HOLLOW UNIT BOND WRENCH
TEST TRIALS
FINAL REPORT
W. Mark McGinley Ph. D., PE
Professor, Architectural Engineering
Civil Architectural and Agricultural Engineering
Department
North Carolina A & T State University
Greensboro NC 27411
Email - [email protected]
Web site - http://www.ncat.edu/~mcginley
Phone - 336-334-7575
Fax -336-334-7126
December, 2004
1
TABLE OF CONTENTS
1.0 Introduction ............................................................................................................. 4
2.0 Testing Program...................................................................................................... 4
3.0 Test Results ........................................................................................................... 10
4.0 Discussion ............................................................................................................. 16
5.0 Conclusions........................................................................................................... 20
6.0 References ............................................................................................................. 20
2
Abstract
In an effort to determine the effectiveness of the hollow unit bond wrench test apparatus and
provide input towards improvements of this apparatus, a pilot testing program was proposed.
This testing program constructed four high, stack bonded prisms with face shell bedding using 6
or 8 x 16 concrete masonry units (Certified C 90 Block from a single source) and cured in plastic
bags in lab air. A number of two high (couplet) stack bonded prisms with face shell bedding
were also constructed using 6 or 8 x 16 concrete masonry units (Certified C 90 Block from a
single source) and cured in plastic bags in lab air. After a minimum of 28 days of bag cure, the
four high prisms were tested using the procedures in ASTM E 518. Each joint of the couplet
prisms were also tested with a proposed bond wrench testing apparatus after a minimum of 28
days of curing. These tests were repeated for a total of two unit types and three mortar types.
Based on the results of this pilot test program the following conclusions were made:
1. The flexural tensile bond strengths measured using the bond wrench testing
apparatus are lower than those measured for similar specimens tested using the
ASTM E 518 procedures. The bond wrench test values appear to be about ½
those measured by the E 518 tests
2. The bond wrench test results have slightly higher coefficients of variation than
the ASTM E 518 test procedures.
3. The 6 in. hollow concrete masonry specimens appear to produce high flexural
strengths and lower coefficients of variation that for those made with 8 in. units.
4. Much of the observed differences between the flexural bond strengths measured
by the propose bond wrench testing apparatus and the using the ASTM E 518
procedures may be due to differences produced by the fabrication/curing
procedures and specimen size, and needs further study.
5. There needs to be further investigation of the proposed bond wrench and
specimen fabrication procedures before they can be effectively used to predict
the flexural bond strength of hollow unit masonry.
3
HOLLOW UNIT BOND WRENCH TEST TRIALS
W. MARK MCGINLEY
North Carolina A & T State University
1.0 Introduction
In an effort to determine the effectiveness of the hollow unit bond wrench test apparatus and
provide input towards improvements of this apparatus, a pilot testing program was proposed.
This testing program was to:
1. Construct four high, stack bonded prisms with face shell bedding using 6 or 8 x 16
concrete masonry units (Certified ASTM C 90 Standard Specification for Loadbearing
Concrete Masonry Units) Block from a single source) and cure in plastic bags in lab air.
2. Construct a number of two high (couplet) stack bonded prisms with face shell bedding
using 6 or 8 x 16 concrete masonry units (Certified ASTM C 90 Block from a single
source) and cure in plastic bags in lab air.
3. After a minimum of 28 days of bag cure, test the four high prisms using the procedures
in ASTM E 518 Standard Test Methods for Flexural Bond Strength of Masonry.
4. Also test each joint of the couplet prisms with the proposed bond wrench testing
apparatus after a minimum of 28 days of curing.
5. Repeat the above tests for a total of two unit types and three mortar types.
2.0 Testing Program
A total of fifteen, two high, face shell bedded prisms (couplets) were fabricated for each of the
unit and mortar combinations shown in Table 1. Figures 1 and 2 show the fabrication of the
bond wrench testing prisms using mortar flows of 125 +- 5 and a jig. Five couplets were
fabricated for each mortar batch. In addition, each mortar batch was tested for flow before and
after the prisms were fabricated using the procedures shown in ASTM C 780 Standard Test
Method for Preconstruction and Construction Evaluation of Mortars for Plain and Reinforced
Unit Masonry. Three compression cubes were also fabricated for each mortar batch as per
ASTM C 780.
