Prestressed Hollow Core Slabs Supported On Beams: Finnish Shear Tests On Floors in 1990-2006

VTT WORKING PAPERS 148
Matti Pajari
Prestressed hollow core slabs

supported on beams
Finnish shear tests on floors in 1990–2006
ISBN 978-951-38-7495-7 (URL: http://www.vtt.fi/publications/index.jsp)
ISSN 1459-7683 (URL: http://www.vtt.fi/publications/index.jsp)
Copyright © VTT 2010
JULKAISIJA – UTGIVARE – PUBLISHER
VTT, Vuorimiehentie 5, PL 1000, 02044 VTT

puh. vaihde 020 722 111, faksi 020 722 4374
VTT, Bergsmansvägen 5, PB 1000, 02044 VTT

tel. växel 020 722 111, fax 020 722 4374
VTT Technical Research Centre of Finland, Vuorimiehentie 5, P.O. Box 1000, FI-02044 VTT, Finland
phone internat. +358 20 722 111, fax +358 20 722 4374
Series title, number and
report code of publication
VTT Working Papers 148

VTT-WORK-148
Author(s)
Matti Pajari
Title
Prestressed hollow core slabs supported on beams
Finnish shear tests on floors in 1990–2006
Abstract
Arrangements and results of 20 full-scale load tests on floors, each made of eight
to twelve prestressed hollow core slabs and three beams, are presented. The tests
have been carried out by VTT Technical Research Centre of Finland and Tampere
University of Technology.
ISBN
978-951-38-7495-7 (URL: http://www.vtt.fi/publications/index.jsp)
Series title and ISSN Project number
VTT Working Papers
1459-7683 (URL: http://www.vtt.fi/publications/index.jsp)
Date Language Pages
November 2010 English 674 p.
Name of project Commissioned by
Querkraftfähigkeit von Spannbeton-Fertigdecken Institut für Massivbau der RWTH Aachen
bei biegeweicher Lagerung
Keywords Publisher
Prestress, hollow core slab, flexible support, VTT Technical Research Centre of Finland
shear resistance, floor test, composite P.O. Box 1000, FI-02044 VTT, Finland
construction, concrete Phone internat. +358 20 722 4520
Fax +358 20 722 4374
Preface
The reduction of the shear resistance of hollow core slabs due to deflection of the sup-
porting beams has been studied since 1990. Despite numerous tests, theoretical and nu-
merical analyses and international cooperation, no common European understanding
about the reasons of and solutions for this phenomenon has been achieved. A German
research project “Querkrafttragfähigkeit von Spannbeton-Fertigdecken bei biegewei-
cher Lagerung”, recently completed at Institut für Massivbau, Rheinisch-Westfälische
Technische Hochschule, Aachen, aimed to be a step to that direction. The present report
has been elaborated as a part of this project.
All reported tests have been performed in confidential projects and commissions. The
owners of the results mentioned in the report have permitted the publication of all relevant
data and paid the costs of the information service, which is gratefully acknowledged.
The work has financially been supported by the research team in Aachen, i.e. Prof.
Hegger, Dr. Roggendorf and their coworkers. Without their contribution it would not
have been possible to realise the work. Special thanks are due to them for their patience
in waiting for the completion of the report and for the kind and encouraging atmosphere
before, during and after the project.
4
Contents
Preface ........................................................................................................................... 4
Meaning of abbreviations................................................................................................ 6
1. Introduction ............................................................................................................... 7
2. Summary................................................................................................................... 8
3. Shear tests on floors ................................................................................................. 9
VTT.CR.Delta.265.1990........................................................................................................ 11
VTT.S.WQ.265.1990 ............................................................................................................. 33
VTT.PC.InvT.265.1990 ......................................................................................................... 73
VTT.PC.InvT.400.1992 ....................................................................................................... 101
VTT.S.WQ.400.1992 ........................................................................................................... 133
VTT.PC.InvT-Unif.265.1993................................................................................................ 163
VTT.PC.InvT-Topp.265.1993 .............................................................................................. 199
VTT.PC.Rect-norm.265.1993.............................................................................................. 231
VTT.PC.InvT_Cont.265.1994.............................................................................................. 263
TUT.CR.MEK.265.1994 ...................................................................................................... 293
VTT.RC.Rect-norm.265.1994 ............................................................................................. 317
VTT.CP.LBL.320.1998 ........................................................................................................ 347
VTT.CR.Delta.400.1999...................................................................................................... 375
VTT.CP.Super.320.2002..................................................................................................... 413
TUT.CP.LB.320.2002 .......................................................................................................... 443
VTT.S.WQ.500.2005 ........................................................................................................... 467
VTT.PC.InvT.500.2006 ....................................................................................................... 525
VTT.CR.Delta.500.2005...................................................................................................... 563
VTT.PC.InvT.400.2006 ....................................................................................................... 613
VTT.CR.A-beam.320.2006.................................................................................................. 639
5
Meaning of abbreviations
VTT Technical Research Centre of Finland

TUT Tampere University of Technology
PC Prestressed concrete beam
RC Reinforced concrete beam
S Steel beam
CP Composite, prestressed beam
CR Composite reinforced beam
WQ Top-hat steel beam
InvT Inverted T-beam (concrete)
Rect Rectangular beam (concrete)
A-beam, Delta, MEK, LB, LBL, Super Patented composite beams
Unif Uniformly distributed load over half floor
Topp Reinforced concrete topping
Norm Normal support (slabs on the top of the beam)
Cont Continuous beam
6
1. Introduction
The effect of flexible supports, i.e. reduction of the shear resistance of the hollow core
slabs due to deflection of the supporting beams, has been experimentally studied since
1990. The results and analysis of ten tests carried out by VTT, Finland, have been pub-
lished previously. Due to these tests and parallel tests performed elsewhere it has be-
come clear that the reduction of the shear resistance has to be taken into account in design.
European standard EN 1168. Precast concrete products. Hollow core slabs has been
amended by a sentence stating that the effect of flexible supports on the shear resistance
shall be taken into account. How this can be done, is not specified. Therefore, national
design rules, if any, are applied to meet this requirement. It is obvious that a European
design method has to be developed, but this is not only a question of standardisation;
research is also needed.
In 2005, a research project dealing with the effects of flexible supports was started at
RWTH, Aachen. New floor tests were performed, but the results of the former Finnish
and German tests were also considered. As a part of the project, the test arrangements
and results of twenty Finnish floor tests in 1990–2006 have been elaborated and pub-
lished in this report. The aim has been to provide experimental data which can be re-
ferred to when writing scientific papers or when developing and standardizing European
design rules. No analysis of the results is presented. The aim has also been to make the
data so complete that there is no need to read the original test reports, five of which
have been written in Finnish. Some tabulated characteristics of the tests are given on the
first page. The rest of the report is divided in 20 Chapters, each including the results on
one floor test and the related reference tests.
The German test results are available at
http://www.imb.rwth-aachen.de/Weitere-Informationen/
(Titel „Zum Tragverhalten von Spannbeton-Fertigdecken bei biegeweicher Lagerung“)
7
2. Summary
Basic data about the tests are given in Table 1.
Table 1. Thickness of slabs (hslab), length of core filling (Lfill), span of beams (L), length of slabs
(Lslab), shear resistance / one slab in floor test (Vobs), mean of shear resistances observed in
reference tests (Vref) and last measured deflection of the middle beam before failure ().
Vobs
hslab Lfill L Lslab Vobs Vref 1) L/
Test mm mm m m kN kN Vref mm
VTT.CR.Delta.265.1990 265 50 5,0 6,0 114,6 283,9 0,40 16,3 307
VTT.S.WQ.265.1990 265 50 5,0 6,0 166,1 230,5 0,72 17,6 284
VTT.PC.InvT.265.1990 265 50 5,0 6,0 103,4 230,5 0,45 9,9 505
2)
VTT.PC.InvT.400.1992 400 320 5,0 7,2 252,1 490,3 0,51 5,4 926
VTT.S.WQ.400.1992 400 30 5,0 7,2 293,6 516,3 0,57 14,6 342
VTT.PC.InvT-Unif.265.1993 265 185 5,0 6,0 147,6 251,8 0,59 39 128
VTT.PC.InvT-Topp.265.1993 265 50 5,0 6,0 140,3 193,6 0,72 13,8 362
VTT.PC.Rect-Norm.265.1993 265 50 5,0 6,0 163,8 210,9 0,78 7,7 649
VTT.PC.InvT-Cont.265.1994 265 50 5,0 6,0 191,4 194,6 0,98 5,2 962
TUT.CR.MEK.265.1994 265 50 5,02 6,0 148,2 223,2 0,66 16,7 301
VTT.RC.Rect-Norm.265.1994 265 50 7,2 6,0 106,7 226,2 0,47 30,3 238
VTT.CP.LBL.320.1998 320 50 5,0 7,2 161,9 295,3 0,55 20,9 240
VTT.CR.Delta.400.1999 400 50 5,0 8,4 222,0 419,5 0,53 24 208
VTT.CP.Super.320.2002 320 250 4,8 9,6 127,5 242,8 0,53 17,5 274
TUT.CP.LB.320.2002 320 50 4,8 7,2 149,2 313,3 0,48 21,3 225
VTT.S.WQ.500.2005 500 400 7,2 10,0 269,6 650,7 0,41 21,2 340
VTT.PC.InvT.500.2005 500 400 7,2 10,0 336,4 547,1 0,61 21,8 330
VTT.CR.Delta.500.2005 500 400 7,2 10,0 366,9 529,4 0,69 25,7 280
VTT.PC.InvT.400.2006 400 50 4,8 9,0 282,4 332,7 0,85 6,2 774
VTT.CR.A-beam.320.2006 320 50 4,8 8,0 183,3 284,0 0,65 20,9 230
1)
Last measured deflection before failure
2)
Deflection at failure > 5,4 mm and < 7,2 mm
8
3. Shear tests on floors
9
10
VTT.CR.Delta.265.1990
1 General information
1.1
Identification
VTT.CR.Delta.265.1990 Last update 2.11.2010
and aim
DE265 (Internal identification)
Aim of the test To test the interaction between Delta beam and hollow
core slabs.
1.2
Test type
Fig. 1. Illustration of test setup. Delta beam in the middle, steel I-beams at the ends.
1.3 VTT/FI 6.9.1990

Laboratory
& date of test
1.4
Test report Author(s) Koukkari, H.
(in Finnish) Name Deltapalkin ja ontelolaataston koekuormitus (Load test on Delta beam
and hollow core floor) (in Finnish)
Ref. number RAT01814/90
Date 17.9.1990
Availability Confidential, owner is Peikko Group Oy, P.O. Box 104,
FI-15101 Lahti, Finland
2 Test specimen and loading

2.1
General plan
Fig. 2. View on test arrangements.
11
Fig. 3. Plan. Beams 1) were 1,2 m long cuts of railway rails, There was a small gap
between the ends of consecutive cuts. The beams between beams 1) and 2) were also
cuts of railway rails. Beams 2) were 1,5 m long steel I-profiles. Measures to eliminate
the friction between the stacking steel beams were considered unnecessary.
Fig. 4. Plan and section A-A.

2.2
End beams HE 200 B
15
200
9 15
200
Fig. 5. End beam.
12
2.3
Middle beam The middle beam, see Figs 6–8, comprised
- a prefabricated steel component made of a top plate, bottom plate and two folded and
perforated web plates
- a cast-in-situ part of concrete K30 which filled the empty space between the slab ends
laying on the ledges of the web plates.
The concrete was cast by VTT in laboratory, 24.8.1990.
Structural steel: Raex, see 9.1
Fig. 6. Middle beam. Cross-section and tie reinforcement (hooks made of rebars T16)
for anchoring of slabs to the beam. Depth of beam = 265 mm + 6 mm.
Holes  150 315 300 300
70
220
Fig. 7. Middle beam. Position of holes in the webs. South end.
2.4
Arrangements - Simply supported, span = 5,0 m
at middle - Note the nonsymmetric position of the beam with respect to the supports (Fig. 7). Such
beam a displacement was necessary because there was no web hole in the mid-point of the
beam. The tie reinforcement had to be anchored both to the longitudinal joints between
adjacent hollow core units and to the beam through the web holes.
Tie reinforcement:
- perpendicular to the beam: 3x(1+1) hooks T16 of the type shown in Fig. 6, anchored to
the longitudinal joints of the slabs and inside the beam
- parallel to the beam: In the cast-in-situ concrete, parallel to the Delta beam and
between the edges of the beam and the slab ends, there were two straight rebars T8,
4,75 m long, one on each side of the beam, see Fig. 8.
13
Fig. 8. Middle beam. One end of the Delta beam was free to move in beam’s direction.
60 490
60
Fig. 9. Cast-in-situ concrete within and outside middle beam. Note the tie bars T8.
2.5
Slabs 1160
40
185 265
35
40
152 223 223 223 223 152
1200
Fig. 10. Nominal geometry of slab units.
- Extruded by Partek Oy, Nastola factory, grade of concrete K60

- 10 lower strands J12,5 initial prestress 1100 MPa
J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2m see 9.1 for the strength.
2.6
Temporary No temporary supports below beams.
supports
2.7
Loading See Fig. 11. There was a gypsum layer between the tertiary beams and the top surface
arrangements of the slabs. The primary spreader beams were in direct contact with the secondary
spreader beams and the secondary beams with the tertiary spreader beams. No
attempts were made to eliminate the friction. For this reason it is difficult to evaluate
exactly, to which extent the spreader beams participated to the load-carrying
mechanism.
14
600 600 600 600 600 600 600 600
5 6 7 8
P P
(from slab end)
1
1
1
490 1000
1 1 1 1
North
South
1000
P P
B
1
1
1
1
B
1 2 3 4
A A
B B
Primary A A
P P P
Secondary
Tertiary
Load cells
Fig. 11. Loading arrangement with three layers of spreader beams.
Fig. 12. Arrangements at South end of Delta beam.
15
3 Measurements
3.1
Support The support reaction of the middle beam due to the actuator loads 4P was measured
reactions by load cells below the South end of the middle beam. See Figs 8 and 11. Due to the
eccentric position of the concrete slabs with respect to the supports of the middle beam,
the support reaction below the North end was roughly = 1,08 times the support reaction
below the South end where the reaction was measured.
3.2
Vertical
displacement
Fig. 13. Location of transducers 1 … 29 for measuring vertical deflection as well as the
location of transducers 30 (measuring differential horizontal displacement between slab
edge and Delta beam) and 31 (measuring vertical diff. displacement between slab end
and middle beam).
3.3
Average strain Not measured
3.4
Horizontal. See Fig. 13 for the only horizontal transducer 30.
displacements
3.5
Strain There were strain gauges for measuring the steel strain in the Delta beam, both parallel
to the beam and in transverse direction at the bottom surface of the ledges. Two strain
gauges were glued to the top surface of the top plate and two to the bottom surface of
the bottom plate, all four parallel to the beam.
The soffit and top surface of the hollow core slabs were also provided with strain gauges
in order to measure the strain parallel to the Delta beam. The position of all strain
gauges is given in Fig. 14.
16
Fig. 14. Position of strain gauges. a) Bottom surface of the ledges of Delta beam,
gauges 1–10. b) On the bottom and top flange of Delta beam (gauges 11, 12, 23 and 24)
and on the hollow core slabs (13–22).
4 Special arrangements
-
5 Loading strategy
5.1
Load-time Before starting the test, all measuring devices were zero-balanced. Thereafter, the
relationship actuator loads P were cyclically varied in such a way that three cycles of the type
0  43,2 kN  0 were followed by two cycles of the type 0  86,4 kN  0 (Stage I)
wherafter P was monotonously increased to failure (Stage II).
5.2
After failure
6 Observations during loading
Stage I At P = 38,8 kN longitudinal cracks appeared in the joint concrete

along the Delta beam close to the supports of the beam. The cracks
grew both in length and width with increasing load.
Stage II At P = 210 kN, the first inclined crack appeared at the edge of slab 4
near the support. Before failure there was an inclined crack at the
outermost edge of slabs 1, 4, 5 and 8. At P = 240 kN slab 1 failed
along an inclined crack as shown in Fig. 17. This was followed by the
failure of all slabs on the same side of the Delta beam. The failure
patterns are illustrated in Figs 15–22.
After failure
7 Cracks in concrete
In the following figures, the numbers refer to the value of the actuator loads P in kN.
7.1
Cracks at See Fig. 15.
service load
17
7.2
Cracks after
failure
Fig. 15. Cracks on the top and on the edges of the floor after failure.
Fig. 16. Cracks in the soffit after failure.
18
Fig. 17. Slab 1 after failure.
Fig. 18. Top surface of slabs 1 (on the left), 2 and 3 after failure.
19
Fig. 20. Top surface of slabs 4 (on the right) and 3 after failure.
20
Fig. 22. Soffit of slabs 1–4 after failure.
8 Observed shear resistance
The measured support reaction vs. the total imposed actuator load is depicted in Fig.
23. This relationship and the known eccentricity of the slabs and loads with respect to
the beam supports have been used when calculating the observed shear resistance of
the slabs.
21
Fig. 23. Total imposed actuator load vs. support reaction of middle beam. South end.
The shear resistance of one slab end (support reaction of slab end at failure) due to
different load components is given by
Vobs = Vg,sl + Vg,jc + Veq + V
where Vg,sl, Vg,jc, Veq and VP are shear forces due to the self-weight of slab unit, weight of
joint concrete, weight of loading equipment and actuator forces P, respectively. The test
report does not give all these components but the sum
Veq + VP = 101,9 kN
which is based on measured support reaction below one end of the middle beam and on
the geometry of the test specimen. The self weight of the loading equipment was 420 kg/
one actuator. From this, Veq = 1,8 kN is obtained. Hence VP = 100,1 kN.
From the measured density of the concrete and nominal geometry
Vg,sl + Vg,jc = 12,3 + 0,4 kN = 12,7 kN
follows. The shear resistance Vobs = 114,6 kN (shear force at support) is obtained for
one slab unit with width = 1,2 m. The shear resistance per unit width is vobs = 95,5 kN/m
9 Material properties
9.1
Strength of ReH/Rp0,2 Rm
Component Note
steel MPa MPa
Delta beam
- web plates 355 Nominal (Raex 37-52)
- top plate 390 Nominal (Raex 423)
- bottom plate 390 Nominal (Raex 423)
End beams  350 Nominal (Fe52)
Slab strands J12,5 1570–1630 1770–1860 Nominal (no yielding in test)
Reinforcement Txy 500 Nominal value for reinforcing bars
(no yielding in test)
22
9.2
Strength of # Cores h h d Date of test Note
slab concrete, mm mm
d
floor test
12 50 50 6.9.1990 Upper flange of slabs 1–4. 3
from each, = 2393 kg/m3
Mean strength [MPa] 74,2 (0 d)1) Vertically drilled
St.deviation [MPa] 6,2 Tested as drilled2)
9.3
Strength of Not measured, assumed to be the same as that in the floor test
slab concrete,
reference tests
9.4
Strength of a
cast-in-situ # a a Date of test Note
concrete mm
a
3 150 31.8.1990 Kept in laboratory in the same
Mean strength [MPa] 33,8 (-6 d)1) conditions as the floor specimen
St.deviation [MPa] - = 2247 kg/m3
1)
Date of material test minus date of structural test (floor test or reference test)
2)
After drilling, kept in a closed plastic bag until compression
10 Measured displacements and strains

In the following figures, P stands for the actuator force plus load due to loading
equipment per one actuator. The cylic stage (Stage I) is not shown. The first point on
each curve corresponds to the start of the monotonous loading stage (Stage II). Due to
the abrupt failure, the measured results at the last load step are missing.
10.1
Deflections
Fig. 24. Mid-point deflection of middle beam (transducer 15) and end beams
(transducers 1 and 29).
23
Fig. 25. Deflection of middle beam measured by transducers 12–17.
Fig. 26. Deflection of slabs 1–4 measured by transducers 2–6.
24
25
Fig. 30. Deflection along longitudinal line of symmetry measured by transducers 1, 4, 9,

15, 21, 26 and 29.
10.3
Average strain -
26
10.4
Differential
displacement
Fig. 31. Differential vertical displacement between middle beam and end of slab 7
measured by transducer 31.
Fig. 32. Differential horizontal displacement between middle beam and edge of slab 4
measured by transducer 30.
27
10.5 Strain
Fig. 33. Longitudinal (gauge 4) and transverse strain (gauges 1–3 and 5–7) measured at
the bottom surface of the ledge of Delta beam.
Fig. 34. Longitudinal (gauge 9) and transverse strain (gauges 8 and 10) measured at the
bottom surface of the ledge of Delta beam.
28
Fig. 35. Strain parallel to the beam

- on the top surface (gauges 19 and 20) and on the bottom surface
(gauges 21 and 22) of the hollow core slabs
- on the top surface of the top plate (gauge 23) and on the bottom
surface of the bottom plate of the Delta beam (gauge 12).
Fig. 36. Strain parallel to the beam

- on the top surface (gauges 13–15) and on the bottom surface
(gauges 16–18) of the hollow core slabs
- on the top surface of the top plate (gauge 24) and on the bottom
surface of the bottom plate of the Delta beam (gauge 11).
The tensile strain of the order 0,001 measured at the bottom of the bottom plate and in
the ledges of the Delta beam confirm that the steel in the beam was far from yielding.
29
11 Reference tests
After the load test on the floor, slabs 3 and 7 were taken as reference test specimens.
Their ends which had been supported by the end beams, where loaded in shear as
shown in Fig. 37. The concrete tie beam partly outside the slab end, partly in the hollow
cores, was not removed before loading.
Fig. 37. Layout of reference test. For Lj, see the next table.
Table. Reference tests. Span of slab, shear force Vg at support due to the self weight of
the slab, actuator force Pa at failure + weight of loading equipment Peq, total shear force
Vobs at failure and total shear force vobs per unit width.
Span Vg Pa+Peq Va+eq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
R3 >6.9.1990 4938 10,3 348,5 277,9 288,2 240,2 Web shear failure
R7 >6.9.1990 4810 10,0 340,4 269,6 279,6 233,0 Web shear failure
Mean 283,9 236,6
12 Comparison: floor test vs. reference tests
The observed shear resistance (support reaction) of the hollow core slab in the floor test
was equal to 114,6 kN per one slab unit or 95,5 kN/m. This is 40% of the mean of the
shear resistances observed in the reference tests.
30
13 Discussion
1. The span of the middle beam was 5,0 m; that of the end beams 4,9 m.
2. The friction between the spreader beams was not intentionally eliminated, which
may have affected the response of the floor test specimen to some extent.
3. The failure took place at an unexpected low load level. Therefore, the load increments
applied were still relatively big and the gap between the failure load, at which no
measurements were made, and the proceeding load level at which the response
was measured, was big, too. The conclusions below about the strains and
deflections at failure are based on the extrapolation of the measured curves.
4. At failure, the net deflection of the middle beam due to the imposed actuator
loads (deflection minus settlement of supports) was 16,3 mm or L/307, i.e. rather
small. It was 3,5–4,3 mm greater than that of the end beams. Hence, the torsional
stresses due to the different deflection of the middle beam and end beams had a
minor effect, if any, on the failure of the slabs.
5. The shear resistance measured in the reference tests was higher than the mean
of the observed values for similar slabs given in Pajari, M. Resistance of
prestressed hollow core slab against a web shear failure. VTT Research Notes
2292, Espoo 2005. This difference may be attributable to the concrete tie beam at
the edge of the sheared end in the reference test. It prevented the deformation of
the end section of the slab and thus equalized the strains in the webs of the slab,
which effectively eliminated the premature failure of any individual web.
6. The beams did not yield in the floor test.
7. The failure mode was web shear failure of edge slabs close to the supports of the
middle beam. Unlike in an isolated hollow core slab unit, the appearance of the
first inclined crack close to the slab end did not mean failure but the loads could
still be increased.
31
32
VTT.S.WQ.265.1990
1.1
Identification VTT.S.WQ.265.1990 Last update 2.11.2010
and aim
WQ265 (Internal identification).
Note that the top-hat steel beam was called HQ-beam
when the floor test was carried out, but later on the
name has been changed. The present name WQ-
beam is used in the following
Aim of the test To study whether or not the shear resistance of the
hollow core slabs is reduced when supported on a
WQ-beam
1.2
Test type
WQ-beam
Fig. 1. Illustration of test setup.
1.3 VTT/FI 11.10.1990

Laboratory
& date of test
1.4
Name Matalien leukapalkkien ja ontelolaataston kuormituskokeet (Load tests
on shallow beams and hollow core floor), in Finnish
Date 19.11.1990
Availability Confidential, owner is Rautaruukki Oyj,
P.O. Box 35, FI-01531 Vantaa, Finland
33
VTT.S.WQ.265.1990

2.1
General plan
Fig. 2. Overview on arrangements.
34
VTT.S.WQ.265.1990
Fig. 3. Plan and longitudinal section along joint between adjacent slab units. Note the
suspension reinforcement 3T8 c/c 1200 in the longitudinal joints of slabs and rebars
(1+2+1)T8 tying the slab ends at the ends of the specimen and on both sides of the
WQ-beam.
T8: Hot rolled, weldable rebar A500HW,  = 8 mm
35
VTT.S.WQ.265.1990
2.2
End beams
Fig. 4. End beam.
- Simply supported, span = 5,0 m

- There was plywood between the slabs and the end beam, see also Fig. 3
- Structural steel: Fe 52, fy  350 MPa (nominal fy)
2.3
Middle beam The beam was designed to carry the support reactions from the slabs, slightly lower
than those corresponding to the estimated shear resistance of the slab ends. The beam
was made by PPTH-Teräs Oy and delivered to VTT on the 8th of August 1990. The
measured camber of the beam was 12,7 mm.
Fig. 5. WQ-beam.
Structural steel: Fe 52C, fy  350 MPa (nominal fy)
2.4
at middle - 4 load cells below support at South end
beam - bearing length of slabs = 60 mm
- see Fig. 3 for the bar reinforcement across the beam and parallel to it
- joint concrete cast 27.9.1990
36
VTT.S.WQ.265.1990
2.5
Slabs 1160
40
185 265
35
40
152 223 223 223 223 152
1200

- Extruded by Parma Oy 29.6.1990
- delivered to VTT, 12.9.1990
- grade of concrete K60
- 10 lower strands J12,5; initial prestress 1100 MPa
J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2
2.6
Temporary -
supports
2.7
attempts were made to eliminate the friction. For this reason it is difficult to evaluate, to
which extent the spreader beams participated in the load-carrying mechanism.
37
VTT.S.WQ.265.1990
600 600 600 600 600 600 600 600
5 6 7 8
(from slab end) P P
1
1
1
1000
1 1 1
490
1
North
South
1000
P P
B
1
1
1
1
B
1 2 3 4
A A
B B
Primary A A
P P P
Secondary
Tertiary
Load cells
Fig. 7. Loading arrangements with three layers of spreader beams. The joint concrete
between the slab ends and middle beam was extended to the ends of the WQ-beam.
38
VTT.S.WQ.265.1990
3 Measurements
3.1
Support There were four load cells below the South end of the WQ-beam for measuring the
reactions support reaction due to the actuator loads.
3.2
Vertical
displacement
Fig. 8. Location of transducers 1 … 33 for measuring vertical deflection, 34 … 39 for

measuring the opening of cracks along the WQ-beam, 40 and 41 for measuring the
differential horizontal displacement between the end of the beam and the joint concrete
(see Fig. 13) and 42 & 43 for measuring differential vertical displacement between the
slab end and the beam.
39
VTT.S.WQ.265.1990
3.3
Average strain -
3.4
Horizontal See Fig. 8, transducers 34 … 41. Transducers 40 and 41 measured the sliding of the
displacements joint concrete along the WQ-beam. Fig. 13 gives an impression of the vertical position of
these transducers.
3.5
Strain
Fig. 9. Position of strain gauges 1 … 48, all parallel to the beams.
40
VTT.S.WQ.265.1990
Fig. 10. Position of strain gauges 49 … 53 below WQ-beam, all transverse to the beam,
and position of strain gauges 42–48, all parallel to the beam.
-
5 Loading strategy
5.1
Load-time Date of test was 11.10.1990
relationship
Before starting the test, all measuring devices were zero-balanced. Thereafter, the actuator
loads P were varied in such a way that after five cycles of the type 0  201,6 kN  0
(Stage I) the loads P were monotonously increased to 293 kN (Stage II). At this point
unloading was necessary due to a leakage in the hycraulic circuit. After having fixed the
leakage, loads P were monotonously increased to failure (Stage III).
5.2
After failure
For the cracks observed during the loading and after the failure, see Figs 11–19.
Stage I Cracks parallel to and along the edges of the WQ-beam were observed in
the joint concrete. Some longitudinal cracks along the strands in the soffit of
the slabs and vertical cracks in the tie beams at the ends of the floor were
discovered.
Stage II The cracks along the edges of the WQ-beam grew gradually and at P = 230 kN
they were continuous from one beam end to the other. At P = 273 kN, an
inclined crack, starting at the mid-depth of slab 4 next to the WQ-beam and
growing upwards, appeared. At P = 283 kN an inclined crack also appeared
at the end of slab1, and at P = 292 kN in slab 8.
Stage III Right before failure, an inclined crack was observed in slab 8, and at the
same time, a similar crack appeared in slab 1. At P = 345 kN, slabs 8 and 7
failed in shear.
41
VTT.S.WQ.265.1990
7.1
Cracks at
service load
7.2
Cracks after
failure
Fig. 11. Inclined cracks in slab 1 after failure of slabs 7 and 8.
Fig. 12. Failure of slabs 8 and 7.
42
VTT.S.WQ.265.1990
Fig. 13. Failure of slabs 8 and 7.
Fig. 14. Failure of slab 8.
43
VTT.S.WQ.265.1990
44
VTT.S.WQ.265.1990
Fig. 16. Cracks on the top and at the edges of the floor after failure. The load values
refer to actuator load P at which the crack was observed.
45
VTT.S.WQ.265.1990
Fig. 17. Cracks on the top and at the edges of the floor after failure.
46
VTT.S.WQ.265.1990
Fig. 18. Cracks after failure in the soffit.
47
VTT.S.WQ.265.1990
Fig. 19. A part of the previous figure in more detail. The position of the webs in the
hollow core slabs is indicated by dashed lines.
The total actuator load = 4P vs. measured support reaction below the South end of the
middle beam is shown in Fig. 20. The theoretical reaction is calculated assuming simply
supported slabs. This comparison shows that the support reaction due to the actuator
forces can be calculated accurately enough assuming simply supported slabs. However,
the failure of the slab ends at the North end of the middle beam resulted in reduction of
support reaction below that end while the actuator force could still slightly be increased.
The maximum support reaction is regarded as the indicator of failure.
48
VTT.S.WQ.265.1990
Fig. 20. Stages I and II. Measured support reaction due to total actuator load 4P.
Fig. 21. Stage III. Measured support reaction due to total actuator load 4P.
49
VTT.S.WQ.265.1990
Vobs = Vg,sl + Vg,jc + Veq + VP
joint concrete, weight of loading equipment and actuator forces P, respectively. The test
report does not give all these components but
VP = 151,6 kN
is obtained from the failure load P = 345 kN using the load-reaction relationship shown
in Fig. 21 [reaction = 0,8844x(2P)].
In the same way
Veq = 1,8 kN
is obtained from the weigth of the loading equipment (= 210 kg / one slab).
From the nominal geometry and measured density of the concrete
Vg,sl + Vg,jc = 12,3+0,4 = 12,7 kN
one slab unit with width = 1,2 m. The shear force per unit width is vobs = 138,4 kN/m.
9.1
Strength of Component ReH/Rp0,2 Rm Note
steel MPa MPa
End beam  350 Nominal (Fe 52, no yielding in test)
WQ-beam  350 Nominal (Fe 52C, no yielding in test)
Slab strands J12,5 1570–1630 1770–1860 Nominal (no yielding in test)
Reinforcement Txy 500 Nominal value for reinforcing bars,
9.2
Strength of
slab concrete, # Cores h h d Date of test Note
floor test mm mm
d
6 50 50 12.11.1990 Upper flange of slabs 5 and
3 (3pc. each),  = 2398 kg/m3
Mean strength [MPa] 65,3 (+1 d)1) vertically drilled
9.3
Strength of Not measured, assumed to be the same as that in the floor test.
slab concrete,
reference tests
50
VTT.S.WQ.265.1990
9.4 Strength of a
grout in joints # a a Date of test Note
mm
a
Mean strength [MPa] 30,5 (-1 d)1) conditions as the floor specimen
St.deviation [MPa]  =2223 kg/m3
a
# a a Date of test Note
mm
a
Mean strength [MPa] 31,7 (+1 d)1) conditions as the floor specimen
1)
2)
10 Results measured during floor test
In the following graphs, P is the actuator load.
10.1
Deflections The measured deflections in Stages II and III are shown in Figs 22–39. The numbers
close to the curves refer to the number of transducer, see Fig. 8.
Fig. 22. Stage II.
51
VTT.S.WQ.265.1990
Fig. 23. Stage III.
Fig. 24. Stage II.
52
VTT.S.WQ.265.1990
Fig. 25. Stage III.
Fig. 26. Stage II.
53
VTT.S.WQ.265.1990
Fig. 27. Stage III.
Fig. 28. Stage II.
54
VTT.S.WQ.265.1990
Fig. 29. Stage III.
Fig. 30. Stage II.
55
VTT.S.WQ.265.1990
Fig. 31. Stage III.
Fig. 32. Stage II.
56
VTT.S.WQ.265.1990
Fig. 33. Stage III.
Fig. 34. Stage II.
57
VTT.S.WQ.265.1990
Fig. 35. Stage III.
Fig. 36. Stage II.
58
VTT.S.WQ.265.1990
Fig. 37. Stage III.
10.2
Crack width The differential displacement measured by horizontal transducers 34–39 reflect the
crack width in the joint concrete next to the WQ-beam. These differential displacements
in stages II and III are shown in Figs 38 and 39.
Fig. 38. Stage II.
59
VTT.S.WQ.265.1990
Fig. 39. Stage III.
10.3 -
Average strain
10.4
Differential A positive value means that the concrete is moving towards the beam end.
horizontal
displacement
Fig. 40. Stage II.
60
VTT.S.WQ.265.1990
Fig. 41. Stage III.
10.5
Differential A positive value means that the slab end is deflecting more than the beam.
vertical
displacement
Fig. 42. Stage II.
61
VTT.S.WQ.265.1990
Fig. 43. Stage III.
10.6
Strain
Fig. 44. Stage II.
62
VTT.S.WQ.265.1990
Fig. 45. Stage III.
Fig. 46. Stage II.
63
VTT.S.WQ.265.1990
Fig. 47. Stage III.
Fig. 48. Stage II.
64
VTT.S.WQ.265.1990
Fig. 49. Stage III.
Fig. 50. Stage II.
65
VTT.S.WQ.265.1990
Fig. 51. Stage III.
Fig. 52. Stage II.
66
VTT.S.WQ.265.1990
Fig. 53. Stage III.
Fig. 54. Stage II.
67
VTT.S.WQ.265.1990
Fig. 55. Stage III.
Fig. 56. Stage II.
68
VTT.S.WQ.265.1990
Fig. 57. Stage III.
Fig. 58. Stage II.
69
VTT.S.WQ.265.1990
Fig. 59. Stage III.
Fig. 60. Stage II.
70
VTT.S.WQ.265.1990
Fig. 61. Stage III.
11 Reference tests
After the floor test, slabs 3 and 5 were taken for reference test specimens. Their ends
which had been supported by the end beams, were loaded in shear as shown in Fig. 62.
The concrete tie beam partly outside the slab end, partly in the hollow cores, was not
removed before loading.
Fig. 62. Layout of reference test. For Lj, see the next table.
Test Date Span Vg Pa+Peq Va+eq Vobs vobs Note

mm kN kN kN kN kN/m
R3 12.10.1990 5940 12,3 285,5 237,4 249,6 208,0 Web shear failure
Mean 230,5 192,1
71
VTT.S.WQ.265.1990
13 Discussion
1. The friction between the spreader beams was not eliminated, which may have
affected the response of the floor test specimen to some extent.
2. The last measured net deflection of the middle beam due to the imposed
actuator loads only (deflection minus settlement of supports) was 17,6 mm or
L/284. It was 4,9–6,2 mm greater than that of the end beams. Hence, the
torsional stresses due to the different deflection of the middle beam and end
beams may have had a minor effect on the failure of the slabs.
3. The shear resistance measured in the reference tests was of the same order
as or slightly higher than the mean of the observed values for similar slabs
given in Pajari, M. Resistance of prestressed hollow core slab against a web
shear failure. VTT Research Notes 2292, Espoo 2005. The concrete tie beam
at the sheared end may have enhanced the resistance. It prevented the
deformation of the end section of the slab and thus equalized the strains in the
webs of the slab, which effectively eliminated the premature failure of any
individual web.
5. The failure mode was web shear failure of edge slabs close to the supports of
the middle beam. Unlike in an isolated hollow core slab unit, in the floor test
the appearance of the first inclined crack close to the slab end did not mean
failure but the loads could still be increased.
72
VTT.PC.InvT.265.1990
1.1
Identification
VTT.PC.InvT.265.1990 Last update 2.11.2010
and aim
PC265 (Internal identification).
Aim of the test To study whether or not the shear resistance of the
hollow core slabs is reduced when supported on a
shallow prestressed concrete beam
1.2
Test type
Prestressed
concrete beam
1.3 VTT/FI 14.–19.11.1990
Laboratory
& date of test
1.4
Name Matalien leukapalkkien ja ontelolaataston kuormituskokeet (Load tests
on shallow beams and hollow core floor), in Finnish
Date 28.11.1990
Availability Confidential, owner is Rakennustuoteteollisuus RTT ry,
P.O. Box 381, FI-00131 Helsinki

2.1
General plan
Fig. 2. Overview on arrangements.
73
Fig. 3. Plan and longitudinal section along joint between adjacent slab units. 8 refers
to rebar T8, see 9.1.Note the tie reinforcement 3T8 c/c 1200 in the longitudinal joints of
slabs and rebars (1+2+1) T8 tying the slab ends at the ends of the specimen and on
both sides of the PC beam.
74
2.2
End beams - Reinforced concrete beam
- Made by Lohja Oy, delivered to VTT on the 2nd of August 1990
- Concrete K40
80 50 Plastic plug
Tie beam in cores
2 T8 Concrete filling
L=4750 Dowel reinforcement
End beam
Hoops T8 300
c/c 150
5 T20, L=5150
300
Fig. 4. Arrangements at end beam. Txy refes to a rebar with diameter xy mm, see 9.1.
The size and spacing of the dowel reinforcement has not been given in the report.
2.3
Middle beam The beam was designed to carry the support reactions from the slabs, somewhat lower
than those corresponding to the estimated shear resistance of the slab ends on non-
flexible support. The beam was manufactured by Lohja Oy in Nummela factory and it
was delivered to VTT on the 2nd of August 1990. The measured camber of the beam
was 20,6 mm.
Shear keys 150x150x25 c/c 300

A1 A1
45
45
265 A1 4 12,5
p o = 1 100 M P a
A1
40
45 100
46
150 480 150
14 12,5 p o = 1 100 M P a
Fig. 5. Cross-section of PC beam. 12,5 refers to a prestressing strand J12,5 and A1 to

a rebar T12, see 9.1. Concrete K60.
75
Fig. 6. PC beam. Manufacturing drawing by Oy Lohja CAD Ab.
J12,5: 7 indented wires,  =12,5 mm, Ap = 93 mm2, low relaxation (<2,5% 100 h)
76
2.4
Arrangements 50 50 Plastic plug 3 T8 L=2000 c/c 1200
at middle
beam 40
185 3 holes c/c 1200
40 30
Neoprene 30x5
25 30 45 50 2 T8 L = 4750
Fig. 7. Arrangements at middle beam for T8 see 9.1.

- 4 load cells below support at South end
- joint concrete K30 cast 7.11.1990, maximum aggregate size 8 mm, consistency
1–2 VBs, rapidly hardening cement
2.5
Slabs 1160
40
185 265
35
40
152 223 223 223 223 152
1200
- Extruded by Parma Oy 29.6.1990, the same bed and casting lot as for the slabs in test
VTT.S.WQ.265
- delivered to VTT, 18.10.1990
- concrete K60
J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2, low relaxation (<2,5% 100 h),
see 9.1
77
2.6
Temporary -
supports
2.7
attempts were made to eliminate the friction. For this reason it is difficult to evaluate, to
which extent the spreader beams participated in the load-carrying mechanism.
600 600 600 600 600 600 600 600
5 6 7 8
P P
(from slab end)
1
1
1
1000
1 1 1 1
580
North
South
1000
P P
B
1
1
1
B
1 2 3 4
A A
Primary A A
P P
Secondary
Tertiary
Load cells
Fig. 9. Loading arrangement with three layers of spreader beams.
78
3 Measurements
3.1
Support reactions There were four load cells below the South end of the PC beam for measuring the
support reaction due to the actuator loads.
3.2
Vertical
displacement
Fig. 10. Location of transducers 1 … 33 for measuring the vertical deflection, 40 & 41 for
measuring the differential horizontal displacement between the end of the beam and the
joint concrete (see Figs 11 & 17) and 42 & 43 for measuring differential vertical
displacement between the slab end and the beam.
79
3.3
Average strain -
3.4
Horizontal See Fig. 8, transducers 34 … 41. Transducers 40 and 41 measured the sliding of the
displacements slab along the WQ-beam. Figs 11 and 17 give an impression of the position of these
transducers.
Fig. 11. Position of horizontal transducers 39 and 41at South end of middle beam.
80
3.5
Strain
Fig. 12. Position of strain gauges 1 … 44, all parallel to the beams.
-
81
5 Loading strategy
5.1
Load-time The exact date of the floor test is not mentioned in the test report but it has been before
relationship 19.11.1990 and most likely not before 14.11.1990.
Before starting the test, all measuring devices were zero-balanced. Thereafter, the
actuator loads P were varied in such a way that after five cycles of the type 0  185,3 kN
 0 (Stage I), loads P were monotonously increased to failure load 205,6 kN (Stage II).
5.2
After failure
For the cracks observed during the loading and after the failure, see Figs 12–23.
Stage I Cracks parallel to and along the edges of the PC beam were observed in
the joint concrete. Cracks between the tie beams and the slab ends were
also observed above the end beams.
Stage II At P = 180 kN, inclined cracks appeared in the upper corners of the
outermost webs of slabs 1, 4, 5 and 8 next to the supports of the middle
beam. At P = 200 kN new inclined cracks below the first inclined cracks
appeared and at P = 205,6 kN a failure took place along these new cracks.
7.1
Cracks at
service load
7.2
Cracks after
failure
82
Fig. 15. Cracks parallel to beam. South end of middle beam.
83
Fig. 16. Failure of slabs 1 (on the right) and 5 (on the left).
Fig. 17. Cracks after failure in slab 8 and in the joint concrete.
84
85
Fig. 20. Cracks parallel to beam in joint concrete. North end of middle beam.
Fig. 21. Cracks on the top and at the edges of the floor after failure. The load values
refer to the value of actuator load P at which the crack was observed.
86
Fig. 22. Failure cracks and other major cracks on the top and at the edges of the floor
after failure.
87
88
The actuator load = P vs. measured support reaction below the South end of the middle
beam is shown in Fig. 24.
Fig. 24. Measured support reaction below South end of PC beam vs. actuator load P.
joint concrete, weight of loading equipment and actuator forces P, respectively.
VP = 88,9 kN
is obtained from the failure load P = 205,6 kN using the load-reaction relationship shown
in Fig. 31 [VP = 0,8645x(0,5P)].
In the same way
Veq = 1,8 kN
is obtained from the weigth of the loading equipment (= 210 kg / one slab).
From the nominal geometry and measured density of the concrete
Vg,sl + Vg,jc = 12,3+0,4 = 12,7 kN
89
9.1
steel Component Note
MPa MPa
Slab strands J12,5 1570–1630 1770–1860 Nominal (no yielding in test)
Reinforcement 500 Nominal value for reinforcing bars
Txy (=xy mm) A500HW (no yielding in test)
9.2
Strength of
floor test mm mm
d
Upper flange of slabs 4 and

6 50 50 14.–19.11.1990?
8 (3pc. each),  = 2418 kg/m3
Mean strength [MPa] 63,8 (? d)1) vertically drilled
9.3
Strength of -
slab concrete,
reference tests
9.4
Strength of a
grout in joints # a Date of test Note
a mm
a

Mean strength [MPa] 26,8 (? d)1) conditions as the floor specimen
1)
2)
90
10 Results measured during floor test
In the following graphs, P is the actuator load.

10.1
Deflections The measured deflections in Stage II are shown in Figs 25–41. The numbers close to
the curves refer to the number of transducer, see Fig. 10.
Fig. 25. Measured deflection.
91
92
93
10.2
Crack width The differential displacement measured by horizontal transducers 34–39 reflect the
crack width in the joint concrete next to the middle beam. These measured
displacements in Stage II are shown in Fig. 33.
94
Fig. 33. Opening of joint in Stage II.
10.3
Average strain -
10.4
Differential A positive value means that the concrete is moving towards the beam end.
horizontal
displacement
Fig. 34. Differential horizontal displacement between South end of middle beam and
edge of slab in Stage II.
95
10.5
Differential A positive value means that the slab end is deflecting more than the beam.
vertical
displacement
Fig. 35. Differential vertical displacement between middle beam and slab end. Stages I
and II.
10.6
Strain
Fig. 36. Strain measured by gauges 1–6. Stage II.
96
97
Fig. 40. Strain measured by gauges 25–28, 33–36 and 42–43. Stage II.
98
Fig. 41. Strain measured by gauges 29–32 and 38–41. Stage II.
11 Reference tests
No reference tests were performed for the slabs in the present floor test. Instead, the
reference tests carried out for two slabs taken from floor test VTT.S.WQ.265.1990 are
regarded as applicable to this case, too, because
- the slabs for these two floor tests were taken from the same casting lot and bed
- the slabs were cast as early as 29.6.1990 and the change in concrete strength must
have been very small in October – November 1990 when the tests were carried out.
The results presented below are taken from report VTT.S.WQ.265.1990.
Span Vg Pa+Peq Va+eq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
Mean 230,5 192,1
99
13 Discussion
1. The friction between the spreader beams was not eliminated, which may have
affected the response of the floor test specimen to some extent. This additional
stiffness reduced the deflection of the floor but it is difficult to evaluate whether
the net effect on the observed shear resistance was positive or negative.
2. The net deflection of the end beams (deflection minus settlement of supports
was very small beams, apparently < 2 mm. The original idea was to reduce the
horizontal interaction between the end beam and the slab ends above it, but due
to some misunderstanding, the beams were provided with dowel reinforcement
which was not specified in the drawings. These dowels made the laboratory
personnel believe that the slab ends and the end beam must be cast together. In
this way the resulting composite beam became far too stiff to deflect like the
middle beam, which was the primary design criterion for the end beam.
3. The last measured net deflection of the middle beam due to the imposed
actuator loads only (deflection minus settlement of supports) was 9,9 mm or
L/505. It was ≈8 mm greater than that of the end beams. Hence, the torsional
stresses due to the different deflection of the middle beam and end beams may
have had a minor effect on the failure of the slabs.
On one hand, the torsion in the slab elements due to the different deflection of
the middle beam and end beams reduced the deflection of the middle beam but
increased the torsional shear stresses in the webs of the outermost slab
elements. The net effect of the torsion on the observed shear resistance may
have been positive or negative.
4. The shear resistance measured in the reference tests was of the same order as
or slightly higher than the mean of the observed values for similar slabs given in
Pajari, M. Resistance of prestressed hollow core slab against a web shear
failure.VTT Research Notes 2292, Espoo 2005. The concrete tie beam at the
sheared end may have enhanced the resistance in the reference tests. It
prevented the deformation of the end section of the slab and thus equalized the
strains in the webs of the slab, which effectively eliminated the premature failure
of any individual web.
the middle beam. Unlike in an isolated hollow core slab unit, in the floor test the
appearance of the first inclined crack close to the slab end did not mean failure
but the loads could still be increased.
the middle beam. Unlike in an isolated hollow core slab unit, in the floor test the
appearance of the first inclined crack close to the slab end did not mean failure
but the loads could still be increased.
100
1.1
Identification VTT.PC.InvT.400.1992 Last update 2.11.2010
and aim
PC400 (Internal identification)
Aim of the test To study the shear resistance of thick hollow core
slabs supported on beams.
1.2
Test type
Prestressed Prestressed
concrete beam Prestressed concrete beam
concrete beam
1.3 Laboratory VTT/FI 24.2.1992
& date of test
1.4
Test report Author(s) Pajari, M.
Name Loading test for 400 mm hollow core floor supported on prestressed
concrete beams
Ref. number RAT-IR-3/1993
Date 15.4.1993
Availability Public, available on request from VTT Expert Services,
P.O. Box 1001, FI-02044 VTT.
Financed by Lohja Oy, Finland; NCC Prefab AB, Sweden; Parma Oy,
Finland; Oy Partek Concrete Ab, Finland; Skanska Prefab AB,
Sweden and AB Strängbetong, Sweden. The Finnish companies
were financially supported by TEKES, Finland.

