Ugan

the testing with its surface covered with a wet cloth.
;-2.a--r
@
('l
LLis0mm_J
~80mm~
Fig. l. Shape and loading method of the specimen

2.1.3 Experimental parameters and the number of test specimens
Experiments are conducted with parameters fixed at fJ values of the Wedge
Theory, with 2a value in Fig.1 varied. The identification of each
(called "Series"), /3, 2a, S value (loading area), and the number of
specimens ("Numbers") are shown in Table 1 below. The number of
specimens were fixed within the range of 38 to 40, which would
a
narrow variation of the t-value. 6 pieces of specimens for compression
testing were prepared per each lot, with their dimensional size 100 X 200H.
Table 1. Experimental parameter
Series
A
B
c
D
E
F
G
H
I
{J=H/2a
20.00
14.30
10.00
8.00
5.00
4.00
3.45
2.94
2.50
2.00
80
S cm2
8.0
11.2
16.0
20.0
32.0
40.0
46.4
54.4
64.0
100
80.0
2a mm
10
14
20
25
40
50
58
68
327
Numbers
40
40
40
40
40
38
38
38
38
38
2.1.4 Loading test

We conducted this experiment for Series A through J per each lot from April
1993 until December 1994, in which 40 specimens were tested for each testing
Series A through E, and 38 for Series F through J. The standard specimen age
was determined to be 28 days. Compression testing was conducted
to
the loading experiment, so that the compression strength of the specimen in
each lot was measured.
2.2 Experiment results
2.2.1 Wedge failure and compressive slip failure
Prior to the analysis of the testing results, the failure types for wedge failure
and compressive slip failure will be explained as different loading conditions
different failure types. The Wedge Theory is quoted again . from
Suenaga and Ishimaru (1974a) as below, which means that the solution of
compressive strength "p" in a compression test is given in Eq.1 under
the parameter t=cos a .
p-
At 2 +Bt+C
(1)
t-
In this equation,
A= T 0-Ft
f3
C= r 0 + f3Ft
To-
v1+
(2)
(3)
(4)
-
Fe
=tan
/3 =!!_
a
which gives a minimum value of Eq.l, will

and as
within the range of t>
dt
(5)
Such a geometrical condition that gives a wedge failure is;
/3 ~tO
(6)
In the range of Eq.6, wedge-typed failure occurs, and partial compressive

is to be determined as below, with Eq.5 substituted for Eq.l
C'ti",an .....-r-n
+2V(A) 2 +AB+AC
(7)
the f3 value is also determined as below, with Eq.5 substituted for
328
K+vfK 2 +4A 2
f3 ~
2A
{8)
In this equation,
K=2A + 2 Ft+Ft
{9)
The minimum /3 value which causes wedge failures will be calculated with
and Ftt as parameters, as shown below in Table 2.
Table 2. Limit values of /3 {h/ a)
deg.
37
0.04
2.24
2.36
2.50
2.66
2.83
3.03
3.27
3.54
3.88
4.30
4.85
39
41
43
45
47
49
51
53
55
57
0.05
2.31
2.45
2.60
2.77
2.97
3.21
3.49
3.83
4.26
4.83
5.63
Ftt (Ft/Fe)
0.06
2.39
2.54
2.71
2.90
3.14
3.41
3.75
4.18
4.75
5.57
6.86
0.07
2.47
2.64
2.83
3.05
3.32
3.65
4.07
4.64
5.44
6.70
9.04
0.08
2.56
2.75
2.96
3.22
3.54
3.95
4.48
5.24
6.42
8.58
14.00
0.09
2.66
2.87
3.12
3.42
3.80
4.30
5.01
6.08
7.91
12.30
35.00
Secondly, in the case of tO< f3, the failure types would be the compressive
failures shown in Fig.3, as was previously mentioned in Asakawa et al.
{1994), and the p value is calculated in Eq.10 as shown below;
T 0 cos </J
e-
p- cos sin(
)
The denominator of
becomes the highest value (1-sin )/2 in the case
of 2 () - =90 , where the lowest value of p will be as follows based on
Mohl-Coulomb's failure criterion;
2 To COS efJ
Pmm
1 . ,,.r..
Fe
-sm 'P
And the slip angle under this condition will be given by;
() =45
+_!/!__
(12)
329
w~
p
y
~
p
!T
llLl
'l
(a-1)
(a-2)
.J
~4
2
12
16
20
(3
(b)
Fig.2. Mode of fracture in

partial compression according
to "Wedge Theory" and stress
Fig.3. Failure mechanism in pure