A total of five, face shell bedded, four high stack bonded prisms were fabricated using mortar
batched to flows of 125 +- 5 and a jig for each of the unit and mortar combinations shown in
Table 1. Two or three prisms were fabricated from a single batch of mortar and mortar flows
were taken at the start and at the end of the fabrication, as per ASTM C 780. Mortar
compression cube specimens were also formed for each mortar batch and tested for
compression as per ASTM C 780. Figure 3 shows these prisms during fabrication.
After bag curing in Lab air for 28 days each prism was tested for flexural bond strength. The
five high prisms were tested using the procedures described in ASTM E–518. The couplets
were tested using the procedures described in ASTM C 1072 Standard Test Method for
Measurement of Masonry Flexural Bond Strength and the proposed hollow unit bond wrench
testing apparatus Figures 4, 5, 6 and 7 show the prism couplet specimens being tested by the
bond wrench. Dry stacked block was used to support the specimens and the wrench arm after
failure.
4
Figures 8 and 9 show the four high stack bond prisms in the E-518 testing apparatus before and
after testing. A span of 24” was used and the top rollers were configured to create point loads at
the span third points. The test prisms were seated on the support rollers using a gypsum
capping compound that was allowed to harden before testing.
Table 1 Proposed Test Matrix.
Series
1
2
3
4
Unit Size
8
8
8
6
Mortar Type
Type N Masonry Cement
Type PCL N
Type S PCL
Type N Masonry Cement
Note PCL indicates Portland Cement Lime Mortars
Figure 1 Lower Block Buttered with Mortar
5
Figure 2 Upper block lowered on mortar and aligned with jig.
Figure 3 Four High Prism Fabrication
6
Figure 4 Couplet Being Placed in Bond Wrench
Figure 5 Upper Head Being Attached to the Couplet
7
Figure 6 Couplet Being Tested
Figure 7 Typical Bond Failure of Couplet
8
Figure 8 Prism in E 518 Testing Apparatus
Figure 9 Typical E 518 Prism Failure
9
3.0 Test Results
The results of the testing program are summarized in the following tables.
Table 2 shows the results of the mortar tests. Tables 3 and 4 show the cross-sectional
properties measured for each of the couplet and four high block prisms using the
procedures in ASTM C 140 Standard Test Methods for Sampling and Testing Concrete
Masonry Units and Related Units. In all cases, the average face shell thickness was
assumed to be constant across the width of the block and defined the minimum mortar
bedded area. The moment of inertia, area and section modulus was calculated for the
minimum net bedded mortared area for both the couplets and 4 high prisms and these
values are also summarized in Tables 3 and 4.
Table 2 Mortar Test Results
Type N Masonry Cement used for Masonry Cement Mortars
Type S lime Used
Type I cement used for PCL Mortars
Weight Weight Weight Weight
Flow
Mix
Flow
Sand
Cement
Lime
Water
Start
End
(%)
(lb)
(lb)
(lb)
(lb)
(%)
PCLS-1(C1-10)
PCLS-2(C11-15)
PCLN-1(C1-5)
PCLN-2(C6-10)
PCLN-3(C11-15)
MASN -1(C1-5)
MASN-2(C6-10)
MASN-3(C11-15)
MASN-4(6 in C1-5)
MASN-5(6 in C6-10)
MASN-6(6 inC11-15)
PCLS-3 (P1-3)
PCLS-4 (P4-5)
PCLN-4 (P1-3)
PCLN-5 (P4-5)
NMAS-7 (P1-5)
NMAS-8 (6 in. P1-5)
150
100
73.3
73.3
73.3
73.3
73.3
73.3
73.3
73.3
73.3
75
75
73.3
73.3
73.3
73.3
47
31.3
15.7
15.7
15.7
23.3
23.3
23.3
23.3
23.3
23.3
23.5
23.5
15.7
15.7
23.3
23.3
10
6.7
6.7
6.7
6.7
0
0
0
0
0
0
5
5
6.7
6.7
0
0
31.4
21.3
16.3
17.8
17.8
14.4
14.2
13
13
14.6
14.9
18.5
18
18.3
18.5
15.5
17.5
124.3
135.3
121
123.2
125.4
132
127.6
128.7
128.7
119.9
118.8
118.8
127.3
119.9
121
132
132
95.7
129
113.3
110
125.4
132
127.6
125.4
126.5
110
114.4
101.2
126.5
103.4
121
118.8
113.3
Mortar tests
Cube Compression Test
Results (psi)
A
B
C
Ave.