(see also Appendix A)
2.1
General plan 7200 480 7200
50 50
4
8 4
44
7 3 5000
6 2
5 1
1260 1260
Fig. 2. Plan.
101
2.2
End beams Simply supported, prestressed concrete beams. Span = 5,0 m
Concrete: K60
A
5600
40 100 100 40 Strands 3 J12,9

40 p0
= 1250 MPa
380
13
42
40 100 40 Strands 8 J12,9
50 50 p0
= 1250 MPa
A A
Fig. 3. End beam. For J12,9 see 2.3.
80 50
Plastic plug
2 T10 
L = 4750
Tie beam 400
Plywood
30 30 3 T12 c/c 1200

End beam 350 1100
380
140 140 Fig. 5. Tie reinforcement in joints

between adjacent slabs. T12 refers
to a reinforcing bar, see 2.4.
Fig. 4. Arrangements at end beam. T10 refers
102
2.3
Middle beam Concrete: K60
Tendons:
J12,9: Prestressing strand, 7 wires,  =12,9 mm, Ap = 100 mm2, see 9.1
1600 1200 1200

Holes 150 150
A
150
30
A
5600
40 400 40 Strands 2+2 J12,9

= 1210 MPa kN
40 po
Shear keys 13
150x150x15
400 c/c 300
Strands 1+1 J12,9
= 1210 MPa
po
150 13 115
42
40 150 50 50 50 50 150 40 Strands 12+12 J12,9
50 50 100 50 50 = 1310 MPa
880 po
A A Debonding: 4 J12,9 , L=1300

at both ends
Fig. 6. Elevation and section of middle beam.
2.4
Arrangements 320 mm of all hollow cores where filled by concrete. Neither slots nor holes were made
at middle in the hollow core slabs.
beam
320 50 480 50 320
4 T10 L = 5550 Hard mineral wool
40
320
40
10
Neoprene 20x10
150
60 140 550 200
Fig. 7. Section along hollow cores.
103
30
3 T12 c/c 1200 L = 2500
Fig. 8. Tie reinforcement in joints between adjacent slabs.
Txy: Hot rolled, weldable rebar A500HW,  = xy mm, see 9.1.
2.5
Slabs 1158
40
62,1
R125
216
400 197,9
60
60 18
35 40 1200
Fig. 9. Nominal geometry of slab units (in scale).
- Extruded by Partek Betoniteollisuus Oy, Hyrylä factory 15.1.1992

1156
48,7 Max measured bond slips:
2,0 in slab 5; 1,4 in slab 4;
1,3 in slab 7 and 0,9 in slabs
306,4 bw,i b w,i = 290,9 3 and 4
Measured weight of slab

units = 5,49 kN/m
38,5 40,2 1195
Fig. 10. Mean of most relevant measured geometrical

characteristics.
2.6
Temporary Temporary supports below beams (Yes/No)
supports - No
104
2.7
Loading There were two separate, manually controlled hydraulic circuits, one for actuators P1
arrangements and the other for actuators P2, see Fig. 11. Attempts were made to keep P1 ≈ P2.
The primary spreader beams on the top of the floor were railway rails cut in pieces
slightly shorter than 1,2 m. The friction between the secondary and primary spreader
beams was eliminated by teflon plates.
580 7200
7200
B
4
8 4
44
7 2P 2P 5000
1 2 2P 3
1
6 2P 2
2
5 1
1260 1260 B
A A
2P 4P2 2P
1 1
A A
Fig. 11. Plan. P1 and P2 refer to vertical actuator forces.
P P P P Det A F F
2 2 2P 2P 2 2
1 1
P2 Steel plates
F
Det A Teflon
plates
5000 Gypsum
B B F
Fig. 12. Section B-B, see previous Fig. 13. Detail A, see previous figure.
figure.
105
3 Measurements
3.1
Support
reactions 1
3 load
cells
South
5
Fig. 14. Load cells below the South end of the middle beam.
3.2
Vertical 7200 580 7200
displacement 24 North
9 14 19 29 34 39
45 444
8 13 18 23 28 33 38
6 3
7 12 17 22 27 32 37
7 2
6 11 16 21 26 31 36
8 1
5 10 15 25 30 35
3600 South 3600
20
1400 1400
Fig. 15. Location of transducers 5 … 39 for measuring vertical deflection along lines I … VII.
3.3 North
Average strain 47 46
8 4 Steel bar Transducer
55 53 51
7 (61) (59) (57)
3
54 52 50 L
6 (60) (58) (56)
2
Fig. 16. Apparatus for measuring average
strain. L =1120mm and 1060 mm for the
5 1 bottom and top side transducers,
respectively.
49 48
South
70 70
Fig. 17. Position of device (transducers 50–61) measuring average strain parallel to
the beams. Transducers 46–49 measured the sliding of the outermost slabs along the
beam. Numbers in parentheses refer to the soffit of the floor, others to the top side.
106
3.4
Horizontal. North
displacements
41 40
8 4
7 3
43 42
6 2
5 1
45 44
South
Fig. 18. Transducers 40–45 measuring crack width on the top of the floor.
3.5
Strain To detect longitudinal cracks along strands, the soffit of slabs 2, 3, 6 and 7 was provided
with strain gauges as shown in Figs 19 and 20. The measuring length of the gauges
was 67 mm.
250
40 20
39 19
38 18
50 37 17
50 36 16 3
7
35 15
34 14
33 13
67
32
12
80 30 31 10
11
29 9
28 8
Fig. 19. Strain gauges below strands. 27 7
6 26 6 2
25 5
24 4
23 3
22 2
21 1
Fig. 20. Position of strain gauges. The webs

of the slabs are indicated by dot-and-dash
lines below strands.
-
107
5 Loading strategy
5.1
Load-time Date of the floor test was 24.2.1992
relationship
When the actuator forces Pi were equal to zero but the weight of the loading equipment
was on, all measuring devices were zero-balanced.
The loading history is shown in Fig.21. Note, that the number of load step, not the time,
is given on the horizontal axis.
In the following, the loading until failure (steps 1–29) is called Stage 1 and the post-
failure stage with a support below failed slab 1 (steps 30–38) Stage 2. In Stage 3 the
support was still under slab 1 and the load on slab 2 was removed.
400 Stage 1 Stage 2 Stage 3

350
300 P1
250 P2
Pi [kN]
200
150
100
50
0
0 5 10 15 20 25 30 35 40
Load step
Fig. 21. Development of actuator loads P1 and P2.
The weight of loading equipment per actuator was 1,2 kN and 5,6 kN for actuators
P1 and P2, respectively. Consequently, the imposed load per slab was
F1 = P1 + 1,2 kN for slabs 2, 3, 6 and 7

5.2
After failure See Stages 2 and 3 on the next page.
108
Stage 1 At P1 = 276,8 kN, P2 = 265,3 kN a shear failure took place in slab 1

between the line load and the support, see Appendix A, Fig. 4.
Stage 2 A support was placed below the line load F1 on slab 1 as shown in Fig.
23. The aim was to continue the loading with seven line loads but the
end of slab 2 failed shortly after the reloading was started as shown in
Fig. 24. This failure was obviously due to the load transfer from slab 1
to slab 2 across the vertical joint because the support under slab 1
was not able to carry load before a certain additional deflection of slab
1 had taken place, and this deflection was not possible before slab 2
had failed.
Stage 3 After the failure of slab 2, the actuator on slab 2 was removed. Now
slab 1 was tightly lying on the support below the line load. A shear
failure took place in slab 5 at P1 = 375,0 kN, P2 = 379,0 kN, see
Appendix A, Figs 5–7.
After failure When demolishing the test specimen it was observed that the core
fillings were perfect and the gap between the soffit of the slabs and the
upper surface of the ledges of the middle beam was completely filled
by the grout, see Appendix A, Figs 8–12.
The middle beam looked intact after the failure.
7.1 -
Cracks at
service load
7.2
Cracks after
failure
48 F F 4
2 2
7 F F
7 1 1 3
6 F F 2
1 1
5 F F 1
2 2
Fig. 22. Stage 1. Cracks on the top and at the edges of the floor after failure.
109
8
48 F F4 4
7
7 F F3
7 3
6
6 F F2 2
5 1
5 F F 1
Temporary support
Fig. 23. Stage 2. Cracks after failure of slab 2.
8
48 F F4 4
7
7 F F3
7 3
6
6 F 2
5 F
5 1
Temporary support
Fig. 24. Stage 3. Cracks after failure of slab 5.
The maximum support reaction is regarded as the indicator of failure. The failure most
likely took place before the load values P1 = 276,8 kN, P2 = 265,3 kN were measured
and when the sum of actuator loads load on half floor was = 1085,0 kN. This is
supported by the fact that the efforts to keep all actuator loads equal succeeded well
except at this point. Due to the softening of slab 1 the pressure in actuators P2 was
reduced and the middle beam rose upwards, which resulted in increasing pressure in
actuators P1. This is in accordance with the difference in loads shown in Fig. 21 at steps
22 and 23. In this way the support reaction of the South end of the middle beam was
reduced at the expense of increased support reaction at the North end.
Fig. 25 shows the relationship between the measured support reaction below the South
end of the middle beam and the sum of actuator loads on half floor. The ratio of the
reaction to the load is shown in Fig. 26 and in a larger scale in Fig. 27. Assuming simply
supported slabs gives the theoretical ratio of 0,835. The measured support reaction
seems to follow the theoretical value rather well until load 1015 kN at which the ratio
0,845 is obtained. The next step gives only 0,810 which suggests that slab 1 has
already lost a part of its shear stiffness and the load on it is partly transferred to slab 2.
After failure of slab 1 it is unclear, how much load was transferred via slab 1 to the
middle beam and how much directly to the floor of the hall. Therefore, no definite values
for the shear resistance are given in Stages 2 and 3.
110
Reaction [kN]
900
800
700
600
500
400
300
200
100
0
0 200 400 600 800 1000 1200
Load on half floor [ kN]
Fig. 25. Stage 1. Support reaction measured below South end of the middle beam vs.
load on half floor = 2(P1 + P2 ).
Reaction / load on half floor

0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0,0
0 200 400 600 800 1000 1200
Fig. 26. Ratio of measured support reaction (below South end of the middle beam) to
load on half floor.
Reaction / load on half floor

0,86
0,85
0,84
0,83
0,82
0,81
0,80
0 200 400 600 800 1000 1200
Fig. 27. A part of the previous figure in a large scale.
111
The observed shear resistance of one slab end (support reaction of slab end at failure)
due to different load components is given by
Vobs = Vg,sl + Vg,jc + Veq + Vp
where Vg,sl, Vg,jc, Veq and Vp are shear forces due to the self-weight of slab unit, weight
of joint concrete, weight of loading equipment and actuator forces Pi, respectively.
It is concluded that the maximum support reaction of the failed slab 1 has been at least
Vp = 0,845 x (actuator loads on half floor) /4 = 0,845 x(276,8+265,5) /2 = 229,1 kN. In
the same way, the support reaction due to the weight of the loading equipment has been
0,845x(1,2+5,6)/2 = 2,87 kN. Vg,jc is calculated from the nominal geometry of the joints
and measured density of the grout. When calculating Vg,sl, the measured weight of the
slabs is used. The values of the shear force components are given in Table 1 below.
Table 1. Components of shear resistance due to different loads.
Shear force
Action Load
kN/slab
Weight of slab unit 5,49 kN/m 19,4
Weight of joint concrete 0,19 kN/m 0,7
Loading equipment (1,2+5,6)/2 kN/slab 2,9
Actuator loads (276,8+265,5)/2 kN /slab 229,1
The observed shear resistance Vobs = 252,1 kN (shear force at support) is obtained for
9.1
Component Note
steel MPa MPa
Slab strands J12,5 1630 1860 Nominal (no yielding in test)
Beam strands J12,5 1630 1860 Nominal (no yielding in test)
Nominal value for reinforcing bars
Reinforcement Txy 500
112
9.2
Strength of
floor test mm mm
d
6 50 50 2.4.19920)? Upper flange of slab 1,
Mean strength [MPa] 81,4 (+? d)1) vertically drilled
Density = 2437 kg/m3
0)
This is the date given in the report. It is most likely too late because it is the same as the date
of the core tests for VTT.S.WQ.400.1992 carried out after the present floor test
# Cores h h d Date of test Note

mm mm
d
6 50 50 2.4.19920)? Upper flange of slab 5,
Mean strength [MPa] 84,3 (+? d)1) vertically drilled
0)
This is the date given in the report. It is most likely too late because it is the same as the date
of the core tests for VTT.S.WQ.400.1992 carried out after the present floor test
9.3
Strength of The slabs for the reference tests were taken from the floor test specimen.
slab concrete,
reference tests
9.4
Strength of
concrete in # Cores h h d Date of test Note
middle beam mm mm
d
6 100 100 18.3.1992 Top surface of beam,

9.4
Strength of a
and core filling a mm
a

St.deviation [MPa] 0,58
1)
2)
9.5
Strength of Not measured, nominal value K60
concrete in
end beams
113
10 Measured displacements
Note that the last points on each curve represent the post failure situation.
10.1
Deflections
300
250
200
P2 [kN]
150 5
6
100 7
8
50
9
0
0 1 2 3 4 5 6 7 8 9 10
Deflection [mm]
Fig. 28. Deflection on line I, Western end beam.
300
250
200
P2 [kN]
150 10
11
100 12
13
50
14
0
0 2 4 6 8 10 12 14
Deflection [mm]
Fig. 29. Deflection on line II.
114
300
250
P2 [kN] 200
150 15
16
100 17
18
50
19
0
0 2 4 6 8 10 12 14
Deflection [mm]
Fig. 30. Deflection on line III.
300
250
200
P2 [kN]
150 20
21
100 22
23
50
24
0
0 1 2 3 4 5 6 7 8 9 10
Deflection [mm]
Fig. 31. Deflection on line IV, middle beam.
300
250
200
P2 [kN]
150 25
26
100 27
28
50
29
0
0 2 4 6 8 10 12 14 16 18
Deflection [mm]
Fig. 32. Deflection on line VI.
115
300
250
P2 [kN] 200
150 30
31
100 32
33
50
34
0
0 2 4 6 8 10 12 14 16 18
Deflection [mm]
Fig. 33. Deflection on line VII.
300
250
200
P2 [kN]
150 35
36
100 37
38
50
39
0
0 1 2 3 4 5 6 7 8 9 10
Deflection [mm]
Fig. 34. Deflection on line VIII along Eastern end beam.
400
350
300
250
P2 [kN]
200 5
150 6
7
100
8
50 9
0
-4 -2 0 2 4 6 8 10
Deflection [mm]
Fig. 35. Deflection on line I along Western end beam, stages 1–3.
116
400
350
300
P2 [kN] 250
200 20
150 21
22
100
23
50 24
0
0 2 4 6 8 10 12
Deflection [mm]
Fig. 36. Deflection on line IV along middle beam, stages 1–3.
400
350
300
250
P2 [kN]
200 35
36
150
37
100 38
50 39
0
0 2 4 6 8 10
Deflection [mm]
Fig. 37. Deflection on line VIII along Western end beam, stages 1–3.
300
250
200
P2 [kN]
150 West end

100 East end
50 Middle
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 38. Stage 1. Net deflection of mid-point of beams (rigid body motion = settlement of
beam supports eliminated).
117
The last measured net deflection of the middle beam was 5,4 mm before the highest
load level and the first measured after it was 7,2 mm. Before failure, the net deflection of
the middle beam was 1,4 mm and after the highest load level 1,7–2,2 mm greater than
that of the end beams.
10.2
Crack width
300
250
40
200
41
P2 [kN]
150 42
43
100 44
45
50
0
0,0 0,5 1,0 1,5 2,0
Differential displacement [mm]
Fig. 39. Differential displacement (  crack width) measured by transducers 40–46.
10.3
Average strain
400
(actually
350 50
differential
displacement) 300 51
52
250
53
P2 [kN]
200 54
150 55
100
50
0
-1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,2
Fig. 40. Differential displacement at top surface of floor measured by transducers 50–55.
118
400
350
300
250
P2 [kN]
56
200
57
150 58
100 59
60
50
61
0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
Fig. 41. Differential displacement at soffit measured by transducers 56–61.
10.4
Shear
displacement 400
350
300
250
P2 [kN]
200
46 150
47
100
48
49 50
0
-1,0 -0,9 -0,8 -0,7 -0,6 -0,5 -0,4 -0,3 -0,2 -0,1 0,0
Fig. 42. Stages 1–3. Differential displacement between edge of slab and middle beam.
A negative value means that the slab is moving towards the end of the beam.
119
10.5 Strain
A gradual growth in the measured strain means that there has been a crack, most likely
attributable to the release of the prestressing force, before the loading. A sudden
increase in crack width indicates a new crack. An example of the former and latter
behaviour are illustrated e.g. by transducers 35 and 34, respectively, see Fig. 49.
300 300
250 250
200 200 6
P2 [kN]
P2 [kN]
1
150 2 150 7
100 3 100 8
4 9
50 50
5 10
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
Strain [10-6 ] Strain [10-6 ]
Fig. 43. Strain measured by gauges 1–5. Fig. 44. Strain measured by gauges 6–10.
300 300
250 250
200 11 200 16
P2 [kN]
P2 [kN]
150 12 150 17
13 18
100 100
14 19
50 50
15 20
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
300 300
250 250
200 21 200 26
P2 [kN]
P2 [kN]
150 22 150 27
23 28
100 100
24 29
50 25 50
30
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
120
300 300
250 250
200 31 200 36
P2 [kN]
P2 [kN]
150 32 150 37
33 38
100 100
34 39
50 35 50
40
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
In Figs 51 and 52 the cracks below the soffit of slab 2, 3, 6 and 7 are shown.
250
W20 W10 Visible crack
Visible crack,
W19 W9 not detected
by strain gauge
W18 W8
7 Crack detected
W17
by strain gauge
W7
3
W16
W6
W15 W5
W14
W4
6 W13
2
W3
W12 W2
W11 W1
Fig. 51. Location of cracks observed visually or by strain gauges under load
P2 = 153,2 kN.
121
250
W20 W10 Crack
W19 W9
W8 3
7 W18
W17 W7
W16 W6
W15 W5
W14 W4
6 W3 2
W13
W12 W2
W11 W1
Fig. 52. Location of cracks observed visually or by strain gauges under load
P2 = 251,3 kN.
11 Reference tests
P 1260 Tie beam
a)
40 40
L V
P 1260
b)
40 40
L V
Fig. 53. Layout of reference test. a) With tie beam, slabs 1 and 5. b) Without tie beams,
slabs 7 and 8. For L, see the Table 2.
122
Table 2. Span L, ultimate load Pobs, ultimate shear force Vobs and failure mode in
reference tests. The weight of the loading equipment = 0,5 kN is included in Pobs.
Slab L Tie Pu Vu Failure mode

mm beam kN kN
1 5985 Yes > 600* > 494*
1260
1150
1050
5 5000 Yes 591 461 1260

1230
990
Anchorage failure
7 7110 No 564 487 1260
750
Shear tension failure

8 7115 No 568 490 1260
1150
700

Mean 483

13 Discussion
1. The net deflection of the middle beam due to the imposed actuator loads only
(deflection minus settlement of supports) was 5,4–7,2 mm or L/926–L/694, i.e.
rather small.
2. The shear resistance measured in the reference tests was slightly higher than
the mean of the observed values for similar slabs given in Pajari, M. Resistance
of prestressed hollow core slab against a web shear failure. VTT Research
Notes 2292, Espoo 2005.
3. Before failure, the net deflection of the middle beam was 1,5 mm and right after
the highest load level 1,8–2,2 mm greater than that of the end beams. This is a
too small difference to cause considerable torsional stresses in the slabs.
The failure mode was web shear failure of edge slab 1 close to the middle beam.
The middle beam seemed to recover completely after the failure.
123
APPENDIX A: PHOTOGRAPHS
A124
A125
A126
A127
A128
A129
A130
A131
A132
VTT.S.WQ.400.1992
1.1
Identification
VTT.S.WQ.400.1992 Last update 2.11.2010
and aim
ST400 (Internal identification)
Aim of the test To study the shear resistance of thick hollow core
slabs supported on steel beams.
1.2
Test type
Steel beam Steel beam Steel beam

1.3 VTT/FI 25.3.1992
Laboratory
& date of test
1.4
Name Loading test for 400 mm hollow core floor supported on steel beams
Date 23.4.1993
P.O. Box 1001, FI-02044 VTT.
Financed by Lohja Oy, Finland; NCC Prefab AB, Sweden; Parma Oy,
Finland; Oy Partek Concrete Ab, Finland; Skanska Prefab AB,
Sweden and AB Strängbetong, Sweden. The Finnish companies were
financially supported by TEKES, Finland.

2.1
General plan 7200 306 7200
50 50 North
4
8 44
4
7 3 5000
6 2
5 1
1260 1260 South
Fig. 2. Plan.
133
VTT.S.WQ.400.1992
2.2
End beams Simply supported box girders made of steel. Span = 5,0 m
80 50
Plastic plug
2 T10 
L = 4750
Tie beam 400
Plywood
3 T12 c/c 1200
350
1100
30 30
End beam
300
Fig. 4. Tie reinforcement in joints
200 between adjacent slabs. T12 refers
Fig. 3. Arrangements at end beam. T10 refers
2.3
Middle beam Stiffening plate Stiffening plate
Holes 3  16 1200 1200 1400
60
100 5000 100
290
40
8
Fe 52
30
140 140
586
Fig. 5. Elevation and section of middle beam.
134
VTT.S.WQ.400.1992
2.4
Arrangements 30 50 290 50 30
at middle
beam 4 T10 L = 5150 Hard mineral wool
40
40
8 320
Plastic plug
Fe52
40
140 30
586
60
3 T12 c/c 1200 L = 2500

2.5
1158
Slabs
40
62,1
R125
216
400 197,9
60
60 18
35 40 1200


1158
49,0 Max measured
bond slips: 1,2 and
1,1 in slab 1; 1,0 in
303,54 bw,i b w,i = 290,8 slabs 3 and 7
Measured weight
40,1
of slab units =
38,5 1195 5,51 kN/m
Fig. 9. Mean of most relevant measured geometrical characteristics.
135
VTT.S.WQ.400.1992
2.6
supports - No
2.7
arrangements and the other for actuators P2, see Fig. 10. Attempts were made to keep P1 ≈ P2 so as to
generate two linear line loads on the floor.
The primary spreader beams on the top of the floor were railway rails cut in pieces
slightly shorter than 1,2 m. The friction between the secondary and primary spreader
beams was eliminated by teflon plates.
406
7200 7200
North
B
4
8 4
44
7 2P 2P 5000
1 2 2P 3
1
6 2P 2
2
5 1
A
1260 1260 South B A
2P 4P2 2P
1 1
A A
Fig. 10. Plan. P1 and P2 refer to vertical actuator forces.
P P P P Det A F F
2 2 2P 2P 2 2
1 1
P2 Steel plates
F
Det A Teflon
plates
5000 Gypsum
B B F
Fig. 11. Section B-B, see previous Fig. 12. Detail A, see previous figure.
figure.
136
VTT.S.WQ.400.1992
3 Measurements
3.1
Support
reactions 1
3 load
cells
South
5
3.2
Vertical 7200 406 7200
displacement 24 North
9 14 19 29 34 39
45 444
8 13 18 23 28 33 38
6 3
7 12 17 22 27 32 37
7 2
6 11 16 21 26 31 36
8 1
5 10 15 25 30 35
3600 South 3600
20
1400 1400
Fig. 14. Location of transducers 5 … 39 for measuring vertical deflection.

3.3 47 North 46
Average strain (63) (62)
8 4 Steel bar Transducer
55 53 51
7 (61) (59) (57) 3
54 52 50 L
6 (60)
(58) (56) 2
5 1 strain. L =1120mm and 1060 mm for the
49 48 bottom and top side transducers,
(65) (64) respectively.
South
70 70
Fig. 16. Position of device (transducers 50–61) measuring average strain parallel to the
beams. Transducers 46–49 and 62–65 measured the sliding of the outermost slabs along
the beam. Numbers in parentheses refer to the soffit of the floor, others to the top side.
137
VTT.S.WQ.400.1992
3.4
Horizontal North
displacements
41 40
8 4
7 3
43 42
6 2
5 1
45 44
South
3.5
Strain To detect longitudinal cracks along strands, the soffit of slabs 2, 3, 6 and 7 was provided
with strain gauges as shown in Figs 18 and 19. The measuring length of the gauges
was 67 mm.
160
40 20
39 19
38 18
50 50 17
37
36 16
7 3
35 15
34 14
67 33 13
32
80 12
30 31 10
11
29 9
28 8
Fig. 18. Strain gauges below strands.
27 7
6 26 6 2
25 5
24 4
23 3
22 2
21 1
Fig. 19. Position of strain gauges. The webs

of the slabs are indicated by dot-and-dash
lines below strands.
-
138
VTT.S.WQ.400.1992
5 Loading strategy
5.1
relationship
All measuring devices were zero-balanced when the actuator forces Pi were equal to
zero but the weight of the loading equipment was on.
The loading history is shown in Fig. 20. Note, that the number of load step, not the time,
is given on the horizontal axis.
In the following, the cyclic stage (steps 1–16) and the monotonous stage until failure
(steps 16–41) are called Stage 1.
Stage 1
400
350
300 P1
250 P2
Pi [kN]
200
150
100
50
0
0 5 10 15 20 25 30 35 40 45
Load step
Fig. 20. Development of actuator loads P1 and P2.

139
VTT.S.WQ.400.1992
5.2
After failure After failure, the failed slab 5 was temporarily supported and the load test was
continued. This is called Stage 2.
Stage 2
450
400
350
P1
300
P2
Pi [kN]
250
200
150
100
50
0
0 5 10 15 20 25 30 35
Load step
Fig. 21. Development of actuator loads P1 and P2 in Stage 2.
Stage 1 At P1 = 300,3 kN, P2 = 297,1 corners of slabs 5 and 8 cracked next

to the support of the beam. At P1 = 319,7 kN, P2 = 318,1 kN a narrow
shear crack appeared in the web of slab 5, see Fig. 22. It grew in width
and length when the loads were increased.
At P1 = 337,9 kN, P2 = 334,2 kN a shear failure took place in slab 5

between the line load and the support, see Appendix A, Figs 3–5.
Stage 2 After the failure, a temporary support was placed below slab 5 as
shown in Fig. 23. The aim was to continue the load test with seven line
loads. When the loads were increased, a corner of slab 1 cracked as
shown in Fig. 23. At P1 = 381,1 kN, P2 = 377,2 kN slabs 6 and 7 failed
as shown in Fig. 23 and in Annex A, Figs 7–11.
After failure When demolishing the test specimen it was observed that the interface
between the webs of the middle beam and the slab ends had cracked
neatly along the edge of the beam, see Appendix A, Fig. 6.
The middle beam looked intact after the failure.
140
VTT.S.WQ.400.1992
7.1
Cracks at Not documented.
service load
7.2
Cracks after
failure F2 = 300 kN
48 F2 F2 4
7 F
1
F
1 3
6 F F
1 1 2
5 F2 F2 1
F2 = 300 kN
F2 = 340 kN F2 = 320 kN
Fig. 22. Stage 1. Cracks after failure on the top and at the edges of the floor specimen.
48 F2 F2 4
7 F1 3
F
1
6 F F 2
1 1
5 F5 = F2 F
2 1
Temporary support
under slab
Fig. 23. Stage 2. Cracks after failure of slabs 6 and 7.
The maximum measured support reaction is regarded as the indicator of failure. The
failure took place at P1 = 337,9 kN, P2 = 334,2 kN or F1 = 339,1 kN, F2 = 339,8 kN.
reaction to the load is shown in Fig. 25 and in a larger scale in Fig. 26. Assuming simply
supported slabs gives the theoretical ratio of 0,825. The measured support reaction
seems to be ≈ 0,805 times the load before failure.
141
VTT.S.WQ.400.1992
Reaction [kN]
1100
1000
900
800
700
600
500
400
300
200
100
0
0 200 400 600 800 1000 1200 1400
load on half floor = 2(P1 + P2).
Reaction / Load on half floor

0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0,0
0 200 400 600 800 1000 1200 1400
Fig. 25. Ratio of measured support reaction (below South end of the middle beam)
to actuator loads on half floor.

0,810
0,805
0,800
0,795
0,790
600 700 800 900 1000 1100 1200 1300 1400
142
VTT.S.WQ.400.1992
where Vg,sl, Vg,jc, Veq and Vp are shear forces due to the self-weight of slab unit, weight of
joint concrete, weight of loading equipment and actuator forces Pi, respectively.
It is concluded that the maximum support reaction due to the imposed load of the failed
slab 5 due has been
Vp = 0,805 x (actuator loads on half floor) /4 = 0,805 x (337,9+334,2) /2 = 270,5 kN.
In the same way, the support reaction due to the weight of the loading equipment has
been 0,805x(1,2+5,6)/2 = 2,74 kN. Vg,jc is calculated from the nominal geometry of the
joints and measured density of the grout. When calculating Vg,sl, the measured weight of
the slabs is used. The values of the shear force components are given in Table 1 below.
Shear force
Action Load
kN/slab
Actuator loads (337,9+334,2) /2 kN /slab 270,5
9.1
Component Note
steel MPa MPa
Nominal value for reinforcing bars,
no yielding in test
End beams ≈ 350 Nominal value for Fe 52, no yielding in test
143
VTT.S.WQ.400.1992
9.2
Strength of
floor test mm mm
d
6 50 50 2.4.1992 Upper flange of slab 4,
Tested as drilled2)
# Cores h h d Date of test Note

mm mm
d

Tested as drilled2)
9.3
Strength of The slabs for the reference tests were taken from the floor test specimen.
slab concrete,
reference tests
9.4
Strength of a
and core filling a mm
a

St.deviation [MPa] -
9.5
Strength of -
concrete in
end beams
1)
2)
144
VTT.S.WQ.400.1992
Note that the last points on each curve represent the post failure situation.
10.1
Deflections
350
300
250
P2 [kN]
200
5
150 6
100 7
8
50 9
0
0 2 4 6 8 10 12 14 16 18 20
Deflection [mm]
350
300
250
P2 [kN]
200
10
150 11
100 12
13
50 14
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Deflection [mm]
145
VTT.S.WQ.400.1992
350
300
250
P2 [kN]
200
15
150 16
100 17
18
50 19
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Deflection [mm]
Fig. 29. Deflection on line III.
350
300
250
P2 [kN]
200
20
150 21
100 22
23
50 24
0
0 2 4 6 8 10 12 14 16 18 20
Deflection [mm]
Fig. 30. Deflection on line IV, middle beam.
350
300
250
P2 [kN]
200
25
150 26
100 27
28
50 29
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Deflection [mm]
Fig. 31. Deflection on line VI.
146
VTT.S.WQ.400.1992
350
300
250
P2 [kN]
200
30
150 31
100 32
33
50 34
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Deflection [mm]
Fig. 32. Deflection on line VII.
350
300
250
P2 [kN]
200
35
150 36
100 37
38
50 39
0
0 2 4 6 8 10 12 14 16 18 20
Deflection [mm]
Fig. 33. Deflection on line VIII along Eastern end beam.
350
300
250
P2 [kN]
200
150 West end

100 Middle
East End
50
0
0 2 4 6 8 10 12 14 16
Deflection [mm]
Fig. 34. Stage 1. Net deflection of mid-point of beams (rigid body motion = settlement of
beam supports eliminated).
147
VTT.S.WQ.400.1992
The last measured net deflection of the middle beam before failure was 14,6 mm.
Before failure, the net deflection of the middle beam was 1,4 mm and after the highest
load level 2,2–2,5 mm greater than that of the end beams.
10.2
Crack width
350
300
250
40
P2 [kN]
200
41
150
42
100 43
44
50
45
0
0,0 0,5 1,0 1,5 2,0 2,5
10.3
Average strain
350
(actually
differential 300
displacement)
250
P2 [kN]
200
50
51 150
52
100
53
54 50
55
0
-1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0
148
VTT.S.WQ.400.1992
350
300
250
56
P2 [kN]
200
57
150 58
100 59
60
50 61
0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
10.4
Shear 350
displacement
300
250
P2 [kN]
200
46
150
47
62 100
63
50
0
-1,8 -1,6 -1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,2
Fig. 38. North end of middle beam. Differential displacement between edge of slab and
middle beam. A negative value means that the slab is moving towards the end of the
beam.
149
VTT.S.WQ.400.1992
350
300
250
200
P2 [kN]
48
49 150
64 100
65
50
0
-1,8 -1,6 -1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,2
Fig. 39. South end of middle beam. Differential displacement between edge of slab and
middle beam. A negative value means that the slab is moving towards the end of the
beam.
10.5 Strain
A gradual growth in the measured strain means that there has been a crack before the
loading, most likely attributable to the release of the prestressing force. A sudden
increase in crack width indicates a new crack. An example of the former and latter
behaviour are illustrated e.g. by transducers 26 and 30, respectively, see Fig. 45.
350 350
300 300
250 1 250
P2 [kN]
P2 [kN]
200 2 200 6
150 3 150 7
100 4 100 8
50 5 50 9
0 0 10
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
350 350
300 300
250 11 250 16
P2 [kN]
P2 [kN]
200 12 200 17
150 13 150 18
100 14 100 19
50 15 50 20
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
150
VTT.S.WQ.400.1992
21
350 350
22
300 300
23
250 250
24
P2 [kN]
P2 [kN]
200 200 26
25
150 150 27
100 100 28
50 50 29
0 0 30
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
31
350 350
32
300 300
33
250 250 36
34
P2 [kN]
P2 [kN]
200 200 37
35
150 150 38
100 100 39
50 50 40
0 0
-500 0 500 1000 1500 2000 -500 0 500 1000 1500 2000
In Figs 48 and 49 the cracks below the soffit of slabs 2, 3, 6 and 7 are shown.
190
Visible crack
W20 W10 Crack detected
by strain gauge
W19 W9
7
W18 W8 3
W17 W7
W16 W6
W15 W5
W14 W4
6
W3 2
W13
W12 W2
W11 W1
Fig 48. Cracking of bottom surface of hollow core units no 2, 3, 6 and 7 along the
strands under force F2 = 153.2 kN. The cracking limit for the strain gauge measurements
is 0,03%. The length of the broken line corresponds to the length of the visible crack.
151
VTT.S.WQ.400.1992
190
W20 W10 Crack detected
by strain gauge
W19 W9
7
W18 W8 3
W17 W7
W16 W6
W15 W5
W14 W4
6
W3 2
W13
W12 W2
W11 W1
Fig. 49. Cracking of bottom surface of hollow core units no 2, 3, 6 and 7 along the
strands under force F2 =330 kN.
11 Reference tests
P 1260 Tie beam
40 40
L V
Fig. 50. Layout of reference tests on slabs 4 and 8 taken from the floor test specimen.
For L, see Table 2.
152
VTT.S.WQ.400.1992
Table 2. Span L, ultimate load Pu, ultimate shear force Vu and failure mode in reference
tests. The weight of the loading equipment = 0,5 kN is included in Pu.
Slab L Pu Vu Failure mode

mm kN kN
4 7000 582 499,8
1260
440
8 7000 622 532,8 1260
350
Mean 516,3
13 Discussion
(deflection minus settlement of supports) was 14,6 mm or L/342
2. The shear resistance measured in the reference tests was somewhat higher
than the mean of the observed values for similar slabs given in Pajari, M.
Resistance of prestressed hollow core slab against web shear failure. VTT
Research Notes 2292, Espoo 2005. This may be due to the tie beam which
made the shear stresses in the webs more uniform than those in the tests
without tie beams.
3. Before failure, the net deflection of the middle beam was 2,2–2,5 mm greater
than that of the end beams. This is a too small difference to cause
considerable torsional stresses in the slabs.
4. The failure mode was web shear failure of edge slab 5 close to the middle
beam. The middle beam seemed to recover completely after the failure.
153
VTT.S.WQ.400.1992
A154
VTT.S.WQ.400.1992
A155
VTT.S.WQ.400.1992
A156
VTT.S.WQ.400.1992
A157
VTT.S.WQ.400.1992
A158
VTT.S.WQ.400.1992
A159
VTT.S.WQ.400.1992
A160
VTT.S.WQ.400.1992
A161
VTT.PC.InvT-Unif.265.1993
1.1
Identification VTT.PC.InvT-Unif.265.1993 Last update 2.11.2010
and aim
PC265E (Internal identification)
Aim of the test To study the effect of core filling with length equal to
the core height.
1.2
Test type
Prestressed
concrete beam
& date of test
1.4
Name 265 mm hollow core floor supported on prestressed concrete beam
under evenly distributed load
Date 7.3.1993
P.O. Box 1001, FI-02044 VTT.
Financed by Parma Oy, Finland; Oy Partek Concrete Ab, Finland;

Skanska Prefab AB, Sweden and AB Strängbetong, Sweden

2.1
General plan 6000 50
5000
5000
750
750
750
375
480
50
50
Fig. 2. Plan.
163
2.2
End beams 240 80 50
2 T8 
12 L = 4750
Tie beam
230 7,5 265
Plywood
12 Plastic plug
30 30 End beam
230
HE240A
Fig. 3. Cross-section of
end beam. 240 Fe 52
Fig. 4. Arrangements at end beam. T8 refers to a

reinforcing bar with diameter 8 mm, see 2.3.
Simply supported, span = 5,0 m,

fy  350 MPa (nominal fy), did not yield in the test.
2.3
Middle beam Prestressed concrete beam, see Figs 5–7.
Concrete: K60
Passive reinforcement and tendons:

T8: Hot rolled, weldable rebar A500HW,  = 8 mm, see 9.1
J12,5: strand with 7 indented wires,  =12,5 mm, Ap = 93 mm2, see 9.1
Tie reinforcement: Straight rebars in the tie beams, across the middle beam and in the
longitudinal joints of the slabs, see Figs 4 and 7.
Fig. 5. Middle beam.
15
Shear keys 125x125
c/c 300
4 J 12,5
3 holes c/c 1200
 = 1100 MPa
po
30
14 J 12,5 45
 =1100 MPa 46
po 55 6 x 50 70 6 x 50 55
Fig. 6. Section of middle beam. See also Fig. 7.
164
2.4
at middle - Joints and tie beams grouted on the 17th of December 1992
beam - The concrete filling in the hollow cores was cast via the end of the slab; neither slots
nor holes in the slab elements were made.
185 50 480 Mineral

185 wool
40
265 265
185
3 holes c/c 1200
40 2 T8 L = 5150 30
100 Neoprene 20x10 2 T8 L = 5150 3 T8 L=2000
c/c 1200
25 2055 50 150
780
2.5
Slabs 1160
40
bw,i
265
185
40 35 1200  b w,i = 235
Fig. 8. Nominal geometry of slab units. Nominal length = 6000 mm.
- 10 slabs extruded 7.12.1992 by Partek Betoniteollisuus Oy, Hyrylä factory,

all from the same casting bed and casting lot
- Grade of concrete K60
- Slabs 1–8 for floor test, slabs 9–10 for reference tests
J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2, see 9.1.
1158 Max measured bond slips:

44,2 1,8;1,7&1,6; 1,5 and 1,4 mm
in slabs 10; 3; 6 and 9,
bw,i respectively
180,6

units = 4,12 kN/m
38,2 41,2  b w,i = 222,4
1194

characteristics.
2.6
Temporary - No
supports
165
2.7
Loading 64 short I-beams (primary spreader beams) were placed on the floor as shown in Fig.
arrangements 10. The length of these beams was 550 mm. There was gypsum mortar between the
slabs and the floor to smoothe the uneven top surface of the slabs. The loads from 16
actuators were distributed to the primary beams with the aid of 16 tertiary spreader
beams and 32 secondary spreader beams as shown in Fig. 10. To eliminate the
contribution of the loading equipment to the load-carrying mechanism of the floor, two
teflon plates were placed below one end of each secondary and tertiary beam, see
Appendix A, Figs 2 and 3.
The actuators were connected to three hydraulic circuits which were controlled
separately to create the same force in each actuator. Eight of them (P1 and P2) were of
long type with a swivel at both ends, eight (P3) were of short type with a swivel on the
top. Attempts were made to keep all actuator forces equal during the test. The total
imposed load due to the actuator force Pi and weight of underlying loading equipment
was
F1 = P1 + 1,4 kN
F2 = P2 + 1,4 kN
F3 = P3 + 1,8 kN
6000 50
P P P P
1 1 2 2
P P P P
3 3 3 3
5000
P P P P
3 3 3 3
P P P P
1 1 2 2
A A
750
375
375
375
375
375
480
50
50
2P + 2 P 2P1 + 2 P 2P + 2 P 2P + 2 P
1 3 3 2 3 2 3
Tertiary
Secondary
Primary
A A
Fig. 10. Plan. P1, P2 and P3 refer to vertical actuator forces.
166
3 Measurements
3.1
Support
reactions 1
3 load
cells
South
5
3.2 III
6000 6000
Vertical
I North IV V
displacement II
9 14 19 24 29
48 4
8 7 13 18 23 28
3
5000
7 12 17 22 27
6 2
6 11 16 21 26
5 1
5 10 15 20 25
30 South 30
3000 3000 3000 3000
Fig. 12. Location of transducers 5 … 29 for measuring vertical deflection along lines I … V.
3.3
Average strain 47 North 46
(63) (62)
Steel bar Transducer
8 150 4
55 53 51
7 (61) (59) (57)
3
L
54 52 50
6 (60)
(58) (56)
2 Fig. 13. Apparatus for measuring average
bottom and top side transducers,
5 1 respectively.
49 48
(65) (64)
South
50 50
beams. Transducers 46–49 and 62–65 measured the sliding of the outermost slabs
along the beam. Numbers in parentheses refer to the soffit of the floor, others to the top.
167
3.4
Horizontal. North
displacements
41 40
8 4
7 3
43 42
6 2
5 1
45 44
South
Fig. 15. Position of transducers 40–45 measuring crack width.
3.5
Strain
-
5 Loading strategy
5.1
Load-time Date of test was 18.1.1993. When the actuator forces Pi were equal to zero but the
relationship weight of the loading equipment was on, all measuring devices were zero-balanced.
The loading history is shown in Fig.16. Note, that the number of load step, not the time,
is given on the horizontal axis. The whole load test took two hours and five minutes.
In the following, the cyclic stage (steps 0–16) is called Stage I, the monotonous stage
from increment 16 to failure is called Stage II.
90
80
70
60
Pi [kN]
50 P1
40 P2
30 P3
20
10
0
0 10 20 30 40 50
Load step [mm]
Fig. 16. Development of actuator forces P1, P2 and P3.
168
5.2
After failure
Stage I -
Stage II At Pi = 40 kN soffit of the middle beam cracked. At Pi = 45 kN cracks in

the corners of slabs 2 and 7 as well as a longitudinal crack along a hollow
core in the soffit of slab 3 were observed, see Figs 17 and 18.
At Pi = 75 kN one corner of slab 4 cracked as shown in App. A, Fig. 6.

