

14 Vi = factored shear force at section due to externally applied loads occurring

15 simultaneously with Mmax

16 Mcre = net cracking moment

17 Mmax = maximum factored moment at section due to externally applied loads

18 It should be noted that the minimum shear reinforcement for slabs is required only if the

19 factored shear force exceeds Vc, rather than Vc/2 as required by ACI 318 in beams.

20

21 3. CSA A23.3-04

22 It worth mentioning that AASHTO LRFD general shear provision adopted the CSA

23 A23.3-04 shear provision with some minor modifications on material strength limits, crack

10
1 spacing parameter, and requirements of minimum shear reinforcement for members with

2 shear reinforcement (Hawkins et al., 2005). The value of Vc is given by:

3 Vc   f c' bw d v (psi, in.) CSAA23.3-04 Eq. 11-2

4 Where

4.8 51
5 
(1  1500 x ) (39  s ze )

6 Sz = lesser of dv or maximum distance between layers of longitudinal crack control

7 reinforcement.

8 √ (fc’) ≤ 8 MPa (1.16 ksi)

9 dv = effective shear depth, taken as the greater of 0.9d or 0.72h

10 εx = longitudinal strain at mid-depth of the member due to factored loads (positive when

11 tensile)

12

13 4. JSCE 2007

14 JSCE 2007 employs an empirical equation for Vc. This empirical formula counts for

15 concrete compressive strength, member depth, percentage of flexure reinforcement, and

16 effect of prestressing force expressed by (Mo/Mud) ratio (CEB-FIP, 1978; and Eisuke et al.,

17 2013). The control section is generally defined at a distance h/2 (where h is the total depth of

18 the slab) from the face of the support. The value of Vc is given by

19 Vc   d . p . n f vcd .bw .d (Mpa, mm) JSCE Eq. 9.2.4

1000 3  M 
20 Vc  0.2463 f cd' 4 100 pv 1  2 0 bw d (ksi, in., kip)
d  M ud 

21 Where

22 fvcd = 0.2 ∛ (f’cd) (MPa) where fvcd ≤ 0.72 (MPa) JSCE Eq. 9.2.5

23 βd = ∜(1000/d) (d in mm) when βd >1.5, βd is taken as 1.5

11
1 βp = ∛ (100 pv) when βp >1.5, βp is taken as 1.5

2 βn = 1+ 2M0/Mud (N'd ≥ 0) when βn > 2, βn is taken as 2

3 βn = 1+ 4M0/Mud (N'd < 0) when βn < 0.0, βn is taken as 0.0

4 f'cd = concrete compressive strength

5 N'd = design axial compressive force

6 Mud = pure flexure capacity without consideration of axial force

7 Mo = flexure moment necessary to cancel stress due to axial force at extreme tension fiber

8 corresponding to design flexure moment Md

9 bw = web width

10 d = effective depth

11 pv = As / (bw .d)

12 As = area of tension reinforcement

13 It should be noted that the minimum shear reinforcement is required only if the factored

14 shear force exceeds Vc, rather than Vc/2 as required by ACI 318 in beams.

15

16 5. fib Model Code 2010

17 The fib Model Code 2010 shear provisions are based on Simplified Modified Compression

18 Field Theory (SMCFT) which is developed by Bentz et al., 2006. Shear provisions in fib

19 model code 2010 are presented in different levels of approximations depending on the level

20 of details needed (e.g. preliminary design, detailed design, or assessment). The higher the

21 level of approximation (LoA), the greater the design effort applied; more conservative results

22 should be obtained from the lower LoA. Level II model provides the base model for members

23 without shear reinforcement, such as slabs, while Level I provides conservative values due to

24 the applied simplifications on the base model. The control section is generally defined at a

12
1 distance d (the effective flexural depth of the beam) from the face of the support. The value