4909
4044
1982
1958
1844
1013
925
1132
1004
910
969
4247
3916
1651
1711
925
672
4954
4084
2096
1963
1681
1013
929
1033
974
929
920
4173
3797
1641
1760
925
811
4543
4034
2200
1785
1819
969
905
1137
1033
910
905
4281
4064
1790
1800
939
747
4802
4054
2093
1902
1781
999
920
1101
1004
916
931
4234
3925
1694
1757
929
743
Note : PCLS-1 (C 1-20) designates PCL Mortar, type S, Batch 1 was used to fabricate
Couplets 1 Through 10 (8 in. Units)
PCLN-1 (C 1-5) designates PCL Mortar, type N, Batch 1 was used to fabricate
Couplets 1 Through 5 (8 in. Units)
MASN-1 (C 1-5) designates Masonry Cement Mortar, type N, Batch 1 was used
to fabricate Couplets 1 Through 5 (8 in. Units)
MASN-1 (6in C 1-5) designates Masonry Cement Mortar, type N, Batch 1 was
used to fabricate Couplets 1 Through 5 (6 in. Units)
PCLS-3 (P1-3) - designates PCL Mortar, type S, Batch 3 was used to fabricate
4 high prisms 1 Through 3 (8 in. Units)
10
Table 3 Couplet Measured Section Properties
Couplet/Prism ID
Average
depth, d
(in)
Average
Width, b
(in)
Average
Face Shell,
FS (in)
Moment
of Inertia,
I (in4)
Section
Modulus,
S (in3)
Area, A
(in2)
8 in. PCLS C1
8 in. PCL S C2
8 in. PCL S C3
8 in. PCL S C4
8 in. PCL S C5
8 in. PCL S C6
8 in. PCL S C7
8 in. PCL S C8
8 in. PCL S C9
8 in. PCL S C10
8 in. PCL S C11
8 in. PCL S C12
8 in. PCL S C13
8 in. PCL S C14
8 in. PCL S C15
8 in. PCL N C1
8 in. PCL N C2
8 in. PCL N C3
8 in. PCL N C4
8 in. PCL N C5
8 in. PCL N C6
8 in. PCL N C7
8 in. PCL N C8
8 in. PCL N C9
8 in. PCL N C10
8 in. PCL N C11
8 in. PCL N C12
8 in. PCL N C13
8 in. PCL N C14
8 in. PCL N C15
7.63
7.63
7.62
7.63
7.63
7.62
7.62
7.63
7.63
7.62
7.63
7.63
7.63
7.64
7.63
7.62
7.63
7.63
7.61
7.62
7.63
7.63
7.59
7.63
7.63
7.62
7.62
7.59
7.61
7.62
15.59
15.57
15.59
15.59
15.60
15.63
15.59
15.58
15.60
15.59
15.61
15.59
15.59
15.59
15.59
15.60
15.59
15.59
15.57
15.55
15.59
15.59
15.56
15.59
15.57
15.56
15.60
15.59
15.60
15.59
1.27
1.26
1.28
1.27
1.28
1.27
1.29
1.27
1.28
1.28
1.28
1.28
1.28
1.28
1.27
1.29
1.28
1.28
1.28
1.27
1.29
1.27
1.27
1.28
1.29
1.28
1.27
1.28
1.29
1.29
405.50
403.26
406.69
405.43
407.84
405.41
408.40
405.25
406.86
405.27
407.62
408.75
408.03
409.44
405.10
407.97
406.68
406.08
405.32
403.91
408.73
405.25
400.57
406.76
407.92
404.78
404.02
403.19
406.75
407.37
106.36
105.77
106.78
106.34
106.97
106.45
107.23
106.29
106.72
106.41
106.92
107.10
107.03
107.17
106.15
107.12
106.67
106.51
106.53
106.05
107.21
106.30
105.50
106.69
106.99
106.28
106.08
106.19
106.91
106.96
39.65
39.34
39.99
39.64
40.00
39.74
40.27
39.63
39.85
39.78
39.96
40.01
40.04
39.97
39.45
40.18
39.83
39.75
39.97
39.61
40.16
39.61
39.54
39.86
40.06
39.74
39.56
39.90
40.14
40.10
11
Table 3 – Couplet Measured Section Properties (cont.)