When increasing the loads, diagonal shear cracks appeared in the webs
of slab units; first in slab 4 (Figs. 7 and 8 in App. A), then in slab 8 (Figs
15 and 16 in App. A) and finally also in slab 1 (Fig. 17 in App. A).
At P1 = 90,0 kN, P2 = 87,6 kN P3 = 86,8 kN a web shear failure took place

in slab 4 close to the middle beam. The cracking patterns after the failure
are shown in Figs 17 and 18 and in App. A, Figs 9–18 and 21–23.
After When demolishing the test specimen it was observed that all core fillings
failure were perfect, see App. A, Figs 19–20.
7.1
Cracks at
service load
7.2
Cracks after The vertical cracking in the joint concrete next to the middle beam typically took place
failure along the edges of the middle beam, not along the joint concrete or along the slab ends.
i P = 75 kN
8 44
7 3
6 2
5 11
Fig. 17. Cracks after failure on the top and at the edges.
169
8 4
7 3
?
6 Pi =45kN 2
5 1
The maximum support reaction is regarded as the indicator of failure. The failure took
place when P1 = 90,0 kN, P2 = 87,6 kN P3 = 86,8 kN or when the sum of actuator loads
load on half floor was = 702,4 kN.
reaction to the load is shown in Fig. 20 and in Fig. 21 in a larger scale. Assuming simply
supported slabs the ratio = 0,756 is obtained, and the measured support reaction seems
to follow this assumption well until at load 646 kN (Pi = 80 kN) the sudden softening of
slab 4 at the North end of the middle beam resulted in load transfer from the North to the
South end of the middle beam. At failure the ratio equalled 0,78, but the failure took
place at the opposite (North) end of the beam where the support reaction must have
been lower. For this reason it seems justified to use the value of 0,756 for calculation of
the shear resistance.
600
500
Reaction [kN]
400
300
200
100
0
0 100 200 300 400 500 600 700 800
Load on half floor [mm]
Fig. 19. Support reaction measured below South end of the middle beam vs. load on
half floor = 2(P1 + P2 ) + 4 P3.
170
1,0

0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0,0
0 100 200 300 400 500 600 700 800
Load on half floor [kN]
load on half floor.
0,79
0,78
0,77
0,76
0,75
0 100 200 300 400 500 600 700 800
joint concrete, weight of topping concrete, weight of loading equipment and actuator
forces Pi, respectively.
The shear force at failure is calculated assuming that the slabs behave as simply
supported beams. For Veq and VP this means that Veq = 0,756x(8x1,4 +8 x1,8)/8 = 2,4 kN
and Vp = 0,756x(4xP1 + 4xP2 +8xP3)/8. Vg,jc is calculated from the nominal geometry of
the joints and measured density of the concrete, other components of the shear force
are calculated from measured loads and weights. The values for the components of the
shear force are given in Table on the next page.
171
Table. Components of shear resistance due to different loads.
Shear force
Action Load
kN
Loading equipment 3,2 kN 2,4
Actuator loads (4x90,0+4x87,6+8x86,8)/8 kN 132,7
9.1
Component Note
steel MPa MPa
End beams  350 Nominal (Fe 52)
Slab strands J12,5 >1570 >1770 Nominal (no yielding in test)
Beam strands J12,5 >1570 >1770 Nominal (may have yielded in test)
A500HW (no yielding in test)
9.2
Strength of
floor test mm mm
d

1)
Mean strength [MPa] 72,9 (+7 d) tested as drilled2)
St.deviation [MPa] 4,5 Density = 2440 kg/m3
9.3
slab concrete,
reference tests
9.4
Strength of a
grout in # a Date of test Note
longitudinal a mm
joints of slab a
units
1)
Mean strength [MPa] 31,3 (+0 d) conditions as the floor specimen
172
9.5
Strength of a
concrete in the # a Date of test Note
middle beam a mm
a

1)
2)
In the following figures, F stands for the average actuator load + the average weight of
loading equipment per actuator.
10.1
Deflections
90
80
70
60 5
F [kN]
50 6
40 7
30 8
20 9
10
0
0 5 10 15 20 25 30 35 40 45
Deflection [mm]
Fig. 22. Deflection measured by transducers 5–9.
90
80
70
60 10
F [kN]
50 11
40 12
30 13
20 14
10
0
0 5 10 15 20 25 30 35 40 45
Deflection [mm]
173
90
80
70
60 15
F [kN]
50 16
40 17
30 18
20 19
10
0
0 5 10 15 20 25 30 35 40 45
Deflection [mm]
90
80
70
60 20
F [kN]
50 21
40 22
30 23
20 24
10
0
0 5 10 15 20 25 30 35 40 45
Deflection [mm]
90
80
70
60 25
F [kN]
50 26
40 27
30 28
20 29
10
0
0 5 10 15 20 25 30 35 40 45
Deflection [mm]
174
90
80
70
60
F [kN]
50 7
40 27
30 17
20
10
0
0 5 10 15 20 25 30 35 40 45
Net deflection [mm]
Fig. 27. Net mid-point deflection (measured deflection – settlement of supports) for end
beams (7 and 27) and middle beam (17).
10.2
Crack width
90
80
70
40
60
41
F [kN]
50
42
40
43
30
44
20
45
10
0
0 5 10 15 20 25
Fig. 28. Differential displacement (  crack width) measured by transducers 40–45, see
Fig. 15.
90
80
70
40
60
41
F [kN]
50
42
40
43
30
44
20
45
10
0
0 1 2 3 4 5
Fig. 29. A part of the previous figure in a larger scale.
175
10.3
Average strain
(actually 90
differential 80
displacement) 70
F [kN] 50 60
51 50
52 40
53 30
54 20
55 10
0
-3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5
Fig. 30. Differential displacement measured by transducers 50–55, see Fig. 12.
90
80
70
60 56
F [kN]
50 57
40 58
30 59
20 60
10 61
0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 5,0
To get a rough estimate of the strains before failure, the measured results on the top
and at the bottom of the middle beam are shown in Fig. 32. The displacements were
actually measured 10–20 mm outside the top and bottom surface of the floor but this is
ignored in Fig. 32.
176
51 54
52 53
10 50
Slabs
265
Beam
56
60
10
58 61 57
100 59
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0-1.5 -2.0 -2.5 -3.0
DIFFERENTIAL DISPLACEMENT [mm]
Fig. 32. Differential displacement measured by transducers 50–61 at failure. The values
measured on the top and at the bottom are connected with a straight line.
Dividing the values given in Fig. 32 on the top by 1060 mm and those at the bottom by
1120, the value of average strain is obtained. Multiplying the average strain by 1,07, the
strain change at the mid-section of the beam due to the actuator loads is obtained.
Hence, the change in concrete strain on the top of the beam is
 c  2 ,55 / 1060 1,07  0 ,26%
and the change in the stress of the lower tendons of the beam is
 p  E p  p  (190GPa)  3,2/1120 1,07  580MPa
Taking into account the low prestress level 1100 MPa, the losses of prestress, the
elastic deformation due to the release of the prestressing force as well as the short span
of the beam and the low self-weight of the slabs, it is clear that the stress in the lowest
tendons must have been below 1000 MPa before the actuator loads were applied. This
means that the stress in the tendons must have been of the order of yield stress or a bit
lower. It can also be concluded that the compression zone of the beam could still carry a
higher load. So, the yielding of the beam could not be the reason to the failure.
10.4
Shear 90
displacement 80
70
60
F [kN]
46 50
47 40
48 30
49 20
10
0
-14 -12 -10 -8 -6 -4 -2 0
Fig. 33. Sliding of upper edge of slabs 1, 4, 5 and 8 along middle beam, see Fig. 12.
177
90
80
70
F [kN] 60
46 50
47 40
48 30
49 20
10
0
-2,0 -1,5 -1,0 -0,5 0,0
90
80
70
60
F [kN]
62 50
63 40
64 30
65 20
10
0
-1,5 -1,0 -0,5 0,0 0,5 1,0
Fig. 35. Sliding of lower edge of slabs 1, 4, 5 and 8 along middle beam. Cracking of the
corner of slab 4 at F = 70 kN makes the curve of transducer 62 meaningless after that
point. The curve of transducer 63 seems to be incorrect from the beginning.
178
90
80
70
F [kN] 60
50
46-62, slab 4 40
47-63, slab 8 30
48-64, slab1 20
49-65, slab 5 10
0
-1,0 -0,8 -0,6 -0,4 -0,2 0,0
Shear displacement [mm]
Fig. 36. Shear displacement (= differential displacement between upper and lower edge
of slabs 1, 4, 5 and 8) in beam's direction, see Fig. 12. After F = 70 kN the curve for slab
4 reflects primarily the rigid body motion of a loose piece of concrete. Due to the errors
in transducer 63, the curve for slab 8 has no physical meaning.
Fig. 36 shows that the core filling could not totally eliminate the transverse horizontal
shear deformation of the slab ends.
10.5 Strain
-
11 Reference tests
Two tests were carried out: R9 (slab 9) and R10 (slab 10).
Teflon
6000
8
6 5 4
a)
P P 1200
2 3 7
1
750
375
375
375
375
375
P P
1 (6) 2 (5) 3 (4)

b)
7 (8)
6000 Teflon
30 30 30 30
Fig. 37. Layout of reference test. a) Plan. b) Elevation. There was gypsum mortar
between the lowest spreader beams ant the top surface of the slab. Transducers 1–6
measured vertical displacement of the top surface, transducers 7 and 8 the slippage of
the outermost strands.
179
180
160
140
120
P [kN]
100
80 R9
60 R10
40
20
0
0 5 10 15 20 25 30 35 40 45
Load step [mm]
Fig. 38. Loading.
Failure
R9: Bending
crack
crack
R10: Failure
crack
Fig. 39. Failure modes in reference tests. The bending crack in test R9 did not
contribute to the actual failure. See also Appendix A, Figs 24–28.
180
160
140
120
P [kN]
100
80 R9
60 R10
40
20
0
0 10 20 30 40 50 60 70
Deflection [mm]
Fig. 40. Midpoint deflection from which the rigid body motion due to the settlement of the
supports has been eliminated. P is the actuator force.
180
180
160
140
120
P [kN]
100 R9/7
80 R9/8
60 R10/7
R10/8
40
20
0
0 1 2 3 4
Slippage [mm]
Fig. 41. Slippage of outermost strands measured by transducers 7 and 8.
180
160
140
120
P [kN]
100 R9/7
80 R9/8
60 R10/7
R10/8
40
20
0
0,00 0,02 0,04 0,06 0,08 0,10 0,12 0,14
Bond slip [mm]
the slab, actuator force P at failure, weight of loading equipment Peq, total shear force
Span Vg P Peq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
R10 22.1.2002 5940 12,2 160,0 1,4 255,1 212,5 Flexural shear failure
Mean 251,8 209,8
181
13 Discussion
1. The failure mode was web shear failure of edge slabs. The prestressed
concrete beam seemed to recover completely after the failure.
2. The obtained shear resistance was 43% higher than that observed in test
VTT.PC.InvT.265.1990. The filled hollow cores and the uniformly distributed
load over half floor made the difference in the present test. It is difficult to say,
to which extent the enhanced resistance was attributable to each of these
differences.
3. The net deflection of the middle beam due to the imposed actuator loads (deflection
minus settlement of supports) was 39 mm or L/128. It was 17–19 mm greater
than the deflection of the end beams. Hence, the torsion due to the different
deflection of the middle beam and end beams has to be taken into account
when analyzing the test result.
4. The measured strains on the top and at the bottom of the middle beam suggest
that the middle beam was not yielding when the failure took place.
182
VTT.CP.InvT-Unif.265.1993
A183
A184
A185
A186
A187
A188
A189
A190
A191
A192
A193
A194
A195
A196
A197
VTT.PC.InvT-Topp.265.1993
1.1 VTT.PC.InvT-Topp.265.1993 Last update 2.11.2010
Identification
PC265T (Internal identification)
and aim
Aim of the test To study the effect of reinforced concrete topping on
the shear resistance of hollow core slabs supported
on a beam. A prestressed concrete beam was
chosen because, according to previous tests, the
interaction between the slabs and concrete beam
was stronger than with steel beam.
1.2
Test type
Prestressed
concrete beam
& date of test
1.4 Author(s) Koukkari, H.
Test report Name Loading test on 265 mm hollow core floor with topping supported on
prestressed concrete beam
Date 15.12.1993
P.O. Box 1001, FI-02044 VTT.
Financed by Finnish Association of Building Industry RTT, the Inter-
national Prestressed Hollow Core Association IPHA, KB Kristianstads
Cementgjuteri, Sweden, Skanska Prefab AB, Sweden

2.1
General plan See Figs 1 and 10 and Appendix A, Figs 3 and 4.
2.2
End beams 240
12
230 7,5
12
Fig. 3. Arrangements at end beam. T8 refers to a

Fig. 2. Cross-section of end beam. reinforcing bar with diameter 8 mm, see 2.3.
199
Simply supported, span = 5,0 m
2.3
Middle beam Prestressed concrete beam, see Figs 4–6 and Appendix A
Concrete: K60
Passive reinforcement and tendons:

Txy: Hot rolled, weldable rebar A500HW,  = xy mm, see 9.1
J12,5: strand with 7 indented wires,  =12,5 mm, Ap = 93 mm2
B5: Cold formed rebar B500K,  = 5 mm, see 9.1
Tie reinforcement: Straight rebars in the tie beams, across the middle beam and in the
longitudinal joints of the slabs, see Figs 3 and 6.
15
Shear keys 125x125
c/c 300
4 J 12,5
3 holes c/c 1200
 = 1100 MPa
po
30
14 J 12,5 45
 =1100 MPa 46
po 55 6x 50 70 6 x 50 55
Fig. 5. Section of middle beam. See also Fig. 6.
2.4
at middle - Joints and tie beams grouted on the 24th of September 1993
beam - The top surface of the slab units and the middle beam were cleaned and moistened
before casting the topping concrete on the 29th of September 1993
- No special measures to enhance the bond between the topping and the underlying
concrete
200
Mesh reinforcement B5 #150

50 50 480 50
Topping concrete K40
60
40
Plastic
plug 265
185
40 2 T 8 L = 5150 30
100 Neoprene 20x10 2 T8 L = 5150
3T 8 L=2000
2520 50 150
c/c 1200
55 780
48 4
7 3
5000
6 2
5 1
Fig. 7. Overlapping of mesh reinforcement.
2.5
Slabs 1160
40
bw,i
265
185
40 35 1200  b w,i = 235
- Extruded 15.9.1993 by Partek Oy, Hyrylä factory, grade of concrete K60

J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2, see 9.1.
201

44,3 2,0 in slab 7; 1,8 in slab 10
and 1,6 in slab 4,
bw,i
179,1
units = 4,18 kN/m
38,8 34,2  b w,i = 222
1194

characteristics, slabs 1–8.
2.6
Temporary -
supports
2.7 There were two separate hydraulic circuits, one for actuator loads P1 and the other for
Loading actuator loads P2. See Fig. 10 and App. A, Figs 3 and 4 for the loading arrangements.
arrangements
Fig. 10. Plan. P1 and P2 refer to vertical actuator forces. det 1: roller bar, det 2: hinge.
The thickness of the topping was 60 mm in those corners of slabs 1, 4, 5 and 8 which
where on the South and North edge of the floor specimen.
202
F F
P2 Steel plates
F
Teflon
plates
Gypsum
F
550
Fig. 11. Bearing above the spreader beams on slabs 1, 4, 5 and 8. The hydraulic
actuators (P2) prevented the lateral motion of the square steel tube beams
(see App. 1, Figs 3 and 4) and the friction at the ends was eliminated by teflon plates.
3 Measurements
3.1
Support The support reaction due to the actuator loads below the Southern end of the middle
reactions beam was measured by three load cells, see Fig. 12.
1
3 load
cells
South
Fig. 12. Load cells below one end of the middle beam.
3.2
Vertical 30 30
5970 5970
displacement
North III IV V
I II
9 14 19 24 29
48 4
8 13 18 23 28
7 3
5000
7 12 17 22 27
6 2
6 11 16 21 26
5 5 10 20 1 25
15
30 2970 3000 South 3000 2970 30
Fig. 13. Location of transducers 5–29 for measuring vertical deflection along lines I–V.
3.3
Sliding of slab The sliding of the slab ends along the middle beam was measured using eight
(horizontal transducers (46–48 & 62–65). Their position is shown in Fig. 16. See also Figs 7 and
displacement) 10–13 in App. A.
203
3.4
Crack width The differential displacement reflecting the width of vertical cracks in the topping was
(horizontal measured by six transducers located as shown in Fig. 14.
displacement)
Fig. 14. Transducers (40–45) measuring crack width between the middle beam and the
slab ends.
3.5
Average strain Average strain was measured using 12 devices of the type shown in Fig. 15. See Fig 16
for the position of the transducers.
Fig. 15. Apparatus for measuring average strain. L = 1120mm and 1060 mm for the top
and soffit transducers (50–55 and 56–61), respectively.
Fig. 16. Position of transducers 50–61 measuring average strain parallel to the beam
and position of transducers 46–48 & 62–65 measuring sliding of slabs along the beam.
The numbers in parentheses refer to lower transducers.
204
- none
5 Loading strategy
5.1
Load-time The date of the floor test was 11.10.1993
relationship
The imposed load Fi on each hollow core slab was equal to Pi + Peq, i
where Pi is the actuator load P1 or P2 shown in Fig. 10 and Peq,i the load due to the self
weight of the loading equipment. So
F1 = P1 + 1,2 kN for slab units 2, 3, 6 and 7

F2 = P2 + 5,6 kN for slab units 1, 4, 5 and 8
When the actuator forces Pi were equal to zero but the weight of the loading equipment
was on, all measuring devices were zero-balanced.
The loading history is shown in Fig.17. Note, that the number of load increment, not the
time, is given on the horizontal axis. The whole test took roughly two hours.
In the following, the cyclic stage (increments 0–16) is called Stage I, the monotonous
stage from increment 16 to failure is called Stage II.
160
140
120
100
F [kN]
80
F1
60
F2
40
20
0
0 10 20 30 40 50
Number of load increment
Fig. 17. Loading history.
5.2
After failure
205

Pi Observations Cracking pattern
kN
Vertical cracks in the flange of the middle topping
55 (1) beam, under slab 6
flan ge
Vertical cracks in the longitudinal edges

55 (1) of the topping, above the middle beam 5 1
ends
One vertical crack in the tie beam, above topp in g

55(1) the end beam tie b ea m
H E-beam
Several vertical cracks in the middle

55(3) beam ends
A vertical crack in the longitudinal edge of

55(5) the topping 5 1
Cracking in the middle beam flange grew to p pin g

60 down to the bottom of the middle beam
fla n ge
Cracks on the surface of the topping

80 along the joint between the middle beam see Fig. 18
and slabs 1–4
A transverse crack in the soffit of the 2
80 middle beam
bottom o f
m id dle be am
5 6
90 The cracks in the end tie beams reached

the surface ot the topping
A vertical crack in the longitudinal edge of Pi =55 kN
90 the topping starting from the corner of the
middle beam 4 8
100 A vertical crack in the concrete tie beam,

between slabs 5 and 6
A vertical crack in the joint between slab 1
100 and the middle beam, next to the beam 5 1
206
A vertical crack in the joint between the

110 slab 5 and the middle beam, 5 1
next to the beam
A vertical crack in the longitudinal edge of

115 the topping starting from the corner of the 5
middle beam
1
The concrete topping became loose

115 above the middle beam, between slabs 5 1
1 and 5
120 Diagonal cracks developed in edges of slabs 1 and 4 near the middle beam
and a vertical crack along the ends of slabs 5–8 appeared in the topping.
After a while diagonal cracks developed in edges of slabs 5 and 8 near the
middle beam.
135 A shear failure took place in slabs 4 and 8.
7.1
Cracks at
service load
7.2
Cracks after
failure
8 4
7 3
6 2
5 1
4 8 5 1
1000 1000 1000 1000
207
8 4
7 3
6 2
5 1
The maximum support reaction is regarded as the indicator of failure.
500
450
400
350
Reaction [kN]
300
250
200
150
100
50
0
0 100 200 300 400 500 600
half floor = 2(P1 + P2).
208
1,0
0,9
0,8
Reaction / Load 0,7
0,6 Monotonous
0,5 Cycle 1
0,4
0,3
0,2
0,1
0,0
0 100 200 300 400 500 600
Load = 2(P1+P2) [kN]
load on half floor = 2(P1 + P2).
Vobs = Vg,sl + Vg,jc + Vg,top + Veq + Vp
where Vg,sl, Vg,jc, Vg,top, Veq and Vp are shear forces due to the self-weight of slab unit,
weight of joint concrete, weight of topping concrete, weight of loading equipment and
actuator forces Pi, respectively.
The shear force due to the the self-weight of the structure is calculated assuming that
the slabs behave as simply supported beams. Veq and VP are calculated using the
measured relationship between the support reaction of the beam and the loads. This
means that Veq = 0,883xPeq and Vp = 0,883x(P1.+ P2)/2. Vg,jc and Vg,top are calculated
from the nominal geometry of the joints, nominal thickness 60 mm of the topping and
measured density of the concrete, other components of the shear force are calculated
from measured loads and weights. The values for the components of the shear force are
given in Table below.
Shear force
Action Load
kN
Loading equipment (1,2+5,6)/2 kN 3,0
Weight of topping 1,58 kN/m 4,7
Actuator loads 271,6 (135,4+136,3)/2 kN 119,9
one slab unit with width = 1,2 m. The shear force per unit width is vobs = 116,9 kN/m
209
9.1
steel: Component Note
MPa MPa
End eams  350 Nominal (Fe 52, no yielding in test)
Strands J12,5 >1570 >1770 Nominal (no yielding in test)
Reinforcement B5 500 Nominal (B500K)
Reinforcement T8 500 Nominal (A500HW)
9.2
Strength of
slab concrete, # Cores
h
h d Date of test Note
floor test
d mm mm
6 50 50 25.10.1993 Upper flange of slabs 1, 3 &
8, two pc. from each
Mean strength [MPa] 78,2 (+14 d)1) Vertically drilled
St.deviation [MPa] 4,3 Tested as drilled2) Density =
2417 kg/m3
9.3
Strength of
h
reference tests
d mm mm
6 50 50 25.10.1993 Upper flange of slab9
1)
Mean strength [MPa] 72,0 (+14 d) Vertically drilled
St.deviation [MPa] 4,9 Tested as drilled2) Density =
2407 kg/m3
9.4
Strength of a
grout in # a a Date of test Note
longitudinal mm
joints of slab a
units and tie 2 150 11.10.1993 Kept in laboratory in the same
beams Mean strength [MPa] 29,3 (+0 d)1) conditions as the floor specimen
St.deviation [MPa] Density = 2200 kg/m3
h
# Cores h d Date of test Note
d
mm mm
3 50 50 25.10.1993 Vertically drilled
Mean strength [MPa] 33,8 (+14 d)1) Tested as drilled2)
210
9.5
Strength of a
concrete in the # a Date of test Note
a
topping
a mm
1)
Mean strength [MPa] 29,3 (+0 d) conditions as the floor specimen
St.deviation [MPa] Density = 2200 kg/m3
h
d mm mm
1)
Mean strength [MPa] 34,2 (+14 d) Tested as drilled2)
9.6
Strength of
concrete in the # Cores
h
middle beam
d mm mm
1)
Mean strength [MPa] 62,9 (+14 d) Tested as drilled2)
1)
2)
In the following figures, F2 = P2 +5,6 kN is the line load on slabs 1, 4, 5 and 8 due to
actuator force P2 and weight of loading equipment. Note that the last six points on each
curve represent the post failure situation.
211
10.1
Deflections
160
140
120
100 5
F2 [kN]
80 6
60 7
40 8
20 9
0
0 2 4 6 8 10 12
Deflection [mm]
160
140
120
100 10
F2 [kN]
80 11
60 12
40 13
20 14
0
0 4 8 12 16 20
Deflection [mm]
212
160
140
120
100 15
F2 [kN]
80 16
60 17
40 18
20 19
0
0 2 4 6 8 10 12 14 16 18 20
Deflection [mm]
160
140
120
100 20
F2 [kN]
80 21
60 22
40 23
20 24
0
0 4 8 12 16 20
Deflection [mm]
213
160
140
120
100 25
F2 [kN]
80 26
60 27
40 28
20 29
0
0 4 8 12 16 20
Deflection [mm]
160
140
120
100
F2 [kN]
80 7
60 17
40 27
20
0
0 2 4 6 8 10 12 14 16
Net deflection [mm]
Fig. 27. Net mid-point deflection (measured deflection – settlement of supports) for end
beams (7 and 27) and middle beam (17).
214
10.2
Crack width
160
140
120
100 40
F2 [kN]
41
80
42
60
43
40
44
20
45
0
0 0,2 0,4 0,6 0,8 1
Fig. 28. Differential displacement (  crack width) measured by transducers 40–45,

see Fig. 14.
10.3
Average strain
(actually 160
differential
displacement) 140
120
50
100
F2 [kN]
51
80
52
60
53
40
54
20
55
0
-1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0
215
160
140
120
56
100
F2 [kN]
57
80
58
60
59
40 60
20 61
0
-0,5 0,0 0,5 1,0 1,5 2,0
10.4
Sliding of slab A negative value means that the slab is moving towards the end of the beam.
along middle
beam
160
140
120
100
F2 [kN]
46
80
47
60
48
40 49
20
0
-2,0 -1,5 -1,0 -0,5 0,0 0,5 1,0 1,5 2,0
Fig. 31. Sliding of upper edge of slabs 1, 4, 5 and 8 along middle beam, see Fig. 16.
No explanation for the positive values of 46 and 47 could be given after the test.
216
160
140
120
100
F2 [kN]
62
80
63
60
64
40
65
20
0
-3,5 -3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5
Fig. 32. Sliding of lower edge of slabs 1, 4, 5 and 8 along the middle beam, see Fig. 16.
10.5 Strain
-
11 Reference tests
Both ends of slabs 9 and 10 were loaded. The results of slab 10 have been ignored
because the slab was not cast on the same bed as slabs 1–9.
6 5 4 8
1 2 7
3
1300
P
1 (6) Gypsum 3 (4)
2 (5)
7 (8)
30 30 L/2 30 30
L
Fig. 33. Layout of reference test. a) Plan. b) Elevation. For L, see the next Table.
217
the slab, actuator force Pa at failure, weight of loading equipment Peq, total shear force
(support reaction) Vobs at failure and total shear force vobs per unit width.
Span L Vg Pa Peq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
R9/1 19.10.1993 5940 12,4 233,7 0,7 208,5 173,8 Web shear failure
Mean 193,6 161,3
12 Comparison: Floor test vs. reference tests
13 Discussion
1. The failure mode was web shear failure of edge slabs. The prestressed
concrete beam seemed to recover completely after the failure.
2. The net deflection of the middle beam due to the imposed actuator loads
(deflection minus settlement of supports) was 13,8 mm or L/360, i.e. rather
small. It was 4,3–4,8 mm greater than that of the end beams. Hence, the
beams had a minor or negligible effect on the failure of the slabs.
3. The mean of shear resistances measured in the reference tests was roughly
10% lower than the mean of the observed values for similar slabs given in
Pajari, M. Resistance of prestressed hollow core slab against a web shear
failure. VTT Research Notes 2292, Espoo 2005.
4. The topping concrete above the middle beam became loose at one end of the
middle beam when the imposed load was 85% of the failure load. The failure
took place at the opposite side
5. Comparing the deflection of the middle beam in the present test with that in test
VTT.PC.InvT-Unif.265.1993 suggests that the middle beam was still far from
yielding when the failure took place.
218
A219
A220
A221
A222
A223
A224
A225
A226
A227
A228
A229
A230
VTT.PC.Rect-norm.265.1993
1.1
Identification
VTT.PC.Rect-norm.265.1993 Last update 2.11.2010
and aim
PC265N (Internal identification)
Aim of the test To study the shear resistance of hollow core

slabs supported on the top of beam.
1.2
Test type
1.3 VTT/FI 14.12.1993

Laboratory
& date of test
1.4
Test report Author(s) Koukkari, H. & Pajari, M.
Name Loading test on 265 mm hollow core floor supported on prestressed
concrete beam
Date 15.2.1994
P.O. Box 1001, FI-02044 VTT
Financed by the Finnish Association of Building Industry RTT

(supported by the Technology Development Centre of Finland);
the International Prestressed Hollow Core Association IPHA;
KB Kristianstads Cementgjuteri, Sweden and Skanska Prefab AB,
Sweden

(see also photographs in Appendix A)
2.1
General plan
231
5000
4 3 2 1
6000
South
North
905
95 5000
290
95
1000
905
8 7 6 5
1200 1200 1200 1200
Fig. 2. Plan.
232
2.2
End beams I-beams made of steel. Span = 5,0 m
HE 240 A
Steel Fe 52
A 5 stiffening plates, thickness = 10 mm
100 1300 1200 1200 1300 100

A
5200
240
12
7,5
206
12
A A
Fig. 3. End beam.
80 50
2 T8
L = 4750
Tie beam 265

Plywood
Plastic plug
30 30 End beam
230
HE240A
Fe 52
240
Fig. 4. Arrangements at end beam. T8 refers to a reinforcing bar, see 2.4.
2.3
Middle beam The middle beam was a composite beam comprising a precast, prestressed concrete
beam and a cast-in-situ component. The grade of the concrete was K60 in the lower,
precast part and K30 in the upper part. See Figs 5 and 6 for the main characteristics of
the beam.
233
5400
22 H8 c/c 100 5 H8 c/c 200 22 H8 c/c 100 Hoops
2100 1200 2100
A-A
Hoop
230
neopren
10 x 20
350
16 J12,5
380  = 1100 MPa
po
45
46
150 180 150
52,5 50 50 50 75 50 50 50 52,5
Fig. 5. Middle beam. Main reinforcements and hoops. H8 and J12,5 refer to rebar hoops
and prestressing strands with diameter 8 and 12,5 mm, respectively, see 9.1.
2.4
Arrangements 50 290
at middle 4 T8 L = 4950
beam 40
Plastic plug
265
40 50
20 3 T8 L=2000
15 60 c/c 1200
380 Neoprene 20x10
480
Fig. 6. Middle beam. Section along hollow cores.
T8: Hot rolled, weldable rebar A500HW,  = 8 mm, in longitudinal joints between
adjacent slabs, see 9.1.
234
2.5
Slabs 40
185 265
35
40
152 223 223 223 223 152
1200
- Extruded by Partek Betoniteollisuus Oy, Hyrylä factory 8.11.1993 and 11.11.1993

1159 Max measured bond

42,1 slips: 2,5 , 2,2 and 1,9
mm in slab 4; 2,3 mm in
bw,i slabs 1 and 6, 1,9 and
180,6
1,8 mm in slab 7

37,9 36,1  bw,i = 216,5 units = 4,09 kN/m
1193
Fig. 8. Floor test. Mean of most relevant measured

geometrical characteristics of slabs 1–8.
1159 Max measured bond

40,4 slips: 2,3, 1,7and 1,4 mm
in slab 10; 1,7, 1,5 and
bw,i 1,4 mm in slab 9
182,6

units = 4,11 kN/m
37,4 35,3 1193  b w,i = 217,9
Fig. 9. Reference tests. Mean of most relevant measured

geometrical characteristics of slabs 9 and 10.
There were 10 slabs cast on the same bed. Three of them were cast on the 8th of
November, seven on the 11th of November. Since the expected failure mode was
failure of the outermost slabs, the slabs were arranged as shown in Fig. 10. In this
way, one slab from the both casting days remained for the reference tests.
235
North
48 11.11. 11.11. 4
7 8.11. 11.11. 3
6 8.11. 11.11. 2
5 11.11. 11.11. 1
South
Fig. 10.Floor test. Position of slabs cast on November 8 and 11.
2.6
supports - No
2.7
arrangements and the other for actuators P2, see Fig. 14. Attempts were made to keep P1 ≈ P2 to
generate two uniform line loads on the floor.
The primary spreader beams on the top of the floor were slightly shorter than 0,6 m.
There was gypsum mortar between the primary spreader beams and the top surface
of the floor. The friction between the secondary and primary spreader beams was
eliminated by teflon plates (beams spreading loads P2) and by a roller bearing
(beams spreading load P1).
236
5000
4 3 2 1
6000
P P
1 1
905
South
North
95 P P 5000 P P
290 2 2 2 2
95
1000
905
P P
1 1
8 7 6 5
Det A
P P 2 P1 2 P1 P P
2 2 2 2
Fig. 11. Plan. P1 and P2 refer to vertical actuator forces. For det A, see Fig. 12.
Det A F F
P2 Steel plates
F
Teflon
plates
Gypsum
F
Fig. 12. Detail A, see the figure above.
3 Measurements
3.1
Support
reactions
237
1
3 load
cells
South
5
3.2
6000 III 6000
Vertical
displacement I North IV V
II
9 14 19 24 29
48 4
8 13 18 23 28
7 3
5000
7 12 17 22 27
6 2
6 11 16 21 26
5 1
5 10 15 20 25
3000 3000 South 3000 3000

3.3
North
(63) (62)
8 53 4
(59)
55 51
7 (61) (57) 3
L
54 50
6 (60) (56) 2 Fig. 15. Apparatus for measuring average
52 bottom and top side transducers,
5 (58) 1 respectively.
49 48
(65) (64)
South
50 50
beams. Transducers 46–49 and 62–65 measured the sliding of the outermost slabs
along the beam. Numbers in parentheses refer to the bottom, others to the top side.
238
3.4
Horizontal North
displacements
41 40
8 4
7 3
43 42
6 2
5 1
45 44
South
3.5
-
Strain
-
5 Loading strategy
5.1
relationship
is given on the horizontal axis. The load test took 3,5 h but in the beginning there was a
break of half an hour due to a system error in the data logger.
In the following, the cyclic stage (steps 1–16) is called Stage I, the remaining part
(steps 16–48) Stage II.
180
160
140
120
Pi [kN]
100
80
P1
60 P2
40
20
0
0 10 20 30 40 50
Load step
Fig. 18. Development of actuator loads Pi.
239

5.2
After failure -
Stage I The cast-in-situ concrete cracked vertically along ends of slabs 5, 7

and 8 at P1 = P2 = 55 kN.
Stage II At P1 = P2 = 70 kN he cast-in-situ concrete cracked vertically along

ends of slabs 1 and 4. At P1 = P2 = 80kN these cracks had grown
together. At P1 = P2 = 90 kN similar cracks on the opposite side of
the middle beam had also grown together.
At P1 = P2 = 168 kN inclined shear cracks appeared in the outermost

webs of slabs 1, 5 and 8 close to the ends of the middle beam. This
was followed by a sudden drop of P2.
P2 could still be increased to the previous value and beyond it.

At P1 = P2 = 168 kN an inclined shear crack appeared in the
outermost web of slab 4 close to the North end of the middle
beam. At P1 = 167,5 kN and P2 = 171,2 kN slabs 5–8 failed in
shear as shown in Fig. 19.
After failure When the slabs were removed, it came out that the joint concrete
had completely filled the space between the slab and the middle
beam, the space under the slab end and the core fillings included.
240
7.1
service load
7.2
Cracks after
failure
8 4
7 3
6 2
5 1
8 4
7 3
6 2
5 1
5 1 4 8
1000 1000 1000 1000
Fig. 19. Cracks after failure on the top, at the bottom and in the longitudinal edges of the
floor.
end of the middle beam and the sum of actuator loads on half floor.
The ratio of the reaction to the load is shown in Fig. 21 and in a larger scale in Fig. 22.
Based on Fig. 22 it is justified to assume that at failure the support reaction due to the
line load is equal to 0,874 times the line load. Assuming simply supported slabs gives
the theoretical ratio of 0,85.
241
600
500
Load on half floor [kN] 400
300
200
100
0
0 100 200 300 400 500 600 700
Load step
half floor = 2(P1 + P2 ).
1,4
1,2
Support reaction [kN]
0,8
0,6
0,4
0,2
0
0 100 200 300 400 500 600 700
Fig. 21. Ratio of measured support reaction (below South end of the middle beam)
to actuator loads on half floor.
242
0,91

0,9
0,89
0,88
0,87
0,86
300 400 500 600 700
Fig. 22. A part of the previous figure in a large scale. The point corresponding to the
highest support reaction has been indicated by an arrow.
joint concrete, weight of loading equipment and actuator forces P1, and P2, respectively.
It is concluded that the maximum support reaction due to the imposed load on the failed
slabs has been Vp = 0,874 x (actuator loads on half floor) /
4 = 0,874x(167,5 + 171,2) /2 = 148,0 kN. In the same way, the support reaction due to
the weight of the loading equipment has been 0,874x(1,2+5,6)/2 = 2,97 kN. Vg,jc is
calculated from the nominal geometry of the joints and measured density of the grout.
When calculating Vg,sl, the measured weight of the slabs is used. The values of the
shear force components are given in Table 1 below.
Shear force
Action Load
kN/slab
243
9.1
steel: MPa MPa
Strands J12,5 1630 1860 Nominal (no yielding in test)
Reinforcement 500 Nominal value for reinforcing bars A500HW,
Txy no yielding in test
9.2
Strength of
slab concrete, # Cores h
floor test mm mm
d
6 50 50 23.12.1993 Upper flange of slabs 1, 2
and 5, two cores from each
Tested as drilled2)
9.3
Strength of
slab concrete, # Cores h
reference tests mm mm
d
Tested as drilled2)
h
d
mm mm
Tested as drilled2)
9.4
Strength of a
and tie beams mm
a
St.deviation [MPa] - Density = 2177 kg/m3
244
9.5
Strength of
concrete in # Cores h
upper part of mm mm
d
middle beam
6 75 75 23.12.1993 Upper surface of beam,
Tested as drilled2)
9.6 Strength of
concrete in # Cores h
lower part of mm mm
d
middle beam
Tested as drilled2)
1)
2)
Note that the last two points on each curve represent the post failure situation.
10.1
Deflections
P1 [kN]
180
160
140
120
100 5
80 6
60 7
40 8
20 9
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
245
P1 [kN]
180
160
140
120
100 10
80 11
60 12
40 13
20 14
0
0 2 4 6 8 10 12 14
Deflection [mm]
P1 [kN]
180
160
140
120
100 15
80 16
60 17
40 18
20 19
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 25. Deflection on line III, middle beam.
246
P1 [kN]
180
160
140
120
100 20
80 21
60 22
40 23
20 24
0
0 2 4 6 8 10 12 14
Deflection [mm]
Fig. 26. Deflection on line IV.
P1 [kN]
180
160
140
120
100 25
80 26
60 27
40 28
20 29
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 27. Deflection on line V, Eastern end beam.
247
P1 [kN]
180
160
140
120
100
80 7, End
60 17, Middle
40 27, End
20
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 28. Net deflection of mid-point of beams (rigid body motion = settlement of beam
supports has been eliminated).
The last measured net deflection of the middle beam at highest load level before failure
was 7,7 mm. This is 3,8–4 mm higher than the net deflection of the end beams.
10.2
Crack width P1 [kN]
180
160
140
120 40
100 41
80 42
60
43
40
44
20
45
0
0,0 0,4 0,8 1,2 1,6 2,0
Crack width [mm]
Fig. 29. Differential displacement ( crack width) measured by transducers 40–45.
248
10.3
Average strain P1 [kN]
(actually 180
differential
displacement
160
between slab 50 140
edges) 120
51
100
52
80
53
60
54
40
55 20
0
-1,0 -0,8 -0,6 -0,4 -0,2 0,0
P1 [kN]
180
160
140
56
120
57
100
58
80
59
60
40 60
20 61
0
0,0 0,2 0,4 0,6 0,8 1,0 1,2
249
10.4
Shear In Figs 32 and 33 the differential displacements measured by transducers 46–49 and
displacement 62–65 are shown. A negative sign means that the slab edge is coming closer to the
between slab fixing point of the transducer.
edge and
middle beam P1 [kN] Top Bottom
180
160
140
120
100
46
80
47
60
62 40
63 20
0
-1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0
middle beam.
P1 [kN] Top Bottom

180
160
140
120
100
48
80
49
60
64 40
65 20
0
-1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0
middle beam.
10.5 Strain
-
250
11 Reference tests
Both ends of slabs 9 and 10 were loaded in shear as shown in Fig. 34. The tests were
performed after the floor test (21.12.1993).
6 5 4 7
P
2 8
1 3
0,5L 1000
P
Gypsum
1 (6) 2 (5) 3 (4)
8 (7)
30 30 L 30 30
Fig. 34. Layout of reference tests. Transducers 2 and 5 are in the mid-span. For L, see
Table 2. The slip of the outermost strands was measured by transducers 7 and 8.
The load-deflection relationship for the mid-point of the slabs is shown in Fig. 35 and the
failure load in Table 2. The measured self-weight of the slabs = 4,11 kN/m has been
used when calculating the shear resistance. Virtually no slip of the outermost strands
was observed. For the failure mode, see also Appendix A.
250
200
150
P [kN]
Test 9.1
100
Test 9.2
Test 10.1
50
Test 10.2
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 35. Net mid-point deflection of slabs. Rigid body motion (= settlement of supports)
has been eliminated.
251
Table 2. Span L, ultimate load Pu, shear force due to self weight Vg, shear force due to
imposed load VP, ultimate shear force Vu and failure mode in reference tests. The weight
of the loading equipment = 1,4 kN is included in Pu.
Slab L Pu Vg VP Vu Failure mode

mm kN kN kN kN
9,1 5936 240,4 12,19 201,1 213,3 860
250
9.2 5000 256,4 10,3 206,7 217,0 600
100
10.1 5935 219,4 12,2 183,5 195,7 400
10.2 5000 257,3 10,3 207,4 217,7 420
Mean 210,9
The observed shear resistance in the floor test was 163,8 kN per one slab unit or
136,5 kN/m. This is 65% of the mean observed in the reference tests.
13 Discussion
(deflection minus settlement of supports) was 7,7 mm or L/649 at the highest
load level.
as the mean of the observed values for similar slabs given in Pajari, M.
Resistance of prestressed hollow core slab against web shear failure. VTT
Research Notes 2292, Espoo 2005.
3. Before failure, the net deflection of the end beams was 3,8–4 mm lower than
that of the middle beam. This is a too small difference to cause considerable
torsional stresses in the slabs.
4. The failure mode was a web shear failure of slabs on one side of the middle
beam. Before failure there were diagonal cracks in all for slab edges next to
the supports of the middle beam.
The failure behaviour of the slabs was similar to that in the other Finnish floor tests
in which the slabs were supported close to the soffit of the middle beam.
252
A253
A254
A255
A256
A257
A258
A259
A260
A261
VTT.PC.InvT_Cont.265.1994
1.1
Identification
VTT.PC.InvT_Cont.265.1994 Last update 2.11.2010
and aim
PC265C (Internal identification)
Aim of the test To study the shear resistance of hollow core slabs
supported on continuous beam.
1.2
Test type
1.3 VTT/FI 3.3.1994

Laboratory
& date of test
1.4
Name Loading test for 265 mm hollow core floor supported on continuous
beams
Ref. number RTE5-IR-4/1994
Date 29.4.1994
P.O. Box 1001, FI-02044 VTT.

(supported by the Technology Development Centre of Finland); the
Finnish Steel Work Association; the International prestressed Hollow
core Association IPHA; KB Kristianstads Cementgjuteri, Sweden;
Stombyggarna i Hudiksvall AB, Sweden and SBUF, Sweden

2.1
General plan
263
Support above beam Support below beam 5000 1000
1000
4 3 2 1
6000
NORTH 1000 SOUTH
50 50
P 1000 480 5000 1000 P
3 3
50
1000
8 7 6 5
1200 1200 1200 1200
Fig. 2. Plan.
2.2
End beams Box girders made of steel. Span = 5,0 m
HE 200 A
Steel Fe 52 7 stiffening plates, thickness = 10 mm
A
100 1000 1300 1200 1200 1300 1000 100

A
7200
200
10
6,5 170
10
A A
Fig. 3. End beam.
264
80 50
2 T8 
L = 4750 Tie beam
265
Plywood
Plastic plug
30 30
Fig. 4. Arrangements at end beam. T8 refers to a reinforcing bar, see 2.4.
2.3
Middle beam The middle beam was a prestressed concrete beam provided with shear keys (indents)
on both sides of the web. The grade of the concrete was K60. See Figs. 5, 6 and 8 for
the dimensions of the cross-section.
Holes  80 1300 1200 1200 1300
40 70 70 70 40
2000 5000 2000
Fig. 5. Middle beam. Elevation.
Shear keys c/c 300

125 175
75
125 265
100
Fig. 6. Middle beam. Shear keys.
70 50 50 70 70 50 50 70
7+7 J12,5 45
45
 =1100 MPa
po 4 J12,5
 =1100 MPa
po
14 J12,5 45
 =1100 MPa 46
po 55 6 x 50 6 x 50
70
Fig. 7. Middle beam. Position of prestressing strands J12,5. For J12,5, see 9.1.
265
2.4
Arrangements 50 50 480
at middle
beam 40 Shear keys 125x125
Plastic plug 15 c/c 300
185 265
40 2 T8 L = 5150
100 Neoprene 20x10 2 T8 L = 5150
20
25 55 50 150
780
3 holes c/c 1200
70
3 T8 L = 2000 c/c 1200
Filling length 100 mm in outermost voids,

50 mm in other voids
Steel bar  60
Concrete
Grout 50
100
Steel profile
a) b)
Fig. 10. To prevent sliding of slab edge along the middle beam (this is not possible if the
continuous beam carries hollow core slabs on both sides of the continuous support), a
steel bar penetrating the web and a steel profile were installed. The empty space around
the bar in the hole as well as the gap between the U-profile and the slab edge were
grouted. a) Elevation. b) Top view (see also Fig. 16). See Also Appendix A, Figs 5 and 7.
266
2.5
Slabs 40
185 265
35
40
152 223 223 223 223 152
1200


43,7 5,9 and 3,0 mm in slab 1;
3,3 and 2,5 mm in slab 6
bw,i and 2,2 mm in slab 7
183,9

units = 3,99 kN/m
34,2 34,9  b w,i = 215,7
1193

Max measured bond slips:

1156
6,5, 4,0, 2,3 and 2,2 mm in
43,5 slab 10 (1,3 mm in slab 9)
bw,i
183,5

units = 4,01 kN/m
34,1 35,9 1194  bw,i = 216,5
Fig. 13. Reference tests. Mean of most relevant

measured geometrical characteristics of slab 9.
There were too large bond slips in strands at one end of slabs 1 and 6 (3,0–5,9 mm)
and at both ends of slab 10. Slab 10 was rejected and the ends of slabs 1 and 6 with too
high slip were placed on the end beams.
2.6
supports - No
267
2.7
Loading There were three separate, manually controlled hydraulic circuits, one for actuators
arrangements P1 the other for actuators P2, and the third for actuators P3, see Fig. 14. Attempts were
made to keep P1 ≈ P2 to generate two uniform line loads on the floor. The aim of loads
P3 was to make the middle beam behave as one with clamped ends (no rotation at
support) as far as possible.
The primary spreader beams on the top of the floor were slightly shorter than 0,6 m. The
friction between the secondary and primary spreader beams was eliminated by teflon plates
(beams spreading loads P2) and by a roller bearing (beams spreading load P1). There was
gypsum mortar between the primary spreader beams and the top surface of the floor.
Support above beam Support below beam

5000 1000
1000
4 3 2 1
6000
Primary Secondary
P1 P1
NORTH 1000 SOUTH
50 50
P P P P
P 1000 480 2 2 5000 2 2 1000 P
3 3
50
1000 Det B
P1 P1
8 7 6 5
Secondary
Primary P2 P2 2 P1 2P 1 P2 P2 Det A
1200 1200 1200 1200
Fig. 14. Plan. P1, P2 and P3 refer to vertical actuator forces. For det A and det B, see
Figs. 15 and 16.
268
Det A 100 mm infill in

F F
outermost core
P2 Steel plates
F
Teflon
plates
Gypsum Shaft through
F beam
Fig. 15. Detail A, see the figure above. Fig. 16. Det B, see the figure above.
Load cell
1000 5000 1000
Fig. 17. Support conditions of end beams.
Hydraulic actuator Hydraulic actuator

(P ) (P )
Load cell 3 3
1000 5000 1000
Fig. 18. Supports below and concentrated loads on middle beam.
3 Measurements
3.1
Support
reactions
1
3 load
cells
South
269
3.2
Vertical 6000 III 6000
displacement V
I II IV
9 14 19 24 29
48 4
8 13 18 23 28
7 3
5000
7 12 17 22 27
6 2
6 11 16 21 26
5 5 10 20 1 25
15
3000 3000 3000 3000
3.3
Average strain 47 North 46 Steel bar Transducer
(63) (62)
8 4
L
55 53 51
7 (61) (59) (57) 3 Fig. 21. Apparatus for measuring average
54 52 50
6 bottom and top side transducers,
(60)
(58) (56) 2 respectively.
5 1
49 48
(65) (64)
South
70 70
Fig. 22. Position of device (transducers 50–61) measuring average strain parallel to
the beams. Transducers 46–49 and 62–65 measured the sliding of the outermost slabs
along the beam. Numbers in parentheses refer to the soffit of the floor, others to the top
side.
270
3.4
Horizontal. North
displacements
41 40
8 4
7 3
43 42
6 2
5 1
45 44
South
3.5
-
Strain
See 2.7 for the support conditions of the beams.
5 Loading strategy
5.1
relationship
is given on the horizontal axis. The loading took 2 h 20 min.
In the following, the cyclic stage (steps 1–16) is called Stage I, the monotonous stage
until P3 = 450 kN Stage II and the final stage with constant P3 until failure Stage III
(steps 34–42).
500
450
P1
400
P2
350
P3
300
P [kN]
250
200
150
100
50
0
0 5 10 15 20 25 30 35 40 45
Load step
271
The weight of loading equipment per actuator was 1,2 kN and 5,6 kN for actuators P1
and P2, respectively. Consequently, the imposed load per slab was

The weight of the loading equipment below actuator loads P3 was 0,5 kN and
F3 = P3 + 0,5 kN on the middle beam.
The bending moment between the supports of the middle beam followed closely the
elastic bending moment distribution until P1 = P2 = 160 kN, P3 = 450 kN. Thereafter P3
was kept constant. The bending moment in the span was roughly equal but opposite to
that at supports, see Fig. 25.
M [kNm]
-600
-400
-200 0 1 2 3 4 5
0
200
400
600
Distance from support [m]
Fig. 25. Bending moment diagram of middle beam between supports at failure.
5.2
After failure -

Stage I The vertical joints between the slab ends and the middle beam
cracked, and at P1 = P2 = 55 kN these cracks had grown through
the whole length of the joint.
Stage II At P1 = P2 = 110 kN, the middle beam cracked in flexure at supports.
Stage III At P1 = P2 = 195 kN it was observed that the outermost hollow core
slabs 1, 4, 5 and 8 were cracked in flexure below the line loads.
At P1 = P2 = 200 kN slabs 1–4 suddenly and simultaneously failed
in shear, see Figs 26 & 27 and Appendix A, Figs 7–16. It was
impossible to say where the first shear crack appeared.
After failure When the slabs were removed, it came out that the joint concrete had
completely filled the space between the slab and the middle beam,
the space under the slab end included.
272
7.1
service load
7.2
Cracks after
failure
195 200 195
110 North
0 48 4 0
7 55 55 3
6 2
5 1
200
1
110 South
200
195 195
Fig. 26. Cracks after failure on the top and at the edges of the floor specimen.
The numbers refer to the actuator loads P1.
North
48 200 4
7 180 3
6 2
5 1
South
Fig. 27. Cracks after failure in the soffit. The numbers refer to the actuator loads P1.
273
end of the middle beam and the sum of actuator loads on half floor. The effect of the
actuator forces P3 have been subtracted from the measured support reaction, and it is
not included in the load on the half floor, either.
line load is 0,88 times the line load. Assuming simply supported slabs gives the
theoretical ratio of 0,84.