2 of Vc is given by

3 Vc  k v f ck bw z (MPa, mm) fib MC 2010 Eq. 7.3-17

180
4 kv ( I )  (z in mm) fib MC 2010 Eq. 7.3-19
(1000  1.25z )

0.4 1300
5 kv ( I )  (z in mm) fib MC 2010 Eq. 7.3-21
(1  1500 x ) (1000  k dg z )

32
6 kdg   0.75 (dg in mm) fib MC 2010 Eq. 7.3-20
16  d g

7 Where

8 z = effective shear depth

9 εx= strain at mid-depth of the effective shear depth

10 dg = specified maximum aggregate size

11 √ (fck) ≤ 8 MPa (1.16 ksi)

12

13 6. AS 3600-2009

14 Minimum shear reinforcement requirements are waived for hollow-core slabs in the

15 Australian code AS 3600-2009. Two types of shear cracking are generally considered and the

16 least value controls the shear capacity of the cross section, similar to the ACI 318-14 detailed

17 method and AASHTO LRFD simplified method. First, flexure-shear cracking, occurs when

18 cross-section under consideration is cracked in flexure. Flexure-shear shear strength is a

19 function of member depth, concrete compressive strength, applied axial force, prestressing

20 force, and shear span. Second, web-shear cracking, the AS 3600 does not provide a specific

21 equation for calculating the web-shear capacity. Instead, the code requires calculating the

22 web-shear strength based on the combination of shear force with the prestressing force and

13
1 other applied loads affects the section that would produce a principal tensile stress of f'ct at

2 either the centroidal axis or the intersection of flange and web, whichever is more critical.

3 The value of flexure shear is given by

1/ 3
 ( Ast  Apt ) 
4 Vuc  1 . 2 . 3 .bv .d o f cv    Vo  Pv AS 3600-09 Eq. 8.2.7.2(1)
 bv d o 

5 Where

6 1= 1.1 (1.6 – do/1000) ≥ 0.8

7 2= 1, for members subject to pure bending; or

8 = 1 - (N*/3.5Ag) ≥ 0 for members subjected to axial tension; or

9 = 1 + (N*/314 Ag) for members subjected to axial compression

10 3 =1, or may be taken as 2do/av but not greater than 2, provided the applied loads and the

11 support are orientated so as to create diagonal compression over the length (av).

12 av = distance from the section at which shear is being considered to the face of the nearest

13 support.

14 fcv = ∛ (f’c) ≤ 4MPa

15 Ast = cross sectional area of longitudinal reinforcement provided in the tensile zone and

16 fully anchored to the cross-section under consideration.

17 Vo = shear force which would occur at the section when bending moment at that section

18 was equal to the decompression moment (Mo) given by

19 Mo = Z σcp.f

20 σcp.f = compressive stress due to prestress, at the extreme fiber where cracking occurs

21 For statically determinate structures:

22 Vo=Mo / |M*/V*| AS 3600-09 Eq. 8.2.7.2(2)

23 f'ct = 0.33√(f'c)

24

14
1 7. European Standard EN 1168-2005

2 The first version of the European standard EN 1168, which is similar to Eurocode EC2,

3 Part 1 (EN 1992-1-1: 2004) provisions, is partially based on Plasticity Theory as developed

4 by Thürlimann (1975, 1983) and by Nielsen (1984). EN 1168 and EC-2 provide a special

5 case of shear tension for members without shear reinforcement such as hollow-core. This

6 equation counts for concrete tensile strength and effective prestressing force which is similar

7 to web-shear cracking equation in ACI 318-14. The control section is generally defined at a

8 distance H/2 (where H is the total depth) from the face of the support. Vc value is given by

bw .I
f ct   l cp f ct
2
10 Vc  (MPa, mm)
S

11 Where

12 I = moment of inertia

13 bw = smallest web width

14 S = first area moment of section about centroidal axis

15 fct = tensile strength of concrete

16 σcp = concrete compressive stress at centroidal axis for due to prestressing

17 αℓ = lx/ lpt2 is the degree of prestressing transmission (αl ≤ 1,0);