Couplet/Prism ID
Average
depth, d
(in)
Average
Width, b
(in)
Average
Face Shell,
FS (in)
Moment of
Inertia, I
(in4)
Section
Modulus, S
(in3)
Area, A (in2)
8 in. MAS N C1
8 in. MAS N C2
8 in. MAS N C3
8 in. MAS N C4
8 in. MAS N C5
8 in. MAS N C6
8 in. MAS N C7
8 in. MAS N C8
8 in. MAS N C9
8 in. MAS N C10
8 in. MAS N C11
8 in. MAS N C12
8 in. MAS N C13
8 in. MAS N C14
8 in. MAS N C15
6 in. MAS N C1
6 in. MAS N C2
6 in. MAS N C3
6 in. MAS N C4
6 in. MAS N C5
6 in. MAS N C6
6 in. MAS N C7
6 in. MAS N C8
6 in. MAS N C9
6 in. MAS N C10
6 in. MAS N C11
6 in. MAS N C12
6 in. MAS N C13
6 in. MAS N C14
6 in. MAS N C15
7.62
7.61
7.62
7.60
7.63
7.63
7.63
7.63
7.63
7.61
7.63
7.63
7.63
7.63
7.63
5.63
5.63
5.63
5.65
5.63
5.62
5.63
5.63
5.63
5.63
5.63
5.63
5.63
5.63
5.63
15.59
15.59
15.59
15.63
15.59
15.59
15.56
15.58
15.61
15.57
15.56
15.56
15.57
15.59
15.58
15.63
15.64
15.67
15.68
15.66
15.67
15.66
15.65
15.64
15.63
15.68
15.69
15.63
15.66
15.63
1.28
1.27
1.27
1.27
1.29
1.28
1.27
1.28
1.27
1.27
1.27
1.27
1.30
1.28
1.28
1.06
1.06
1.07
1.06
1.07
1.06
1.07
1.07
1.06
1.08
1.08
1.06
1.07
1.09
1.07
406.02
402.69
404.87
403.71
408.31
407.13
404.84
407.60
406.34
402.38
405.37
403.89
410.81
407.21
405.88
175.75
176.22
177.02
178.14
177.65
175.84
176.97
177.31
175.67
177.91
178.61
176.69
176.36
178.65
176.43
106.61
105.84
106.30
106.22
107.10
106.79
106.19
106.91
106.58
105.76
106.33
105.94
107.75
106.81
106.46
62.49
62.66
62.94
63.07
63.08
62.61
62.92
63.04
62.46
63.17
63.50
62.82
62.70
63.52
62.73
39.90
39.52
39.72
39.79
40.09
39.92
39.59
40.00
39.76
39.51
39.68
39.45
40.52
39.93
39.73
33.15
33.28
33.49
33.26
33.53
33.30
33.50
33.62
33.10
33.65
34.01
33.36
33.35
34.05
33.36
12
Table 4 E-518 Measured E 518 Prism Section Properties
Couplet/Prism
ID
8 in. PCL S P1
8 in. PCL S P2
8 in. PCL S P3
8 in. PCL S P4
8 in. PCL S P5
8 in. PCL N P1
8 in. PCL N P2
8 in. PCL N P3
8 in. PCL N P4
8 in. PCL N P5
8 in. MAS N P1
8 in. MAS N P2
8 in. MAS N P3
8 in. MAS N P4
8 in. MAS N P5
6 in. MAS N P1
6 in. MAS N P2
6 in. MAS N P3
6 in. MAS N P4
6 in. MAS N P5
Average
depth, d
(in)
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.60
7.62
7.63
7.63
7.63
7.63
5.63
5.63
5.63
5.62
5.63
Average
Width, b
(in)
15.62
15.60
15.58
15.60
15.61
15.59
15.61
15.60
15.59
15.59
15.58
15.71
15.59
15.62
15.61
15.60
15.57
15.63
15.63
15.61
Average
Face Shell,
FS (in)
1.29
1.30
1.29
1.28
1.28
1.28
1.29
1.29
1.29
1.29
1.28
1.28
1.29
1.28
1.28
1.06
1.07
1.08
1.07
1.06
Moment of
Inertia, I
(in4)
409.17
411.16
408.07
407.11
408.29
406.66
410.46
409.56
408.78
405.71
405.99
410.47
408.51
407.92
406.79
175.62
176.40
177.35
176.41
175.88
Section
Modulus, S
(in3)
107.32
107.85
107.04
106.78
107.09
106.66
107.66
107.42
107.22
106.74
106.60
107.66
107.15
107.00
106.70
62.44
62.72
63.06
62.81
62.54
Area, A (in2)
40.19
40.53
40.07
39.89
40.06
39.84
40.40
40.27
40.16
40.15
39.90
40.25
40.12
39.99
39.83
33.14
33.45
33.68
33.56
33.22
The results of the bond wrench tests are summarized in Table 5. At failure, the mortar
joint typically separated from the block unit at either the top of the joint (T), or the bottom
of the joint (B). In a few cases, the mortar joint separated at both the top and bottom
(T/B). In addition, the mortar joints on two couplets fractured as the upper head of the
bond wrench was being attached suggesting that the weight of the upper head was
sufficient to fail the mortar/block bond. The average and coefficients of variation (COV)
for each mortar and unit configuration are also shown in Table 5. It should be noted that
the prematurely broken specimens were excluded from this analysis.