900
800
700
600
500
400
300
200
100
0
0 100 200 300 400 500 600 700 800
Reaction force [kN]
load on half floor = 2(P1 + P2 ).

1,2
0,8
0,6
0,4
0,2
0
0 100 200 300 400 500 600 700 800
actuator loads on half floor.
274

0,900
0,895
0,890
0,885
0,880
0,875
0,870
300 400 500 600 700 800
joint concrete, weight of loading equipment and actuator forces P1, and P2, respectively.
It is concluded that the maximum support reaction due to the imposed load on the failed
slabs has been Vp = 0,88 x (actuator loads on half floor) /
4 = 0,88 x(199,7 + 200,5) /2 = 176,09 kN. In the same way, the support reaction due to
the weight of the loading equipment has been 0,88x(1,2+5,6)/2 = 2,99 kN. Vg,jc is
calculated from the nominal geometry of the joints and measured density of the grout.
When calculating Vg,sl, the measured weight of the slabs is used. The values of the
shear force components are given in Table 1 below.
Shear force
Action Load
kN/slab
275
9.1
steel MPa MPa
Nominal value for reinforcing bars,
no yielding in test
9.2
Strength of
slab concrete, h
floor test mm mm
d
and 5, two cores from each
Tested as drilled2)
9.3
Strength of
slab concrete, h
d
Tested as drilled2)
9.4
Strength of a
and core filling mm
a
9.5
Strength of
concrete in h
middle beam mm mm
d
Tested as drilled2)
1)
2)
276
10.1
Deflections P1 [kN]
220
200
180
160 5
140
120 6
100 7
80 8
60
40 9
20
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
P1 [kN]
220
200
180
160 10
140
120 11
100 12
80 13
60
40 14
20
0
0 2 4 6 8 10 12 14 16 18
Deflection [mm]
277
P1 [kN]
220
200
180
160
140
15
120
100 16
80 17
60
18
40
20 19
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 33. Deflection on line III, middle beam.
P1 [kN]
220
200
180
160 20
140
120 21
100 22
80 23
60
40 24
20
0
0 2 4 6 8 10 12 14 16 18
Deflection [mm]
Fig. 34. Deflection on line IV.
P1 [kN]
220
200
180
160
140
25
120
100 26
80 27
60
28
40
20 29
0
0 1 2 3 4 5 6 7 8 9
Deflection [mm]
Fig. 35. Deflection on line V, Eastern end beam.
278
P1 [kN]
220
200
180
160
140
120
100 7, West end
80
60 27, East end
40 17, Middle
20
0
0 1 2 3 4 5 6 7 8
Deflection [mm]
supports eliminated).
The last measured net deflection of the middle beam before failure was 5,2 mm. This
was 1,7–1,8 mm lower than the net deflection of the end beams.
10.2
Crack width
P1 [kN]
220
200
180
160 40
140
120 41
100 42
80 43
60
40 44
20 45
0
0,0 0,5 1,0 1,5 2,0 2,5
Crack width [mm]
Fig. 37. Differential displacement ( crack width) measured by transducers 40–45.
279
10.3
Average strain P1 [kN]
(actually 220
differential 200
displacement 180
between slab 50 160
edges) 140
51 120
52 100
53 80
60
54
40
55 20
0
-0,6 -0,5 -0,4 -0,3 -0,2 -0,1 0,0
P1 [kN]
220
200
180
160
56
140
120 57
100 58
80
59
60
40 61
20 55
0
-0,1 0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8
10.4
Shear In Figs 41 and 42 the differential displacements measured by transducers 46–49 and
displacement 62–65 are shown. A negative sign means that the slab edge is coming closer to the
between slab fixing point of the transducer. Fig. 40 illustrates how the negative curvature and cracking
edge and due to negative bending moment at support result in positive values of measured
middle beam differential displacements on the top edge of the slab.
280
P3
Fig. 40. Measuring differential displacement on the top of the floor.
P1 [kN]
Bottom 220
200 Top
180
160
140
120
100 46
80 47
60 62
40
20 63
0
-0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0 1,2
middle beam.
P1 [kN]
Bottom220
200 Top
180
160
140
120
100 48
80 49
60 64
40
20 65
0
-0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
middle beam.
10.5 Strain
-
281
11 Reference tests
Both ends of slab 9 were loadedin shear as shown in Fig. 43. The tests were performed
after the floor test (3.3.1994) but before the cores were drilled and loaded (11.4.1994).
The slip of the outermost strands was measured by transducers 7 and 8. No slip was
observed before failure.
6 5 4 7
P
8
1 2 3
0,5L 1000
P
Gypsum
1 (6) 2 (5) 3 (4)
8 (7)
30 30 L 30 30
Fig. 43. Layout of reference tests. For L, see Table 2.
imposed load Vu, ultimate shear force Vu and failure mode in reference tests. The weight

mm kN kN kN kN
9.1 5936 211,0 11,9 176,5 188,4

9.2 5000 237,1 9,6 191,1 200,7
Mean 194,6
282

The observed shea191,4 kN per one slab unit or 159,5 kN/m. This is 98% of the mean
of the shear resistances observed in the reference tests.
13 Discussion
(deflection minus settlement of supports) was 5,2 mm or L/962
of prestressed hollow core slab against web shear failure. VTT Research Notes
2292, Espoo 2005.
3. Before failure, the net deflection of the end beams was 1,7–1,8 mm greater than
that of the middle beam. This is a too small difference to cause considerable
torsional stresses in the slabs.
4. The failure mode was an abrupt web shear failure of slabs on one side of the
middle beam. No shear cracks were observed before the failure. This is typical
of the shear tests carried out on non-flexible supports and different from the
behaviour of the other 19 Finnish floor tests reported elsewhere in this working
paper.
5. Within the accuracy of the measurements, the shear resistance observed in the
floor test was equal to the mean of the resistances observed in the reference
tests. This is also different from the behaviour of the other 19 Finnish floor tests.
283
A284
A285
A286
A287
A288
A289
A290
A291
A292
TUT.CR.MEK.265.1994
1.1
TUT.CR.MEK.265.1994 Last update 2.11.2010
Identification
and aim MEK265 (Internal identification)
Aim of the test To study the interaction between the MEK beam and
hollow core slabs.
1.2 Test type
Fig. 1. Illustration of test setup. MEK beam in the middle, steel beams (square tubes) at
the ends.
1.3 TUT/FI 15.4.1994

Laboratory
& date of test
1.4 Author(s) Iso-Mustajärvi, Pertti
Test report Name MEK-liittopalkin ja ontelolaaston kuormituskoe
(Load test on MEK composite beam and hollow core floor)
Ref. Tutkimusselostus N:o 253/94,
number Tampere University of Technology, Building Construction
Date 20.4.1994
Availability Confidential, owner is Normek Oy, Hiomotie 10, FI-00380 Helsinki, Finland
Note Figures in this paper have partly been modified (e.g. translation of text)
by M. Pajari

2.1
General plan
Fig. 2. Overview.
293
TUT.CR.MEK.265.1994
Fig. 3. Plan.
294
TUT.CR.MEK.265.1994
Fig. 4. Section C-C. Loading arrangements.
2.2
End beams
Fig. 5. End beam. Measured dimensions. Steel Fe 52D.
295
TUT.CR.MEK.265.1994
Fig. 6. End beam, section A-A. Measured dimensions. 8 K 1200 refers to reinforcing
bars T8 c/c 1200 where 8 is the thickness of the bar in mm.
Simply supported, span = 5,02 m, fy  350 MPa (nominal fy)
2.3
Middle beam Simply supported, span = 5,02 m
The beam comprised a steel component, see Figs 7 and 8, which formed a composite
beam with cast-in-situ concrete. The steel component was made of an I-beam, a
horizontal plate (bottom flange) below the I-beam and two vertical zig-zag-shaped
connector elements, welded to the I-beam and to the bottom plate. The cast-in-situ
concrete, cast simultaneously with the grouting of the joints on the 31st of March
1994, filled the empty space between the slab ends laying on the bottom flange.
Concrete: K30, max aggregate size 8 mm
Structural steel: Fe52D
Fig. 7. Steel component of MEK beam. Cross-section.
296
TUT.CR.MEK.265.1994
Fig. 8. Elevation. Note the zig-zag-shaped connector element.
2.4
Arrangements
at middle
beam
Fig. 9. Section B-B. 8 K 1200 refers to reinforcing bars T8 c/c 1200 where 8 is the
thickness in mm.
2.5
Slabs 1160
40
185 265
35
40
152 223 223 223 223 152
1200

- Extruded by Parma Oy, 9 slab elements were delivered to laboratory 11.03.1994
- 10 lower strands J12,5; initial prestress not given in the report but 1100 MPa was
a common value at that time
2.6
supports
297
TUT.CR.MEK.265.1994
2.7
Loading See Figs 2–4.
arrangements
The idea was to create two line loads using four actuators. Two actuator forces loaded
one primary spreader beam which distributed the load to two secondary beams. Each
secondary beam distributed the load to two tertiary spreader beams and each tertiary
beam to two quaternary spreader beams. The quaternary beams were 550 mm long
steel tubes with square cross-section 80 mm x 80 mm x 5 mm. The top surface below
these beams was evened out with gypsum and a soft wood fibre plate was placed onto
the gypsum and below the beams. In this way, a linear line load was created by 8
quaternary spreader beams.
The bearings below the primary, secondary and tertiary spreader beams were hinges
which also allowed longitudinal displacement at one end of each spreader beam. The
fixed hinges were placed symmetrically.
3 Measurements
3.1
Support There were load cells below the South end of the middle beam.
reactions
298
TUT.CR.MEK.265.1994
3.2
Vertical
displacement
Fig. 11. Location of transducers 1–15 for measuring vertical deflection and that of
transducers 16–28 for measuring differential horizontal displacement.
299
TUT.CR.MEK.265.1994
Fig. 12. Section D-D.
3.3
Average strain -
3.4
Horizontal. See Fig. 11 for the position of the horizontal transducers 16–28.
displacements
300
TUT.CR.MEK.265.1994
3.5
Strain The strains in the floor were measured by strain gauges placed as shown in Figs 13 and 14.
Fig. 13. Position of strain gauges 62 … 69 and 80 … 82.
301
TUT.CR.MEK.265.1994
Fig. 14. Location of strain gauges 40 … 48.
- None …
5 Loading strategy
5.1
Load-time Date of the floor test was 15.4.1994.
relationship
The support reaction, displacements and strains due to the installation of the slab
elements, grouting and the weight of loading equipment are given in Sections 10.1
and 10.5.
When the actuator forces P were equal to zero, all measuring devices were zero-
balanced. Thereafter, each actuator force P was increased to 115 kN and reduced back
to zero. This load cycle was repeated for four more times (5 cycles altogether) before
increasing the actuator load monotonously to P = 324 kN which was the failure load.
The cyclic and monotonous stages are called stage I and stage II, respectively.
302
TUT.CR.MEK.265.1994
Fig. 15. Support reaction below South end of MEK beam vs. sum of actuator forces P (= 4P).
5.2
After failure
Stage I At P = 91 kN, the first vertical cracks were observed in the tie beams at
the ends of the floor. At P = 114 kN the joint concrete cracked along the
slab ends next to the MEK beam 1,2 m from the edges of the floor.
Stage II At P = 172 kN the crack on the East side of MEK beam extended over
the whole length of the beam. On the West side the same was true at
P = 302 kN. At P = 302 kN cracks also appeared between the joint
concrete and the edges of MEK beam. These cracks were at the end
of the beam and 1,2 m long.
At P = 313 kN an inclined crack was observed at the end of slabs 4 and

8 close to the MEK beam. At P = 324 kN a new inclined crack appeared
in slab 8. This crack was on the East side of the previous inclined crack
and resulted in shear failure of the floor. The development of the cracks
and the failure mode are illustrated in Figs 16–22.
303
TUT.CR.MEK.265.1994
7.1
Cracks before
failure
Fig. 16. Cracks on the top surface and at the edges before failure. The numbers refer to
the value of actuator load P.
304
TUT.CR.MEK.265.1994
Fig. 17. Cracks in tie beam.
Fig. 18. Inclined cracks in slab 8. The final failure crack on the right.
305
TUT.CR.MEK.265.1994
7.2
Cracks after
failure
306
TUT.CR.MEK.265.1994
307
TUT.CR.MEK.265.1994
joint concrete, weight of loading equipment and actuator forces P, respectively. In this
case the support reaction due to the self weight of the floor and loading equipment was
measured directly. Therefore, the weight of the components was not measured. The
shear force at failure can be calculated from values given in Table 1 on the next page.
308
TUT.CR.MEK.265.1994
Support reaction Shear force

Action
kN kN
Weight of slab unit +
Weight of joint concrete +
Loading equipment (65,53)/4 16,4
Actuator loads 131,8
9.1
MPa MPa
Bottom plate in MEK beam 354 513
350 501
358 518
Other steel in beams (Fe52D)  350 Nominal value
Slab strands J12,5 ≥ 1550 ≥ 1770 Obviously no yielding before failure
Reinforcement T8 500
9.2
Strength of
slab concrete h
d
mm mm
3 50 50 15.4.1994
Mean strength [MPa] 60,0 (+0 d)1)
St.deviation [MPa]
9.3
slab concrete,
reference tests
9.4
Strength of a
of slab units mm
and in MEK a
beam 4 150 15.4.1994 Kept in laboratory in the same
1)
In the following figures, P stands for the actuator load in one actuator. Only the
monotonous loading stage after the cyclic stage is shown.
309
TUT.CR.MEK.265.1994
10.1
Deflections Table 2. Support reaction below South end of MEK beam and deflection measured by
transducers 13–15, both due to installation of hollow core slabs, grouting and loading
equipment.
Reaction 13 14 15 40 41 42
kN [mm] [mm] [mm] [0,001] [0,001] [0,001]
65,53 3,05 4,04 3,26 -0,491 0,129 -0,122
Fig. 23. Mid-point deflection of end beams.
Fig. 24. Deflection of hollow core slabs, transducers 2–6.
310
TUT.CR.MEK.265.1994
Fig. 25. Deflection of hollow core slabs, transducers 7–11.
Fig. 26. Deflection of MEK beam, transducers 13–15.
311
TUT.CR.MEK.265.1994
Fig. 27. Differential vertical displacement between MEK beam and slab end measured
by gauges 28 and 29.
10.2
Crack width
Fig. 28. Opening of joint ( crack width), transducers 16–19.
10.3
Average strain -
(actually
differential
displacement)
312
TUT.CR.MEK.265.1994
10.4
Shear
displacement
Fig. 29. Southern end of middle beam. Differential displacement between edge of slab and
middle beam. A positive value means that the slab is moving towards the end of the beam.
Fig. 30. Northern end of middle beam. Differential displacement between edge of slab and
middle beam. A positive value means that the slab is moving towards the end of the beam.
10.5 Strain
Table 3. Strain in MEK beam due to installation of hollow core slabs, grouting and
loading equipment measured by transducers 40–48.
40 41 42 43 44 45 46 47 48
[0,001] [0,001] [0,001] [0,001] [0,001] [0,001] [0,001] [0,001] [0,001]
-0,491 0,129 -0,122 -0,177 -0,133 0,102 0,204 0,152 0,165
313
TUT.CR.MEK.265.1994
Fig. 31. Strain measured by gauges 40 (transverse to beam) and 42, 43 and 44
(parallel to beam).
Fig. 32. Strain measured by gauges 41 and 45–48.
314
TUT.CR.MEK.265.1994
Fig. 33. Strain measured by gauges 69 and 80–82 in the soffit of hollow core slabs.
Fig. 34. Top surface of thee floor. Strain measured by gauges 62–68.
315
TUT.CR.MEK.265.1994
11 Reference tests
Fig. 35. Layout of reference test, slab 9. Lj = 5,93 m. Note the cast-in-situ concrete at
the loaded end, cast on the 31st of March 1994, which filled 50 mm of the hollow cores.
Table 4. Reference test. Span of slab, sum of actuator force Pa at failure and weight of
loading equipment Peq, total shear force Vobs at failure and total shear force vobs per unit
width.
Span Pa+Peq Vobs vobs

Test Date Note
mm kN kN kN/m
R1 ? 5930 230,7 223,2 186,0 Web shear failure
was equal to 148,2 kN per one slab unit or 123,5 kN/m. This is 66% of the shear
resistance observed in the reference test.
13 Discussion
1. The deflection of the middle beam due to the imposed actuator loads only
(deflection minus settlement of supports) was 16,7 mm or L/301.
as the observed values for similar slabs given in Pajari, M. Resistance of
prestressed hollow core slab against a web shear failure. VTT Research
Notes 2292, Espoo 2005.
3. At failure load, the maximum difference in the mid-point deflection of the

beams was 4,2 mm. Hence, the torsional stresses due to the different
deflection of the middle beam and end beams had a minor effect on the
failure of the slabs.
4. The failure mode was web shear failure of edge slab 8. The measured strains
in the MEK beam show that the beam could not yield in longitudinal bending.
316
VTT.RC.Rect-norm.265.1994
1.1
Identification
VTT.RC.Rect-norm.265.1994 Last update 2.11.2010
and aim
RC265N (Internal identification)
Aim of the test To study the shear resistance of hollow core slabs
supported on the top of reinforced concrete beam.
1.2
Test type
1.3 VTT/FI 11.11.1994

Laboratory
& date of test
1.4
Name Loading test for 265 mm hollow core floor supported on reinforced
concrete beams
Ref. number RTE-IR-7/1994
Date 20.12.1994
P.O. Box 1001, FI-02044 VTT.

(supported by the Technology Development Centre of Finland); the
Finnish Steel Work Association, the International Prestressed Hollow
Core Association IPHA; KB Kristianstads Cementgjuteri, Sweden,
Stombyggarna i Hudisksvall, Sweden and SBUF, Sweden

(see also photographs in Appendix A)
2.1
General plan
317
290 1200
1200
North A
12 6
11 5
10 4
7200
9 3
8 2
7 1
6000 South 6000

12290 A
Fig. 2. Plan.
2.2
End beams The length of the beam was 7,6 m.
Concrete K30 80 50
2 T12 L = 7500 2 T8 
Stirrup L = 7150 Tie beam
265
360 Plywood
3030 Plastic plug

2 T25 L = 7500
3 T25 L = 7500 End beam
300 360
Fig. 3. Section of end beam. TXY refers to

rebars with thickness xy mm, see 9.1. The 300
stirrups were 6 and 8 mm thick rebars in
the middle and at the ends of the beams, Fig. 4. Arrangements at end beam. T8
respectively. refers to a reinforcing bar, see 9.1
2.3
Middle beam The middle beam was a reinforced concrete beam. The grade of the concrete was K60.
See Figs 5 and 6. Txy refers to a reinforcing bar with diameter xy mm, see 9.1.
318
200 A 7200 200
200 A 200
Stirrup A 12 T8 c/c 250
2000 3200 2000
21 * 2 T10 c/c 100 21 * 2 T10 c/c 100
Stirrup B
21 T10 c/c 100 Stirrup C 12 T10 c/c 250 21 T10 c/c 100
Fig. 5. Stirrups. Txy refers to a reinforcing bar with diameter xy mm, see 9.1. Stirrup A is
a hoop (rounded rectangular) surrounding all longitudinal rebars shown in Fig. 6.
150 180 150
Concrete K60
230
Stirrup C T10
4 T12 Stirrup B T10
350
380
Stirrup B T10
4 T25
65
6 T25 55
480
Fig. 6. Middle beam. For the material of Txy and stirrups see 9.1.
2.4
Arrangements 50 290
Concrete K30
at middle 4 T8 L = 7400
beam 40
Plastic plug
265
185
40 50
20
Neoprene 20x10
15 60 5 T8 L=2000
c/c 1200
380 between slabs
480
Fig. 7. Middle beam. Section along hollow cores. T8 refers to rebar A500HW,
d = 8 mm, see 9.1.
319
2.5
Slabs 40
185 265
35
40
152 223 223 223 223 152
1200

- Extruded by Partek Betoniteollisuus Oy, Hyrylä factory10.10.1994
- 10 lower strands J12,5
J12,5: seven indented wires,  =12,5 mm, Ap = 93 mm2, initial prestress = 950 MPa,
see 9.1.

1153
1,6 mm in slab 12; 1,4 and
45,3
1,3 mm in slab 5
bw,i
177,3

units = 4,25 kN/m
38,4 39,7  b w,i = 219,1

1193


1153 1,3 mm in slab 13 and
46,9 1,2 mm in slab 14,
bw,i
175,0

units = 4,33 kN/m
40,0 39,6 1193  b w,i = 221,8
Fig. 10. Reference tests. Mean of most relevant

measured geometrical characteristics of slabs 13 and 14.
2.6
supports - No
320
2.7
Loading The loads P, see Fig. 11, were generated by 6 identical hydraulic actuators, each
arrangements connected to the same hydraulic pressure. The actuator loads were spreaded with the
aid of 6 primary, 12 secondary and 24 tertiary spreader beams to two transverse line
loads on the slabs as shown in Figs 11–13. The reaction forces of the actuators were
carried by a temporary steel frame which was fixed to the floor of the hall by tension
bars. Four holes were drilled through the test floor for these bars. The position of the
tension bars is shown in Fig. 11. After the test it came out that the holes had no effect
on the failure mechanism.
The tertiary spreader beams on the top of the floor were slightly shorter than 0,6 m. The
friction between the teriary and secondary beams was eliminated by roller bearings and
that between the secondary and primary spreader beams by teflon plates.
There was gypsum mortar between the tertiary spreader beams and the top surface of
the floor.
290
Hole through floor 1200 1200
Hydraulic actuator
North A
P
12 6
P
11 5
P
10 4
7200 P
9 3
P
8 2
P
7 1
6000 South 6000
12290 A
Det A
Teflon P P P P P P
sheets
7200
A A
Fig. 11. Loading arrangements.
321
Det A F F
P2 Steel plates
F
Teflon
plates
Gypsum
F
Fig. 12. Detail A, see the Fig. 11.
Fig. 13. Loading frame carrying the reaction of the actuators.
3 Measurements
3.1
Support
reactions
1
3 load
cells
South
322
3.2
Vertical I II III North IV V
displacement 11 18 25 32 39
12 6
10 17 24 31 38
11 5
9 16 23 30 37
10 4
8 15 22 29 36
9 3
7 14 21 28 35
8 2
6 13 20 27 34
7 1
12 19
5 26 33
3000 3000 South 3000 3000
3.3
Average strain 64 69 66 150
North
(65) (67) B
80 104 92
12 (81) (105) (93) 6
11
78 102 90 B
5
(79) (103) (91)
A A
76 100 88
10 (77) (101) (89) 4
74 98 86
9 3
(75) (99) (87)
72 96 84
8 2
(73) (97) (85) B B
70 94 82
7 1
(71) (95) (83)
A 60 62 A South
(61) 68 (63)
Fig. 16. Position of transducers measuring average strain parallel to the beam (70–105),
displacement of the slab edges relative to the beam in beam's direction (60–67) and
sliding of the upper part of the beam relative to the lower, precast part of the beam (68–69).
Numbers in parentheses refer to the bottom, others to the top side.
323
Fig. 17. Apparatus for measuring average strain. L = 1120mm and 1060 mm for the
bottom and top side transducers, respectively.
3.4
Horizontal. North
displacements
45 50
12 6
11 44 49 5
10 4
43 48
9 3
8 42 47 2
7 1
41 46
South
3.5
-
Strain
-
5 Loading strategy
5.1
relationship
Before the onset of the loads, the middle beam and the end beams had cracked due to
the dead weight. The mid-point deflection of the middle beam was 32 mm and that of
the end beams of the same order. The maximum crack widths in the middle beam and
end beams were 0,20 and 0,15 mm, respectively.
All measuring devices were zero-balanced when the actuator forces P were equal to
324
is given on the horizontal axis. The load test took 2 h 35 min.
In the following, the cyclic stage (steps 1–16) is called Stage I, the remaining part
(steps 16–52) Stage II.
250
200
150
P [kN]
100
50
0
0 10 20 30 40 50 60
Load step
The weight of loading equipment per one slab was 3,1 kN. Consequently, the imposed
load per slab was
F = 0,5P + 3,1 kN.
5.2
After failure -
Stage I The cast-in-situ concrete in the upper part of the middle beam cracked
vertically along the slab ends. At P = 60 kN the tie beams above the end
beams started to crack. The maximum crack width in the middle beam
was 0,25 mm.
Stage II At P = 220 kN the first shear cracks were observed at the ends of slabs 1
and 7, at P = 225 kN in slab unit 12 and at P = 233 kN also in slab unit 6.
At P = 238,2 kN slab 7 failed.
After When removing the loading equipment, slab 8 was knocked with a
failure hammer. The sound revealed that the slab end had cracked. This was
confirmed when slab 7 had been removed. A diagonal shear crack was
observed at the end of slab 8 close to the middle beam.
When the slabs were removed, it came out that the joint concrete had
completely filled the space between the slabs and the middle beam, the
space under the slab end and the core fillings included. The concrete infill
in the cores of slab units cracked along the slab ends on one side of the
middle beam while on the other side it remained virtually uncracked, see
Figs. 20 and 21 in App. A.
325
Three days after removal of the loads but when the loading equipment
was still on, the deflection of the middle beam was 35 mm, i.e. only 3 mm
more than the deflection before the test. Such a recovery is not possible
after considerable yielding of the reinforcement in the concrete. This
suggests that the softening behaviour observed in the load – deflection
curve (see Fig. 29) is mainly attributable to effects other than the
plastification of the reinforcement.
7.1
Cracks at See Fig. 20.
service load
7.2
Cracks after 225 kN 233 kN
failure
12 6 110 kN
80 kN 11 5
120 kN 100 kN
130 kN
10 4
120 kN 60 kN
60 kN 80 kN
120 kN
9 3 120 kN
80 kN 2 150 kN
8
7 1
225 kN 220 kN
220 kN 220 kN
Fig. 20. Cracks after failure on the top and in the longitudinal edges of the floor. The
force values refer to the load P at which the crack was observed.

failure took place at P = 238,2 kN.
end of the middle beam and the sum of actuator loads on half floor.
line load is equal to 0,765 times the line load. Assuming simply supported slabs gives
the theoretical ratio of 0,800 ... 0,801.
326
600
500
400
300
200
100
0
0 100 200 300 400 500 600 700 800
half floor (= 3P).
1
0,8
0,6
0,4
0,2
0
0 100 200 300 400 500 600 700 800
actuator loads on half floor. The points corresponding to load P  0 are not shown.
327
0,84
0,83

0,82
0,81
0,8
0,79
0,78
0,77
0,76
0,75
0,74
300 400 500 600 700 800
Fig. 23. A part of the previous figure in a large scale. The point corresponding to the
highest support reaction has been indicated by an arrow.
It can be concluded that the maximum support reaction due to the imposed load on the
failed slabs has been Vp = 0,765 x (actuator loads on half floor) /
6 = 0,765x3x238,2/6 = 91,11 kN. In the same way, the support reaction due to the
weight of the loading equipment has been 0,765 x 3,1 = 2,37 kN. Vg,jc is calculated from
the nominal geometry of the joints and measured density of the grout. When calculating
Vg,sl, the measured weight of the slabs is used. The values of the shear force
components are given in Table 1 below.
Action Load Shear force

kN/slab
Loading equipment 3,1 kN/slab 2,37
Actuator loads 238,2 /2 kN /slab 91,11
9.1
Component Note
steel MPa MPa
Type A500 HW. Nominal value for
reinforcing bars, no yielding in test
328
9.2
Strength of
slab concrete, h
floor test mm mm
d
and 12, two cores from each,
Tested as drilled2)
9.3
Strength of
slab concrete, h
d
Mean strength [MPa] 67,8 (< 7 d)1) vertically drilled
Tested as drilled2)
h
d
mm mm
Mean strength [MPa] 67,8 (< 7 d)1) vertically drilled
Tested as drilled2)
9.4
Strength of
grout in joints, h
tie beams and mm mm
d
in the upper
part of the 6 75 75 24.11.1994 Upper surface of middle
middle beam Mean strength [MPa] 25,3 (+13 d)1) beam, vertically drilled
Tested as drilled2)
9.5
Strength of
concrete in h
lower part of mm mm
d
middle beam
Tested as drilled2)
1)
2)
329
10.1
Deflections
250 250
200 200
5 12
6 150 13
150
P [kN]
P [kN]
7 14
100 8 100 15
9 16
50 10 50 17
11 18
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Deflection [mm] Deflection [mm]
Fig. 24. Deflection on line I, Western end Fig. 25. Deflection on line II.
beam.
250 250
200 200
19 26
20 27
150 150
P [kN]
P [kN]
21 28
100 22 100 29
23 30
50 24 50 31
25 32
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Fig. 26. Deflection on line III, middle Fig. 27. Deflection on line IV.
beam.
250
200
33
34
150
P [kN]
35
100 36
37
50 38
39
0
0 5 10 15 20 25 30 35
Deflection [mm]
Fig. 28. Deflection on line V, Eastern end

beam.
330
250
200
150
P [kN] 8
100 22
36
50
0
0 5 10 15 20 25 30 35
Deflection [mm]
supports has been eliminated).
The last measured net deflection of the middle beam at highest load level before failure
was 30,3 mm. This is 3,8–7 mm higher than the net deflection of the end beams. It is
possible that the twist of the outermost slabs has affected the shear resistance.
10.2
Crack width
250 250
200 200
41 46
150 150
P [kN]
P [kN]
42 47
100 43 100 48
44 49
50 45 50 50
0 0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0
Displacement [mm] Displacement [mm]
Fig. 30. Differential displacement ( crack Fig. 31. Differential displacement ( crack
width) measured by transducers 41–45. width) measured by transducers 46–50.
10.3
Average strain
250 250
(actually
differential 200 200
displacement 70 82
150 72 150 84
between slab
P [kN]
P [kN]
74 86
edges) 100 100
76 88
78 90
50 50
80 92
0 0
-0,8 -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 -0,8 -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6
Fig. 32. Differential displacement at top Fig. 33. Differential displacement at top
surface of slabs 7–12 measured by surface of slabs 1–6 measured by
transducers 70, 72,...,80. transducers 82, 84,...,92
331
250 250
200 200
94 71
150 96 150 73
P [kN]
P [kN]
98 75
100 100 100 77
102 79
50 50
104 81
0 0
-2,0 -1,5 -1,0 -0,5 0,0 0,5 1,0 -0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0 1,2
Fig. 34. Differential displacement at top Fig. 35. Differential displacement at soffit
surface of middle beam measured by of slabs 7–12 measured by transducers
transducers 94, 96,...,104. 71, 73,...,81.
250 250
200 200
83 95
150 85 150 97
P [kN]
P [kN]
87 99
100 89 100 101
91 103
50 50
93 105
0 0
-0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0 1,2 0,0 0,5 1,0 1,5 2,0 2,5 3,0
Fig. 36. Differential displacement at soffit Fig. 37. Differential displacement at soffit
of slabs 1–6 measured by transducers of the middle beam measured by
83, 85,...,93. transducers 95, 97,...,105.
10.4
Shear In Figs 38 and 39 the differential displacements between the lower and upper part of the
displacement slab edges relative to the middle beam in beams’s direction are shown. A negative sign
between slab means that the slab edge is approaching the end of the beam.
edge and
middle beam Bottom
250 Bottom 250
200 200
60 Top 64 Top 150

150
P [kN]
P [kN]
61 65
62 100 66 100
63 67
50 50
0 0
-3,5 -3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5 -3,5 -3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5
Fig. 38. South end of middle beam. Fig. 39. North end of middle beam.
Differential displacement measured by Differential displacement measured by
transducers 60–63. transducers 64–67.
332
In Fig. 40 the longitudinal, differential displacements between the upper and lower part
of the middle beam are shown. A negative sign means that the upper part is approaching
the end of the lower part (moving outwards more than the lower part).
250
200
150
P [kN]
68
69 100
50
0
-0,4 -0,3 -0,2 -0,1 0,0
Displacement [mm]
Fig. 40. South end of middle beam. Differential displacement between upper and lower
part of the middle beam at South end (transducer 68) and at North end (transducer 69).
10.5 Strain
-
11 Reference tests
Slabs 13 and 14 were loaded in shear as shown in Fig. 41. The tests were performed
after the floor test (11.11.1994) but before 18.11.1994.
6 5 4 7
P
1 2 3 8
5940/2 1200
P
Gypsum
1 (6) 2 (5) 3 (4)
8 (7)
30 30 5940 30 30
Fig. 41. Layout of reference tests. The slippage of the outermost strands was measured
by transducers 7 and 8.
The load-deflection relationship for the mid-point of the slabs is shown in Fig. 42.
The change in the slope is attributable to flexural cracking below the load. Despite
the cracking, the failure mode in both tests was tensile failure in the web. Virtually
no slippage of the outermost strands occurred before failure.
333
300
250
P [kN] 200
150
13
100 14
50
0
0 5 10 15 20 25
Deflection [mm]
Fig. 42. Net mid-point deflection of slabs 13 and 14. Rigid body motion (= settlement of
supports) has been eliminated.
The failure modes and failure loads are given in Table 2. The measured self-weight of
the slabs = 4,33 kN/m has been used when calculating the shear resistance.
imposed load VP, ultimate shear force Vu and failure mode in reference tests. The weight

mm kN kN kN kN
13 5940 264,9 12,86 212,7 225,6
14 5940 266,5 12,86 214,0 226,9
Mean 226,2
334
The observed shear resistance in the floor test was 106,7 kN per one slab unit or
88,9 kN/m. This is 47% of the mean of the shear resistances observed in the reference
tests.
13 Discussion
(deflection minus settlement of supports) was 30,3 mm or L/238 at the highest
load level.
2. The failure mode was a web shear failure of slab at the edge of the tested floor
next to the middle beam. Before failure there were diagonal cracks in all four
slab edges next to the supports of the middle beam.
2292, Espoo 2005.
4. Before failure, the net deflection of the end beams was 3,8–7 mm higher than
the net deflection of the end beams. It is possible that the twist of the outermost
slabs has affected the shear resistance. On the other hand, when the first
diagonal cracks were observed at load P = 220 kN (92% of the failure load), the
net deflection of the end beams was less than 1,0 mm smaller than that of the
middle beam.
5. The failure behaviour of the slabs was similar to that in the other Finnish floor
tests in which the slabs were supported close to the soffit of the middle beam.
335
A336
A337
A338
A339
A340
A341
A342
A343
A344
A345
A346
VTT.CP.LBL.320.1998
1.1
Identification
VTT.CP.LBL.320.1998 Last update 2.11.2010
and aim
LBL320 (Internal identification)
Aim of the test To quantify the interaction between the LBL beam and
hollow core slabs.
1.2
Test type
Fig. 1. Overview on test arrangements. LBL beam in the middle, steel beams (square
tubes) at the ends.
1.3 VTT/FI 25.3.1998

Laboratory
& date of test
1.4
Name Load test on hollow core floor
Ref. number RTE30146/98
Date 3.8.1998
Availability Confidential, owner is
Lujabetoni Oy
Harjamäentie 1
FI-71800 Siilinjärvi
Finland

(see also Appendices A and B)
The tie beams at the ends of the slabs and the enlargements at the edges of the slab
near the middle beam were cast and the joints grouted on the 16th of March 1998.
347
VTT.CP.LBL.320.1998
2.1
General plan East
B 5000 (span of beam)
4 3 2 1
7200
North
1200 1200 1200 1200
50
(from slab end)

(from slab end)
In-situ concrete K30

1500
44 44 1200
E
D C
5000 (span of beam)
432
South
D C
E
1200
8 7 6 5
7200
5000 (span of beam)

West
Fig. 2. Plan. For sections B-B, C-C, D-D and E-E see Figs 3, 4, 5 and 6.
348
VTT.CP.LBL.320.1998
2.2
End beams See Fig. 3. Simply supported, span = 5,0 m, roller bearing below Northern end
50 50 Concrete K30
2  8
L=4750
Plywood
Steel plate
40 40 Plastic plug
12.5 100 200
RHS
Fe52
200
B B
Fig. 3. Arrangements at end beam (section B-B in Fig. 2). 8 refers to a reinforcing bar
T8 with diameter 8 mm.
2.3
Middle beam The beam, see illustration in Fig. 4 and App. A for more details, comprised
- a prefabricated steel component with a bottom plate, two upper chords made of
reinforcing bars and bent reinforcing bars welded both to the chords and to the
bottom plate (lattice girders) as shown in App. A.
- a precast and prestressed concrete component, see Fig 5 and App. A, provided with
shear keys as shown in Fig. 4.
Precast prestressed
concrete
Shear keys
50x220x10 c/c 100
Steel plate
Fig. 4. Illustration of LBL beam.
The prestressed part was cast by Lujabetoni Oy
Concrete: K80, max aggregate size 16 mm, rapidly hardening cement
349
VTT.CP.LBL.320.1998
Passive reinforcement and tendons in LBL beam:

Txy: Hot rolled, weldable rebar A500HW,  = xy mm
J12,9: Prestressing strand, 7 wires,  =12,9 mm, Ap = 100 mm2,
prestress = 1350 MPa, low relaxation (nominal value 2,5% at  = 0,7fpu)
Structural steel: S355J0, fy  355 MPa (nominal value) unless otherwise specified
Tie reinforcement across the beam: A500HW, see above.
2.4
Arrangements 450
at middle
beam 260
symm. 300
2  12 A500HW
150
Bars welded to inserted steel plate 1000
60
112
2  12 A500HW L = 2520
one on the other
1000 520 1000
C C
Fig. 5. Suspension and tie reinforcement in joints between adjacent hollow core units
(section C-C in Fig. 2).  12 refers to a reinforcing barT12 with diameter 12 mm.
50 44 432 In-situ concrete K30
40
320
Plastic
240
plug
4    L= 4900
40
8 80 20
680
D D
Fig. 6. Arrangements at middle beam (section D-D in Fig. 2). 10 refers to a reinforcing
bar T10 with diameter 10 mm.
350
VTT.CP.LBL.320.1998
450
symm. 300 260

1  12 A500HW
150
Bar welded to inserted steel plate 1000
112
2  8 L = 1400, 400 mm bent into 1  12 A500HW L = 2520

the grout, parallel to the beam
1000 520 1000
E E
Fig. 7. Suspension and tie reinforcement at the outermost edges of the hollow core floor
(section E-E in Fig. 2). 8 and 12 refer to reinforcing bars T8 and T12 with diameters 8
and 12 mm.
2.5
1162
Slabs
17 8
38 38
50 70
23
R1
320 63 222 52 52
106 R1
23
70
10
10 36 40 40 40 40 40 40 65
10 10
187 274 137 25 224 25 25 230 25 44
1196

- Extruded by Lujabetoni Oy, Hämeenlinna factory 14.1.1998
- 13 lower strands J12,5, initial prestress 1100 MPa
1163 (at mid-depth)

54,3
2,5 and 2,1 mm in slab 8
61,0
222,8
61,7 61,7 61,7 61,0 2,4 mm in slab 6
2,2 and 2,1 mm in slabs
2 and 3
37,5 42,3
units = 5,13 kN/m
characteristics.
351
VTT.CP.LBL.320.1998
2.6
Temporary No
supports
2.7
Loading
arrangements
4 3 2 1
300 300 300 300 300 300 300 300

North 300 300 300 300 300 300 300 300
Pa2 Pa2
1200
South
44
432 Pa1 Pa1 Middle beam Pa1 P
a1
44
1200
Pa2 Pa2
A 8 7 6 5 A
West
Det 1
Pa1 Pa1 2 Pa2 Pa2 Pa1 Pa1
Load cells 1, 2 ja 3
under support of beam
A A
Fig. 10. Plan. Pa1 and Pa2 refer to vertical actuator forces.
Det 1 F F
Pa1 Steel plates
F
Teflon
plates
Gypsum
F
550
Fig. 11. Detail 1, see previous figure.
352
VTT.CP.LBL.320.1998
3 Measurements
3.1
Support
reactions 3 7
4 8
1 2
3
Fig. 12. Load cells below the Northern support of the middle beam.
3.2 15 14 13 12 11
Vertical 40 40 40 40 40
30 30 30 30 30
displacement
3600
4 3 2 1
II 30 30 30 30 30
20 19 18 17 16
Transducer for measuring vertical displacement

3600
30 30 30 30 30
3600
III 25 20 20 24 20 23 20 22 20 21
216
IV 30 29 28 27 26
216
V
35 34 33 32 31
8 7 6 5
3600
40 39 38 37 36
VI
3600
45 44 43 42 41
VII
353
VTT.CP.LBL.320.1998
3.3
Average strain -
3.4
Horizontal Measures differential displacement between slab end and beam (crack width)
displacements
Measures differential displacement between steel flange and concrete component
of beam in beam's direction
Measures differential displacement of slab edge and beam in beam's direction,
numbers in parentheses refer to bottom fibre of slab, others to top fibre
50 4 3 2 1 46
(51) (47)
56
7 6 5
54
40 40 40
8 9 10 55
57
52 H 48 H
(53) 8 7 6 5 (49)
Fig. 14. Transducers measuring crack width (5–10), shear displacement at the ends
of the middle beam (46–53) and differential displacement between bottom plate and
concrete component of the middle beam. Transducers 47, 49, 51 and 53 are below
transducers 46, 48, 50 and 52, respectively.
L-profile, glued to slab 48 48 46

J
49
49 47
J
L-profile, glued to slab
J J
H H
Fig. 16. Section J-J, see Fig. 16.

Fig. 15. Section H-H, see Fig. 15. Only
Only transducers 46–49 shown.
transducers 48 and 49 shown.
55 54
55
J J J
H H
Fig. 17. Section H-H, see Fig. 15. Only Fig. 18. Section J-J, see Fig. 18.
transducers 55 is shown. Only transducers 54 and 55 shown.
3.5
-
Strain
354
VTT.CP.LBL.320.1998
-
5 Loading strategy
5.1
relationship
All measuring devices were zero-balanced when the actuator forces Pa1 and Pa2 were
equal to zero but the weight of the loading equipment and the self-weight of the
structure were acting.
The weight of the loading equipment below loads Pa1 was equal to 5,6 kN and that
below loads Pa2 equal to 1,2 kN. To create two uniform line loads to the floor, attempts
were made to keep Pa2 equal to Pa1 + 4,4 kN.
The loading history for Pa1 and Pa2 is shown in Fig. 20 and the measured support
reaction in Fig. 21.
175
150 Pa1
Pa2
125
100
P [kN]
75
50
25
0
0 50 100 150 200 250
Time [min]
Fig. 19. Actuator forces Pa1 and Pa2 vs. time.
355
VTT.CP.LBL.320.1998
600
550
500
450
400
Rp,obs [kN]
350
300
250
200
150
100
50
0
0 25 50 75 100 125 150 175
Pm=(Pa1+Pa2)/2 [kN]
Fig. 20. Support reaction RP,obs below Northern end of middle beam. Note the slight
increase in RP,obs despite the reduction of Pm at the end of the curve.
1
0,9
0,8
0,7
0,6
Vp,obs / Pm
0,5
0,4
0,3
0,2
0,1
0
0 25 50 75 100 125 150 175
Pm=(Pa1+Pa2)/2 [kN]
Fig. 21. Development of observed support reaction VP,obs per one slab as function of
mean actuator force Pm.
5.2
After failure -
356
VTT.CP.LBL.320.1998

Stage I The joint concrete between the middle beam and the slab units
(Cyclic) started to crack along the slab ends.
Stage II At Pa1 = 70 kN the cracks in the joint concrete between the middle
(Monotonous) beam and the slab units were continuous. At Pa1 = 80 kN the tie
beams above the end beams started to crack. At Pa1 = 115 kN
flexural cracks were observed in slab unit 1 (see Fig. 23).
At Pa1 = 150 kN first shear cracks were observed in slab unit 5 and
later in slab units 8, 1 and 4. At Pa1 = 168,3 kN and Pa2 = 174,0 kN
a shear failure took place in slab units 5–8.
The cracking pattern after failure is shown in Fig. 23 and in App. B.
Figs 6–15.
After failure The concrete infill in the cores of slab units remained virtually uncracked
as can be seen in App. B, Figs 16–17. The cracking took typically
place along the surface of the core filling and along the slab ends.
7.1
Cracks at
service load
7.2
Cracks after 160
failure
145 40 40 145
8 4
125
7 3
80 70 80
70
6 2
125 145
5 1
135 40 90 160
125 150 135 125 115
Fig. 22. Cracking pattern after failure at top surface and at edges of floor. The force
values refer to the actuator force Pa1.