compression of a rectangular disk.
(a-1,2) slip failure line at
relations along fracture lines.
tan-\h/ a)<
(b) p/Fc- f3 relation curve
The relation between p /Fe and f3 ( f3 =l.5 through 20) under the condition of
=55 and Ftt=0.5 is described in Fig.3, for which the calculation of Eq.5 is
employed in the case of f3 ~ f3 o, and Eq.10 in the case of f3 < f3 o. In Eq.10,
the p /Fe value increases drastically when /3 falls below 2.5, however, any
phenomenon which represents this kind of data could not be detected in the
experiments thus far. Additionally, Noguchi and Tomosawa (1995) confirm
that the value shows a gradual increase while /3 stays between 2 and 1.5. If
e,
tan-1 fJ becomes smaller than a slip angle

which is to be calculated based
on Eq.12, the
angle becomes constant ( 8 ), and p/Fc=l is obtained as
shown in
based on the assumption of introducing a slip curve as shown
in Fig.3 (a-1). The relation as explained above is kept unchanged until it
reaches the f3 value as shown in Fig.3 (a-2), and in case of a lower f3 value,
the load amount
for the failure increases and p /Fe value is
considered to increase as the slip lines on both sides do not exist in a certain
portion of the
cut area. In the case of =55 and Ftt=0.05, p
=1.59 is found in
Therefore, in the analysis of actual values obtained in
the experiment,
should be employed in the case of f3 ~ f3 o, Eq.10 in the
case of f3 <po, and in the case of tan-1 f3 <
calculating the theoretical values.
330
e, p/Fc=l should be employed for
2.2.2 Experimentresults
The 10 Series of compression tests were conducted on 390 pieces of
specimens, which were taken from 26 lots.
For the analysis of the
compression the test results, the means of the test values of four pieces out of
six in total were adopted, with the maximum and minimum values
neglected. The results of the compression test are shown in Fig.4(a) through
Fig.4(e),. with the values of p/Fc indicated in the vertical axis and f3 in the
horizontal axis respectively. The dots on each figure indicate the distribution
of test results. Thicker lines represent the highest and lowest values of the
confidence interval around the mean of experiment results, which was
calculated by the t-test with its confidence coefficient at 95%. Thinner lines
indicate the calculated values based on the theoretical equation as mentioned
before. The combination of parameters, which were employed for the
comparison with the test results are as follows; internal friction angle; 11
different values by the step of 2 between 37 and 57 . Tensile strength
ratio; 9 different values by the step 0.01 between Ftt=0.01 and 0.09 for each
above, which results in 99 combinations in total. Of all
only one
combination of =55 and Ftt=0.05 proved to make all its f3 values above
the 95% of confidence coefficient in the t-test of experimental valu~s. In this
case, the F-test conducted on the experimental values, which were compared
with theoretical values, showed a significant difference narrower than 5%
regardless of any /3 value .
Consequently, the following remarks were obtained;
1) Estimated values of population mean are most suitable for those based on
=55 and Ftt=0.05.
2) As an internal friction angle becomes closer =57.5 , the solution
given based on the Wedge Theory, and the estimated distribution of
population mean of experimental values become closer to theoretical
solutions.
Fig.4(f) shows the comparison study of values based on =55 and Ftt=0.05
and the means of experimental values, which indicated significant
conformity.
Fig.5 (a) through Fig.5 Q) show the result of normal distribution, into which
the histogram of experimental values together with the means of
experimental values and their distribution were calculated, with the relative
frequency employed as the horizontal axis.
331
Ftt=O. 08
(.)
.r:: 2 f-'.,,.-'!;/~:1"'9!;.i-'--""~-:-r.-l-L~U<.-U-<-;
0.
10
20
15
/3
(a)
10
25
20
25
fj =H/2a
=H/2a
</>
15
=37deg
(b)
</>
=45deg
Ftt=O. 08
Ftt=O. 08
(.)
.r:::_ 2 f------------'.---------.----'~:>~~M0'--------------:----------------;
0.
10
(c)
20
15
/3
10
25
<f>
15
20
25
f3 =H/2a
=H/2a
=5 ldeg
(d)
</>
=55deg
i Ftt=O. OS
>-----------------<----------------------------:----------,,~-d ..~-H~N~
Ftt=O. 05
Ftt=O. 04
(.)
~ 2 ................................................ ~--------------<
0.
20
15
10
25
10
(e)
</>
15
20
fj =H/2a
fj =H/2a
=57deg
(f)
Fig.4 Experimental results
332
</>
=55deg, Ft=O. 05Fc
25
0.4
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.1
0.1
0.1
l.9
2.2
2.5
2.8
3.1
2.0
P/Fc
A series
(a)
2.6
3.2
P/Fc
to
B series
(b)
0.J
1.2 1.4 !.6 l.8 2.0 2.2 2.4
(c)
i.s
1.0 1.2 I.4 1.6
1.s
2.0 2.2
P/Fc
P/Fc
C series
(d)
D series
0.6------
0.4
0.5
0.3
0.3
0.3
0.2
0.2
0.2
0.1
0.1
0.1
0.4
0.3
0.2
0.1
i.6
o.9
1.2
1.s
1.5
o.i.7 0.8 0.9 l.O I.I 1.2 1.3 1.4
P/Fc
(e)
1.0
P/Fc
E series
F series
(f)
(g)
1.1
1.2
1.3
1.0
P/Fc
G series
(h)
I.I
l.2
1.3
P/Fc
H series
0.8-----0.5
0.6
0.4
0.3
0.4
0.2
0.1
i.1 o.8 o.9 1.0
I.I
1.2
I.3
o.i.7
0.8
0.9
P/Fc
(i)
I series
l.O
I.I
1.2
1.3
P/Fc
J series
(j)
Fig. 5 Results of normal distribution
3. Theoretical Background
For the calculation of tensile strength of concrete, split tests of concrete
cylinders are normally conducted, and our intention was to introduce our
Wedge Theory to this testing method. It is assumed that the bearing surface
2a becomes smaller, such results as to be similar to those of cylinder split
tests should be obtained in the light of the Wedge Theory.
is changed
into;
~=(-~)Ft+ 2A: +2
(A: +( Af _~)Ft+ A;
20
~t
(13)
Therefore,
lim( ~)=Ft (iS~oo)
(14)
is obtained , which means that the relation as below approximately exists