13
Table 5 – Couplet Flexural Bond Wrench Test Results
Couplet/Prism ID
8 in. S PCL C1
8 in. S PCL C2
8 in. S PCL C3
8 in. S PCL C4
8 in. S PCL C5
8 in. S PCL C6
8 in. S PCL C7
8 in. S PCL C8
8 in. S PCL C9
8 in. S PCL C10
8 in. S PCL C11
8 in. S PCL C12
8 in. S PCL C13
8 in. S PCL C14
8 in. S PCL C15
8 in. PCL N C1
8 in. PCL N C2
8 in. PCL N C3
8 in. PCL N C4
8 in. PCL N C5
8 in. PCL N C6
8 in. PCL N C7
8 in. PCL N C8
8 in. PCL N C9
8 in. PCL N C10
8 in. PCL N C11
8 in. PCL N C12
8 in. PCL N C13
8 in. PCL N C14
8 in. PCL N C15
Max. Load (lb)
372.28
264.61
139.15
344.43
226.28
150.38
131.67
27.7
0
54.36
252.9
131.03
118.82
227.15
277.76
316.27
79.15
45.7
172.49
113.14
129.3
109.49
142.18
186.46
58.84
265.73
238.95
193.75
72.72
162.35
Max. Stress (psi)
95.17
69.95
39
88.49
59.75
41.8
37.07
12.34
0
18.75
66.15
36.93
34.05
59.83
72.56
81.19
24.64
16.64
47.11
32.98
36.5
32.01
40.21
50.36
19.72
69.63
63.29
52.39
23.25
44.49
14
Break Type Ave Stress (psi)
COV (%)
T
T
B
B
T
T/B
T
B
Broke when head applied
B
T
T
T
B
T
48.79
51.57
T/B
T
T
T/B
B
B
B
T
T
T
T
T
T/B
B
T
42.29
37.94
Table 5 – Couplet Flexural Bond Test Results (cont.)
Couplet/Prism
ID
8 in. MAS N C1
8 in. MAS N C2
8 in. MAS N C3
8 in. MAS N C4
8 in. MAS N C5
8 in. MAS N C6
8 in. MAS N C7
8 in. MAS N C8
8 in. MAS N C9
8 in. MAS N C10
8 in. MAS N C11
8 in. MAS N C12
8 in. MAS N C13
8 in. MAS N C14
8 in. MAS N C15
6 in. MAS N C1
6 in. MAS N C2
6 in. MAS N C3
6 in. MAS N C4
6 in. MAS N C5
6 in. MAS N C6
6 in. MAS N C7
6 in. MAS N C8
6 in. MAS N C9
6 in. MAS N C10
6 in. MAS N C11
6 in. MAS N C12
6 in. MAS N C13
6 in. MAS N C14
6 in. MAS N C15
Max. Load
(lb)
49.1
0
27.7
63.68
58.43
32.99
26.23
32.97
85.62
90.54
47
158.8
98.1
36.5
66.9
79.45
81.55
136
61.74
82.09
87.29
33.3
66.59
9.93
81.4
65.46
43.58
11.21
129.53
55.1
Max. Stress
(psi)
17.45
0
12.35
21.03
19.6
13.49
12.01
13.54
26.21
27.61
17
44
28.9
14.4
21.7
42.86
43.6
65.37
35.99
43.53
45.97
23.98
37.34
14.65
43.21
36.63
28.16
15.08
62.23
32.86
Break Type
Ave Stress (psi)
COV (%)
T
Broke when head applied
T
T
T
T
B
T
T
T
T
T
T
T/B
T
T
T
T/B
T
T
T/B
T
T/B
B
T/B
T
T/B
B
T
T
19.29
53.77
38.10
39.10
Table 6 shows the flexural tensile bond test results for each E 518 test specimen. Also
shown are the average and COV for the five identical tests for each mortar and unit
configuration evaluated during this investigation. Note that one of the test specimens
broke during placement in the testing apparatus and two of the specimens broke at
mortar joints, outside of the central maximum moment region, and were discarded in
subsequent analysis as directed by the procedures in E 518.