The ratio (measured support reaction below one end of the middle beam)/ (theoretical support
reaction due to actuator forces on half floor) is shown in Fig. 23. The theoretical reaction is
calculated assuming simply supported slabs. This comparison shows that the support
reaction due to the actuator forces can be calculated accurately enough assuming simply
supported slabs. However, the failure of the slab ends first at the South end of the middle
beam resulted in reduction of support reaction below that end and increase at the North end
while the symmetrically positioned actuator forces were reduced. Therefore, the measured
support reaction under the maximum actuator forces, not the maximum measured support
reaction, is regarded as the indicator of failure. See also Chapter 10, Fig. 29.
357
VTT.CP.LBL.320.1998
RP,obs / RP,th
1,2
1,0
0,8
0,6
0,4
0,2
0,0
0 20 40 60 80 100 120
Pa1 [kN]
Fig. 23. Ratio of measured support reaction of the middle beam (Rp,obs) to theoretical
support reaction (Rp,th) vs. actuator force Pa1. Only actuator loads Pa1 and Pa2 are taken
into account in the support reaction.
Assuming simply supported slabs and calculating the support reaction of the actuator
loads from equilibrium of forces, gives support reaction which is 83,7% of the actuator
loads. On the other hand, just before the failure, the measured support reaction under
the North end of the middle beam was 81,68% of the loads on half floor. Using this
relationship for the weight of loading equipment, and assuming that the weight of the
slabs and jointing concrete was distributed to both ends of the slab units as if the slabs
were simply supported beams, the shear resistance of one slab end (support reaction of
slab end at failure) due to different load components can be calculated as shown below.
Support reaction due to

weight of slab unit (Vg,sl) 3810  3750  3770  3760 18,5 kN
9 ,82 N
2 4
weight of cast-in-situ 2  0 ,52  3  0 ,75 0,8 kN
concrete (Vg,isc) kN
4
loading equipment (Veq) 5 ,6  1,2 2,8 kN
0 ,8168 kN
2
actuator loads (Vp) 168 ,3  174 ,0 139,8 kN
0 ,8168 kN
2
The shear resistance of one slab end due to imposed load

Vobs,imp = Vp + Veq = 139,8 kN + 2,78 kN = 142,6 kN
and the total shear resistance
Vobs = Vobs,imp + Vg,sl + Vg,isc = 142,6 kN + 18,5 kN + 0,8 kN = 161,9 kN
are obtained.
358
VTT.CP.LBL.320.1998
9.1
MPa MPa
Bottom plate  355 Nominal (S355J0)
Lattice girders 500 Nominal (A500HW)
Nominal value for reinforcing bars A500H
9.2
Strength of
slab concrete, h
floor test mm mm
d
Upper flange of slab1
6 50 50 1.4.2009
(11–13) and 5 (51 –53),
St.deviation [MPa] 5,3 Tested as drill2460 kg/m3
9.3
Strength of
slab concrete, h
d
Upper flange of slab 9
6 50 50 1.4.2009
(91–93) and 10 (101 –103),
Mean strength [MPa] 61,8 (0 d)1) vertically drilled
Tested as drilled2)
9.4
Strength of a
longitudinal mm
joints of slab a
units 3 150 25.3.1998 Kept in laboratory in the same
9.5
Strength of
concrete in h
LBL beam mm mm
d
3 75 75 1.4.2009 Upper flange, vertically
Mean strength [MPa] 84,8 (+7 d)1) drilled
Tested as drilled2)
St.deviation [MPa] -
1)
2)
359
VTT.CP.LBL.320.1998
In the following Figs 26–32, Vp stands for the shear force of one slab end due to
imposed actuator loads, more accurately 1/4 of the measured support reaction below
the North end of the middle beam. On the other hand, in Fig. 33 the mean actuator force
= (Pa1+Pa2)/2 is used as load parameter.
Comparison between Figs 29 and 33 shows that the maximum load carrying capacity
was achieved before the maximum support reaction below the North end was measured.
This suggests that there was load transfer from the South to the North while the shear
cracks propagated in the slab units.
10.1
Deflections In Fig. 29 (middle beam) the net deflection (= measured deflection – rigid body motion
due to settlement of supports) is given, other figures show the measured deflection.
160 160
140 140
120 120
100 100
Vp [kN
Vp [kN
80 11 80 16
12 17
60 60
13 18
14 40 19
40
15 20
20 20
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Fig. 24. Deflection on line I along end Fig. 25. Deflection on line II in the middle
beam. of slabs 1–4.
160 160
140 140
120 120
100 100
Vp [kN
Vp [kN
80 80 26
21 27
60 22 60
28
23 29
40 40
24 30
25 20
20
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Fig. 26. Deflection on line III close to the Fig. 27. Net deflection on line IV along the
line load, slabs 1–4. middle beam.
360
VTT.CP.LBL.320.1998
160 160
140 140
120
120
100
Vp [kN
100
36
Vp [kN
80
80 31 37
32 60
38
60
33 40 39
40 34 40
20
35
20 0
0 5 10 15 20 25 30
0
0 5 10 15 20 25 30 Deflection [mm]
Deflection [mm]
Fig. 29. Deflection on line VI in the middle
Fig. 28. Deflection on line V close to the of slabs 5–8.
line load, slabs 5–8.
160
160
140 140
120 120
100 Vp [kN] 100

28, middle
Vp [kN
80
80 41 13
42 60 43
60
43 40
40 44
20
45
20 0
0 5 10 15 20 25 30
0
Deflection [mm]
0 5 10 15 20 25 30
Deflection [mm]
Fig. 31. Net deflection of midpoint of
Fig. 30. Deflection on line VII along end middle beam (28) and those of end
beam, slabs 5–8. beams (13, 43). (Settlement of supports
eliminated.)
160 160
10.2
140
Crack width 140
120
120
100
100
Vp [kN]
5
Vp [kN
80
5 6
80
60 7
6
60 8
7 40 9
40 8 10
20
9
20 10 0
0,0 0,5 1,0 1,5 2,0
0
Crack width [mm]
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Crack width [mm]
Fig. 33. Same as previous figure but the
Fig. 32. Differential displacement ( crack cyclic loading phase is not shown.
width) measured by transducers 5–10.
361
VTT.CP.LBL.320.1998
10.3
Average strain -
(actually
differential
displacement)
10.4
Shear 160 160
displacement
140 140
120 120
100 100
Vp [kN]
Vp [kN]
80 80
46 slab 1 50 slab 4
60 60
47 slab 1 51 slab 4
48 slab 5 40 52 slab 8 40
49 slab 5 53 slab 8
20 20
0 0
-4 -3 -2 -1 0 1 -4 -3 -2 -1 0 1
Fig. 34. Southern end of middle beam. Fig. 35. Northern end of middle beam.
Differential displacement between edge Differential displacement between edge
of slab and middle beam. A negative of slab and middle beam. A negative
value means that the slab is moving value means that the slab is moving
towards the end of the beam. towards the end of the beam.
160
140
120
100
Vp [kN]
80
46-47 slab1
48-49 slab 5 60
50 - 51 slab 4 40
52 - 53 slab 8
20
0
Fig. 36. Shear displacement = differential -2,0 -1,5 -1,0 -0,5 0,0
displacement at upper edge – differential Displacement [mm]
displacement at lower edge of slab.
362
VTT.CP.LBL.320.1998
10.5 Relative displacement between bottom flange and concrete component of middle
beam
160 160
140 140
120 120
100 100
Vp [kN]
Vp [kN]
80 80
54 60 60
56 55
57 40 40
20 20
0 0
-0,06 -0,05 -0,04 -0,03 -0,02 -0,01 0,00 0,01 -1,00 0,00 1,00
Fig. 37. Differential displacement Fig. 38. Differential displacement measured

measured by transducers 54, 56 and 57. by transducer 55 (obviously incorrect).
11 Reference tests
6 5 4 8
P
7
1 2 3
3560 1200
P
Gypsum
1 (6) 2 (5) 3 (4)
8 (7)
40 40 7118 (Slab unit 9) 40 40
7120 (Slab unit 10)
Fig. 39. Layout of reference test. a) Plan. b) Elevation. The displacements were measured
by transducers 1–6, the bond slip of the outermost strands by transducers 7 and 8.
363
VTT.CP.LBL.320.1998
350
300 Slab 9
Slab 10
250
200
P [kN]
150
100
50
0
0 5 10 15 20 25 30 35 40 45 50
Deflection [mm]
Fig. 40. Actuator force – time relationship. R1 : Slab 9; R2:Slab 10.
350 350
300 300
250 250 1
1 Pa [kN]
Pa [kN]
200 2 200 2
150 3 150 3
4 4
100 100
5 5
50 6 50 6
0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
R1 R2
Fig. 41. Displacements measured by transducers 1–6. Pa is the actuator force.
350 350
300 300
250 250
Pa [kN]
Pa [kN]
200 200
7 7
150 150
8 8
100 100
50 50
0 0
0 0,01 0,02 0,03 0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08
Slippage [mm] Slippage [mm]
R1 R2
Fig. 42. Slippage of outermost strands measured by transducers 7–8. Pa is the actuator
force.
364
VTT.CP.LBL.320.1998
Span Vg Pa Peq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
Mean 295,3 246,1
13 Discussion
1. At maximum load, the net deflection of the middle beam due to the imposed
actuator loads (deflection minus settlement of supports) was 20,9 mm or L/240,
i.e. rather small. It was 4,5–7,5 mm greater than that of the end beams. See Fig.
33 for the difference. Hence, the torsional stresses due to the different deflection
of the middle beam and end beams may have had a minor effect on the failure
of the slabs.
2. The shear resistance measured in the reference tests was typical of the similar
slabs produced in Finland, see Pajari, M. Resistance of prestressed hollow core
slab against a web shear failure. VTT Research Notes 2292, Espoo 2005.
3. The bond between the cast-in-situ concrete and the edges of the hollow cores
was weak. This can be seen in the photographs in App. B in which the hardened
hollow core fillings remained almost intact in most cases.
4. The sliding of the edge slabs along the middle beam was negligible before 85%
of the failure load was achieved. At failure the differential displacement between
the bottom flange of the beam and the soffit of the edge slabs was of the order
of 0,2 … 0,5 mm. This reduced the negative effects of the transverse actions in
the slab and had a positive effect on the shear resistance.
5. The transverse shear deformation of the edge slabs was considerable which can
be seen in Figs 47–49.
6. The failure mode was web shear failure of edge slabs. The LBL beam seemed
to recover completely after the failure.
365
VTT.CP.LBL.320.1998
APPENDIX A: STEEL COMPONENT OF LBL BEAM
Fig. 1. Steel component of LBL beam. Elevation (lattice girders) and plan (bottom plate). See Fig. 2
for sections A, B and C.
A366
VTT.CP.LBL.320.1998
Fig. 2. Sections A, B and C, see Fig. 1.
A367
VTT.CP.LBL.320.1998
Fig. 3. Splicing of lattice girders.
A368
VTT.CP.LBL.320.1998
Fig. 4. Bottom flange and position of lattice girders.
A369
VTT.CP.LBL.320.1998
Fig. 5. Design of lattice girders.
A370
VTT.CP.LBL.320.1998
APPENDIX B: PHOTOGRAPHS
Fig. 1. Overview. Fig. 2. Loading equipment.
Fig. 3. Arrangements at end beams. Fig. 4. Transducers at the end of middle beam.
Fig. 5. Failure pattern of slab unit 5. Fig. 6. Detail of slab unit 5 after failure.
B371
VTT.CP.LBL.320.1998
Fig. 7. Slab unit 5 seen from above after failure. Fig. 8. Failure pattern of slab unit 8.
Fig. 9. Cracking pattern of slab unit 1 after failure of Fig. 10. Detail of slab unit 1. Note the transverse
slab units 5–8. crack in the soffit.
Fig. 11. Cracking pattern of slab unit 4 after failure

of slab units 5–8.
Fig. 12. Failure pattern of slab units 5–8 seen from
above.
B372
VTT.CP.LBL.320.1998
Fig. 13. Failure pattern of slab units 5–8 seen from Fig. 14. Failure pattern of slab units 5–7 seen from
below. below.
Fig. 15. Edge of middle beam after removing slab Fig. 16. Edge of middle beam after removing slab
units 1–4. units 5–8.
Fig. 17. Detail of middle beam after removing slab Fig. 18. Failure pattern of slab unit 9 in reference
units. test.
B373
VTT.CP.LBL.320.1998
Fig. 19. Failure pattern of slab unit 9 in reference Fig. 20. Failure pattern of slab unit 10 in reference
test. test.
Fig. 21. Failure pattern of slab unit 10 in reference

test.
B374
1.1
VTT.CR.Delta.400.1999 Last update 2.11.2010
Identification
and aim DE400 (Internal identification)
Aim of the test To quantify the interaction between the Delta beam
and 400 mm thick hollow core slabs.
1.2
Test type
Fig. 1. Overview of test arrangements. Delta beam in the middle, steel beams at the
ends.
1.3 VTT/FI 2.12.1999

Laboratory
& date of test
1.4
Date 29.3.2000
Availability Confidential, owner is Peikko Group Oy,
P.O. Box 104, FI-15101 Lahti, Finland

(see also Appendices A)
375
2.1
General plan B 5000 (span of beam) West
8400 (Length of slab)

4 3 2 1
1200 1200 1200 1200

1200 (from slab end)
50
(from slab end)

1700
South
E
D C
540 5000 (span of beam)
North
D C
E
1200
8 7 6 5
A A
East
Fig. 2. Plan.
376
2.2
End beams B B
50 50 Concrete K30
2 T8 Plastic plug
L=4750
Plywood
Steel plate
50 50 Steel rod, free to roll

10 75 250
RHS
St355
150
Fig. 3. Arrangements at end beam. T8 refers to a reinforcing bar A500HW with diameter
8 mm, see 2.3.

2.3
Middle beam The beam, see Fig. 4 and App. A, comprised
- a steel box girder with inclined, perforated webs
- a concrete component, cast by VTT in laboratory on the 19th of November 1999,
which filled the empty space between the slab ends laying on the bottom flangee of
the beam.
Concrete: K30
Reinforcement:
Txy: Hot rolled, weldable rebar A500HW, diameter = xy mm
Structural steel in Delta beam: Raex 420, fy  420 MPa (nominal fy) unless otherwise
specified
Holes
 150
225 300 300
130
(vertical
distance)
End plate 225

385x490x6
Ledges flame 30 30 Ledges flame

cut cut
5250
Fig. 4. Delta beam. The diameter of the holes was 150 mm.
377
278
25
385
Rebars 2 32
65
15 480 6
100 500 100
Fig. 5. Cross-section of Delta beam. The rebars were welded to the webs of the beam.
2.4
Arrangements - Simply supported, span = 5,0 m, roller bearing at one end
at middle
beam C C
Rebar 12 L = 3800 in each longitudinal joint between adjacent slab units
540
278
Fig. 6. Section along joint between adjacent slab units, see Fig. 2.
378
D D
540
Rebars 212 Rebars 212
278
25
400
65
100 15 480 385
Concrete K30
Plastic plug Rebars32
Fig. 7. Section along hollow cores, see Fig. 2.
E E
Rebar 8 L = 3800, in cast-in-situ concrete outside longitudinal edges of

outermost slab units 540
278
Fig. 8. Section along outer edge of floor specimen, see Fig. 2.
379
2.5
Slabs

- Extruded by Parma Betonila Oy, Hyrylä factory 29.10.1999
- 13 lower strands J12,5; initial prestress 1000 MPa
1158 (at mid-depth) Max measured bond slips:

1,9 in slab 7; 1,8 in slab 8;
46,1 1,7 in slabs 2, 4, and 7

308,4 units = 5,39 kN/m
b w,i
38,6 42,1
 b w,i = 297,6

characteristics.
46,1
r=125
60,5
400 193 b w,i
215
62
38,4 42
62 91 62  b w,i = 297,6
1158
Fig. 11. Cross-sectional geometry based on nominal

and measured geometry.
2.6
supports
380
2.7
Loading
arrangements
4 3 2 1
(from slab end) South 300 300 300 300 300 300 300 300
300 300 300 300 300 300 300 300
Pa2 Pa2
1200
North
E
Pa1 Pa1 Pa1 Pa1
540
E
1200
Pa2 Pa2
A 8 7 6 5 A
A A
Det 1
Pa1 Pa1 2 Pa2 2Pa2 Pa1 Pa1
Load cells 2, 3 and 4 under support of beam
Det 1 F F
Pa1 Steel plates
F
Teflon
plates
Gypsum
F
550
381
3 Measurements
3.1
Support 1
reactions 3 load cells
North
5
3.2 15 14 13 12 11
Vertical 40 40 I 40 40 40
30 30 30 30 30
displacement
4 3 2 1 4800
30 30 30 30 30
II
20 19 18 17 16
3600
Transducers for measuring vertical displacement

- Transducers 27, 28, 29 on the continued centrelines
of the longitudinal joints between the slabs
- Transverse measures given with respect to the
centrelines of the joints between the slabs
30 30 30 30 30
25 20 20 24 23 22 21
III 20 20 20
270 540
30 29 IV 28 27 26
270
35 34 V 33 32 31
8 7 6 5
3600
40 39 38 37 36
VI
4800
45 44 VII 43 42 41
Fig. 15. Location of transducers 11 … 45 for measuring deflection along lines I … VII.
382
3.3
Average strain Not measured
3.4
Horizontal. Measures differential displacement between slab end and beam (crack width)
displacements Measures differential displacement between slab edge and beam in beam's direction,
numbers in parentheses refer to the soffit of the floor, others to the top
50 4 3 2 1 46
(51) (47)
20
7 6 5
130
40 40 40
Middle beam
10 9 130 8
20 H H
52 8 7 6 5 48
(53) (49)
Fig. 16. Transducers measuring crack width (5–10) and shear displacement at the ends
of the middle beam (46–53). Transducers 47, 49, 51 and 53 are below transducers 46,
48, 50 and 52, respectively.
H H J J
(Only transducers 48 and 49 shown) (Only transducers 46 - 49 shown)
J 48 46
L-profile, glued to slab 48 20
30 (from slab end)
49
J
Fig. 17. Section H-H, see Fig. 20.
383
3.5
Strain and The strain was measured using strain gauges at mid-span of the middle beam. Their
differential position is shown in Fig. 19 The contact between the soffit of the slabs and the ledges of
vertical the middle beam was monitored by two transducers shown in Fig. 19.
displacement
Plan
CL
7 3
55 57 59 56 54
A CL 58
A
61 60
6 2
A A
59 58
57 61 60 56
55 CL 54
Fig. 19. Horizontal strain gauges 56–61 and transducers 54–55 for measuring
differential vertical displacement.
384
- None
5 Loading strategy
5.1
Load-time Date of the floor test was 2.12.1999. The loading took 2 h 35 min. The actuator loads at
relationship different time steps are given in Fig. 20.
250
200
Actuator load [kN]
150
100 Pa1
Pa2
50
0
0 10 20 30 40 50
Time step
Fig. 20. Actuator forces Pa1 and Pa2 at different time steps.
5.2
After failure
385
Before test Some longitudinal cracks along the strands were discovered in the
soffit of the slabs before the test, see Fig. 22. They were below the
webs (strands) of the hollow core units, parallel to the strands and
obviously caused by the transfer of the prestressing force. Their
width was less than 0,1 mm before the test, but it grew when the
floor was loaded. They were obviously caused by the release of the
prestressing force.
Stage I The joint concrete started to crack along the web of the Delta beam,
see Fig. 23. In Figs 22 and 23 as well as in all figures in App. A
presenting cracking, the numbers refer to the value of Pa1 in kN
when the crack was first observed.
Stage II At Pa1 = 80 kN the tie beams at both ends of the specimen cracked
vertically. There were also continuous, visible cracks between the
joint concrete and the Delta beam, see Fig. 21. From Pa1 = 80 kN
on, an increasing number of inclined cracks were observed in the
corners of the outermost slab units close to the ends of the Delta
beam.
At Pa1 = 200 kN, diagonal shear cracks were observed in the
corners of slab units 1 and 5, see App. 1, Figs 8 and 9. These
cracks grew in width and length until at Pa1 = 238 kN, Pa2 = 233 kN
slab units 5–8 failed in shear between the line load and the middle
beam. The failure mode is illustrated in Fig. 23 and in App. A, Figs
8–24.
After failure The joint between the slab ends and Delta beam opened along the
webs of the beam. The joint concrete as well as the interface
between the joint concrete and the slab ends remained virtually
uncracked, see Fig. 21 and App. A, Figs 23–28.
Fig. 21. Cracking mode between joint concrete and Delta beam.
The ledge of the Delta beam was in tight contact with the bottom
surface of the slab units until failure. After the failure, the collapsing
slab units deformed the ledges as shown in App. 1, Figs 14–16 and
25–28.
386
7.1
Cracks at
service load 60 0
8 0 4
60 60
60
7 0 3
0
0 0
60 60
6 0 60 2
60 60
0
5 1
0 60
60
Fig. 22. Cracks in the soffit at service load (Pai = 60 kN). The initial cracks observed
before loading are indicated with 0. The figures give the value of the actuator force at
which the crack was observed.
7.2
Cracks after
failure
235 210 100 210 235
225 80 225
South
8 60 80 4
7 3
225
80 225
80
6 2
5 1
190
East North200 200 80 170 100120 200
West
387
In Fig. 24 the reaction force under one end of the Delta beam, measured by load
cells 2–4, is shown as a function of the load on half floor. The relationship is slightly
nonlinear.
700
600
500
400
300
200
100
0
0 200 400 600 800 1000
Fig. 24. Reaction force under support 1 of middle beam vs. load on half floor
(=2Pa1 +2Pa2).
Fig. 25 illustrates the ratio of measured support reaction of one slab end Vp due to the
actuator loads (= one fourth of measured support reaction under one end of Delta beam)
to the imposed loads on one slab unit (= (4xPa1+4xPa2)/8) as a function of Pa1. The
dashed line in Fig. 25 indicates what the response for a simply supported slab would
have been.
1.0
0.9
0.8
0.7 Simply supported behaviour
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 50 100 150 200 250
P1 [kN]
Fig. 25. Ratio of measured support reaction of a half floor to the load on the half floor.
Only imposed load considered.
388
Assuming simply supported slabs and calculating the support reaction of the actuator
loads from equilibrium of forces gives a support reaction which is 86,1% of the actuator
loads. On the other hand, just before the failure, the measured support reaction under
end 2 of the Delta beam was 82,9% of the loads on half floor. Using this relationship
also for the weight of loading equipment, and assuming that the weight of the slabs and
cast-in-situ concrete was distributed to both ends of the slab units as if the slabs were
simply supported beams, the shear resistance of one slab end (support reaction of slab
end at failure) due to different load components can be calculated as shown in Table 2.
Support reaction due to

weight of slab unit (Vg,sl) 4600  4630  4590  4630 22,65 kN
9,82 kN
24
weight of cast-in-situ 2  0,72  3  1,4 1,4 kN
concrete (Vg,isc) kN
4
loading equipment (Veq) 5,6  1,2 2,82 kN
0,8291 kN
2
actuator loads (Vp) 354,6  263,2  162,7 195,1 kN
kN
4
The shear resistance of one slab end due to imposed load
Vu,imp = Vp + Veq = 195,1 kN + 2,8 kN = 197,9 kN
and the total shear resistance
Vu = Vu,imp + Vg,sl + Vg,isc = 197,9 kN + 22,7 kN + 1,4 kN = 222,0 kN
are obtained.
The strong deviation of the support reaction from the simply supported behaviour of the
slab units at Pa1 < 100 kN cannot be explained by the negative bending moment carried
by the tie reinforcement penetrating the Delta beam. The bending moment
corresponding to the yield stress 500 MPa of the tie reinforcement is of the order of
37 kNm per floor and 37/4 = 9,3 kNm per slab unit. This increases the support reaction
of one slab unit at the Delta beam by 1,1 kN which is far too small to explain the
behaviour. It is more likely that the extra support moment is due to the joint concrete.
The tie reinforcement obviously helps in mobilizing vertical friction, dowel action and
aggregate interlocking along the inclined webs of the Delta beam, particularly in the web
holes. With increasing crack width along the Delta beam these effects fade out.
At Pa1 = 200 kN diagonal shear cracks at the ends of slab units 1 and 5 started to
change the load-carrying mechanism of these slab units. The loads were more and
more transferred to the neighbouring slab units and less directly to the beam. As a
result, the support reactions of the Delta beam became different in such a way that the
reaction force at the North end was smaller than that at the opposite end. This effect can
be seen in Fig. 25 when Pa1 is greater than 200 kN.
389
9.1
MPa MPa
Delta beam  420 Nominal (Raex 420)
9.2
Strength of
slab concrete, h
# Cores h d Date of Note
floor test mm mm test
d
6 50 50 10.12.1999 Upper flange of slabs 5 and 6,
vertically drilled. Tested as
Mean strength [MPa] 68,6 (+8 d)1)
drilled2)
9.3
Strength of
slab concrete, h
d
6 50 50 10.12.1999 Upper flange of slab9, vertic-
Mean strength [MPa] 67,3 (+2 d)1) ally drilled. Tested as drilled2)
9.4
Strength of a
concrete in # a a Date of test Note
Delta beam mm
a
and joints
1)
2)
390
In the following figures, Vp stands for the average shear force of one slab end due to
imposed actuator loads, calculated assuming simply supported slabs.
10.1
Deflections Vp [kN] Vp [kN]
200 200
180 180
160 160
140 140
120 120
11 16
100 100
12 17
80 80
13 18
60 60
14 19
40 40
15 20
20 20
0 0
0 5 10 15 20 25 0 5 10 15 20 25 30 35
Fig. 26. Deflection on line I along Western Fig. 27. Deflection on line II in the middle
end beam. of slabs 1–4.
Vp [kN] Vp [kN]
200 200
180 180
160 160
140 140
120 120
21 26
100 100
22 27
80 80
23 28
60 60
24 29
40 40
25 30
20 20
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Fig. 28. Deflection on line III close to the Fig. 29. Deflection on line IV along the
Delta beam, slabs 1–4. middle beam.
Vp [kN] Vp [kN]
200 200
180 180
160 160
140 140
120 120
31 36
100 100
32 37
80 80
33 38
60 60
34 39
40 40
35 40
20 20
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30 35 40
Fig. 30. Deflection on line V close to the Fig. 31. Deflection on line VI in the middle
Delta beam, slabs 5–8. of slabs 5–8.
391
Vp [kN]
200
200
180
180
160
160
140
140
Vp [kN]
120
120
41 100
100
42 80 13,West end
80
43 60
60 28, Delta
44 40
40
45 43,East end
20 20
0 0
0 5 10 15 20 25 0 2 4 6 8 10 12 14 16 18 20 22
Deflection [mm] Displacement [mm]
Fig. 32. Deflection on line VII along Fig. 33. Net deflection of midpoint of middle
Eastern end beam, slabs 5–8. beam (28) and those of end beams (13, 43).
(Settlement of supports eliminated.)
10.2
Crack width Vp [kN] Vp [kN]
200 200
180 180
160 160
140 140
120 5 120 5
100 6 100 6
80 7 80 7
60 8 60 8
40 9 40 9
20 10 20 10
0 0
0 1 2 3 4 0 1 2 3 4
Fig. 34. Differential displacement ( crack Fig. 35. Same as previous figure but the
width) measured by transducers 5–10. cyclic loading phase is not shown.
10.3
Opening of
200
gap between
slab and ledge
150
of Delta beam
Vp [kN]
100
54, slabs 2 - 3 50
55, slabs 6 - 7
0
-1,00 -0,75 -0,50 -0,25 0,00 0,25 0,50 0,75 1,00
Growth of gap [mm]
Fig. 36.
392
10.4
Shear 200 200
displacement
150 150
Vp [kN]
Vp [kN]
46, slab 1 50, slab 4
100 100
47, slab 1 51, slab 4
48, slab 5 52, slab 8
50 49, slab 5 50 53, slab 8
0 0
-0,5 0,0 0,5 1,0 1,5 2,0 2,5 3,0 -0,5 0,0 0,5 1,0 1,5 2,0 2,5 3,0
Fig. 37. Northern end of middle beam. Fig. 38. Southern end of middle beam.
200
180
160
140
Vp [kN]
120
100
80 46-47, slab 1
60 48-49, slab 5
40 50-51, slab 4
20 52-53, slab 8
0
-0,5 0,0 0,5 1,0 1,5 2,0 2,5 3,0
Displacement [mm]
Fig. 39. Shear displacement = differential

displacement at upper edge – differential
displacement at lower edge of slab.
10.5 Strain
200 200
150 150
Vp [kN]
Vp [kN]
56, ledge, slabs 2 - 3 100 100

60
57, ledge, slabs 6 - 7
50 61
58, top flange 50
59, top flange
0 0
-2000 -1500 -1000 -500 0 0 250 500 750 1000 1250 1500
-6
-6
Strain [10 ] Strain [10 ]
Fig. 40. Strain measured by gauges Fig. 41. Strain measured by gauges 60
56–59. and 61 parallel to Delta beam.
393
11 Reference tests
6 5 4 8
a)
P
1 2 7
3
1200
4800 (Test R9/1)
4150 (Test R9/2)
P
1 (6) Gypsum 3 (4)
2 (5)
b)
8 (7)
50 50 8290 (Test R9/1) 50 50
7150 (Test R9/2)
Fig. 42. Loading arrangements and position of transducers for measuring displacement
(1–6) and slip of strand (7 and 8). a) Plan. b) Elevation.
600
500
400
P [kN]
300
200 R9/1
R9/2
100
0
0 5 10 15 20 25 30 35 40 45 50 55
Step
Fig. 43. Actuator force – time relationship.
600 600
500 500
400 400 1
1
P [kN]
P [kN]
300 2 300 2
3 3
200 200 4
4
5
100 5 100
6
6
0 0
-5 0 5 10 15 -5 0 5 10 15
R9/1 R9/2
Fig. 44. Displacements measured by transducers 1–6. P is the actuator force.
394
600 600
500 500
400 400
P [kN]
P [kN]
300 300
7 7
200 8 200 8
100 100
0 0
0,00 0,04 0,08 0,12 0,16 0,0 0,0 0,1 0,1 0,2
R9/1 R9/1
Fig. 45. Slippage of outermost strands measured by transducers 7–8. P is the actuator force.
the slab, actuator force P at failure, weight of loading equipment Peq, total shear force
Span Vg P Peq Vobs vobs

Test Date Note
mm kN kN kN kN kN/m
Mean 419,5 349,6
13 Discussion
2292, Espoo 2005.
3. The maximum difference in the net mid-point deflection of the beams was
4,0 mm. Hence, the torsional stresses due to the different deflection of the
middle beam and end beams had a negligible effect on the failure of the slabs.
4. The edge slabs slided 0,13 … 0,23 mm along the beam before failure. This
reduced the negative effects of the transverse actions in the slab and had a
positive effect on the shear resistance.
5. The failure mode was web shear failure of edge slabs. The plastic deformation
of the ledges of Delta beam was considerable on the failed side. Otherwise the
Delta beam seemed to recover completely after the failure.
395
Fig. 1. Delta-beam (middle beam in floor test).
Fig. 2. Detail of Delta-beam.
A396
Fig. 3. Overview of test arrangements.
A397
Fig. 5. Arrangements at support of middle beam.
Fig. 6. Measuring equipment on slab units 2, 3, 6 and 7.
A398
Fig. 7. Arrangements at support of end beam.
Fig. 8. Failure pattern of slab unit 5.
A399
A400
Fig. 12. Cracking pattern of tie beam between slab units 2 and 3 at failure of the floor.
A401
Fig. 13. Cracking pattern of tie beam at end of slab units 6 and 7 at failure of the floor.
Fig. 14. Deformation of ledge of Delta-beam under slab units 5–8 after failure of the floor.
A402
Fig. 15. Deformation of ledge of Delta-beam under slab units 5–8 after failure of the floor.
Fig. 16. Deformation of ledge of Delta-beam under slab unit 5 after failure of the floor.
A403
Fig. 17. Top surface of the floor after removal of loads.
Fig. 18. Top surface of slab unit 5 after failure.
A404
A405
A406
Fig. 23. Cracks along middle beam.
Fig. 24. Cracks along middle beam.
A407
Fig. 25. Joint concrete between slab unit 5–8 and midde beam after removal of slab units.
Fig. 26. Detail of the joint concrete after failure.
A408
Fig. 27. Web and deformed ledge of middle beam after removal of slab units and joint concrete. The
tie bars penetrating the beam have been flame-cut after the test.
Fig. 28. Detail of middle beam after failure.
A409
Fig. 29. Reference test R9/1. Failure pattern.
A410
A411
VTT.CP.Super.320.2002
1.1
Identification
VTT.CP.Super.320.2002 Last update 2.11.2010
and aim
SUP320 (Internal identification)
Aim of the test There were essential differences between Super beam
and beams in previous floor test. A test was needed to
quantify the interaction between the Super beam and
hollow core slabs.
1.2
Test type
HE200B HE200B
Super beam
1.3 VTT/FI 17.1.2002

Laboratory
& date of test
1.4
Date 3.4.2002
Availability Confidential, owner is Betset Oy,
P.O. Box 14, FI-43701 Kyyjärvi, Finland

413
2.1 B 4800 (span of beam) West

General plan
B
South
North
9600 (Length of slab )
4 3 2 1
1200 1200 1200 1200

40
(from slab end)
(from slab end)

1700
1300
E
C D
4800 (span of beam)
680
E C D
8 7 6 5
9600
A A
4800 (span of beam)

East
Fig. 2. Plan.
414
2.2
End beams HE 200 B
B B
15
200
9 15 Concrete K30
40 50
Plastic plug
200 Tie beam
26
Fig. 3. End beam. L=4700
Plywood 40 40
Steel plate Steel rod,

free to roll
HE 200 B
Fig. 4. Arrangements at end beam. 6 refers to a

reinforcing bar TW6, see 2.3.
Simply supported, span = 4,8 m, roller bearing below Southern end

2.3
Middle beam The beam, see Fig. 5 and App. A, comprised
- a prefabricated part with two L-shaped steel profiles, a folded and perforated steel
plate welded to these profiles and a precast and prestressed concrete slab with ribbed
reinforcement
- a cast-in-situ part of concrete which filled the empty space between the slab ends
laying on the L-profiles.
- Lower, prestressed part cast by Betset Oy, 13.12.2001

- Upper part cast by VTT in laboratory, 4.1.2002
Concrete: K80 in the prefabricated part, K40 in the upper part
Passive reinforcement and tendons in Super beam:

TWxy: Hot rolled, weldable rebar A500HW,  = xy mm
J12,9: Prestressing strand, 7 wires,  =12,9 mm, Ap = 100 mm2, prestress = 1380 MPa
Kz: Cold formed rebar B500K,  = z mm
Structural steel: S355J2G3, fy  355 MPa (nominal fy) unless otherwise specified
Tie reinforcement: Straight, nonprestressed prestressing strands across the middle

beam and in the longitudinal joints of the slabs or outside the outer slabs, see Figs 7- 9.
J12,5: 7 indented wires,  =12,5 mm, Ap = 93 mm2

J9,3: 7 indented wires,  =9,3 mm, Ap = 52 mm2
415
2.4
Arrangements - Simply supported, span = 4,8 m, roller bearing at both ends
at middle
beam C C D D
250 Concrete K40 Concrete 2 J12,5 L = 4000
40 K30 Concrete K40 Concrete K30
240 230
110
40
10 Concrete K80
70 30
Plastic plug 600 100
10 10
1 J12,5 L = 3000
Fig. 6. Section along hollow cores, see Fig. 7. Section along joint between
Fig. 2. adjacent slab units, see Fig. 2.
416
E E
Concrete 1 J12,5 L = 4000
K30 Concrete K40 Concrete K30
1 J9,3 L = 3000
Fig. 8. Section along outer edge of floor

specimen, see Fig. 2.
Fig. 9. Tie reinforcement across middle
beam.
2.5
Slabs

- Extruded by Betset Oy, Kyyjärvi factory 21.11.2001
- 2 upper strands J12,5 initial prestress 900 MPa

42,3 21,3 2,5; 2,4; 2,3 and 2x2,1 mm,
all in slab unit 6.
44,3 55,9 55,9 55,9 44,3 Measured weight of slab
317
units = 4,40 kN/m
86
34,4 b w,i = 256 43

characteristics.
417
35 141 21 b w,i = 256

21,6 35
72,5
r126
r126
71 b w,i
223 229
320 72,5 105
6
r12
r12
67,3 72,5 86
43 35 43 35
1162
Fig. 12. Geometry for calculation of cross-sectional

characteristics.
2.6
Temporary There were temporary supports at mid-span of all three beams during erection, see Fig.
supports 13. They carried the weight of the slabs and cast-in-situ concrete as well as the weight
of the loading equipment, and were removed during the first stage of the floor test.
Fig. 13. Temporary support

below mid-point of middle
beam.
418
2.7
Loading
arrangements 4 3 2 1
600 300 300 300 300 300 300

300 300 300 300 300 300 600
Pa2 Pa2
1300
(from slab
end)
E
C D
340 Pa1 Pa14800 (span of middle beam)
340 Pa1 Pa1
E C D
Pa2 Pa2
A 8 7 6 5 A
A A
Det 1
P 2 Pa2 2Pa2
a1 Pa1 Pa1 Pa1
Det 1 F F
Pa1 Steel plates
F
Teflon
plates
Gypsum
F
550
419
3 Measurements
3.1
Support
reactions 1
3 load cells
North
5
3.2
Vertical 11
displacement 15 14 13 12
40 40 40
30 30 I 40 30 40 30 30
5100
4 3 2 1
30 30 30 30 30
II
20 19 18 17 16
Transducers for measuring vertical displacement 4500
- Transducers 27, 28, 29 on the continued centrelines of the
longitudinal joints between the slabs
- Transverse measures given with respect to the centrelines of
the joints between the slabs
South
North
30 30 30 30 30
25 20 20 24 23 22 21
III 20 20 20
340 680
30 29 IV 28 27 26
340
V
35 34 33 32 31
8 7 6 5
4500
40 39 38 37 36
VI
5100
45 44 VII 43 42 41
420
3.3
55 (61) 54 (60) L
South
North
440 57 (63) 56 (62)
440
L
59 (65) 58 (64)
8 7 6 5
L L
58 56 54 20
60
64 62 20 30 20
beams. Numbers in parentheses refer to the soffit of the floor, others to the top.
Fig. 19. Apparatus for measuring average strain. L =1040 mm and 1150 mm for
the top and soffit transducers (54–59 and 60–65), respectively.
3.4
displacements Measures differential displacement between slab edge and beam in beam's direction,
numbers in parentheses refer to the soffit of the floor, others to the top
50 4 3 2 1 46
(51) (47)
20
7 6 5
130
40 40 40
Middle beam
10 9 130 8
20 H H
52 8 7 6 5 48
(53) (49)
421
H H J J
(Only transducers 48 and 49 shown) (Only transducers 46 - 49 shown)
48 46
L-profile, glued to slab 48 J 20
30 (from slab end)
49 20
J 47
49
Fig. 21. Section H-H, see Fig. 20. Fig. 22. Section J-J, see Fig. 21.
3.5
Strain
L CL
3 50 50 2
66 30 110
69 20 L L
67 68
L
Longitudinal
strain gauges
600
L 30 20 Transverse 20 5
72
70 5050 strain gauges
20
30 110
7 73 71 6
L
Fig. 23. Position of strain gauges 66–73 at mid-section of middle beam.
5 Loading strategy
5.1
relationship
Before removal of the temporary supports and when the actuator forces Pai were equal
to zero but the weight of the loading equipment was on, all measuring devices were
zero-balanced. Thereafter, the temporary supports were removed. Fig 24 shows the
effect of this operation to the support reaction of the middle beam. The loading history is
shown in Fig. 25.
422
Pa1 [kN]
20 120,0
18
100,0
16
14 80,0
12
R [kN]
10 60,0
8
6 40,0
4
20,0
2
0 0,0
0 5 10 15 20 25 30 0 100 200 300 400 500
Time [min] R [kN]
a) b)
Fig. 24. Development of support reaction R below Northern end of middle beam a)
When removing the temporary supports. b) Later as function of Pa1.
150
Stage I Stage II Stage III
130
110
Pa2
90 Pa1
Pa [kN]
70
50
30
10
-10
0 20 40 60 80 100 120 140 160 180
Time [min]
5.2
After failure
Stage I Some longitudinal cracks along the strands were discovered in the soffit
of the slabs, see Fig. 30. Their width was of the order of 0,08 … 0,10
mm. They were obviously caused by the release of the prestressing
force.
Stage II The interface between the cast-in-situ concrete and slab ends cracked
along slabs 1–4 at Pa1 = 30 kN and along slabs 5–8 at Pa1 = 50 kN. The
tie beams at the ends of the floor specimen failed between slabs 2 and 3
at Pa1 = 30 kN and between slabs 6 and 7 at Pa1 = 50 kN.
Pai = 50 kN corresponds to the expected service load when the shear

resistance of the slabs was assumed to be critical in the design.
At the end of Stage II, some cracks in the soffit had increased in length
and their maximum width was of the order of 0,10–0,12 mm , see Fig. 30.
423
Stage III At Pa1 = 115 kN an inclined crack was observed outside slab 1 at
Northern support of middle beam and soon after that a similar crack
appeared outside slab 5.
At Pa1 = 120 kN, similar cracks appeared outside slabs 4 and 8, see Figs
8–11 in App. B.
The next aimed load step Pa1 = 120 kN, Pa2 = 124,4 kN could not be
achieved because new inclined cracks appeared below the previous
ones in slabs 1 and 5 Consequently, slab 1, followed by slabs 4 and 5
failed in shear. The highest support reaction was obtained at load
combination Pa1 = 112,6 kN, At Pa2 = 123,7 kN.
After When demolishing the test specimen it was observed that all core fillings
failure were not perfect. Due to the cracking mechanism, see Fig. 26, the core
fillings could be measured, see Figs 27–29 and App. B, Figs 22–25.
Fig. 26. Cracking took typically place along the surface of the core filling.
Fig. 27. Super beam after removal of slab units.
424
b cf
d 72.5
cf
105
72.5
Chamfer
30x30 250
a) b)
Fig. 28. a) Core filling seen in the beams’s direction. b) Nominal
dimensions of a hollow core.
4d 4c 4b 4a 3d 3c 3b 3a 2d 2c 2b 2a 1d 1c 1b 1a
d 3 7 8 10 6 3 2 3 5 10 11 5 6 5 7
cf
b 20 18 12 4 12 17 18 17 18 10 12 16 19 15 14
cf
Cracked or not? Thin crack

Wide crack Thin crack
b
cf 13 15 13 8 14 14 12
South North
d
cf 10 8 7 9 8 6 9
8d 8c 8b 8a 7d 7c 7b 7a 6d 6c 6b 6a 5d 5c 5b 5a
Fig. 29. Core fillings seen from above after removal of slabs. The core
fillings are identified by the number of the slab and letters a–d. dcf and bcf,
see Fig. 42, are given in centimetres.
7.1
Cracks at
service load 4 3 2 1
50
50 50
0 0 50
0
South
North
50 0
50 0 0
50
B 50
0
50
50
0 50
50
8 7 6 5
Fig. 30. Cracks in the soffit at service load (Pai = 50 kN). The initial cracks observed
before loading are indicated with red colour. The figures give the value of the actuator
force at which the crack was observed.
425
7.2
Cracks after
failure
4 3 2 1
120
115
115
120
8 7 6 5
4 3 2 1
North
South
8 7 6 5

The ratio (measured support reaction below one end of the middle beam)/ (theoretical
support reaction due to actuator forces on half floor) is shown in Fig. 33. The theoretical
reaction is calculated assuming simply supported slabs. This comparison shows that the
support reaction due to the actuator forces can be calculated accurately enough
assuming simply supported slabs. However, the failure of the slab ends at the North end
of the middle beam resulted in reduction of support reaction below that end while the
actuator force could still slightly be increased. The maximum support reaction is
regarded as the indicator of failure.
426
RP,obs / RP,th
1,2
1,0
0,8
0,6
0,4
0,2
0,0
0 20 40 60 80 100 120
Pa1 [kN]
support reaction (Rp,th) vs. actuator force Pa1. Only actuator loads Pa1 and Pa2 are taken
into account in the support reaction.
Vobs = Vg,sl + Vg,jc + Veq + Va
where Vg,sl, Vg,jc, Veq and Va are shear forces due to the self-weight of slab unit, weight of
joint concrete, weight of loading equipment and actuator forces Pai, respectively. All
components of the shear force are calculated assuming that the slabs behave as simply
supported beams. For Veq and Va this means that Veq = 0,8677xPeq and Va =
0,8677x(Pa1 + Pa2)/2. Vg,jc is calculated from the nominal geometry of the joints and
density of the concrete, other components of the shear force are calculated from
measured loads and weights. The values for the components of the shear force are
given in Table below.
Shear force
Action Load
kN
Actuator loads (112,6+123,7)/2 kN 102,52
427
9.1 ReH/Rp
Rm
Strength of Component 0,2 Note
steel MPa MPa
L-profiles  355 Nominal (S355J2G3)
Folded plate  355 Nominal (S355J2G3)
Tie strands J12,5 and J9,3 1630 1860 Nominal (no yielding in test)
Reinforcement Txy and Kz 500
9.2
Strength of
slab concrete, h
floor test mm mm
d
6 50 50 25.01.2002 Upper flange of slabs 5 and
Mean strength [MPa] 62,1 (+8 d)1) 6, vertically drilled, tested as
St.deviation [MPa] 4,6 drilled2), density =2328 kg/m3
9.3
Strength of Not measured, assumed to be the same as that in the floor test
slab concrete,
reference tests
9.4
Strength of a
longitudinal mm
joints of slab a
St.deviation [MPa] 0,90 density = 2152 kg/m3
9.5
Strength of a
concrete in the # a a Date of test Note
upper part of mm
the beam and a
in the core 6 150 17.1.2002 Kept in laboratory in the same
filling Mean strength [MPa] 33,3 (+0 d)1) conditions as the floor specimen
1)
2)
428
In the following figures, Va stands for the shear force of one slab end due to imposed
actuator loads, calculated assuming simply supported slabs. Note that the last three
points on each curve represent the post failure situation for which the real shear force
has been lower than that shown in the figures. This note is based on the measured
support reaction, see Fig. 25.b.
10.1
Deflections 120 120
100 100
80 80
Va [kN]
Va [kN]
60 11 60 16
12 17
40 13 40 18
14 19
20 20
15 20
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Fig. 34. Deflection on line I along western Fig. 35. Deflection on line II in the middle
120 120
100 100
80 80
Va [kN]
Va [kN]
60 21 60 26
22 27
40 23 40 28
24 29
20 20
25 30
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
120 120
100 100
80 80
Va [kN]
Va [kN]
60 31 60 36
32 37
40 33 40 38
34 39
20 20
35 40
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
line load, slabs 5–8. of slabs 5–8.
429
120 120
100 100
80 80
Va [kN]
Va [kN]
60 41
60
42 13, End beam
40 43 40 28, Super
44 43, End beam
20 20
45
0
0
0 5 10 15 20 25 30
0 5 10 15 20 25
Fig. 40. Deflection on line VII along end Fig. 41. Net deflection of midpoint of
beam, slabs 5–8. middle beam (28) and those of end beams
(13, 43). (Settlement of supports
eliminated.)
10.2
Crack width 120 120
100 100
80 80
Va [kN]
Va [kN]
5 5
60 60
6 6
40 7 40 7
8 8
20 9 20 9
10 10
0 0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0
Crack width [mm] Crack width [mm]
Fig. 42. Differential displacement ( crack Fig. 43. Same as previous figure but the
width) measured by transducers 5–10. cyclic loading phase is not shown.
10.3
(actually
100 100
differential
displacement) 80 80 60
54 61
Vp [kN]
Va [kN]
55 60 60 62 beam
56 beam 63 beam
57 beam 40 40
64
58 65
20 20
59
0 0
-2,0 -1,5 -1,0 -0,5 0,0 0,0 0,5 1,0 1,5 2,0 2,5
Differential displacement [mm] Differential displacement [mm]
surface of floor measured by transducers measured by transducers 60–65.
54–59.
430
10.4
Shear 120 120
displacement 100 100
80 80
Vp [kN]
Vp [kN]
46 Slab 1 60 50 Slab 4 60
47 Slab 1 51 Slab 4
48 Slab 5 40 52 Slab 8 40
49 Slab 5 53 Slab 8
20 20
0 0
-2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5
120
100
80
Va [kN]
60 46-47 Slab 1
48-49 Slab 5
40
50-51 Slab 4
20 52-53 Slab 8
Fig. 48. Shear displacement = differential 0

displacement at upper edge – differential 0,0 0,5 1,0 1,5 2,0 2,5
displacement at lower edge of slab. Shear displacement [mm]
10.5 Strain
120 120
100 100
80 80
Va [kN]
Va [kN]
60 60
66 68
40 67 69 40
72
20 70
73 20
71
0 0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0
Strain [10-3] Strain [10-3]
Fig. 49. Strain measured by gauges 66, Fig. 50. Strain measured by gauges 68,
67, 70 and 71. 69, 72 and 73.
431
11 Reference tests
6 5 4 8
a) P
1 2 7
3
1300
P
1 (6) Gypsum 3 (4)
2 (5)
8 (7)
b)
35 35
35 35 3565
8500 (Test R9/1)
9524 (Test R9/2)
Fig. 51. Layout of reference test. a) Plan. b) Elevation. Test R9/2 was carried out first.
300
250 R9/2
R9/1
200
P [kN]
150
100
50
0
0 5 10 15 20 25 30 35 40
Time [min]
300 300
250 250
200 1 200
1
2
P [kN]
P [kN]
150 150 2
3 3
100 4 100 7 4
5 5
50 6 50
6
0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
R9/2 R9/1
Fig. 53. Displacements measured by transducers 1–6. P is the actuator force.
432
250 350
300
200
250
150 200
P [kN]
P [kN]
7 150 7
100 8
8
100
50
50
0 0
0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35
R9/2 R9/1
Fig. 54. Slippage of outermost strands measured by transducers 7–8. P is the actuator force.