while the /3 value is rather high.
p/ {3= Ft
(15)
Eq.15 is further changed into,
/3 =H/2a, H=d, P=2ap
333
Therefore,
(16)
p/ ,B=2ap/H=P/d= Ft
is obtained.
On the other hand, as the tensile strength of concrete, obtained in the split
is;
(17)
Ft'=2P / n d
P/d is deleted from Eq.16 and Eq.17, and
(18)
= n Ft'/2Ft
In this equation, if it is assumed that Ft' is equal to Ft,
(19)
=n /2
is obtained, and for the internal friction angle
(20)
=tan-\ n /2)=57.5
This proves that the internal friction angle =55 , which was obtained in
the experiment, is very close to the theoretical solution.
4. Conclusion
As a result of the compression test conducted on plain concrete specimens,

the internal friction angle and the ratio between tensile strength and
uniaxial compressive strength Ftt of concrete (or mortar) proved to become
very close to the solution obtained based on the Wedge Theory in the case of
=55 and Ftt=0.05.
5. References
Asakawa, T., Ishimaru, R., Yoshida, H. and Suenaga, Y. (1994) Study on shear
capacity of RC columns confined with rectangular steel pipe used for the
precast concrete wall structure reducing to columns at the ground story.
J. Str. Const. Eng., AIJ. 460, 143-152
Noguchi, T. and Tomosawa, F (1995) Effect of size and shape of specimen on
mechanical properties of high strength concrete in compression. J. Str.
Const. Eng., AIJ. 473, 19-28
Suenaga, Y. and Ishimaru, R. (1974a) Dynamic analysis on concrete in
combined stress (Part 3). Trans. AIJ. 220, 1-7
Suenaga, Y. and Ishimaru, R. (1974b) Dynamic analysis on concrete in
combined stress (Part 4). Trans. AIJ. 221, 9-16
334

Ugan

Uploaded by

Copyright:

Available Formats

Ugan

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ugan

Uploaded by

Copyright:

Available Formats

the testing with its surface covered with a wet cloth.

Fig. l. Shape and loading method of the specimen

2.1.4 Loading test

which gives a minimum value of Eq.l, will

within the range of t>

Such a geometrical condition that gives a wedge failure is;

In the range of Eq.6, wedge-typed failure occurs, and partial compressive

the f3 value is also determined as below, with Eq.5 substituted for

Fig.2. Mode of fracture in

Fig.3. Failure mechanism in pure

relations along fracture lines.

tan-1 fJ becomes smaller than a slip angle

e, p/Fc=l should be employed for

Fig.4 Experimental results

=55deg, Ft=O. 05Fc

1.2 1.4 !.6 l.8 2.0 2.2 2.4

1.0 1.2 I.4 1.6

o.i.7 0.8 0.9 l.O I.I 1.2 1.3 1.4

i.1 o.8 o.9 1.0

Fig. 5 Results of normal distribution

lim( ~)=Ft (iS~oo)

is obtained , which means that the relation as below approximately exists

As a result of the compression test conducted on plain concrete specimens,

You might also like