15
Table 6 Maximum Flexural Tensile Bond Test Results for the E 518 tests
Couplet/Prism
ID
8 in. S PCL P1
8 in. S PCL P2
8 in. S PCL P3
8 in. S PCL P4
8 in. S PCL P5
8 in. PCL N P1
8 in. PCL N P2
8 in. PCL N P3
8 in. PCL N P4
8 in. PCL N P5
8 in. MAS N P1
8 in. MAS N P2
8 in. MAS N P3
8 in. MAS N P4
8 in. MAS N P5
6 in. MAS N P1
6 in. MAS N P2
6 in. MAS N P3
6 in. MAS N P4
6 in. MAS N P5
Peak Load (lb)
Peak Stress (psi)
3339.00
128.03
4385.00
166.22
992.00
40.72
1937.45
75.69
2428.80
94.27
2546.43
99.04
2353.88
90.96
Broke on Set Up
2206.35
85.83
2321.60
90.57
952.62
39.32
1352.37
53.78
1474.69
54.59
1039.36
42.40
586.89
25.58
845.50
59.17
801.93
56.12
567.69
42.88
939.24
64.80
1006.20
69.36
Average Stress
(psi)
COV (%)
100.99
47.76
91.60
5.98
Broke outside load region
43.32
31.74
Broke outside load region
58.47
17.27
4.0 Discussion
Examination of Tables 5 and 6 result in the following observations:
1. The flexural tensile bond strengths measured using the proposed bond wrench
testing apparatus are lower than those measured for similar specimens tested
using the E 518 procedures.
2. The bond wrench test results have slightly higher coefficients of variation than
those measured for the E 518 test procedures and both sets of coefficients of
variation are relatively high for lab tests.
3. The 6 in. hollow concrete masonry units appear to produce higher flexural
strengths and lower coefficients of variation that for 8 in. units.
The difference in bond strengths measured by each of the test methods is more clearly
seen in Figure 10. For the 8 in. CMU units, the ratio of flexural bond measured by the E
518 method and the bond wrench method varies from 2.07 to 2.25. This ratio
decreases to 1.6 for the flexural bond strengths measured for the 6 in. CMU specimens.
16
Flexural Tensile Bond Strength (psi)
120.00
100.00
80.00
Bond Wrench
E 518
60.00
40.00
20.00
0.00
8" S PCL
8" N PCL
8" N Mas
6" N Mas
Mortar Type and Units size
`Figure 10 A Comparison of Measured Flexural Tensile Bond Strengths
It appears that the C1072 and E 518 test methods evaluated during this investigation do
not produce equivalent flexural tensile strength values and both methods produced
relatively high coefficients of variation.
In effort to determine why these differences occurred, a review of some previous
investigations of bond between masonry units and mortar was conducted. [Thomas,
1994], [Hamid and Hakam,1996], [Gosh, 1990], [Matthys, 1990], [Hughs and Zsembery
1980].
A report by R. Thomas [Thomas, 1994] outlined the results of an extensive investigation
conducted at the National Concrete Masonry Association (NCMA) Lab. This
investigation evalauted the relationship between the flexural strength measured by a
“bond wrench” testing apparatus and that produced by large wall tests conducted using
the procedures in ASTM E 72. In these tests it was found that ten identical (4 ft x 8 ft- 8
in. CMU – type S – PCL Mortar) ASTM E 72 wall tests produced an average flexural
bond strength of 162 psi (COV - 32%) and thirty companion, two high prisms tested
using a “bond wrench” produced an average 169 psi flexural bond strength (COV 33%). These results are quite similar and suggest that the bond wrench gives similar
results to full sized wall tests. Reasonably good agreement was also found for 4 in. and
12 in. unit specimens, although the 4” specimens appeared to give higher bond strength
results (similar to the higher strengths found for the 6 in prisms evaluated in the current
investigation). The flexural bond strengths measured in the current investigation for the
same mortar type and unit size were significantly lower that the NCMA test values, 49
psi for the bond wrench and 101 psi for the E 518 tests.