Test Date Note
mm kN kN kN kN kN/m
R9/1 24.1.2002 8500 19,0 288,3 0,7 264,4 220,3 Flexural shear failure
Mean 242,8 202,4

13 Discussion
1. The net deflection of the middle beam due to the removed temporary supports and
imposed actuator loads (deflection minus settlement of supports) was 17,5 mm or
L/274, i.e. rather small. It was 3,1–3,3 mm greater than that of the end beams.
3. The shear resistance measured in the reference tests was 10% lower than the mean of
the observed values for similar slabs given in Pajari, M. Resistance of prestressed hollow
core slab against web shear failure. VTT Research Notes 2292, Espoo 2005.
4. The maximum difference in the net mid-point deflection of the beams was less than
3,3 mm. Hence, the torsional stresses due to the different deflection of the middle
beam and end beams had a negligible effect on the failure of the slabs.
5. The bond between the cast-in-situ concrete and the edges of the hollow cores was weak.
6. The edge slabs slided 0,13 … 0,23 mm along the beam before failure. This reduced
the negative effects of the transverse actions in the slab and had a positive effect
on the shear resistance.
7. The transverse shear deformation of the edge slabs was considerable which can be
seen in Figs 47–49.
8. The failure mode was web shear failure of edge slabs. The Super beam seemed to
recover completely after the failure even though it obviously had cracked in flexure.
433
APPENDIX A: DETAILS OF SUPER BEAM
A434
A435
A436
Fig. 1. Tie reinforcement at the edge of the floor. Fig. 2. Overview on arrangements.
Fig. 3. Loading arrangements. Fig. 4. Transducers measuring average strain in

beam’s direction.
B437
Fig. 5. Arrangements at end beam. Fig. 6. Transducers measuring sliding of slabs

along middle beam.
Fig. 7. Arrangements for line loads. Fig. 8. Failure of slab 5.
Fig. 9. Failure of slab 1. Fig. 10. Diagonal crack in slab 8 after failure.
B438
Fig. 11. Failure of slab 4. Fig. 12. Vertical cracking of end beam between
slabs 6 and 7.
Fig. 13. Wide crack along the western edge of middle Fig. 14. Failure cracks in slab 2.
beam next to slab 2.
Fig. 15. Cracks in slab 4 after failure. Fig. 16. Cracks in slabs 2–4 after failure.
B439
Fig. 17. Cracks in slab 5 after failure. Fig. 18. Cracks in slab 1 after failure.
Fig. 19. Soffit of slabs 3 and 4 after failure. Fig. 20. Soffit of slabs 1–2 after failure.
B440
Fig. 22. Western side of middle beam. Note the

bond failure along the vertical interface of the cast-
in-situ concrete and the precast beam as well as the
Fig. 21. Failed ends of slabs 3–4 after failure. intact core fillings.
Fig. 23. Eastern side of middle beam after demolition. Fig. 24. Western side of middle beam after
demolition.
B441
Fig. 25. Intact core fillings after failure. Note the Fig. 26. Web shear failure in reference test R9/2.
geometric imperfections at the end.
Fig. 27. Web shear failure in reference test R9/2. Fig. 28. Flexural shear failure in reference test R9/1.
Fig. 29. Flexural shear failure in reference test R9/1.
B442
TUT.CP.LB.320.2002
1.1
TUT.CP.LB.320.2002 Last update 2.11.2010
Identification
and aim LB320 (Internal identification)
Aim of the test To quantify the interaction between LB beam and

hollow core slabs.
1.2
Test type
End beam Middle beam (LB) End beam
Fig. 1. Overview on test arrangements. LB beam in the middle, steel beams (square
tubes) at the ends.
1.3 TUT/FI (Tampere University of Technology) 19.6.2002

Laboratory
& date of test
1.4
Test report Author(s) Suonio, M., Taskinen A.
Name LUJABEAM-palkin (LB) laatastokoe (Test on hollow core floor with
LUJABEAM (LB), in Finnish)
Ref. number -
Date 28.8.2002
Availability Confidential, owner is
Lujabetoni Oy
Harjamäentie 1
FI-71800 Siilinjärvi
Finland

(see also Appendices A)
443
TUT.CP.LB.320.2002
2.1
General plan
Fig. 2. Plan. Span of beams 4,8 m.

2.2
End beams
Fig. 3. Arrangements at end beam. 8 refers to a reinforcing bar T8, see 2.3.
2.3
Middle beam The beam, see Figs 4–7, comprised a precast and prestressed concrete component
provided with shear keys and two L-shaped steel ledges.
444
TUT.CP.LB.320.2002
Precast prestressed
concrete
Shear keys
50x220x10 c/c 100
L-profile, steel
Fig. 4. Illustration of LB beam.
Fig. 5. Design of LB beam. Txy refers to reinforcing bar A500HW with diameter xy, see 2.3.
445
TUT.CP.LB.320.2002
Fig. 6. Sections A-A, B-B and C-C. Txy refers to reinforcing bar A500HW, see 2.3.
Fig. 7. Detail 1. Design of ledges.
446
TUT.CP.LB.320.2002
The prestressed part was cast by Lujabetoni Oy in their Siilinjärvi factory 25.4.2002
Concrete: K80, max aggregate size 16 mm, rapidly hardening cement

Passive reinforcement and tendons in LBL beam:
J12,9: Prestressing strand, 7 wires,  =12,9 mm, Ap = 100 mm2, prestress = 1250 MPa,
low relaxation (nominal value 2,5% at  = 0,7fpu)
Structural steel: S355J2G3, fy  355 MPa (nominal value) unless otherwise specified
2.4
at middle
beam
Fig. 8. Section along hollow cores, see Fig. 2.
Fig. 9. Section along joint between Fig. 10. Section along outer edge of floor
adjacent slab units, see Fig. 2. The rebars specimen, see Fig. 2. The rebars are
are made of A500 HW. made of A500 HW.
447
TUT.CP.LB.320.2002
2.5
Slabs

- Extruded by Lujabetoni Oy, Siilinjärvi factory, date not given in the report

1,9 mm in slab unit 4,
59,4 1,8 mm in slab units 1 and
6, 1,7 in slab units 4, 5, 8
bw,i and 2 x 1,7 in slab unit 6
226,3
units = 5,33 kN/m
37,8 bw,i = 311,7 39,3

characteristics.
2.6
Temporary There were no temporary supports below beams.
supports
448
TUT.CP.LB.320.2002
2.7
Loading
arrangements
Fig. 13. Section A-A in Fig. 2. Palkin laipan katkaisu reunavalun tasalle = the ledge of
the beam was cut along the edge concrete.
3 Measurements
3.1
Support
reactions
Fig. 14. Load cells 60, 63, 68 and 69 measuring the force in the tension bars,
see also Fig.13. Load cells 61, 62, 67 and 64–66 below the supports of the LB beam.
449
TUT.CP.LB.320.2002
3.2
Vertical
displacement
Fig. 15. Location of transducers 1–18 for measuring vertical deflection of floor.
3.3
Average strain
in mid-span
and crack
width along
edges of
middle beam
Fig. 16. Position of device (transducers 26, 27, 28 and 30) measuring average
strain parallel to the beams. Numbers in parentheses refer to the soffit of the
floor, others to the top. Position of transducers 19–24 for measuring crack width
between LB beam and ends of hollow core slabs.
450
TUT.CP.LB.320.2002
3.4
Horizontal. See 3.3
displacements
3.5
Strain
Fig. 17. Strain gauges 40, 41, 46–49 measuring longitudinal strain of beam and strain
gauges 42–45 measuring transverse strain of ledges.
On top of the end beams shims were placed to simulate the camber of the middle beam
in such a way that when the hollow core slabs were installed, all corners of the end
beams touched the supporting beams.
All cast-in-situ concrete (K30, max aggregate size 8 mm) was taken from the same
batch. The date of casting was 4.6.2002.
The ledges of the LB.beam were flame-cut along the outer edges of the outermost
hollow core slabs and holes for the tension bars were drilled through the hollow cores.
451
TUT.CP.LB.320.2002
5 Loading strategy
5.1
relationship
Before starting, all measuring devices were zero-balanced. The loading history is shown
in Fig. 18.
Fig. 18. Actuator force P vs. time.

In the figures below, P refers to the force in the hydraulic actuators.
Stage I At P = 75 kN the joint concrete between the middle beam and the slab
(Cyclic) units started to crack along the slab ends, and at P = 80 kN there was
a continuous crack in the joint concrete on both sides of the beam
along the whole length of the beam.
At P = 100 kN the cast-in-situ concrete cracked along the outer edge of

slabs 1 and 5 and along the edge of slab 8.
At P = 120 kN first vertical cracks were observed in the tie beams

above the end beams of the test specimens.
Stage II At P = 140 kN the cast-in-situ concrete (edge concrete) cracked along

(Monotonous) the outer edge of slab 8.
At P = 240 kN the joint concrete cracked along the edges of the middle
beam in the mid-span of the beam.
At P = 260 kN a longitudinal crack was observed in slab 4, see Fig. 20.
At P = 283,4 a web shear failure took place in slab 8.
The cracking pattern after failure is shown in Figs 20–21 (top surface
and edges) and in Fig. 19 (soffit after failure).
After failure See App. A, Figs 3–7. LB beam seemed intact after demolishing the slabs.
452
TUT.CP.LB.320.2002
7.1
Cracks at See Fig. 20
service load
7.2 Cracks
after failure
453
TUT.CP.LB.320.2002
Fig. 20. Cracks on the top and at the edges of the floor at load P = 260 kN.
The numbers give the value of actuator load P when the crack appeared.
454
TUT.CP.LB.320.2002
Fig. 21. Failure mode on the top and at the South edge (slabs 7 and 8).
455
TUT.CP.LB.320.2002

The ratio (measured support reaction below one end of the middle beam)/ (load due to
actuator forces on half floor) is shown in Fig. 22 and the ratio (measured support
reaction below one end of the middle beam) / (theoretical support reaction due to
actuator forces on half floor) in Fig. 23. The theoretical reaction is calculated assuming
simply supported slabs, which means that the support reaction force is 84,2% of 2P.
The measured shear force is 89,4% of 2P before failure. This comparison shows that
the support reaction due to the actuator forces cannot be calculated accurately enough
assuming simply supported slabs.
Fig. 22. Measured support reaction of the middle beam vs. actuator forces 2P. Only
actuator loads are taken into account in the support reaction.
Fig. 23. Measured support reaction of one slab vs. theoretical support reaction due to
actuator forces.
456
TUT.CP.LB.320.2002
VP is calculated from the measured support reactions below the middle beam. Vg,sl and
Vg,jc are calculated assuming that the slabs behave as simply supported beams, but Veq
is obtained from Veq = 0,8938xFeq because the measured support proved to be 89,38%
of the imposed line load on one slab unit. The values for the components of the shear
force are given in Table below.
Shear force
Action Load
kN
Weight of slab unit 18,9
Weight of joint concrete 0,8
Loading equipment 2,8
Actuator loads (283,4)/2 kN 126,7
9.1
MPa MPa
L-profiles  355 Nominal (S355J2G3C)
Reinforcement Nominal value for reinforcing bars
500
A500HW (Txy) (no yielding in test)
9.2
Strength of
h
floor test
d mm mm
6 50 50 ? 2) Upper flange of slab 8,
Mean strength [MPa] 65,7 vertically drilled
9.3
Strength of
h
reference tests
d mm mm
3 50 50 ? 2) Upper flange of slab,
Mean strength [MPa] 72,9 vertically drilled
St.deviation [MPa] - Tested as drilled2)
457
TUT.CP.LB.320.2002
9.4
Strength of a
a
mm
9.5
Strength of a
concrete in # a a Date of test Note
middle beam mm
a
1)
2)
Date is not given in the report, but it is very likely 19.–22.6.2002
In the following figures, P stands for the actuator load.
10.1
Deflections
Fig. 24. Deflection in mid-span of slabs Fig. 25. Deflection in mid-span of slabs
1–4 measured by transducers 2–6. 5–8 measured by transducers 7–11.
Fig. 26. Net deflection of midpoint of LB

beam and those of end beams (1, 12).
Settlement of supports is eliminated from
deflection of LB beam.)
458
TUT.CP.LB.320.2002
10.2
Crack width
10.3
Average strain
(actually
differential
displacement)
Fig. 28. Differential displacement at top and bottom surface of floor measured by
transducers 26, 27, 28 and 30.
10.4
Differential
displacement
Fig. 29. Differential vertical displacement between top surface of LB beam and the end
of hollow core slabs measured by transducers 13 and 14. A positive value means that
the slab end deflects more than the beam.
459
TUT.CP.LB.320.2002
10.5 Strain
Fig. 30. Longitudinal strain in ledges measured by gauges 40 and 41.
Fig. 31. Transverse strain in ledges measured by gauges 42, 43 and 45. The strain
measured by gauge 44 is not shown because it was clearly erroneous.
Fig. 32. Longitudinal concrete strain in LB beam measured by gauges 47, 48 and 49.
The strain measured by gauge 46 is not shown because it was clearly erroneous.
460
TUT.CP.LB.320.2002
11 Reference tests
Fig. 33. Layout of reference test and load – mid-point deflection curve.
See also App. A, Fig. 8.
461
TUT.CP.LB.320.2002
Table 2. Reference test. Span of slab, shear force Vg at support due to the self weight of
Span Vg Pa+Peq Vobs vobs

Test Date Note
mm kN kN kN kN/m
Web shear failure
R1 17.6.2002 7120 20,9 343,0 313,3 261,1
(flexural shear failure)
App. A, Fig. 8 shows that the lower end of the inclined failure crack is at a distance of
500 mm from the support. At this distance, assuming the losses of prestress equal to
10% and the transfer length of the prestressing force equal to 600–800 mm, the axial
stress in the soffit = -2,9 … -0,1 MPa is obtained. This suggests that the failure mode
could not be initiated by a flexural crack. Hence, the failure mode has been web shear
failure rather than flexural shear failure.

was equal to 149,2 kN per one slab unit or 124,3 kN/m. This is 48% of the shear
resistance observed in the reference test. Note that in this case the sheared end of the
reference slab was provided with cast-in-situ concrete simulating the grouting outside
the beam end in floor test, see App. A, Fig. 8.
13 Discussion
2. The shear resistance measured in the reference test was a bit higher than the
mean of the observed values for similar slabs given in Pajari, M. Resistance of
2292, Espoo 2005. This may be attributable to the cast-in-situ concrete at the
sheared end in the reference test.
3. The maximum difference in the net mid-point deflection of the beams was of
the order of 5–6 mm. An estimated value is given because the settlement of
the supports of the end beams was not measured. Hence, the torsional
stresses due to the different deflection of the middle beam and end beams had
a negligible effect on the failure of the slabs.
the middle beam (LB beam). The LB beam seemed to recover completely after
the failure.
462
TUT.CP.LB.320.2002
Fig. 2. Equipment for measuring average strain in hollow core slab.
A463
TUT.CP.LB.320.2002
Fig. 3. A step between LB beam and slab 4.
Fig. 4. Failure in slab 8.
A464
TUT.CP.LB.320.2002
Fig. 5. Soffit after failure. Slab 6 in the front, slabs 7 and 8 in the rear.
Fig. 6. Slab 8 after failure. The loose top part has been removed.
A465
TUT.CP.LB.320.2002
Fig. 7. Slab 8 after failure. All loose concrete material has been removed.
Fig. 8. Failure mode in reference test.
A466
VTT.S.WQ.500.2005
1.1
Identification
VTT.S.WQ.500.2005 Last update 2.11.2010
and aim
WQ500 (Internal identification)
Aim of the test To study the shear resistance of 500 mm slab supported
on steel beams.
1.2
Test type
End beam (HE 340A) Middle beam (WQ beam) End beam (HE 340A)
Fig. 1. Illustration of test setup. The end beams were hot-rolled steel beams.
1.3 VTT/FI 16.8.2005

Laboratory
& date of test
1.4
Name Load test on hollow core slab floor with steel beams
Date 14.12.2005
Availability Available at www.rakennusteollisuus.fi
467
VTT.S.WQ.500.2005
2.1
General plan West 7200 (span of beam)

6 5 4 3 2 1

Line load
South
250 250
North
7200 (span of beam)
1500
Line load
12 11 10 9 8 7
East 7200 (span of beam)
Fig. 2. Plan.
A468
VTT.S.WQ.500.2005
2.2
End beams B B Simply supported, span = 7,2 m,
roller bearing below one end
50 50
Concrete K30
Rebars 2 d 10
L = 7200 500
Plywood
Steel
Roller bar 50 50
330
HE340A 9,5
16,5
300
Fig. 3. Arrangements at end beam.
2.3
Middle beam - Manufactured by PPTH Oy and transported to VTT Research Hall 1 23.5.2005
- Simply supported, span = 7,2 m, roller bearing at one end
Steel S355
Camber 4,0 mm
A
190 Two transverse plate stiffeners 440x480x20 inside the beam, welded to the inner surface of the box 190
20 Seven holes, d = 48 mm 20
230 1170 1200 1200 1200 1200 1170 230

A
7600
A A
500
10 40
Steel pipes through the beam,
inner diameter = 48 mm,
c/c 1200
62
80 30
150 100
150 500
800
Fig. 5. Cross-sectional geometry.
469
VTT.S.WQ.500.2005
2.4 C C
Arrangements
at middle
beam Rebars 4 d 12
L = 7600 500 50 400
Rebars 5 d 20, L = 3500, 120

one in each longitudinal 40
joint between
500
adjacent slabs
80
100 30
150 500 30 150
Circular hole through beam
c/c 1200
Fig. 6. Section C-C (see Fig. 13) along a joint between adjacent hollow core units.
D D
EPS plug 50 400
Fig. 7. Section D-D (see Fig. 13) along hollow cores. Plugs made of expanded polystyrene.
E E
H H
500
80
30
50 100
Bent rebar d 20
L = 800
H H
Hole through beam
Slab Slab
50
Cast-in-situ
concrete 150
50 100
Bent rebar d 20
L = 800
Fig. 8. Tie reinforcement in the cast-in-situ concrete outside the edge of the outermost
slabs (section E-E in Fig. 13). See also App. A, Fig. 9.
Tie reinforcement: Horizontal rebars A500HW through the beam and parallel to the beam.
A470
VTT.S.WQ.500.2005
2.5
Slabs
45 - Extruded by Parma Oy,
Hyrylä factory 28.4.2005
R50 R125
- Strands: 16 J 12,5,
410
205
500 Ap = 93 mm2/strand, initial
prestress = 1050 MPa
R50 R120
20
35 45
1200
1153 (1161 at mid-depth) Max measured bond

slippage: 1,9 mm
48,0 (in slab 11);
2x1,7 mm
(in slabs 10 and 11);
b w,i 400,4 2x1,6 mm
(in slabs 3 and 8)
- Measured weight = 6,49

40,1
kN/m
41,8 67,1  b w,i = 339,7
1195
Fig. 10. Most relevant measured geometrical properties.
2.6
Temporary No
supports
2.7 An auxiliary loading frame was built above the test floor. It was tied by tension bars to
Loading the floor of the research hall, see Figs 11 and 12. Six tension bars penetrated the floor
arrangements through the hollow cores outside the line loads, the rest were outside the test specimen.
The loads were generated by 12 actuators. See Fig. 13 for the position of the actuators.
Taking into account the weight of the loading equipment Peq = 6,14 kN/slab unit, the
following relationship can be written for the line load on one slab:
F = Pa + 6,14 kN (1)
Pa is the load in the actuator. The line loads were applied to the slab units by 24 tertiary
steel beams of the type HE 120 A (a hot-rolled I-beam with depth = 114 mm, width =
120 mm and thickness of flange = 8 mm), each 550 mm in length. The top surface of the
slabs under these beams was evened out by gypsum. On the top of the tertiary beams,
secondary spreader beams were placed, each on two bearings. On the top of the
secondary beams, primary spreader beams were installed. The friction between the
primary and secondary spreader beams was eliminated with teflon sheets, and that
471
VTT.S.WQ.500.2005
between the secondary and tertiary spreader beams with a freely rolling circular steel
bar, see Figs 14–16 The drifting of the primary spreader beams was prevented by the
friction at the upper and lower end of the actuators. The upper end of each actuator was
provided with a swivel (ball bearing).
+4,5 m
2Pa 2Pa 2Pa 2Pa 2Pa 2Pa
420 780 220 980 20 1020 180
Fig. 11. Auxiliary loading frame. Elevation. North edge on the right.
1000
1000
Fig. 12. Auxiliary loading frame. Plan. The position of the tension bars is indicated by
white circles above the blue and green beams. North edge on the right.
A472
VTT.S.WQ.500.2005
B 7200 (span of beam)

16377 16377 16378 16377 16378 16377
6 5 4 3 2 1
600 600 600 600 600 600 600 600 600 600 600 600

E
D C
250 250
South
North
7200 P1
(span of
beam)
D C C
E
Primary
1500
P
2
A 8 A
Tertiary Secondary
12 11 10 9 8 7
16377 16377 16378 16378 16377 16377
7200 (span of beam)
Fig. 13. Loading arrangements. The slabs are from two casting lots: 16377 and 16378.
The position of the actuator loads is indicated by black circles above the middle beam.
A A
Det 1
2P 2P 2P 2P 2P 2P
a a a a a a
Primary
Secondary
Tertiary

7200
Fig. 14. Section A-A, see Fig. 13.
473
VTT.S.WQ.500.2005
Det 1 F F
Steel plate Pa Pa
F
Stiffening Stiffening
plates plates
L-profile Teflon
plates
Gypsum Roller bar
Fig. 15. Detail 1, see Fig. 14.
3 Measurements
3.1
Support Below one end of the middle beam, the support reaction due to the actuator loads was
reactions measured by three load cells as shown in Figs 14 and 16.
1
3 load cells
A474
VTT.S.WQ.500.2005
3.2
Vertical
50
30 30 30 30 30 30 30
displacement
17 15 14 I 13 11
24 23 22 21 II 20 19 18
2600
600 600 600 600 600 600 600 600 600 600 600 600
31 30 29 28 27 26 25
III
(from slab end)

1700
6 5 4 3 2 1
E
G
250 250
38 37 36 35 IV 34 33 32
1700
12 11 10 9 8 7
V
45 44 43 42 41 40 39
2600
8
52 51 50 49 VI 48 47 46
50
59 57 56 VII 55 53
475
VTT.S.WQ.500.2005
3.3–3.4
Average strain
and horizontal 6 5 4 3 2 1
displacements J
G
83 (84) 69 (70) 67 (68) 81 (82)
I 63 I 61
150
62 30
250 250
60 250 60
73 (74) 71 (72)
250
65
150
66 64
87 (88) 77 (78) 75 (76) 85 (86)
G
J
12 11 10 9 8 7
Fig. 18. Position of transducers 61-66, 67–78 and 81–88 measuring crack width,
average strain parallel to the beams and displacement of the slab edges relative
to the beam, respectively.
G G
250 250 50 150

20 69 73 77
Inductive transducer
20
20 70 74 78
50 150
250 250

strain. L =1100 mm. Fig. 20. Section G-G, see Fig. 18.
I I J J
83 83 87
20
20
84 88
20 20
L-profile, glued to slab 84
Fig. 21. Section I-I, see Fig. 18. Fig. 22. Section J-J, see Fig. 18.
A476
VTT.S.WQ.500.2005
- None
5 Loading strategy
5.1
Load-time The loads were applied in two stages: Stage I (cyclic) and Stage II (monotonous) as
relationship shown in Fig. 23.
Cyclic Monotonous
300
250
200
Pa [kN]
150
100
50
0
0 20 40 60 80 100 120 140 160 180
Time [min]
Fig. 23. Actuator force Pa vs. time.
Date of test: 16.8.2005
5.2
After failure -
6 Observations during loading (see also the photographs in App. A)
Before loading All measuring devices were zero-balanced when the actuator
forces Pa were equal to zero but the weight of the loading
equipment was on.
The cracks in the slabs were visually inspected and found to be

the same as those observed before the slabs were installed.
They are shown in Fig. 32 in which the initial cracks in the cast-
in-situ concrete are also indicated. The maximum crack width in
the soffit of the slabs was of the order of 0,06–0,08 mm. The
initial longitudinal crack on the top of slab 12 was above the
midmost web. It was not deep. As can be seen later, the initial
cracks on the top of slabs 3, 4, 9 and 12 did not affect the failure.
Cycling loading Pa = 160 kN corresponds to the shear force due to the expected
service load when the shear resistance of the slabs is supposed
to be prevailing in the design.
477
VTT.S.WQ.500.2005
During the cyclic stage, the vertical interface between the cast-in-
situ concrete of the middle beam and the sawn slab ends
gradually cracked. At Pa = 109 kN, during the first load increase,
the soffit of slab 10 cracked over the whole slab length, see Fig.
32 and App. A, Fig. 24. This crack may have been initiated near the
50 mm hole drilled for the vertical tension bars through the floor.
When the load was last time at the expected service level
(Pa = 160 kN), the soffit of the slabs between the line loads and
the middle beam was inspected visually. The visible cracks are
shown in Fig. 32. There were some new cracks along the strands
and some initial cracks had grown in length. The maximum crack
width in these cracks was of the order of 0,08 … 0,10 mm. No
difference between the widths in the initial and new cracks could
be observed. The width of the long longitudinal crack observed at
Pa = 109 kN in the soffit of slab 10 was of the order of 1,0 mm.
Pa = 213 kN An inclined crack was observed in slab 7 next to the middle

beam.
Pa = 257 -259 kN A similar crack was observed in slabs 1 and 12.
Pa = 272 kN The crack in slab 12 grew in width and resulted in an abrupt

shear failure. See Fig. 33 for all cracks and App. A, Figs 25–27
for the shear cracks in slabs 1, 7 and 12.
Observations after failure
When slab 12 failed, the loads on it were transferred to slab 11 over the longitudinal
joint. The strength of this joint and the elastic energy stored in the loading frame
made the floor fail in a complicated manner illustrated in Fig. 33 and in App. A, Figs
26–33. Despite the complexity of the crack pattern after the test, the origin of the
failure was the shear crack in slab 12 next to the middle beam.
About core filling
After the test, the concrete filling of the slab ends next to the middle beam was
investigated. In all hollow cores there was an empty space above the infill in the
upper outer corner as shown in App. A, Figs 48–50. This was observed first after the
end of slab 12 and the cast-in-situ concrete around it was broken during demolition.
Fig. 24 shows the average geometry of the core infill for slabs 1 and 12. To illustrate
the scatter, the geometry of the core infill for the individual cores is shown with
dashed lines for slab 1.
A478
VTT.S.WQ.500.2005
b cf =118 b cf =144
Slab 1 Slab 12
50 400
Fig. 24. a) Void filling seen in the beams’s direction. On the left: average infill in the
cores measured from slab 1 and the measured infill in the individual cores shown
with dashed lines. On the right: average measured infill in the cores of slab 12.
The dimensions of the core infill for slab 1 were measured using a method shown in
Fig. 25. The length of incomplete core filling was measured with a tape measure and
subtracted from 400 mm to get bcf, the length of the “complete” infill. Due to the
vertical dimension of the measuring tape, the zone of complete infill includes both the
zone of 100 % filling and the zone where the open space above the infill was less
than 10 mm in depth. The same 10 mm tolerance was also applied when measuring
bcf for slab 12. As illustrated in Fig. 26, the open space above the infill, which is
10 mm or a bit more in depth, does not weaken the performance of the core infill
against transverse shear.
Tape measure b cf
10 400
Fig. 25. Measuring the length of Fig. 26. Incomplete core infill with
complete infill using a tape measure. 10 mm gap above it.
It is obvious that the incomplete infill shown in Fig. 26 and a 100% infill are equally
effective in eliminating the transverse deformation of the webs. This justifies the
measuring method.
Fig. 50 in App. A shows how the pressure of the grout moved the EPS plug, glued
with polyurethane to the concrete, forward along the core. There was a risk of
collapse of the plug. For this reason, a more efficient vibration when compacting the
grout could not be a solution for a better filling.
When demolishing the floor it was observed that the weakest average core infill was
in slab 1. The hollow cores in slab 12 were slightly more effectively filled than in
some slabs and slightly less effectively filled than in other slabs. In this sense the
core filling in slab 12 was representative to the whole floor.
479
VTT.S.WQ.500.2005
Observations on support conditions of slabs
The soffit of the slab units was not in complete contact with the middle beam when
the slabs were installed and grouted, see Fig. 27. There were two reasons for this
phenomenon. Firstly, the middle beam was stiffer than the end beams and had a
precamber of 4 mm while the end beams were straight and deflected downwards due
to their self weight. Due to these effects the slab ends were laying on non-parallel
supports. Secondly, the soffit of the slabs was not completely planar but slightly
curved downwards in transverse direction. Fig. 28 illustrates the typical joint between
the outermost slabs and the ledge of the WQ beam. This gap was partly filled with
the cement paste, and where wide enough, was also filled with the grout including
aggregate. This is shown in App. A, Figs 53–59. At the support of the beam, the
source of the grout below the slab was either the edge grouting as shown in Fig. 28
and App. A, Figs 41 and 44–47, or end grouting.
After the test, the maximum gap width max was measured for the outermost slabs
where possible. The results are given in the caption of Fig. 29.
17
25

Fig. 27. Typical gap between slab and

beam flange. max = 1,3 mm, 3,8 mm and
5,5 mm for slabs 1, 6 and 7, respectively. Fig. 28. Concrete cast outside the
For slab 12 the gap could not be measured outermost edges of slabs intruded
after the failure. below the slab to some extent.
Fig. 29 shows the horizontal dimensions of a relatively thick grouting below the
corners of the slab. These were measured using a steel wire which was 1 mm thick.
Slab 1 was not checked because the wire was too thick for the gap. Slab 12 could
not be measured because the concrete broken in the failure had filled the gap.
A direct contact between the soffit of the slab end and the ledge of the beam or grout
on the ledge represents a favourable support condition. In this test also other
mechanisms to transmit the support reaction of the slab to the beam may have been
present.
The slab units were saw-cut but there was 10 mm deep zone at the bottom of the
slab cross-section, which was not sawn but broken, see App. A, Fig. 52. The rough
surface of this zone could work as a dowel, see Fig. 30. Even more important may
have been the fact that the reinforced cast-in-situ concrete formed a beam which with
the aid of the concrete in the cores may have transmitted the loads from the slabs to
the bottom flange of the beam. This load-carrying mechanism may have been
effective enough to transmit the loads even without any contact between the flange of
the beam and the precast slab unit as shown in Fig. 31.
A480
VTT.S.WQ.500.2005
6 1
12 7
Fig. 29. The space between slab and the flange of the beam was partly filled with the
jointing concrete. The filled areas were measured for slabs 6 and 7. They are drawn
in scale and shown in red color.
A
B
C
Grout below slab end Rough slab end Gap below slab end
Fig. 30. Details affecting the transmission of Fig. 31. Possible load-carrying
support reaction between slab and beam. On mechanisms through cast-in-situ
the left: Grout below slab end. On the right: concrete when precast slab unit
Rough slab end. and ledge of beam are not in
contact.
481
VTT.S.WQ.500.2005
7.1 160
Cracks at A A A A A
service load A
A A 160 A
A 160
119 140
A A
A A
A A
257
South
North
6 5 4 160 3 160 2 1
160 160
A A A
160 A160160
160 160 A A
A A A A A A A
A A A A A A A
A A A A A A A
160
160
160 109 160

160
A A
12 11 10 9 8 7
A
109
A
129 109 129 A A A A A
149 109 109 A A
A AA A
A A
A
Fig. 32. Cracks at service load (Pa = 140 kN). The initial cracks observed before loading
are indicated with red colour and letter A. A dashed line indicates a crack in the soffit.
The numbers give the value of the actuator force at which the crack (other than initial
crack) was observed.
A482
VTT.S.WQ.500.2005
7.2 177
Cracks after 160 191
A
A
failure A A A A
A A 160 A
A 160
191 119 140
A A
A A
A A
257
6 5 4 3 2 1
257
160 160 160
A 160 A 160
A A 160 160
160 160 A A
A A A A A A A
A A A A A A A
272 A A A A A A A
213
272 160 213
259
160
272
North
272 160
160 109
South
160 272 A A
272 272
12 11 10 9 8 7
A
272
272
272
109
272
272 272
A 272 243 253 257

272 122 129149 109 109 129 A A A A A
109 A A
A AA A
257 A A
193 257 A
245 253
Fig. 33. Cracks after failure. The initial cracks are indicated with red colour and letter A.
A dashed line indicates a crack in the soffit. The numbers give the value of the actuator
483
VTT.S.WQ.500.2005
The ratio (measured support reaction below one end of the middle beam) /
(actuator forces on half floor) is shown in Fig. 34 and in a larger scale in Fig. 35.
1,2
0,8
Rp,obs/(6*Pa)
0,6
0,4
0,2
0
0 50 100 150 200 250 300
Pa [kN]
Fig. 34. Ratio of measured support reaction of the middle beam (Rp,obs) to actuator loads
on half floor.
0,848
0,846
0,844
0,842
0,840
Rp,obs/(6*Pa)
0,838
0,836
0,834
0,832
0,830
0,828
0,826
0,824
0 50 100 150 200 250 300
Pa [kN]
Fig. 35. A detail of the previous figure in a larger scale.
joint concrete, weight of loading equipment and actuator forces Pa, respectively. Vg,sl and
Vg,jc, are calculated assuming that the slabs behave as simply supported beams. Vg,jc is
calculated from the nominal geometry of the joints and density of the concrete, other
components of the shear force from measured loads and weights. For Veq and Vp
the relationship given in Fig. 35 is applied. This means that Veq = 0,847xPeq and
Vp = 0,847xPa. The values for the components of the shear force are given in the
Table on the next page.
A484
VTT.S.WQ.500.2005
Action Load Shear force

kN
Loading equipment 6,14 kN 5,20
Actuator loads 272,00 kN 230,38
one slab. The shear force per unit width is 224,7 kN/m.
9.1
Strength of Thickness Strength Re1)
steel Middle beam
mm MPa
Top flange 40 388
Bottom flange 30 388
Web 10 407
1)
Measured yield strength according to certificate of compliance
ReH/Rp0,2 Rm
Component Note
MPa MPa
End beam 350 Nominal (no yielding)
Strands J12,5 1630 1860 Nominal (no yielding)
Reinforcement 500 Nominal (A500HW,no yielding)
9.2
Strength of
h
floor test
d mm mm
6 50 50 30.8.2005 Upper flange of slabs 9
1)
Mean strength [MPa] 82,0 (+14 d) vertically drilled, tested as
St.deviation [MPa] 1,1 drilled2), density = 2363 kg/m3
9.2
Strength of
h
floor test
d mm mm
6 50 50 30.8.2005 Upper flange of slabs 12
1)
Mean strength [MPa] 84,4 (+14 d) vertically drilled, tested as
485
VTT.S.WQ.500.2005
9.3
Strength of Date of test
slab concrete, h
# Cores h d From
reference test mm mm
d
Mean strength [MPa] 86,8 (+1 d)1) vertically drilled, tested as
9.4
Strength of a
grout # a a Date of test Note
a
mm
St.deviation [MPa] 1,3 Density =2195 kg/m3
1)
Date of material test minus date of test (floor test or reference test)
2)
kept in a closed plastic bag after drilling until compression
In the following figures, Pa is the actuator force.
10.1
300 300
Deflections
250 250
200 200 18
Pa [kN]
Pa [kN]
11 19
150 13 150 20
14 21
100 100
15 22
50 17 50 23
24
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
300 300
250 250
200 200 32
25
Pa [kN]
33
Pa [kN]
26
150 150 34
27
35
100 28 100
36
29
50 37
50 30
38
31
0
0
0 5 10 15 20 25 30
0 5 10 15 20 25 30
A486
VTT.S.WQ.500.2005
300 300
250 250
46
200 39 200 47
Pa [kN]
Pa [kN]
40 48
150 41 150
49
100 42 50
100
43 51
50 44 50 52
45
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
300 300
250 250
200 200
Pa [kN]
Pa [kN]
53
150 150 14
55
35
56
100 100 56
57
50 59 50
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25
Deflection [mm]
Deflection [mm]
Fig. 42. Deflection on line VII along Fig. 43. Net deflection of midpoint of
Eastern end beam, slabs 7–12. middle beam (35) and those of end beams
(14, 56) (Rigid body motion due to
settlement of supports eliminated).
10.2
Crack width
300
250
200 61
Pa [kN]
62
150
63
100 64
65
50 66
0
0,0 0,5 1,0 1,5 2,0 2,5
Fig. 44. Differential displacement parallel to the slab between the middle beam and
the slabs (≈ crack width). See Fig. 18 for the location of the transducers.
487
VTT.S.WQ.500.2005
10.3
250 250
200 200
68
Pa [kN]
Pa [kN] 67
69 150 150 70
71 72
73 100 100 74
75 76
50 50
77 78
0 0
-1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,0 0,5 1,0 1,5 2,0
surface of floor measured by transducers of floor measured by transducers 58, 60,
57, 59, 61, 63, 65 and 67. 62, 64, 66 and 68.
10.4 300 300

Shear
200 200
Pa [kN]
Pa [kN]
150 150 83, slab 6, top

81, slab 1 top
84, slab 6 bot
82, slab 1 bot 100
100 87, slab 12 top
85, slab 7 top
88, slab 12 bot
50 86, slab 7 bot 50
0 0
0,0 0,5 1,0 1,5 2,0 2,5 0,0 0,5 1,0 1,5 2,0 2,5
Differential displacement between edge of Differential displacement between edge of
slab and middle beam. A positive value slab and middle beam. A positive value
means that the slab is moving towards the means that the slab is moving towards
end of the beam. the end of the beam.
300
250
200
Pa [kN]
150
81-82, slab 1
100 83-84, slab 6
85-86, slab 7
50
87-88, slab 12
0
0,0 0,2 0,4 0,6 0,8 1,0
Shear displacement [mm]
Fig. 49. Differential displacement between

top and bottom edge of slabs 1, 6, 7 and 12.
A488
VTT.S.WQ.500.2005
11 Reference tests
One end of slab 13 was loaded in shear as shown in Fig. 50. Test slab 13 was taken
from the same casting lot as slab 12 which failed in the floor test. In the reference
test the span of the slab was 9900 mm. At actuator load Pa = 250 kN (Vp = 427 kN) a
flexural crack was observed below the line load. A web shear failure took place at
Pa = 362,5 kN. No remarkable slippage of the strands was observed before the failure.
The loading strategy is shown in Fig. 51. The failure mode is shown in Fig. 52and in
App. A, Figs 62 and 63. The aim was to load both ends of slab 13. This was not possible
because the other slab end deteriorated in the first test. The measured displacements
are given in Fig. 53.
The shear resistance 650,7 kN observed in the test is higher than the average value for
normal production.
2Pa
R 1500 R
50
3 (4)
1 (2) Gypsum 5 (6)
Neoprene
Steel plate
50 50
L
R R
2 4 6 50
300
Pa
650
Pa
300
1 50
3 5
Fig. 50. Layout of reference test. For L, see the next table.
489
VTT.S.WQ.500.2005
280
400
350
300
Pa [kN]
250
200 1200
150
100
50 Pa
0
0 5 10 15 20 25 30 35 40
Time [min]
Pa

1090
200
Fig. 52. Failure mode
400
350
300
250 1
Pa [kN]
2
200 3
150 4
5
100
6
50
0
0 5 10 15 20 25 30 35 40
Displacement [mm]
Fig. 53. Displacements measured by transducers 1–6.
Table. Span of slab L, shear force Vg at support due to the self weight of the slab, actuator
force Pa at failure, weight of loading equipment Peq, shear force Vp due to imposed load at
failure, total shear force Vobs at failure and total shear force vobs per unit width.
L Vg Pa Peq Vpa+eq Vobs vobs

Test Date
mm kN kN kN kN kN kN/m
R1 29.8.2005 9900 31,63 362,5 0,29 619,06 650,7 542,2
A490
VTT.S.WQ.500.2005

The observed shear resistance (support reaction) of the hollow core slab in the floor
test was equal to 269,6 kN per one slab unit or 224,7 kN/m. This is 41% of the shear
resistance observed in the reference test.
13 Discussion
1. At failure, the net deflection of the middle beam due to the imposed actuator load
small.
2. The shear resistance measured in the reference test was considerably higher
than the mean of observed values for similar slabs given in Pajari, M. Resistance
2292, Espoo 2005.
3. The shear resistance observed in the floor test was 41% of that in the reference
test.
4. The core fillings in the floor tests were not perfect. This may have affected the
shear resistance.
5. The torsional stresses due to the different deflection of the middle beam and end
beams had a negligible effect on the failure of the slabs because the maximum
difference in the net mid-point deflection was less than 4,7 mm.
6. The bond between the smooth edges of the middle beam and the grout was
weak.
7. The bond between the soffit of the slab and the grout below it was also weak.
Due to the weak bond, the edge slabs slided along the beam before failure.
This reduced the negative effects of the transverse actions in the slab and
had a positive effect on the shear resistance
491
Fig. 1. WQ beam.
Fig. 2. Initial crack in slab 1.
A492
Fig. 3. Initial cracks in slab 2.
A493
A494
A495
A496
Fig. 11. Slabs 1,2,6,7,8 and 12 in their final position.
Fig. 12. Slab 12 on WQ beam and a tie bar 20 mm.
A497
Fig. 13. A short, bent tie bar, outer edge of slab 1.
Fig. 14. Support conditions above end beam.
A498
Fig. 15. Initial bending crack in slab 3 at a distance of 1050 mm from slab end.
Fig. 16. Initial bending crack in slab 9 at a distance of 1350 mm from slab end.
A499
Fig. 17. Overview on test arrangements.
Fig. 18. View on the loading frame and spreader beams.
A500
Fig. 19. Actuators above primary spreader beams.
Fig. 20. Arrangements between spreader beams. Note the white teflon sheets between the primary and
secondary spreader beams.
A501
Fig. 21. Three orange load cells below one support of WQ beam.
Fig. 22. North end of WQ beam before loading.
A502
Fig. 23. North end of WQ beam before loading.
Fig. 24. Longitudinal crack in the soffit of slab 10 at Pa = 160 kN. The cracking took place at Pa =
109 kN.
A503
Fig. 25. Inclined crack in slab 7 at Pa = 213 kN.
Fig. 26. Failure of slab 12 at Pa = 272 kN.
A504
A505
Fig. 30. Failure of slabs 11 and 12 at Pa = 272 kN.
A506
Fig. 31. Failure of slabs 11 and 12 at Pa = 272 kN.
Fig. 32. Slabs 11 and 12 after removing the loading equipment.
A507
Fig. 33. Slabs 11 and 12 after removing the loading equipment.
Fig. 34. Slab 7 after test. Cracks in tie beam. The read line and capital A indicate an initial crack.
A508
Fig. 35. Slab 8 after test. Cracks in tie beam. The read line and capital A indicate an initial crack. The
nonuniform colour is due to a mortar treatment carried out after demolding.
Fig. 36. Slab 9 after test. Cracks in tie beam.
A509
A510
Fig. 40. Slab 1 after test.
A511
A512
A513
A514
Fig. 48. Failure surface at end of slab 1. Note the incomplete filling of the cores.
A515
Fig. 49. Concrete filling taken from one core of slab 12.
Fig. 50. Concrete filling in one core of slab 2. Note that the polystyrene plug is inclined due to the
casting pressure. Consequently, the length of the filling at the bottom is greater than 400 mm.
A516
Fig. 51. Failure surface of slabs 11 and 12.
Fig. 52. Rough surface at slab end below even, saw-cut surface.
A517
Fig. 53. Changes in colour on the flange below slab 1 due to the concrete or cement paste intruding
into the gap between the soffit of the slab and the steel flange. See also the vertical stripes along the
vertical edge of the flange.
Fig. 54. Thin layers of grout as well as changes in colour on the flange below slab 1 due to the grout
intruding into the gap between the soffit of the slab and the steel flange.
A518
A519
Fig. 57. A detail of grout below slab 6.
Fig. 58. Grout below slab 5.
A520
Fig. 59. Steel flange below slab 7. No clear sign of grout can be seen below the slab and above the
flange.
Fig. 60. Overview on arrangements in reference test.
A521
Fig. 61. Loading arrangements in reference test.
Fig. 62. Failure pattern in reference test. South edge.
A522
Fig. 63. Failure pattern in reference test. North edge.
A523
1.1
Identification
and aim
PC500 (Internal identification)
Aim of the test To study the shear resistance of 500 mm slab supported
on concrete beams.
1.2 Test type
Middle beam (Prestressed concrete beam)

End beam (HE 340B) End beam (HE 340B)
Fig. 1. Illustration of test setup. The end beams were hot-rolled steel beams.