18 lx is the distance of the considered section from the starting point of transmission length;

19 lpt2 is the upper value of transmission length (see Equation (8.18) of EN 1992-1-1:2004);

20

21 8. Yang’s method

22 The maximum stress theory has been adopted by Yang as a failure criterion. The major

23 difference in Yang’s method than most code provisions is in calculating the critical shear

24 section. Principal stresses are typically calculated at the section’s centroid. However, in this

15
1 method, critical point location is variable and depend on the void geometry and is located on

2 a line extending from the center of support at an angle β = 35° with the bottom surface of the

3 hollow-core slab. For hollow-core with circular voids, the critical point is more likely to be

4 located close to the section’s centroid. While, for hollow-core slabs with noncircular voids, it

5 can simply be predicted as the intersection of the narrowest web width of hollow-core slab

6 and the inclined line as mentioned before (Yang, 1994).

 b S cp  e A  dN p 
 f ct xcp z cp    cp   
 2 S cp b  I y AS cp  dx 
bI y  
7 Vt  V z . max   1 / 2 
S cp  b 2
  e Acp   1 z cp e   
 f ct xcp z cp    
 I
 f ct xcp z cp   

 N f  f ct2  
 A I  p ct
 2
 cpS   y AS cp   y   
 

8 Where

9 b = sum of individual web widths

10 Scp = first area moment of section about critical-point axis

11 fct = tensile strength of concrete in the presence of biaxial stress

12 xcp = horizontal distance from center of support to critical point

13 zcp = vertical distance from centroid to critical point

14 e = eccentricity of prestressing strand measured from the centroid of the cross section

15 Acp = area of section above the critical point

16 Ag = gross area of concrete section, excluding voids

17 dNp/dx = prestress force gradient

18 Np = prestressing force in strands

19

20

21

22

16
 1 REFERENCES

2 Bentz, Evan C., Frank J. Vecchio, and Michael P. Collins. “Simplified modified compression

3 field theory for calculating shear strength of reinforced concrete elements.” ACI

4 Structural Journal 103.4 (2006): 614-624.

5 CEB, “CEB-FIP Model Code for Concrete Structures,” CEB Bulletin No. 125E, 1978, pp.

6 348

7 Collins, M. P., Mitchell, D., Adebar, P., and Vecchio, F. J. “A General Shear Design

8 Method”, ACI Structural Journal, February, Vol. 93, No.1, 1996, pp. 36-45.

9 Cumming, D. A., Shield, C. K., and French, C. E., “Shear Capacity of High-Strength

10 Concrete Pre-Stressed Girders”, MN/RC-1998/12, Center for Transportation Studies,

11 University of Minnesota, Minneapolis, MN, 1998, pp. 361.

12 Eisuke, N., Avendaño, A. R., and Bayrak, O., “Shear Database for Prestressed Concrete

13 Members”, ACI Structural Journal, V. 110, No. 6 (November-December) 2013, pp. 909-

14 918.

15 Kuchma, D. A., Hawkins, N. M., et al. “Simplified Shear Provisions of the AASHTO LRFD

16 Bridge Design Specifications”. PCI Journal, v 53 n 3, p 53–73, May-June 2008.

17 Nielsen, M. P., “Limit Analysis and Concrete Plasticity”, Prentice Hall, 1984, pp. 420

18 Thürlimann, B., and Frey, R., “Fatigue tests on reinforced concrete beams with and without

19 reinforcement”. Institute fur Baustatik und Konstruction ETHZ, 1983, Zurich,

20 Switzerland. (in German)

21 Thürlimann, B., Grob, J., and Lüchinger, P., “Torsion, Biegung und Schub in

22 Stahlbetonträgem, Institut für Baustatik und Konstruktion”, ETHZ, 1975. Autographie zu

23 Fortbildungskurs für Ingenieure aus der Praxis. (In German)

17

View publication stats