& date of test
1.4 Test report
Author(s) Pajari, M.
Name Load test on hollow core slab floor with concrete beam
Ref. number VTT-S-2303-06
Date 9.3.2006
525
2.1 West 7200 (span of beam)

General plan

6 5 4 3 2 1

Line load
240 240
South
North
1500
12 11 10 9 8 7
Line load
Fig. 2. Plan.
2.2
End beams 50 50
Concrete
K30
Rebars
2 T16 500
L = 7100
Plywood
Steel
Roller bar 50 50
HE340B 340
12
21,5
300
Simply supported, span = 7,2 m, roller bearing below Northern end

526
2.3
Middle beam - Cast by Betonimestarit Oy, 4.8.2005
- Concrete K80
- Simply supported, span = 7,2 m, roller bearing at both ends
200 480 200 47 4J12,5 + 2 A2
7 holes through Chamfer

beam, c/c 1200 20x20
500
Chamfer 100 2 A1
20x20 60
13
43
280 13 48
49 48 48 48 48 48 48 49 36J12,5
880
28 96 66 48 74 48 66 96
Fig. 4. Cross-sectional geometry. Fig. 5. Longitudinal reinforcement. J12,5

refers to a prestressing strand; A1 and A2
to ribbed reinforcing bars, see below.
J12,5: Prestressing strand  =12,5 mm, Ap = 93 mm2, p0 = 1300 MPa, nominal
strength fp0,2/fp = 1630/1860 MPa, did not yield in the test
A1: Rebar A500HW,  12 mm, L = 7500 mm, nominal yield strength fy = 500 MPa, did
not yield in the test
A2: Rebar A500HW,  25 mm, L = 7500 mm, nominal yield strength fy = 500 MPa, did
not yield in the test
200 1200 1200 1200 1200 1200 1200 200
7600
527
528
2.4
Arrangements Rebars 4 d 10 L = 7600
at middle
beam
Rebars 5 d 20, L = 3500,
one in each longitudinal
joint between slabs
100
60
Circular holes d 100
c/c 1200
Fig. 8. Section along a joint between adjacent hollow core units.
400 60 480 60 400
EPS
plug
500
10
100
Neoprene
60 20 280
10x20
200 200
880
Fig. 9. Section along hollow cores. Plugs made of expanded polystyrene.
E E
60 100
H H
500
100
60 10
Rebar d 20, L = 800
H H
Hole through beam
Slab Slab
50
Cast-in-situ
concrete 150
60 100
Rebar d 20, L = 800
Fig. 10. Tie reinforcement in the cast-in-situ concrete outside the edge of the outermost
slabs (section E-E in Fig. 15). See also App. A, Fig. 9.
529
Tie reinforcement: Straight horizontal rebars A500HW through the beam and parallel to
the beam, nominal yield strength fy = 500 MPa
2.5
Slabs
45 - Extruded by Parma Oy,
R50 R125
410 500
205
R50 R120
20
35 45
1200
1156 (1164 at mid-depth) In slabs 1–2, 5–8 and 12,

one strand below the
47,5 second web in the bottom
layer was weakly bonded.
Its slippage in all slab ends
was 1,8–3,8 mm.
b w,i 401,6
In other strands the max
measured slippage was:
2,1 mm in slab 11
39,1 2 x 1,9 mm (in slab 5);
41,9 67,1 3 x 1,8 mm in slabs 2, 7 & 8
 b w,i = 332,5
1195
- Measured weight =
6,54 kN/m
Fig. 12. Most relevant measured geometrical properties
(average values).
2.6
Temporary No
supports
2.7
Loading An auxiliary loading frame was built above the test floor. It was tied by tension bars to
arrangements the floor of the research hall, see Figs 13 and 14. Six tension bars penetrated the floor
through the hollow cores outside the line loads, the rest were outside the test specimen.
The loads were generated by 12 actuators. Taking into account the weight of the loading
equipment Peq = 6,22 kN / slab unit, the following relationship can be written for the line
load on one slab:
F = Pa + 6,22 Kn (1)
Pa is the load in the actuator. The line loads were applied to the slab units by 24 tertiary
530
steel beams of the type HE 120 A ( a hot-rolled I-beam with depth = 114 mm, width =
120 mm and thickness of flange = 8 mm), each 550 mm in length. The top surface of the
slabs under these beams was evened out by gypsum. On the top of the tertiary beams,
secondary spreader beams were placed, each on two bearings. On the top of the
secondary beams, primary spreader beams were placed. The friction between the
primary and secondary spreader beams was eliminated with teflon sheets, and that
between the secondary and tertiary spreader beams with a freely rolling circular steel
bar, see Figs 15–17. The drifting of the primary spreader beams was prevented by the
friction at the upper and lower end of the actuators. The upper end of each actuator was
provided with a swivel (ball bearing).
+4,5 m
420 780 220 980 20 1020 180
Fig. 13. Auxiliary loading frame. Elevation.
1000
1000
Fig. 14. Auxiliary loading frame. Plan. The position of the tension bars is indicated by
white circles above the blue and green beams.
531
West B 7200 (span of beam)

16379 16379 16378 16378 16379 16379
6 5 4 3 2 1
600 600 600 600 600 600 600 600 600 600 600 600

E
D C
240 240
7200 (span of beam)

South
P1
North
D C C
E
1500
P
2
A 8 A
12 11 10 9 8 7
16379 16379 16378 16378 16379 16379
Fig. 15. Loading arrangements. The slabs are from two casting lots: 16377 and 16378.
The position of the actuator loads is indicated by black circles above the middle beam.
A A
Det 1
2P 2P 2P 2P 2P 2P
Tertiary a a a a a a
Secondary
Primary

7200
532
Det 1 F F
Steel plate Pa Pa
F
plates plates
L-profile Teflon
plates
Gypsum Roller bar
3 Measurements
3.1
Support
reactions
1
3 load cells
533
3.2
Vertical 30 30 30 30 30
50
displacement
17 16 14 I 12 11
24 23 22 21 II 20 19 18
2600
600 600 600 600 600 600 600 600 600 600 600 600
31 30 29 28 27 26 25
III
(from slab end)

1700
6 5 4 3 2 1
E
G
240 240
38 37 36 35 IV 34 33 32
1700
12 11 10 9 8 7
V
45 44 43 42 41 40 39
2600
8
52 51 50 49 VI 48 47 46
50
59 58 56 VII 54 53
534
3.3
Average strain
6 5 4 3 2 1
G
69 (70) 67 (68)
200
60 62
240
73 (74) 71 (72)
240
60 65
200
77 (78) 75 (76)
G
12 11 10 9 8 7
Fig. 20. Position of device (transducers 67–78) measuring average strain parallel to the beams.
G G
240 240 60 200

20 69 73 77
Inductive transducer
20 70 78
Fig. 21. Apparatus for measuring average 74 20

60
strain. L = 1100 mm. 440 440
Fig. 22. Section G-G, see Fig. 20.
3.4
Horizontal
displacements 6 5 4 3 2 1
E
83 (84) 81 (82)
H 63 H 61
62 30
240 240
60 60
65
66 64
87 (88) 85 (86)
E
12 11 10 9 8 7
Fig. 23. Transducers measuring crack width (61–66) and shear displacement at the
ends of the middle beam (81–88). Transducers 82, 84, 86 and 88 are below transducers
81, 83, 85 and 87, respectively.
535
H H
E E
83
L-profile, glued to slab 87
83 20
220
84 88
20 20
84

- None
5 Loading strategy
5.1
Load-time The loads were applied in two stages: Stage I and Stage II as shown in Fig. 26: Stage II
relationship represents the post-failure loading. Measured results are given only for Stage I.
Stages I and II
400
I II
350
300
250
Pa [kN]
200
150
100
50
0
0 30 60 90 120 150 180 210
Time [min]
Date of test 29.9.2005
536
5.2 -
After failure
6 Observations during loading (see also the photographs in App. A)
Before loading All measuring devices were zero-balanced when the

actuator forces Pa were equal to zero but the weight of the
loading equipment was on.
The cracks in the slabs were visually inspected and found to

be the same as those observed before the slabs were
installed. They are shown in Fig. 27 in which the initial
cracks in the cast-in-situ concrete are also indicated. The
maximum crack width in the soffit of the slabs was of the
order of 0,06–0,08 mm. As can be seen later, the initial
cracks on the top of slabs 3 and 12 did not affect the failure.
Pa = 140 kN corresponds to the shear force due to the

expected service load when the shear resistance of the
slabs is supposed to be prevailing in the design.
Cycling loading The interface between the joint concrete and the sawn slab
ends gradually cracked vertically on both sides of the middle
beam. At Pa = 140 kN (3rd cycle) the soffit of the floor was
inspected visually. The observed cracks are shown in Fig.
27. There were diagonal cracks in the corners of slabs 4, 5
and 11, and some initial cracks had grown in length. The
maximum width of these cracks was of the order of
0,08–0,10 mm.
Pa = 199 kN A sudden increase in deflection was observed in the

Western part of the test floor. Simultaneously a vertical
crack was observed in the Western tie beam between slabs
4 and 5.
Pa = 308 kN An inclined shear crack appeared in slab 1 next to the

middle beam.
Pa = 353 kN Additional inclined shear cracks appeared in slabs 1 (next to
the previous crack) and in slab 7. Since slabs 1 and 7
seemed unable to carry more load, the loads were quickly
reduced to prevent the possible collapse of the loading
equipment. Soon after the unloading, the test floor was
reloaded (Stage II) to check, whether the load before the
unloading was really the failure load. This proved to be the
case. In stage II the maximum actuator load was
Pa = 321,5 kN. The cracks observed after the failure are
shown in Fig. 28.
After failure When removing the slabs, it came out that the bond
between the slab ends and the underlying grout was weak,
see App. A, Figs 27–30.
537
7.1 140
Cracks at A A A A A A A A
A A A A
service load A
A
A A A AA
A A A A A A A A A
140 119
West
A A
A A
6 5 4 3 2 1
South
A A
A A
A A
North
A
A A A
12 11 10 9 8 7
East
A A A A A
A109 A AA A A
A A
A A A A
A A A A
A A A A
115 A
115 135
Fig. 27. Cracks at service load (Pai = 140 kN). The initial cracks observed before loading
The figures give the value of the actuator force at which the crack (other than initial
538
7.2
210 210
Cracks after
140 260 199 260
A A A A A A A
A A A A A
failure A
260 A 199 A
A A A A AA
A A A A A A A 140 119 A
West
A A
A A
353
6 5 4 3 2 1
South
353
308
A A
A A
A A
A
North
A
353
A A A
A
353
12 11 10 9 8 7
East
260 290 178
A A A 260 A A
A109 A AA A A
A A
A A A
A A A A A
A A A A
115 A 260
115 187
135
A dashed line indicates a crack in the soffit. The figures give the value of the actuator
539

The ratio (measured support reaction below one end of the middle beam)/ (actuator
forces on half floor) is shown in Fig. 29 and in a larger scale in Fig. 30.
1,0
0,9
0,8
0,7
Rp,obs /(6*Pa)
0,6
0,5
0,4
0,3
0,2
0,1
0,0
0 50 100 150 200 250 300 350 400
Pa [kN]
Fig. 29. Ratio of measured support reaction of the middle beam (Rp,obs) to actuator loads
on half floor.
0,846
0,845
0,844
Rp,obs /(6*Pa)
0,843
0,842
0,841
0,840
0,839
0,838
0 50 100 150 200 250 300 350 400
Pa [kN]
Fig. 30. A detail of the previous figure in a larger scale.
joint concrete, weight of loading equipment and actuator forces Pa, respectively. Vg,jc is
calculated from the nominal geometry of the joints and density of the concrete, Vg,sl from
measured average weight of slabs, both assuming that the slabs behave as simply
supported beams. For Veq and Vp the relation to the load is taken from the measured
support reaction at failure, in other words, Veq = 0,842xPeq and Vp = 0,842xPa. The
values for the components of the shear force are given in Table on the next page.
540
Shear force
Action Load
kN
Loading equipment 6,22 kN / slab 5,24
Actuator loads 353,0 kN / slab 297,20
one slab width = 1,2 m. The shear force per unit width is 280,3 kN/m
9.1
MPa MPa
Slab strands 1630 1860 Nominal (no yielding)
Beam strands 1630 1860 Nominal (no yielding)
9.2
Strength of
slab concrete, h
floor test mm mm
d
9.3
Strength of Date of test
slab concrete, h
# Cores h d From
d
Mean strength [MPa] 89,2 (+3 ... +4 d)1) vertically drilled, tested as
9.4
Strength of a
a
mm
541
9.5 a
Strength of # a a Date of test Note
beam concrete mm
a
1)
2)
In the following figures, Pa is the actuator force.
10.1
Deflections 400 400
350 350
300 300
250 11 250 18
Pa [kN]
Pa [kN]
12 19
200 200
14 20
150 150
16 21
100 100 22
17
50 50 23
0 0 24
0 5 10 15 20 25 0 5 10 15 20 25 30
400 400
350 350
300 25 300
32
250 26 250 33
Pa [kN]
Pa [kN]
200 27 200 34
150 28 150 35
29 36
100 100
30 37
50 50
31 38
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25
542
400 400
350 350
300 300
39 46
250 250 47
Pa [kN]
Pa [kN] 40
200 41 200 48
150 42 150 49
43 100 50
100
44 51
50 50
45 52
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30 35
400 400
350 350
300 300
250 250
Pa [kN]
200 53 Pa [kN]
200 14
56 35
150 150
58 56
100 100
59
50 50
0 0
0 5 10 15 20 25 0 5 10 15 20 25
Fig. 37. Deflection on line VII along Fig. 38. Net deflection of midpoint of
Eastern end beam, slabs 7–12. middle beam (35) and those of end beams
(14, 56) (Rigid body motion due to
settlement of supports eliminated).
10.2
Crack width
400
350
300
250
Pa [kN]
61
200
62
150 63
100 64
50 65
66
0
0,0 0,5 1,0 1,5 2,0 2,5
Fig. 39. Differential displacement parallel to the slab between the middle beam and
the slabs (≈ crack width). See Fig. 23 for the location of the transducers.
543
10.3
300 300
250 250
200 200
Pa [kN]
Pa [kN] 67 68
69 150 150 70
71 (beam) 72 (beam)
73 (beam) 100 100 74 (beam)
75 50 50 76
77 78
0 0
-2,0 -1,5 -1,0 -0,5 0,0 0,0 0,5 1,0 1,5 2,0 2,5 3,0
57, 59, 61, 63, 65 and 67. 62, 64, 66 and 68.
10.4
Shear 350 350
250 250
200
Pa [kN]
200
Pa [kN]
150 150
81, slab 1 top 83, slab 6 top
100 82, slab 1 bot 100
84, slab 6 bot
85, slab 7 top 50 87, slab 12 top
50
86, slab 7 bot 88, slab 12 bot
0 0
-0,5 0,0 0,5 1,0 1,5 -0,5 0,0 0,5 1,0 1,5 2,0
Differential displacement between edge of Differential displacement between edge of
slab and middle beam. A positive value slab and middle beam. A positive value
means that the slab is moving towards the means that the slab is moving towards
end of the beam. the end of the beam.
544
11 Reference tests
Both ends of slab 13 were loaded in shear as shown in Fig. 44.
2Pa
R 1500 R
50
3 (4)
1 (2) Gypsum 5 (6)
Neoprene
Steel plate
50 50
L
R R
2 4 6 50
300
Pa
650
Pa
300
1 50
3 5
350 350
300 300
250 250
Pa [kN]
Pa [kN]
200 200
150 150
100 100
50 50
0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30
Time [min] Time [min]
R1 R2
Fig. 45. Tests R1 and R2. Actuator force – time relationship.
545
350 300
300 250
250 200
Pa [kN]
Pa [kN]
200 1 1
150
150 2 2
3 100 3
100 4 4
5 50 5
50 6 6
0 0
0 5 10 15 20 25 -2 0 2 4 6 8 10 12 14
R1 R2
Fig. 46. Tests R1 and R2. Displacements measured by transducers 1–6. In test R1, at
Pa = 170 - 250 kN, transducer 4 was accidentaly disconnected from the data logger.
Table. Span of slab, shear force Vg at support due to the self weight of the slab, actuator
force Pa at failure, weight of loading equipment Peq, shear force VPa+eq due to imposed
load (=actuator load and weight of loading equipment) at failure, total shear force Vobs at
failure for one slab and total shear force vobs per unit width.
Test Date Span Vg Pa Peq VPa+eq Vobs vobs

R1 7.9.2005 9867 32,02 318,7 0,29 544,00 576,02 480,0
R2 8.9.2005 8370 27,16 296,8 0,29 491,00 518,17 431,8
Mean 547,1 455,9
13 Discussion
1. The net deflection of the middle beam due to the imposed actuator load
(deflection minus settlement of supports) was 21,8 mm or L/330, i.e. rather small
2. The shear resistance measured in the reference tests was higher than the
mean of observed values for similar slabs given in Pajari, M. Resistance of
prestressed hollow core slab against web shear failure. VTT Research Notes
2292, Espoo 2005.
3. The torsional stresses due to the different deflection of the middle beam and
end beams had a negligible effect on the failure of the slabs because the
maximum difference in the net mid-point deflection was less than 5,8 mm
4. The bond between the smooth edges of the middle beam and the grout was weak.
6. Due to the weak bond, the edge slabs slided along the beam before failure.
This reduced the negative effects of the transverse actions in the slab and had
a positive effect on the shear resistance
546
Fig. 1. Failure mode in reference test R1.
Fig. 2. Failure mode in reference test R2.
A547
Fig. 3. Installing slabs on PC beam.
Fig. 4. Uneven surface of concrete beam due to air bubbles.
A548
A549
A550
Fig. 9. A short tie bar through beam at support.
A551
A552
Fig. 14. Support arrangement at end beam.
A553
Fig. 15. Shear crack at Pa = 308 kN. The black line drawn 200 mm below the shear crack is a misprint
which does not refer to a crack.
Fig. 16. Shear crack in slab 7 at Pa = 353 kN.
A554
Fig. 17. Failure in slab 1. Photographed after stage II. In stage I, the failure crack was the same but
much thinner.
Fig. 18. Failure mode in slab 1.
A555
Fig. 19. Failure mode in slab 1.
Fig. 20. Cracks in soffit of slabs 4 and 5 after failure.
A556
Fig. 21. Crack in soffit of slab 11 after failure.
Fig. 22. Failure mode.
A557
Fig. 23. Core filling in slab 1.
Fig. 24. Grout at support after removal of slab. Note the perfect filling of the gap below the slab end.
A558
Fig. 25. Perfect filling of hollow core. Note the lack of bond between the cast-in-situ and precast
concrete.
Fig. 26. The only observed incomplete filling in hollow core.
A559
Fig. 27. Cast-in-situ concrete below slab end. Good bond with slab, weaker bond with beam.
Fig. 28. Cast-in-situ concrete below slab end. Good bond with beam, weaker bond with slab.
A560
Fig. 29. Vertical cracking at slab ends took place along the web ob the beam.
Fig. 30. Vertical cracking at slab ends took place along the web ob the beam.
A561
1.1
VTT.CR.Delta.500.2005 Last update2.11.2010
Identification
and aim DE500 (Internal identification)
Aim of the test To quantify the interaction between the Delta beam
and 500 mm thick hollow core slabs.
1.2
Test type
End beam (HE 340A) Middle beam (Delta beam) End beam (HE 340A)
1.3 VTT/FI 11.11.2005

Laboratory
& date of test
1.4
Name Load test on hollow core slab floor with Delta beam
Date 10.8.2006
Availability Confidential, owner is Peikko Finland Oy,
P.O. Box 104, FI-15101 Lahti, Finland

563
2.1
General plan West B 7200 (Span of beam)

17485 17485 17486 17486 17485 17485
6 5 4 3 2 1
600 600 600 600 600 600 600 600 600 600 600 600

South
E
D C
7200 (Span of beam)

430
North
D C
E
1500
12 11 10 9 8 7
A 8 A
17485 17485 17486 17486 17485 17485
East 7200 (Span of beam)
Fig. 2. Plan.
2.2
End beams B B
50 50
Concrete K30
Rebars 2 T10
L = 7200 500
Plywood
Steel
Roller bar 50 50
330
HE340A 9,5
16,5
300
Fig. 3. Arrangements at end beam. T10 refers to a reinforcing bar A500HW with
diameter 10 mm.
564
2.3
Middle beam The beam, see Fig. 4 and App. B, comprised
- a steel box girder with inclined, perforated webs
- a concrete component filling the empty space between the slab ends laying on the
bottom flange of the beam, cast by VTT in laboratory on the 27th of October 2005.
Concrete: K30, max aggregate size 8 mm, consistency S4/16–21 cm
Structural steel in Delta beam: S355
Steel S355
Camber 0,0 mm
200 25 holes d 150 c/c 300 in each web plate 198
20
7610
A
Fig. 4. Delta beam. The beam was provided with end plates, 20 mm in thickness.
430 Circular holes

d 150 c/c 300
Circular holes 50
d 150 c/c 300
145
8 500
305
30
200 200
1100
a) b)
Fig. 5. Cross-section A-A of Delta beam. a) At web hole. b) Between web holes.
565
2.4
Arrangements - Simply supported, span = 7,2 m, roller bearing at South end
at middle
beam Rebars Txy: Hot rolled, weldable rebar A500HW, diameter = xy mm
Rebars 5 d 20 L = 3500
6 5 4 3 2 1
Rebar d 20 L = 1050
Rebar d 20 L = 1050
Rebars 1 + 1 d 12 L = 7600
Rebars 1 + 1 d 12 L = 7600
12 11 10 9 8 7
Fig. 6. Tie reinforcement across and along the middle beam.
C C
Rebars 2 d 12
L = 7600 430
Rebars 5 d 20, L = 3500, 50
50
one in each longitudinal
joint between 120
145
adjacent slabs
500
305
30 140
180 20 20 180
1100
Fig. 7. Arrangements at middle beam, section C-C along a joint between adjacent
hollow core units, see Fig. 2. Rebars are made of steel A500HW.
566
D D
430
400
50
30
180 20 20 180
1100
Fig. 8. Section D-D along hollow cores, see Fig. 2.
E E
H H
Rebar d 20 L = 1050
H H
Slab Slab
50
150
Cast-in-situ
concrete
Rebar d 20 L = 1050
Fig. 9. Tie reinforcement in the cast-in-situ concrete outside the longitudinal edge of the
outermost slabs (section E-E in Fig. 2). See also App. A, Fig. 3. Rebars are made of
steel A500HW.
567
2.5
Slabs
45
R50 R125
410 500
205
R50 R120
20
35 45
1200
Fig. 10. Nominal geometry of slab section. For more detailed data see Fig. 11.
Fig. 11. Nominal geometry of concrete section (in scale). (The number and position of
strands were different in the present test, see Fig. 10.)
- Extruded by Parma Oy, Hyrylä factory 15.9.2005

- No upper strands
568

1,7 in slab unit 8; 1,6 in slab
50,1 unit 11; 1,5 in slab units 6, 8
and 11
bw,i Measured weight of slab
403,3 500
206,3 units = 6,43 kN/m
37,7
 bw,i = 333,9 42,5 64,5

characteristics.
2.6
Temporary No
supports
2.7
+4,5 m
Loading
arrangements

X X
180 1020 20 980 220 780 420
Fig. 13. Auxiliary loading frame, fixed to the floor of the hall by tension bars, six of which
penetrated the hollow core floor.
569
1000
1000
Fig. 14. Auxiliary loading frame, plan.
West B 7200 (span of beam)

17485 17485 17486 17486 17485 17485
6 5 4 3 2 1
600 600 600 600 600 600 600 600 600 600 600 600
South
E
D C
215 215
7200 P1
(span of
North
beam)
D C C
E
Primary
1500
P
2
A 8 A
Tertiary Secondary
12 11 10 9 8 7
17485 17485 17486 17486 17485 17485
Fig. 15. Loading arrangements. Section X-X in Fig. 13. The slabs were from two casting
lots: 17485 and 17486.
570
A A
Det 1
2P 2P 2P 2P 2P 2P
a a a a a a
Primary
Secondary
Tertiary

7200
Fig. 16. Section A-A. Pa refers to vertical actuator force.
Det 1 F F
Steel plate Pa Pa
F
plates plates
L-profile Teflon
plates
Gypsum Roller bar
Fig. 17. Detail 1, see previous figure. The stiffening plates are below the loads Pa.
3 Measurements
3.1
Support
reactions
1
3 load cells
571
3.2 30
50
30 30 30 30
Vertical 15 11
17 14 I 13
displacement
24 23 22 21 II 20 19 18
2600
600 600 600 600 600 600 600 600 600 600 600 600
31 30 29 28 27 26 25
III
(from slab end)

South
1700
6 5 4 3 2 1
E
215 215
38 37 36 35 IV 34 33 32
North
E
1700
12 11 10 9 8 7
V
45 44 43 42 41 40 39
2600
8
52 51 50 49 VI 48 47 46
50
59 57 56 VII 55 53
3.3
Average strain 6 5 4 3 2 1
J
and
G
83 (84) 69 (70) 67 (68) 81 (82)
3.4 I 63 I
150 61
Horizontal. 62 30
215 215
60 215 71 (72) 60
73 (74)
displacements 215
65
150
66 64
87 (88) 77 (78) 75 (76) 85 (86)
G
J
12 11 10 9 8 7
beams, position of transducers 61–66 measuring crack width and position of transducers
81–88 measuring measuring shear displacement of the slab ends. Numbers in
parentheses refer to the soffit of the floor, others to the top.
572
L
Fig. 21. Apparatus for measuring average strain. L = 1100 mm.
G G
385 215 215 385

20 69 73 20 77
20 70 20 78
74 50
50 550 550
Fig. 22. Section G-G along hollow cores.
I I J J
83
83 87
20
20
84 88
Fig. 23. Section I-I. Fig. 24. Section J-J.
3.5
Strain
573
None
5 Loading strategy
5.1
relationship
Before test all measuring devices were zero-balanced. The loading history is shown in
Fig. 25. It comprised the following stages:
- Stage I: cyclic loading with three cycles up to Pa = 160 kN and back to zero
- Stage II: monotonous loading close to failure followed by unloading which was
necessary due to the restricted stroke of the actuators
- Stage III: after shimming of actuators, monotonous loading until failure.
Pa = 160 kN corresponds to the shear force due to the expected service load when the
shear resistance of the slab is supposed to be critical in the design.
Stages I, II and III

450
I II III
400
350
300
Pa [kN]
250
200
150
100
50
0
0 50 100 150 200 250 300 350
Time [min]
5.2 -
After failure
574

Before test No longitudinal cracks along the strands were discovered in the soffit of
the slabs.
Stage I During stage I, the joint concrete gradually cracked along the webs of the
middle beam. Here the first visible cracks were observed at Pa = 88 kN.
At the same load new vertical cracks were also observed both in the
Western and Eastern tie beam. The number of the new cracks in the tie
beams increased with increasing load.
Stage II At Pa = 160 kN during stage II, the soffit of the slabs between the line
loads and the middle beam was inspected visually. The observed visible
cracks are shown in Fig. 26. They were all below the webs of the slabs.
When increasing Pa beyond 160 kN, new vertical cracks appeared in the
tie beams.
At Pa = 334 kN an inclined crack was observed in slab 7 accompanied by

a new inclined crack at Pa = 336 kN. At Pa = 337 kN an inclined crack,
similar to the two shear cracks in slab 7, was observed in slab 1.
Simultaneously, a sudden increase in the deflection, as well as a new
vertical crack in the tie beam between slabs 4 and 5, were observed in
the Western part of the test floor.
Despite these shear cracks on opposite sides of the middle beam, the
actuator loads could still be increased until the maximum stroke of the
actuators was achieved at Pa = 379 kN. After this, short prefacricated
steel tubes were placed close to the actuators to keep the floor in
deflected position when unloading and shimming the actuators.
Stage III After completing the shimming in stage III, the floor was reloaded.
A new inclined crack in slab 1 resulting in shear failure was observed
at Pa = 382,4 kN The cracks after the failure are shown in Fig. 28.
The initial crack on the top surface of slab 9 did not contribute to the
failure as can be concluded from the crack pattern.
After When demolishing the test specimen it was observed that that the core
failure fillings were perfect, see App. A, Fig. 22.
Delta beam seemed to recover completely after the test.
The slabs were placed directly on the bottom flange of the Delta beam.
Due to the uneven surfaces of the slab soffit and the steel flange, there
were thin gaps between the slab and the flange of the beam. When
demolishing the floor it came out that these gaps were filled with grout
(thicker gaps) or with cement paste (thinner gaps), see App. A, Fig. 23.
575
7.0 A
Cracks before A A A A A A A
A
A A
test A A
A
A
A
A
A A A A A A
6 5 4 3 2 1
12 11 10 9 8 7
A A
A A A AA A
A A A
A A
A
A A AA
Fig. 26. Initial cracks.
576
160 112
7.1
A 112
A A A A A A 156 160 A
Cracks at A
A
A
A
A A A
A
service load A A
A A A A A 160
156 160 15687 112
6 5 4 3 2 1
160 160
160 160
88
160
160 160
12 11 10 9 8 7
A A
A 156 127 A A AA A
156 160 88 A
A A A
88 A
160 A
A A AA
88 160
Fig. 27. Cracks at Pa = 160 kN. The dashed lines indicate thin ( 0,1 mm) cracks in the
soffit.
577
7.2 304 250

112 176112 160 A
Cracks after A A A A A 156 160 A A
A
A
A A
failure
A
A A A A
A A A A 304 A 160
A 304
160 15687 112 156
South
382
6 5 4 3 2 1
337
160 160
349
160 160
330
88
351
263
349
336
334
160
160 160
12 11 10 9 8
North
7
A A
156 196 195

A 127 A A AA A
156 160 88
A
A A A
88 A
304 160 A
A A AA
88 160 275
Fig. 28. Cracks until Pa = 160 kN in the soffit (dashed lines) and after failure on the top
and at the edges of the slabs.
578

In Fig. 29 the reaction force Rp,obs at one end of the middle beam, measured by three load
cells, vs. actuator load Pa is shown. The relationship is slightly nonlinear indicating some
hysteresis in the cyclic stage.
2000
1800
1600
1400
Rp,obs [kN]
1200
1000
800 Stages I & II
600 Stage III
400
200
0
0 50 100 150 200 250 300 350 400
Pa [kN]
Fig. 29. Measured reaction force Rp,obs at one end support of middle beam vs. actuator
load Pa. Rp,obs includes only the reaction due to the actuator loads Pa.
Figs 30 and 31 illustrate the ratio of measured support reaction Rp,obs to the theoretical
support reaction Rp,th calculated from six actuator loads assuming simply supported slabs.
Thus Rp,th is equal to (9900–1450)/9900x6xPa = 0,8535x6xPa. Before failure the maximum
difference is -0,59%, i.e. Rp,th is less than 1% smaller than the measured support reaction.
The assumption of simply supported slabs is accurate enough to justify the calculation of
the experimental shear resistance based on it.
2,0
1,8 Stages I & II
1,6 Stage III
1,4
Rp,obs/Rp,th
1,2
1,0
0,8
0,6
0,4
0,2
0,0
0 50 100 150 200 250 300 350 400
Pa [kN]
support reaction (Rp,th) vs. actuator force Pa. Only actuator loads Pa are taken into account
in the support reaction.
579
1,2
1,0
Rp,obs/Rp,th 0,8
Stage III
0,6
0,4
0,2
0,0
0 50 100 150 200 250 300 350 400
Pa [kN]
Fig. 31. As Fig. 27 but only stage III is shown.
joint concrete, weight of loading equipment and actuator forces Pa, respectively. All
supported beams. For Veq and Vp this means that Veq = 0,8535xPe and
Vp = 0,8535xPa. Vg,jc is calculated from the nominal geometry of the joints and density of
the concrete, other components of the shear force are calculated from measured loads
and weights. The values for the components of the shear force are given in Table 1.
Shear force
Action Load
kN
Weight of slab unit 6,53 kN/m Vg,sl 32,19
Weight of joint
0,39 kN/m Vg,jc 1,92
concrete
Loading equipment 6,22 kN Veq 5,33
Actuator loads 382,1 kN Vp 327,46
one slab unit and the shear force per unit width is vobs = 305,8 kN.
9.1 ReH/Rp0,2 Rm
Strength of Component Note
MPa MPa
steel:
Delta beam  355 Nominal (S355J2G3)
Nominal value for reinforcing
Reinforcement Txy (A500HW) 500
bars (no yielding in test)
580
9.2
Strength of
slab concrete, h
floor test mm mm
d
9.3
Strength of
slab concrete, h
d
Mean strength [MPa] 75,9 (+1-2 d)1) vertically drilled, tested as
9.4
Strength of a
cast-in-situ # a a Date of test Note
concrete mm
a
1)
2)
In the following figures, VP stands for the shear force of one slab end due to imposed
actuator loads, calculated assuming simply supported slabs.
10.1
Deflections
350 350
300 300
250 250
18
Vp [kN]
Vp [kN]
200 200 19
11
150 150 20
12
21
100 14 100 22
16
50 50 23
17
24
0 0
0 5 10 15 20 25 0 5 10 15 20 25 30 35 40
581
350 350
300 300
250 25 250 32
Vp [kN] 26 33
Vp [kN]
200 200
27 34
150 28 150 35
29 36
100 100
30 37
50 31 50 38
0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35
350 350
300 300
250 39 250 46
40 47
Vp [kN]
Vp [kN]
200 200
41 48
150 42 150 49
43 50
100 100
44 51
50 45 50 52
0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
350 350
300 300
250 250
53
Vp [kN]
200
Vp [kN]
200
54
150 56 150
58 14, End
100 100
59 35, Delta
50 50 56, End
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Fig. 38. Deflection on line VII along end Fig. 39. Net deflection of midpoint of
beam, slabs 7–12. middle beam (35) and those of end beams
(14, 56). Settlement of supports
eliminated.
582
10.2
Crack width
350
300
250
Vp [kN]
200 61
62
150
63
100 64
65
50
66
0
0,0 0,5 1,0 1,5 2,0 2,5
10.3
Average strain Transducers 67–78 shown in Figs 20–22 measured the differential displacement between
two points fixed to the test floor in order to determine the strains in beam’s direction. Figs
41–42 depict the results.
The differential displacement divided by the mutual distance of the two fixing points
represents the average strain between those points. Assuming linear variation for the
strain in vertical direction, the average strain at any depth between the measured top and
bottom values can be calculated from the measured results. The average strain at the top
and bottom fibre of the slabs and of the middle beam obtained in this way are shown in
Figs 43–45 and in Figs 46–48 at service load and at failure load, respectively.
Assuming parabolic curvature due to the uniformly distributed load on the beam and
having the span of the beam equal to 7,2 m, the maximum strain within each of the slabs
3, 4, 9 and 10 is 4% higher than the average strain measured between the edges of the
slab. In this way, the maximum strain is obtained from the average strains by multiplying
the latter by 1,04.
- On the top, the strains in the slabs were essentially smaller than those in the
beam. This can be explained by the low transverse shear stiffness of the slabs.
The scatter of the slab strains on the top was small.
- At the bottom, the strains in the slabs were also essentially smaller than those in
the beam at the same level. The support reaction of the slabs was partly carried by
the inclined webs of the middle beam. The interaction between the soffit of the
slabs and the bottom flange of the beam was too weak to make the strains
compatible. The tie reinforcement also tried to keep the transverse strain of the
slab ends as small as possible. This can clearly be seen in Fig. 45. At Vp = 7 kN
some kind of slippage took place
- At Vp = 137 kN the strain in the soffit of the slabs had a large scatter.
At Vp = 326 kN the strain at the bottom of slab 9 was very small when compared
with that of the other slabs. In slabs 3, 4 and 10 the scatter in the measured strains
was relatively small. In slab 9, the strain at the bottom was much smaller than in
the other slabs.
583
350 350
300 300
250 250
Vp [kN]
Vp [kN] 67 200 200 68
69 70
150 150
71 (beam) 72 (beam)
100 100 74 (beam)
73 (beam)
75 50 76
50
77 78
0 0
-1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
67, 69, 71, 73, 75 and 77. 72, 74, 76 and 78.
350 350
300 300
250 250
Vp [kN]
Vp [kN]
200 200 68
67
69 150 70
150
71 (beam) 72 (beam)
100 100 74 (beam)
73 (beam)
75 50 76
50
77 78
0 0
-1,4 -1,2 -1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
Average strain [mm/m] Average strain [mm/m]
Fig. 43. Top fibre of floor. Average strain Fig. 44. Bottom fibre of floor. Average
calculated from the differential strain calculated from the differential
displacements shown in Figs 41–42. displacements shown in Figs 41–42.
20
67-68 (Slab 3)
600 69-70 (Slab 4)
18
16 500 71-72 (Beam)
14 73-74 (Beam)
Depth [mm]
12 400 75-76 (Slab 9)

Vp [kN]
68
10 300 77-78 (Slab 10)
70
8
72 (beam)
6 200
74 (beam)
4
76 100
2
78
0 0
0,00 0,01 0,02 0,03 0,04
-400 -200 0 200 400 600
Average strain [mm/m] Strain [m/m]
Fig. 45 Bottom fibre of floor. Initial part of Fig. 46. Average strain of slabs 3, 4, 9
the previous figure. and 10 as well as that of the middle beam
at estimated service load Pa = 160 kN
(Vp = 137 kN).
584
67-68 (Slab 3) Slabs 3, 4,9 & 10)

600 69-70 (Slab 4) 600
71-72 (Beam) Beam
500 500
73-74 (Beam)
Depth [mm]
Depth [mm]
400 75-76 (Slab 9) 400
300 77-78 (Slab 10) 300
200 200
100 100
0 0
-2000 -1000 0 1000 2000 -2000 -1000 0 1000 2000
Strain [m/m] Strain [m/m]
Fig. 48. Mean of average strains

Fig. 47. Average strain of slabs 3, 4, 9 and
measured for slabs 3, 4, 9 and 10 as
10 as well as that of the middle beam at
well as mean of average strains
failure load Pa =382 kN (Vp = 326 kN).
measured for the middle beam at failure
load Pa = 382 kN (Vp = 326 kN).
10.4
Shear A positive value means that the slab is moving towards the end of the beam. The
displacement measured curves do not look logical, and there may have been some mess in numbering
the transducers. There are two puzzles:
- why do the upper flange of slab 1 and lower flange of slab 7 in Fig. 49 move together
and vice versa?
- why does the upper flange move less than the lower flange in Fig. 51 and partly in Fig. 49?
350 150
300 125
250
100
Vp [kN]
Vp [kN]
200
75
150
81 (slab 1) 81 (slab 1)
50
100 82 (slab 1) 82 (slab 1)
50 85 (slab 7) 25 85 (slab 7)
86 (slab 7) 86 (slab 7)
0 0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,0 0,1 0,2 0,3 0,4
Fig. 49. North end of middle beam. Fig. 50. Same as previous figure, but only
Differential displacement between edge of the initial part of the curves is shown.
slab and middle beam in beam's direction.
585
350 150
300 125
250
100
Vp [kN]
Vp [kN]
200
75
150
83 (slab 6) 83 (slab 6)
50
100 84 (slab 6) 84 (slab 6)
87 (slab 12) 25 87 (slab 12)
50
88 (slab 12) 88 (slab 12)
0 0
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,0 0,1 0,2 0,3 0,4
Fig. 51. South end of middle beam. Fig. 52. Same as previous figure, but only
Differential displacement between edge of the initial part of the curves is shown.
slab and middle beam in beam's direction.
10.5 Strain
-
11 Reference tests
Slab 13 was taken from the same casting lot 17485 as four other slabs in the floor test.
2Pa
R 1500 R
50
3 (4)
1 (2) Gypsum 5 (6)
Neoprene
Steel plate
50 50
L
R R
2 4 6 50
300
Pa
650
Pa
300
1 50
3 5
Fig. 53. Layout of reference tests. For L, see Table 3. Test R1 was carried out first.
586
350
300
Pa [kN] 250
200
150
R1
100
R2
50
0
0 10 20 30 40 50
Time [min]
In both reference tests flexural cracks were observed below the loads before the shear
tension failure took place in the webs close to the support. No visible slippage of the
strands was observed before the failure. The failure modes are illustrated in Fig. 56 and in
App. A, Figs 38–41.
The time dependence of the load and measured load – deflection relationship are shown
in Figs 55.
Reference test R1 Reference test R2
600 600
500 500
400 1 400 1
Vp [kN]
Vp [kN]
2 2
300 300
3 3
200 4 200 4
5 5
100 6 100 6
0 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Fig. 55. Vertical displecements measured by transducers 1–6. VP is the support reaction
due to actuator forces Pa.
587
300
Pa Pa
650
Pa Pa
300
End 1 End 2
Fig. 56. Failure mode in tests R1 (End 1) and R2 (End 2).
Table 2. Span of slab, shear force Vg at support due to the self weight of the slab, actuator
force Pa at failure, weight of loading equipment Peq, shear force Vp due to imposed load at

Test Date Note
mm kN kN kN kN kN/m
Mean 529,4 441,1
588

13 Discussion
small. It was 2,4–2,6 mm greater than that of the end beams. Hence, the
beams had a negligible effect on the failure of the slabs.
the observed values for similar slabs given in Pajari, M. Resistance of
2292, Espoo 2005.
3. The observed shear resistance in the floor test was 69% of that in reference
tests.
4. The edge slabs slided 0,6 … 1,4 mm along the beam before failure.This reduced
the negative effects of the transverse actions in the slab and had a positive
effect on the shear resistance.
5. The failure mode was web shear failure of edge slabs. The Delta beam seemed
to recover completely after the failure.
589
Note: In Figs. 23–33 the cracks marked with red colour and letter A refer to initial cracks which
existed before the onset of the loading.
Fig. 1. Delta beam as installed.
Fig. 2. Delta beam as installed.
A590
Fig. 3. Slab placed on Delta beam.
Fig. 4. Tie reinforcement at support of beam.
A591
Fig. 5. End of middle beam (Delta beam) after grouting.
A592
Fig. 7. Loading equipment.
Fig. 8. Longitudinal view on the loading equipment.
A593
Fig. 9. Actuators on the primary spreader beam and below the temporary loading frame.
A594
Fig. 11. Transducers at end of middle beam.
Fig. 12. Transducer, fixed to the bottom flange of the middle beam, measuring sliding of slab along beam.
A595
Fig. 13. Bottom flange of middle beam supporting slabs.
Fig. 14. Equipment for measuring average transverse strain of the soffit.
A596
Fig. 15. Initial crack in slab 3 close to end beam.
Fig. 16. Shear cracks in slab 7 at Pa = 336 kN.
A597
Fig. 17. Shear crack in slab 1 at Pa = 337 kN.
Fig. 18. Shear cracks in slab 1 after failure at Pa = 382 kN.
A598
Fig. 19. Shear cracks in slab 1 after removing the loading equipment.
Fig. 20. Failure pattern on the top of slab 1 after removing the loading equipment and drilling the cores.
A599
Fig. 21. Slab 1 after removing slabs 2–6 and 8–12.
Fig. 22. Perfect filling in a hollow core of slab 1.
A600
Fig. 23. Cast-in-situ concrete after removing the slabs. Note the grout layer between the slab and the
flange of the beam.
Fig. 24. Western end beam. Cracks in cast in-situ concrete at the end of slabs 1 and 2.
A601
Fig. 26. Western end beam. Cracks in cast in-situ concrete at the end of slabs 2–4.
A602
A603
Fig. 30. Eastern end beam. Cracks in cast in-situ concrete at the end of slabs 7–9.
A604
A605
A606
Fig. 35. Eastern end beam. No crack in cast in-situ concrete at the end of slab 12.
Fig. 36. Arrangements in reference test R1.
A607
Fig. 37. Arrangements in reference test R1.
Fig. 38. Failure in reference test R1.
A608
A609
‘
A610
APPENDIX B: DELTA BEAM
Fig. 1. Delta beam.
B611
1.1
Identification
and aim
PC400U (Internal identification)
Aim of the test In test VTT.PC.InvT.400.1993 the edges of the beam were
provided with square indents which, after grouting, served
as dowels between the slab ends and the beam. In the
present test the indents were missing and the longitudinal
tie reinforcement was placed near to the mid-depth of the
slabs. It was anticipated that these changes would
enhance the shear resistance of the slabs.
1.2
Test type
HE 260A Prestressed concrete beam HE 260A
1.3 VTT/FI 31.5.2006

Laboratory
& date of test
1.4
Name Load test on hollow core slab floor with prestressed concrete beam
Date 27.12.2006
613
2.1
General plan West 4800 (span of beam)
North
South

4 3 2 1
1200 1200 1200 1200 1200 (from slab end)
Line load
240
4800 (span of beam)

240
1200
8 7 6 5
Line load
9000
Fig. 2. Plan.
2.2
End beams Concrete 50 50
K30
HE260A
Rebars
12,5 2 d 12 60 400
250 L = 4800
7,5 60
12,5 Plywood
260 Steel plate
Roller bar 50 50
HE260A
Fig. 3. End beam made
of steel.
Simply supported, span = 4,8 m, roller bearing below Northern end

614
2.3
Middle beam - Cast by Parma Oy, 13.4.2006
- Simply supported, span = 4,8 m, roller bearing at both ends
P 5 holes d 60
400
200
180
200 P 200
1200 1200 1200 1200
5200
480
47 4J12,5
60 400
Chamfer 20x20 170 2 A1
180 13 43
48
200 200 49 48 48 48 48 48 48 49 26J12,5
880
96 66 48 74 48 66 96
Fig. 6. Cross-sectional geometry. Fig. 7. Main reinforcement.
Concrete: K80
J12,5: Prestressing strand  =12,5 mm, Ap = 93 mm2, p0 = 1300 MPa,

nominal strength fp0,2/fp = 1630/1860 MPa, did not yield in the test
A1: Rebar A500HW,  12 mm, L = 5130 mm, nominal yield strength fy = 500 MPa,
did not yield in the test
615
616
2.4
Arrangements 4 rebars d 12, L = 5200 50
60 480 60 50
at middle
beam
60 Plastic
plug 400
400
80 170 10
60 100 20
10 Neoprene 180
180 10x20
200 200
3 rebars d 16, L = 2800 880
one in each longitudinal joint of slabs
Fig. 9. Section along joint between Fig. 10. Section along hollow cores.
adjacent slab units.
E E
Grout 60 480 60 50
H E H 50 H
80
Slab 400
H End of
beam 80
10
E 60 100
20 180
200 200
880
Fig. 11. Cast-in-situ concrete (grout) Fig. 12. Section E-E, see Fig. 11.
outside hollow core slabs.
H H
Hole through beam
Slab Slab
50 (width on
Cast-in-situ the top)
concrete 150
60
Rebar d 16, L = 800 100
Tie reinforcement: Straight horizontal rebars A500HW through the beam and parallel to
the beam, nominal yield strength fy = 500 MPa
617
Rebars 3 d 16 L = 2800
4 3 2 1
Rebar 1 d 16 L = 800
Rebar 1 d 16 L = 800
Rebars 1 + 1 d 12 L = 5200
Rebars 1 + 1 d 12 L = 5200
8 7 6 5
Fig. 14. Overview on tie reinforcement at middle beam. d xy refers to a reinforcing bar
A500HW with diameter xy mm. For the vertical position of the bars, see Figs 9 and 12.
2.5
Slabs 40 - Extruded by Parma Oy,
- Measured weight =
220 R135 5,65 kN/m
400
65 90 65
45o 30
40 36
1200
1164 (at mid-depth)

Max measured bond
slippage: 3x2,0 mm
53,9 (in slabs 1, 3 and 6;
61,4 67,2 67,2 67,2 61,4 3x1,9 mm
397
(in slabs 1, 4 and 7)
20,2
36,8 bw,i = 324,3 39,7
Fig. 16. Most relevant measured geometrical

properties.
618
2.6
Temporary No
supports
2.7

Loading 600 600 600 600 600 600 600 600
arrangements
P P
a1 a1
60 480 60
South
4800
North
(span of beam)
2P 2P
a2 a2
1200
P
P 2 P
a1 a1
A A
A A
Det 1
Primary 2Pa2 2P
a2
Secondary 2P 2P
a1 a1
Tertiary

Det 1
F F
Steel plate P P
a2 a2
F F
plates plates
L-profile Teflon
plates
Gypsum Roller bar
F
550 550
600
619
3 Measurements
3.1
Support
reactions 1
3 load cells
North
5
3.2
Vertical 30 30 30 30 30
50
displacement 14 12 11
15 I 13
20 19 II 18 17 16
2500
25 24 23 22 21
III
(from slab end)

1400
4 3 2 1
E
240 240
30 29 IV 28 27 26
E
1400
8 7 6 5
South
North
V
35 34 33 32 31
2500
40 39 VI 38 37 36
50
45 44 VII 43 42 41
Fig. 21. Location of transducer 11 … 45 for measuring vertical deflection along lines I … VII.
620
3.3
Average strain
4 3 2 1
G
59 (60) 57 (58)
200
60
240
63 (64) 61 (62)
240
60
200
67 (68) 65 (66)
G
South
North
8 7 6 5
Fig. 22. Position of device (transducers 57–68) measuring average strain parallel to the beams.
G G
Steel bar Transducer 240 240 60 200

20 59 63 67
Fig. 23. Apparatus for measuring 20 60 68

average strain. L =1100 mm. 64 20
64
440 440 60
Fig. 24. Section G-G, see Fig. 22.
3.4
Horizontal.
displacements
4 3 2 1
J
73 (74) 71 (72)
H 53 H 52 51
30
240 240
60 60
56 55 54
77 (78) 75 (76)
J
8 7 6 5
Fig. 25. Transducers measuring crack width (51–56) and shear displacement at the
ends of the middle beam (71–78). Transducers 72, 74, 76 and 78 are below transducers
71, 73, 75 and 77, respectively.
621
H H J J
73
L-profile, glued to slab 73 77
20
220
74 78
20 20
74
L-profile, glued to slab Fig. 27. Section J-J, see Fig. 25.
- None
5 Loading strategy
5.1
Load-time
Stage I Stage II
relationship
300
250
200
Pai [kN]
150
100 Pa1
Pa2
50
0
0 20 40 60 80 100 120 140 160
Time [min]
Date of test 31.5.2006
5.2
After failure -
622
Before loading All measuring devices were zero-balanced when the

actuator forces Pai were equal to zero but the weight of the
loading equipment was on. The loading history is shown in
Fig. 28.
Pai = 125 kN corresponds to the shear force due to the

expected service load when the shear resistance of the
slabs is supposed to be prevailing in the design.
Stage I The joint concrete gradually cracked along the webs of the
middle beam on the Eastern side (slabs 5–8). Here the
first visible cracks were observed at Pa1 = 38 kN. On the
opposite (Eastern) side of the middle beam, a similar
crack was observed at Pa1 = 70 kN.
Pa1 = 125 kN The soffit of the slabs between the line loads and the
during stage II middle beam was inspected visually. The observed two
visible cracks are shown in Fig. 29. They were both below
the webs of the slabs.
Pa1 = 189 – 200 kN New vertical cracks were observed both in the Western
and Eastern tie beam. Both the number and the length of
the cracks in the tie beams (= concrete connecting the
slab ends above the supporting end beams, see Fig. 30)
increased with increasing load.
Pa1 = 295,1 kN and The floor suddenly failed in shear on the Eastern side of
Pa2 = 292,2 kN the middle beam, see App., Figs 12–14 and 20–23. In
slabs 5–8 an inclined crack appeared between the middle
beam and the line load so rapidly that it was impossible to
say, which slab failed first. The cracks after the failure are
shown in Fig. 30.
The joint between the grout and middle beam cracked

neatly along the smooth edges of the middle beam as
shown in App. A, Figs 24 an 25.
623
7.1 A
Cracks at
service load A
A A A A A A
4 3 2 1
125
70
38
125
South
North
8 7 6 5
Fig. 29. Cracks at service load (Pai = 125 kN). The initial cracks observed before loading
The figures give the value of the actuator force at which the crack (other than initial
624
7.2
Cracks after 200 250 190 220 206
failure A
200
A A A A A A
A
190 189 190
4 3 2 1
220
125
220
70
38
125
296 296
South
North
296
296
296 296 296
8 7 6 5
200 208 221 200

206
A
206
200
A dashed line indicates a crack in the soffit. The figures give the value of the actuator
625

The ratio (measured support reaction below one end of the middle beam) / (load on half
floor) is shown in Fig. 31.
0,900
0,890
0,880
0,870
Rp,obs/ Fai
0,860
0,850
0,840
0,830
0,820
0,810
0,800
0 50 100 150 200 250 300
Pa2 [kN]
Fig. 31. Ratio of measured support reaction of the middle beam (Rp,obs) to load on half
floor vs. actuator force Pa2. Only actuator loads Pai are taken into account in the support
reaction.
joint concrete, weight of loading equipment and actuator forces Pai, respectively. The
components of the shear force due to the self-weight are calculated assuming that the
slabs behave as simply supported beams. For Veq and Vp the relation is Veq = 0,8619xPe
and Vp = 0,8619x(Pa1.+ Pa2)/2 at failure. Vg,jc is calculated from the nominal geometry of
the joints and density of the concrete, other components of the shear force are
calculated from measured loads and weights. The values for the components of the
shear force are given in Table below.
Shear force
Action Load
kN
one slab unit wit width = 1,2 m. The shear force per unit width is 235,3 kN/m.
626
9.1
steel MPa MPa
Slab strands 1630 1860 Nominal (no yielding)
Beam strands 1630 1860 Nominal (no yielding)
9.2
Strength of
slab concrete, h
floor test mm mm
d
9.3
Strength of
slab concrete, h
# Cores h d Date of test From
d
9.4
Strength of a
a
mm
9.5
Strength of
beam concrete h
# Cores h d Date of test From
d
mm mm
6 100 100 1.6.2006 Upper part, vertically drilled
Mean strength [MPa] 72,0 (+1 d)1) Tested as drilled2)
St.deviation [MPa] 1,84 2378 Density = 2378 kg/m3
1)
2)
627
In the following figures, Vp stands for the shear force of one slab end due to imposed
10.1
Deflections 300 300
250 250
200 200
Vp [kN]
Vp [kN]
150 11 150 16
12 17
100 13 100 18
14 19
50 50
15 20
0 0
0 1 2 3 4 5 6 0 2 4 6 8 10 12 14 16
Fig. 32. Deflection on line I along western Fig. 33. Deflection on line II in the middle
end beam vs. support reaction Vp of one of slabs 1–4.
slab due to actuator loads.
300 300
250 250
200 200
Vp [kN]
Vp [kN]
150 21 150 26
22 27
100 100 28
23
24 29
50 50
25 30
0 0
0 2 4 6 8 10 12 14 0 1 2 3 4 5 6 7 8
300 300
250 250
200 200
Vp [kN]
Vp [kN]
150 31 150 36
32 37
100 33 100 38
34 39
50 50
35 40
0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 16 18
628
300 300
250 250
200 200
Vp [kN]
Vp [kN]
150 41 150
42 13
100 43 100 28
44 43
50 50
45
0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7
Deflection [mm] Net deflection [mm]
Fig. 38. Deflection on line VII along end Fig. 39. Net deflection of midpoint of middle
beam, slabs 5–8. beam (28) and those of end beams (13, 43).
10.2
Crack width 300
Fig. 40. Differential displacement
250
parallel to the slab between the
middle beam and the slabs (≈ crack 200 51
width). See Fig. 25 for the location of 52
Vp [kN]
the transducers. Transducer 56 has 150 53

given erroneus results because the 100
54
crack width equal to 2,0 mm was not 55
visually observed during the cyclic 50 56
stage
0
0,0 0,5 1,0 1,5 2,0 2,5 3,0
10.3
250 250
200 200
57 58
Vp [kN]
Vp [kN]
59 150 150 60
61 beam 62 beam
63 beam 100 100 64 beam
65 66
67 50 50
68
0 0
-1,0 -0,8 -0,6 -0,4 -0,2 0,0 0,0 0,2 0,4 0,6 0,8 1,0
Fig. 41. Differential displacement at top Fig. 42. Differential displacement at soffit of
surface of floor measured by transducers floor measured by transducers 58, 60, 62,
57, 59, 61, 63, 65 and 67. 64, 66 and 68.
629
10.4
Shear 300 300
displacement
250 250
200 200
Vp [kN]
Vp [kN]
150 71 Slab1 150 75 Slab 5
72 Slab 1 76 Slab 5
100 100
73 Slab 4 77 Slab 8
74 Slab 4 78 Slab 8
50 50
0 0
-0,2 0,0 0,2 0,4 0,6 0,8 1,0 -0,2 0,0 0,2 0,4 0,6 0,8 1,0
of slab and middle beam. A positive value of slab and middle beam. A positive value
means that the slab is moving towards means that the slab is moving towards
the end of the beam. the end of the beam.
11 Reference tests
2Pa
R 1200 200 R
50
5 (2)
4 (3) Gypsum 6 (1)
Neoprene
Steel plate
50 50
L
R R
3 2 1
300 50
Pa
650
Pa
300 50
4 5 6
630
180 200
160 180
140 160
120 140
120
Pa [kN]
Pa [kN]
100
100
80
80
60
60
40 40
20 20
0 0
0 5 10 15 20 25 0 5 10 15 20 25
Time [min] Time [min]
R1 R2
Fig. 46. Tests R1 and R2. Actuator force – time relationship.
350 350
300 300
250 1 250 1
Vp [kN]
Vp [kN]
200 200 2
150 3 3
150
4 4
100 100
5
50 50 5
6
0 6
0
-2 0 2 4 6 8 10 -2 0 2 4 6 8
R1 R2
Fig. 47. Tests R1 and R2. Displacements measured by transducers 1–6. Vp is the shear
force due to the actuator force Pa.
Table. Span of slab, shear force Vg at support due to self weight of slab, actuator force
Pa at failure, weight of loading equipment Peq, shear force Vp due to imposed load at
Span Vg Pa Peq Vp Vobs vobs

Test Date
R1 June 9, 2006 8894 25,2 173,9 0,29 303,1 328,3 273,6
R2 June 9, 2006 7700 21,9 185,1 0,29 315,2 337,0 280,9
Mean 332,7 277,2
631

13 Discussion
1. The net deflection of the middle beam due to the imposed actuator load
(deflection minus settlement of supports) was 6,2 mm or L/774, i.e. very small
2. The shear resistance measured in the reference tests was a bit lower than the
mean of observed values for similar slabs given in Pajari, M. Resistance of
prestressed hollow core slab against web shear failure. VTT Research Notes
2292, Espoo 2005.
3. The torsional stresses due to the different deflection of the middle beam and
end beams had a negligible effect on the failure of the slabs because the
maximum difference in the net mid-point deflection was less than 1,3 mm
4. The bond between the smooth edges of the middle beam and the grout was
weak.
6. The position of the tie reinforcement 170 mm above the slab soffit was
favourable.
7. Due to the weak bond and position of the tie reinforcement, the edge slabs did
slide 0,17 … 0,21 mm along the beam before failure, see Figs 43 and 44. This
reduced the negative effects of the transverse actions in the slab and had a
positive effect on the shear resistance
8. The transverse shear deformation of the edge slabs was considerable which
can be seen from the relative displacement between the top flange and bottom
flange, see Figs 43 and 44. At failure, the transverse horizontal displacement of
the top flange was 0,52 … 0,67 mm greater than that of the bottom flange
9. The observed shear resistance was considerably higher than that in previous
test VTT.PC.InvT.400.1993. This can be explained by the improvements in the
middle beam, i.e. by elimination of indents at the edges of the beam and by the
higher position of longitudinal reinforcement.
632
Fig. 1. Slab installed on middle beam. Fig. 2. Middle beam.
Fig. 3. Reinforcement at the end of middle beam. Fig. 4. Reinforcement at the end of middle beam.
One rebar parallel to the beam on the top of the One rebar parallel to the beam on the top of the
joint is still missing. joint is still missing.
Fig. 5. Loading arrangements. Fig. 6. Transducers measuring deflection and

differential horizontal displacement in the middle of
the test floor.
A633
Fig. 7. Transducers measuring differential Fig. 8. Loading of slabs 2 and 6 with two actuators
horizontal displacement at one end of the middle
beam.
Fig. 9. Loading arrangements. Fig. 10. Transducers at one end of middle beam
Fig. 11. An overview on test arrangements. Fig. 12. Failure of slab 5.
A634
Fig. 13. Failure of slab 8. Fig. 14. Soffit of slabs 5–8 after failure.
Fig. 15. Cracks in Western tie beam after failure. Fig. 16. Cracks in Western tie beam after failure.
Slabs 1–3. Slabs 2–4.
Fig. 17. Cracks in Western tie beam after failure. Fig. 18. Cracks in Eastern tie beam after failure.
Slabs 3–4. Slabs 5–7.
A635
Fig. 19. Cracks in Eastern tie beam after failure. Fig. 20. Failure in slab 5.
Slabs 7–8.
Fig. 21. Failure in slabs 5–7. Fig. 22. Failure in slabs 6–8.
Fig. 23. Failed ends of slabs 5–8. Fig. 24. Western side of middle beam. Note the
failure of the bond along the vertical interface of the
cast-in-situ concrete and the precast beam as well as
the perfect filling of the gap below the slab end.
A636
Fig. 25. Western side of middle beam. Fig. 26. Eastern side of middle beam.
Fig. 27. Overview on arrangements in reference Fig. 28. Reference test R1. Northern side of slab
tests. The actuators in the rear were not used. after failure.
Fig. 29. Reference test R1. Southern side of slab Fig. 30. Reference test R2. Northern side of slab
after failure. after failure.
A637
Fig. 31. Reference test R2. Southern side of slab

after failure.
A638
VTT.CR.A-beam.320.2006
1.1
Identification
VTT.CR.A-beam.320.2006 Last update 2.11.2010
and aim
A320 (Internal identification)
Aim of the test To quantify the interaction between the A-beam and
hollow core slabs.
1.2
Test type
HE200A HE200A
A-beam
1.3 VTT/FI 17.11.2006

Laboratory
& date of test
1.4
Name Load test on hollow core slab floor with A-beam
Date 15.3.2007
Availability Confidential, owner is Anstar Oy, Erstantie 2,
FI-15540 Villähde, Finland
www.anstar.fi

639
2.1 WEST
General plan B 4800 (span of beam)
4 3 2 1
8000
1200 1200 1200 1200
40
(from slab end)

(from slab end)
1500
1250
E
C D
305 4800 (span of beam)
305
SOUTH E C D NORTH
8 7 6 5
4800 (span of beam)
Fig. 2. Plan. Position of line loads shown in blue color.
2.2
End beams B B
50 50 Concrete K30
Plastic plug
Tie beam
28
L=4800
Plywood 50 50
Steel plate Steel rod,

10
6,5 190 free to roll
End beam 10
HE 200 A
200
Fig. 3. Arrangements at end beam. 8 refers to a reinforcing bar TW8, see 2.3.
640

2.3
Middle beam The beam, see Figs 4–9 and App. A, comprised
- a prefabricated steel box girder with perforated top plate and lontitudinal rebars
welded onto the top plate
- a precast concrete component which filled the box girder
- a cast-in-situ concrete component on the top of the box girder
- The precast concrete cast by Anstar Oy

- Upper part cast by VTT in laboratory together with the joint grouting, 6.11.2006
Concrete: K40 in the prefabricated part, K30 in the upper part
A-BEAM:
End plates: Raex 460 M ( nominal fy  460 MPa)
Other structural steel: S355J2G3, fy  355 MPa (nominal fy)
Passive reinforcement in A-beam:

Tie reinforcement:
Txy: Hot rolled, weldable rebar A500HW,  = xy mm, see Figs 4–8.
Fig. 4. A-beam when one slab element has been installed.
641
Fig. 5. Illustration of steel component of middle beam. End plates and hoop
reinforcement not shown.
Steel S355
Camber 0,0 mm
U 10 T Five holes 52x72 mm2
200 1200 1200 1200 1200 200

5200
U T
Fig. 6. A-beam. Elevation.
T T
3 rebars d 32
490
Rebar d 40
130 8
8
Hoops d 6
Steel tube 245
c/c 200 60
4x60x80
45 70
110 4 rebars d 32 580 110 10
800
Fig. 7. A-beam. Section T-T. d xy refers to rebars Txy.
642
U U
500
End plate
270
Rebars welded to end plate

600
Fig. 8. A-beam. Section U-U.
135
190 200 200
Fig. 9. A-beam. Holes in the top flange.
643
2.4
Arrangements - Simply supported, span = 4,8 m, roller bearing at North end
at middle
beam D D
500
210
210
o
60
Suspension
1460 reinforcement d 16 1460
Rebars 2 d 10 L = 5200
275 320
95 15 10 15 95
800
Tie reinforcement d 12
Fig. 10. Arrangements at middle beam, section D-D along a joint between adjacent
hollow core units in Fig. 2. The suspension reinforcement along the edges of the floor
was similar to that shown here but the length of the tails was 1260 mm instead of
1460 mm. d xy refers to rebar Txy, see 2.3.
C C
50
320
95 15 10 15 95
800
Fig. 11. Section C-C along hollow cores in Fig. 2.
644
Rebars 3 d 12 L = 3000
4 3 2 1
Rebar 1 d 8 L = 2400
Rebar 1 d 8 L = 2400
Rebars 1 + 1 d 10 L = 5200
Rebars 1 + 1 d 10 L = 5200
8 7 6 5
Fig. 12. Overview on tie reinforcement at middle beam. d xy refers to a reinforcing bar
Txy, see 2.3.
Rebars 3 d 16 L = 4000
4 3 2 1
Rebar 1 d 12 L = 3600 Rebar 1 d 12 L =3600
8 7 6 5
Fig. 13. Overview on suspension reinforcement at middle beam. d xy refers to a

reinforcing bar Txy, see 2.3.
645
2.5
Slabs 1158
41
35 35
26
26
71
R1
R1
111,5 114,5 229 320
R1
R1
26
26
35 35 1200
Fig. 14. Illustration of nominal slab cross-section drawn in scale. Concrete K60.
- Extruded by Parma , Hyrylä factory 6.10.2006


43,1 43,1 1,9 and 1,6 mm in slab unit
6, 2x1,4 mm in slab unit 8
b Measured weight of slab

w,i
238,9 238,9
units = 4,62 kN/m
33,5 33,5 38,0 73,4

b w,i = 256,0

characteristics.
2.6
Temporary No
supports
646
2.7
Loading 4800 (span of beam) WEST
B
arrangements
B
4 3 2 1
8000
300 300 600 600 600 600 300 300

300 300 300 300 40
P2 P
2
1250 In-situ concrete K30
1500
(from slab
end)
E
C D
305 P1 P1 4800 (span of beam) P1 P
1
305
SOUTH E C D NORTH
P2 P2
A 8 7 6 5 A
4800 (span of beam) EAST
Fig. 16. Plan. P1 =Pa1 and P2 =Pa2 refer to vertical actuator forces.
A A
Det 1
P P1 2 P2 2P2 P1 P1
1
647
Det 1 F F
P1 Steel plates
F
Teflon
plates
Gypsum
F
550
3 Measurements
3.1
Support
reactions
1
3 load cells
North
648
3.2
Vertical 11
displacement 15 14 13 12
50 50 I 50 50 50
30 30 30 30 30
4 3 2 1
4500
30 30 30 30 30
II
20 19 18 17 16
Inductive transducer for measuring vertical displacement

- Transducers 32, 33, 34 on the continued centrelines of the
longitudinal joints between the slabs
- Transverse measures given with respect to the centrelines of
the joints between the slabs
30 30 30 30 30
III
25 24 23 22 21
1400
30 30 30 30 30
30 20 20 29 28 27 26
IV 20 20 20
305 610
35 34 V 33 32 31
305
VI
40 39 38 37 36
8 7 6 5
1400
45 44 VII 43 42 41
2100
50 49 48 47 46
VIII
4500
55 54 IX 53 52 51
Fig. 20. Location and numbering of transducers 11–55 for measuring vertical
displacement at lines I–IX.
649
3.3
65 (75) 64 (74) L
440 67 (77) 66 (76)

440
L
69 (79) 68 (78)
8 7 6 5
L L
68 66 64 20
74
78 76 20 40 20
Fig. 21. Transducers measuring average strain in beam’s direction. Numbers in

parentheses refer to transducers below the floor.
Fig. 22. Apparatus for measuring average strain. L =1100mm.
3.4
displacements
Measures differential displacement between slab edge and beam in beam's direction
58 4 3 2 1 56
(59) (57)
20
7 6 5
130
40 40 40
South Middle beam North
10 9 130 8
20 H H
62 8 7 6 5 60
(63) (61)
650
J J
H H
L-profile, glued to slab 60 J 60 56

20
30 (from slab end)
61
20
J
61 57
Fig. 24. Section H-H, see Fig. 24. Fig. 25. Section J-J, see Fig. 25.
3.5
Strain -
-
5 Loading strategy
5.1
relationship
Before loading all measuring devices were zero-balanced. The loading history is shown
in Fig. 27.
Stage I Stage II
200
160
120
Pai [kN]
80
Pa1
Pa2
40
0
0 40 80 120 160 200
Time [min]
651
700
600
Rp,obs [kN] 500
400
300
200
100
0
0 100 200 300 400 500 600 700 800
2(Pa1+Pa2) [kN]
Fig. 27. Development of support reaction R below Northern end of middle beam
as function of actuator loads 2(Pa1 + Pa2) on half floor.
5.2
After failure -
Before test The soffit of the slabs next to the middle beam was visually inspected
and some cracks along the strands were observed, see Fig. 29.
Stage I The joint concrete at the middle beam gradually cracked along the ends
of slabs 5–8. The first visible cracks were observed at Pa2 = 38 kN. On
the opposite (Western) side of the middle beam, a similar crack was
observed at Pa2 = 70 kN. The first flexural cracks in the tie beams
above the end beams appeared at Pa2 = 61 kN. Their number and
length increased with increasing load, see Figs 29 and 30.
Stage II At Pa2 = 70 kN, the soffit of the slabs between the line loads and the
middle beam was again inspected visually. The observed visible cracks
are shown in Fig. 29. Most of them were below the webs with four
strands, i.e. below the strands, but none below the midmost or
outermost webs where the number of strands was only three or one per
web, respectively, see. Fig. 14.
With increasing load new inclined cracks gradually appeared in the

cast-in-situ concrete at the outermost edges of the slabs next to the
ends of the middle beam, see Fig. 30. Before failure there were such
cracks in the corners of slabs 1, 4, 5 and 8. In slab 8 the cracks were
particularly wide before failure as can be seen in App. B, Fig. 23.
The tie beams at the ends of the floor also cracked in flexure. The first
flexural cracks appeared in the middle of the beams at Pa2 = 61 kN, i.e.
during the cyclic load stage. Thereafter, the Eastern tie beam proved to
be much stronger against cracking than the Western one. The Eastern
tie beam was uncracked next to the joint between slabs 7 and 8 until
failure. No reason for this nonsymmetrical behaviour could be detected.
652
The failure mode was web shear failure in slabs 8 and 7, see Fig. 30
and App. B, Figs 26, 27, 36 and 39–42.
After failure When demolishing the test specimen it was observed that the core
fillings were perfect.
The slabs were placed on the ledge of the middle beam without any
intermediate material. Due to the curved soffit of the slabs and uneven
top surface of the ledge, small gaps remained somewhere between the
soffit of the slab units and the underlying ledge of the middle beam, see
e.g. App. B, Fig. 17. After demolishing the floor it was observed that
cement paste or grout had gone into these gaps thus making the
supporting ledge more even and able to distribute the support reaction
to all webs. This is illustrated in App. B, Figs 43 and 44.
When demolishing the floor it was observed that at the middle beam the
cast-in-situ concrete had cracked vertically along the slab ends. This is
illustrated in App. B, Figs 40–42.
7.1
61
Cracks at
service load
4 3 2 1
70 70 70
70 70
70 70 70
0 70 0 70 0
0 0 0
70
70
38 38
70
0 70 0
70 0 70
70
8 7 6 5
0 0 00 0 0 0 0 0 0
61 71
Fig. 28. Cracks at service load (Pai ≈ 70 kN). The initial cracks are indicated with red
colour. A dashed line indicates a crack in the soffit. The figures give the value of the
actuator force at which the crack was observed.
653
7.2
145 125130
Cracks after 120
failure 61 130
4 3 2 1
70 70 70
70 70
175 70 70 70 175
0 70 0 70 0
170 0 0 0 170
70
185
100 70
105 38 38 80
155 100
145 140 145
185
175 145 70
70 0 165
155 0 0 70 175
165 70
115
70
8 7 6 5
0 0 00 0 0 0 0 0 0
190 177
61
190 71 80 185
140
Fig. 29. Cracking pattern after the failure. The dashed lines illustrate the cracks in the
soffit.
The ratio (measured support reaction below one end of the middle beam)/ (theoretical
support reaction due to actuator forces on half floor) is shown in Fig. 31. The theoretical
reaction is calculated assuming simply supported slabs. This comparison shows that the
support reaction due to the actuator forces can be calculated accurately enough
assuming simply supported slabs. However, the failure of the slab ends at the North end
of the middle beam resulted in reduction of support reaction below that end while the
actuator force could still slightly be increased. The maximum support reaction is
regarded as the indicator of failure.
654
1,6
1,4
Rp,obs / Rp,th 1,2

1,0
0,8
0,6
0,4
0,2
0,0
0 100 200 300 400 500 600 700 800
2(Pa1+Pa2) [kN]
support reaction (Rp,th) vs. actuator forces on half floor.
The theoretical support reaction Rp,th is calculated from four actuator loads assuming
simply supported slabs. Thus Rp,th is equal to (7900–1200)/7900x4xPam = 0,8481x4xPam.
where Pam = (Pa1 + Pa2)/2. Before failure the assumption of simply supported slabs is
accurate enough to justify the calculation of the experimental shear resistance based on it.
Vobs = Vg,sl + Vg,jc + Veq + Vp (1)
joint concrete, weight of loading equipment and actuator forces Pai, respectively. All
supported beams. For Veq and Vp this means that Veq = 0,8481xPe and
Vp = 0,8481x(Pa1.+ Pa2)/2. Vg,jc is calculated from the nominal geometry of the joints and
density of the concrete, other components of the shear force are calculated from
measured loads and weights. The values for the components of the shear force are
given in Table 1.
Shear force
Action Load
kN
Weight of joint
0,24 kN/m 0,95
concrete
one slab unit. The shear force per unit width is vobs = 152,8 kN/m
655
9.1
MPa MPa
A-Beam
- End plates  460 Nominal Raex 460
- Other structural steel  355 Nominal (S355J2G3)
End beams  355 (HEA 120)
9.2
Strength of
slab concrete, h
# Cores h d Date of Note
floor test mm mm test
d
6 50 50 24.11.2006 Upper flange of slab 7 (3 pc.)
and slab 8 (3 pc.), vertically
1)
Mean strength [MPa] 74,7 (+7 d) drilled, tested as drilled2)
9.3
Strength of
slab concrete, h
d
St.deviation [MPa] 3,96 drilled2), density =2416 kg/m3
9.4
Strength of a
longitudinal mm
joints of slab a
9.5
Strength of
concrete h
inside the mm mm
d
A-beam
6 50 50 24.11.2006 Vertically drilled through the
Mean strength [MPa] 45,6 (+7 d)1) holes of the top flange, tested
St.deviation [MPa] 5,24 as drilled2) density=2260 kg/m3
1)
2)
656
In the following figures, VP stands for the shear force of one slab end due to imposed
10.1
Deflections
180 180
160 160
140 140
120 120
Vp [kN]
Vp [kN]
100 100
11 16
80 80
12 17
60 13 60 18
40 14 40 19
20 15 20 20
0 0
0 5 10 15 20 25 0 5 10 15 20 25 30 35
180 180
160 160
140 140
120 120
Vp [kN]
Vp [kN]
100 100
21 26
80 80
22 27
60 23 60 28
40 24 40 29
20 25 20 30
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Fig. 33. Deflection on line III close to the Fig. 34. Deflection on line IV next to the
line load, slabs 1–4. middle beam, slabs 1–4.
657
180 180
160 160
140 140
120 120
Vp [kN]
Vp [kN]
100 100
31 36
80 80
32 37
60 33 60 38
40 34 40 39
20 35 20 40
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Fig. 35. Deflection on line V along the Fig. 36. Deflection on line VI next to the
middle beam. middle beam, slabs 5–8.
180 180
160 160
140 140
120 120
Vp [kN]
Vp [kN]
100 100
41 46
80 80
42 47
60 43 60 48
40 44 40 49
20 45 20 50
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Fig. 37. Deflection on line VII next to the Fig. 38. Deflection on line VIII in the
line loads, slabs 5–8. middle of slabs 5–8. Transducer 46 gave
erroneous results.
180 180
160 160
140 140
120 120
Vp [kN]
Vp [kN]
100 100
51
80 80 13 end
52
60 53 60 33 middle
40 54 40 53 end
20 55 20
0 0
0 5 10 15 20 0 5 10 15 20 25
Fig. 39. Deflection on line IX along Fig. 40. Net deflection of midpoint of middle
Eastern end beam, slabs 5–8. beam (33) and those of end beams (13, 53).
658
10.2
Crack width
180
160
140
120
5
Vp [kN]
100
6
80
7
60 8
40 9
20 10
0
0,0 0,5 1,0 1,5 2,0 2,5 3,0
Fig. 41. Differential displacement parallel to the slab between the middle beam
and the slabs (≈ crack width).
10.3
Average strain
(actually 180 180
differential 160 160
displacement) 140 140
120 120
64 74
Vp [kN]
Vp [kN]
100 100
65 75
80 80
66 beam 76 beam
60 60
67 beam 77 beam
68 40 40 78
69 20 20 79
0 0
-2,0 -1,5 -1,0 -0,5 0,0 0,0 0,5 1,0 1,5 2,0 2,5
Fig. 42. Differential displacement at top surface Fig. 43. Differential displacement at soffit
of floor measured by transducers 64–69. of floor measured by transducers 74–79.
The differential displacement divided by the mutual distance of the two fixing points
represents the average strain between those points. Assuming linearly changing strain
in vertical direction, the average strain at any depth between the measured top and
bottom values can be calculated from the measured results. The average strain at the
top and bottom fibre of the slabs and the middle beam obtained in this way are shown in
Figs 45–46.
Assuming parabolic curvature due to the uniformly distributed load on the beam, and
having the span of the beam equal to 4,8 m, the maximum strain within each of the
slabs 2, 3, 6 and 7 is 9% higher than the average strain measured between the edges of
the slab. In this way, the maximum strain is obtained from the average strains by
multiplying the latter by 1,09.
659
180 180
160 160
140 140
Vp [kN] 120 120
Vp [kN]
Slab 2 100 100 Slab 2
Slab 3 Slab 3
80 80
66 beam 76 beam
60 60
67 beam 77 beam
Slab 6 40 40 Slab 6
Slab 7 20 20 Slab 7
0 0
-0,0015 -0,0010 -0,0005 0,0000 0,0000 0,0005 0,0010 0,0015 0,0020
Average strain Average strain
Fig. 44. Top fibre of floor. Average Fig. 45. Bottom fibre of floor. Average
strain calculated from the differential strain calculated from the differential
displacements shown in Figs 41–42. displacements shown in Figs 41–42.
The variation of the average strain with the depth is illustrated in Figs 47 and 48 at the
evaluated service load (Vp = 57 kN) and failure load, respectively. The mean of the
average strain both for the slabs and for the beam is shown in Figs 49 and 50 at two
load levels.
64-74 slab 2 64-74 slab 2

350 350
65-75 slab 3 65-75 slab 3
66-76 beam 300 66-76 beam 300
67-77 beam 250 67-77 beam 250
68-78 slab 6 68-78 slab 6
Depth [mm]
Depth [mm]
200 200
69-79 slab 7 69-79 slab 7
150 150
100 100
50 50
0 0
600 400 200 0 -200 -400 2000 1000 0 -1000 -2000
-50 -50
Strain [10-6] Strain [10-6]
Fig. 46. Average strain of slabs 2, 3, 6 and Fig. 47. Average strain of slabs 2, 3, 6
7 as well as that of the middle beam, all in and 7 as well as that of the middle beam,
beam’s direction, at estimated service load all in beam’s direction, before failure at
(Vp = 57 kN). Vp = 161 kN.
660
350 350
Slabs 300 Slabs 300

Beam 250 Beam 250
Depth [mm]
Depth [mm]
200 200
150 150
100 100
50 50
0 0
600 400 200 0 -200 -400 2000 1000 0 -1000 -2000
-50 -50
Strain [10-6] Strain [10 ]
-6
Fig. 48. Mean of average strains measured Fig. 49. Mean of average strains measured
for slabs 2, 3, 6 and 7 as well as mean of for slabs 2, 3, 6 and 7 as well as mean of
average strains measured for the middle average strains measured for the middle
beam at estimated service load (Vp = 57 kN). beam before failure at Vp = 161 kN.
Both at service load and at failure, the strains in the top of the slabs were essentially
smaller than those in the beam. This can be explained by the low transverse shear
stiffness of the slabs. The scatter of the slab strains on the top was small.
At service load, the strains in soffit of the slabs were on average of the same order as
those in the beam at the same level. The friction force between the soffit of the slabs
and the bottom flange of the beam was strong enough to make the strains compatible.
This can clearly be seen in Fig. 49.
At failure load, the strain in the soffit of the slabs was greater than that in the beam.
This may be explained by the weak interaction between the beam and the slabs and
longitudinal cracking of the slabs, the first signs of which were observed already before
loading and at service load, see Figs 29 and 35.
10.4
Shear 180 180
140 56 slab 1 140
120 57 slab 1 120
58 slab 4
Vp [kN]
Vp [kN]
100 100 56-57 slab 1

59 slab 4 58-59 slab 4
80 80
60 slab 5 60-61 slab 5
60 60
61 slab 5 62-63 slab 8
40 62 slab 8 40
20 63 slab 8 20
0 0
0,0 0,5 1,0 1,5 2,0 2,5 0,0 0,5 1,0 1,5 2,0 2,5
Differential displacement [mm] Shear displacement [mm]
Fig. 50. Differential displacement between Fig. 51. Transverse shear displacement
edge of slab and middle beam. A positive of slab end calculated from displacements
value means that the slab is moving towards shown in Fig. 51.
the end of the beam. At VP = 140 kN the
end of slab 8 broke into pieces.
661
11 Reference tests
S S S
P
1200 P
R R 300 300
50
3 (4)
5 (6)
1 (2) Gypsum
Wood fibre
plate
Steel plate
50 S
50
L
R R
2 4 6 50
600
600
1 50
3 5
Fig. 52. Layout of reference test. For L see Table 2.
350 350
300 300
250 250
Pa [kN]
200
Pa [kN]
200
150 150
100 100
50 50
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time [min]
Time [min]
Fig. 53. Test R1. Actuator force – time Fig. 54. Test R2. Actuator force – time
relationship. relationship.
662
350 350
300 300
250 250
Pa [kN]
Pa [kN] 200 200 1
1
2 150 2
150
3 3
100 100 4
4
5 50 5
50
6 6
0 0
0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12
Fig. 55. Test R1. Displacements measured Fig. 56. Test R2. Displacements
by transducers 1–6. measured by transducers 1–6.
Table 2. Reference tests. Span of slab, shear force Vg at support due to the self weight
of the slab, actuator force Pa at failure, weight of loading equipment Peq, total shear force
Test Date Span Vg Pa Peq Vobs vobs Note

mm kN kN kN kN kN/m
R2 24.11.2006 6800 15,4 329,5 0,66 289,7 241,4 Flexural shear failure
Mean 283,9 236,6
13 Discussion
(deflection minus settlement of supports) was 20,9 mm or L/230, i.e. rather small.
It was 2,1–5,4 mm greater than that of the end beams. Hence, the torsional
stresses due to the different deflection of the middle beam and end beams had a
negligible effect on the failure of the slabs.
2292, Espoo 2005.
3. The edge slabs slided 0,4 … 0,5 mm along the beam before failure. This reduced
the negative effects of the transverse actions in the slab and had a positive effect
on the shear resistance.
4. The transverse shear deformation of the edge slabs was considerable which can
be seen in Figs 51–52.
5. The failure mode was web shear failure of edge slabs. The A-beam seemed to
recover completely after the failure.
663
APPENDIX A: A-BEAM
Fig. 1. Plan and section B-B.
664
Fig. 2. Sections A-A and C-C and list of steel parts.
A665
Fig. 1. A-beam. Fig. 2. A-beam. Suspension bars temporarily

stored on the top of the beam.
Fig. 3. Straight tie bar and bent suspension bar in Fig. 4. Detail of the previous figure.
their final position.
Fig. 5. Hollow core slabs temporarily supported on Fig. 6. Middle beam before grouting of joints.
end beam.
B666
Fig. 7. End of floor before grouting. Fig. 8. Test floor before grouting. Note the wedges
in the joints to facilitate the demolishing of the floor
after the test. To eliminate the contact between the
floor and the loading frame, the outermost webs of
the slabs were made thinner at the legs of the frame.
Fig. 9. Measuring devices at the end of the middle beam. Fig. 10. Measuring devices at the end of the middle beam.
Fig. 11. View on the loading arrangements. Fig. 12. Loading on outermost slabs.
B667
Fig. 13. Device for measuring strain parallel to the Fig. 14. Device for measuring crack width between
beams. slab end and middle beam.
Fig. 15. Arrangements at end beam. Fig. 16. A general view on test arrangements.
Fig. 17. Gap between soffit of slab 1 and bottom Fig. 18. Good contact between soffit of slab 4 and
flange of middle beam. bottom flange of middle beam.
B668
Fig. 19. Good contact between soffit of slab 5 and Fig. 20. Good contact between soffit of slab 8 and
bottom flange of middle beam. bottom flange of middle beam.
Fig. 21. Service load (Pa2 = 70 kN). Cracks below Fig. 22. Service load (Pa2 = 70 kN). Cracks below
soffit of slabs 2 and 3. soffit of slabs 6 and 7.
Fig. 23. Pa2 = 155 kN. Wide cracks in slab 8. Fig. 24. Cracks in slab 1 after failure.
B669
Fig. 25. Cracks in slab 5 after failure. Fig. 26. Cracks in slab 8 after failure.
Fig. 27. Cracks in slab 8 after failure. Fig. 28. Cracks in Western tie beam after failure.
Slabs 1 and 2.
Fig. 29. Cracks in Western tie beam after failure. Fig. 30. Cracks in Western tie beam after failure.
Slabs 2 and 3. Slabs 3 and 4. The red colour and letter A refer to
an initial crack.
B670
Fig. 31. Cracks in Eastern tie beam after failure. Fig. 32. Cracks in eastern tie beam after failure.
Slabs 5 and 6. The red colour and letter A refer to Slabs 5, 6 and 7. The red colour and letter A refer
an initial crack. to an initial crack.
Fig. 33. Cracks in Eastern tie beam after failure. Fig. 34. Cracks in Eastern tie beam after failure.
Slabs 6, 7 and 8. The red colour and letter A refer Slabs 7 and 8. The red colour and letter A refer
to an initial crack. to an initial crack.
Fig. 35. Cracks in slab 4 after failure. The red Fig. 36. Cracks in slab 8 after failure.
colour and letter A refer to an initial crack.
B671
Fig. 37. Cracks in slab 1 after failure. The red Fig. 38. Cracks in slab 5 after failure. The red
colour and letter A refer to an initial crack. colour and letter A refer to an initial crack.
Fig. 39. Failed ends of slabs 7 and 8. Fig. 40. Failed ends of slabs 7 and 8.
Fig. 41. Longitudinal cracking along joint between Fig. 42. End of slab 8 after removal of the top
slabs 6 and 7. flange.
B672
Fig. 43. A-beam after failure. The reinforcing bars Fig. 44. A-beam after failure. Note the cement
have been cut after the test. Note the cement paste paste which has partly filled the gap between the
which has partly filled the gap between the soffit soffit of slab 1 and the ledge of the beam.
of slab 1 and the ledge of the beam.
Fig. 45. Arrangements in reference tests. Fig. 46. Reference test R1. Northern side of slab
after failure.
Fig. 47. Reference test R1. Southern side of slab Fig. 48. Reference test R2. Northern side of slab
after failure. after failure.
B673
Fig. 49. Reference test R2. Southern side of slab

after failure.
B674
VTT CREATES BUSINESS FROM TECHNOLOGY
Technology and market foresight • Strategic research • Product and service development • IPR and licensing
• Assessments, testing, inspection, certification • Technology and innovation management • Technology partnership
• • • VTT WORKING PAPERS 148

Arrangements and results of 20 full-scale load tests on floors, each
made of eight to twelve prestressed hollow core slabs and three beams,
are presented. The tests have been carried out by VTT Technical
Research Centre of Finland and Tampere University of Technology.
PRESTRESSEd HOLLOW CORE SLABS SUPPORTEd ON BEAMS
ISBN 978-951-38-7495-7 (URL: http://www.vtt.fi/publications/index.jsp)

ISSN 1459-7683 (URL: http://www.vtt.fi/publications/index.jsp)

Prestressed Hollow Core Slabs Supported On Beams: Finnish Shear Tests On Floors in 1990-2006

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Prestressed Hollow Core Slabs Supported On Beams: Finnish Shear Tests On Floors in 1990-2006

Hochgeladen von

Copyright:

Verfügbare Formate

VTT WORKING PAPERS 148

Prestressed hollow core slabs

JULKAISIJA – UTGIVARE – PUBLISHER

VTT, Vuorimiehentie 5, PL 1000, 02044 VTT

VTT, Bergsmansvägen 5, PB 1000, 02044 VTT

VTT Working Papers 148

3. Shear tests on floors ................................................................................................. 9

VTT.PC.InvT.400.1992 ....................................................................................................... 101

VTT.S.WQ.400.1992 ........................................................................................................... 133

VTT.PC.InvT-Topp.265.1993 .............................................................................................. 199

TUT.CR.MEK.265.1994 ...................................................................................................... 293

VTT.RC.Rect-norm.265.1994 ............................................................................................. 317

VTT.CP.LBL.320.1998 ........................................................................................................ 347

TUT.CP.LB.320.2002 .......................................................................................................... 443

VTT.S.WQ.500.2005 ........................................................................................................... 467

VTT.PC.InvT.500.2006 ....................................................................................................... 525

VTT.PC.InvT.400.2006 ....................................................................................................... 613

VTT Technical Research Centre of Finland

DE265 (Internal identification)

1.3 VTT/FI 6.9.1990

2 Test specimen and loading

Fig. 2. View on test arrangements.

Fig. 4. Plan and section A-A.

Fig. 5. End beam.

The concrete was cast by VTT in laboratory, 24.8.1990.

Structural steel: Raex, see 9.1

Holes  150 315 300 300

Fig. 7. Middle beam. Position of holes in the webs. South end.

Fig. 10. Nominal geometry of slab units.

- Extruded by Partek Oy, Nastola factory, grade of concrete K60

600 600 600 600 600 600 600 600

Fig. 11. Loading arrangement with three layers of spreader beams.

Fig. 12. Arrangements at South end of Delta beam.

Stage I At P = 38,8 kN longitudinal cracks appeared in the joint concrete

Fig. 16. Cracks in the soffit after failure.

Fig. 17. Slab 1 after failure.

Fig. 19. Slab 4 after failure.

Fig. 21. Slab 8 after failure.

Fig. 22. Soffit of slabs 1–4 after failure.

8 Observed shear resistance

From the measured density of the concrete and nominal geometry

Vg,sl + Vg,jc = 12,3 + 0,4 kN = 12,7 kN

10 Measured displacements and strains

Fig. 25. Deflection of middle beam measured by transducers 12–17.

Fig. 26. Deflection of slabs 1–4 measured by transducers 2–6.

Fig. 27. Deflection of slabs 1–4 measured by transducers 7–11.

Fig. 28. Deflection of slabs 5–8 measured by transducers 19–22.

Fig. 29. Deflection of slabs 5–8 measured by transducers 24–28.

Fig. 30. Deflection along longitudinal line of symmetry measured by transducers 1, 4, 9,

Fig. 35. Strain parallel to the beam

Fig. 36. Strain parallel to the beam

Span Vg Pa+Peq Va+eq Vobs vobs

Mean 283,9 236,6

12 Comparison: floor test vs. reference tests

6. The beams did not yield in the floor test.

Fig. 1. Illustration of test setup.

1.3 VTT/FI 11.10.1990

2 Test specimen and loading

Fig. 2. Overview on arrangements.

T8: Hot rolled, weldable rebar A500HW,  = 8 mm

Fig. 4. End beam.

- Simply supported, span = 5,0 m

Structural steel: Fe 52C, fy  350 MPa (nominal fy)