17
The NCMA investigation also evaluated a variety of specimen curing conditions and
their affect on measured bond strengths. The NCMA specimen curing conditions that
gave the good agreement described above involved spraying the face of the 24hour old
specimens with water until water was observed to flow down the face of the specimen.
Immediately after the specimen was sprayed, the bag in which the specimen was
constructed was sealed to prevent moisture from escaping. The bags were stripped two
days prior to testing. Other bond wrench tests were also conducted using lab air curing
with multiple water sprays, lab air curing (no bags) and curing in outdoor conditions.
The results of these tests showed similar results for lab air testing with multiple water
sprays and the saturated bag cure. However, the lab air cured specimens and
specimens cured outdoors produced average bond strengths of 46 psi (COV 64%) and
40 psi (COV 72%), respectively showing curing conditions can have a significant effect
on the average bond strengths. It should be noted that these results are very similar to
those observed in this investigation.
In an experimental comparison of bond strengths measured by different test methods,
Hamid and Hakam [Hamid and Hakam, 1996] showed good agreement between bond
wrench tests of grouted 8 in. concrete masonry prisms and grouted full scale walls
tested using the procedures in ASTM E 72.
Another investigation conducted by John Matthys [Matthys, 1990] at the University of
Texas at Arlington measured the flexural bond strength of full scale 4 ft x 8 ft walls
(tested using ASTM E 72 procedures), individually built stack bonded prisms, and
running bond prisms cut from the full scale walls (tested using a “bond wrench”). All
these specimens were constructed with hollow 8 x 8 x 16 units, type S PCL or type S
Masonry Cement mortar. All the stack bond prisms were 9 joints high and all the
specimens (walls and prisms) were cured in lab air. For the type S PCL mortar
specimens, the average flexural strength measured by the bond wrench on the stack
bonded prisms was 23.9 psi (COV 61.6%), 42.5 psi (COV 26.7 %) for the full sized wall
tests and 44.9 psi (COV 41 %) for the running bond prisms cut from the wall. The ratio
of stack bond prism strength to wall tests was about 1.8 for the specimens constructed
with type S PCL Mortar.
Another investigation conducted at the Construction Technology Laboratory by the
Portland Cement Association [Gosh, 1990] also looked at flexural bond strengths
measured with both clay and hollow concrete masonry units for a number of mortar
types using both ASTM C 1072 and E 518 testing procedures. This investigation
observed an average flexural strength of 91 psi for 8 in. hollow block (type S – PCL
Mortar) prisms tested according to ASTM E 518 procedures and using a “bagged” cure.
This compares well with the 101 psi measured during this investigation.
Examination of the bond wrench apparatus used in each of the various investigations
shows similar devices were used. The bond wrench used in the NCMA investigation
and bond wrench used in the current investigation are quite similar, having
approximately the same lever arm (30 in and 28.5 in, respectively) and approximately
the same percentage of axial stress to flexural stress (~11%).
18
Further a comparison of the bond wrench and the E 518 test procedures suggest that all
these tests are applying similar peak flexural stress distributions in low to zero shear
stress regimes. Theoretically, they should produce similar results even with the small
amount of axial stress present in the bond wrench tests. There appears be something
else affecting the bond on the bond wrench couplet prisms since average strengths
measured by the proposed bond wrench testing apparatus are much lower that the E
518 tests.
As was discussed previously, specimen fabrication and curing procedures can
significantly affect the measured bond. This has been found by a number of
investigations and is acknowledged by Note 4 in ASTM C 1072 which reads as follows;
”Workmanship during fabrication, temperature of the materials during fabrication, curing
conditions, time between removal from moist curing to test, and other factors may affect
the bond strengths measured by this test. Standardized specimen fabrication and curing
procedures that attempt to control these variables are prescribed in Test Methods C
1357.”
This suggests that the fabrication procedures may be affecting the measured results
and is consistent with the finding of the other investigations. In fact, examination of the
prism fabrication procedures used in this investigation with respect to those used by
others result in the following observations:
1. Both this investigation and the NCMA [Thomas, 1994] investigation used prism
couplets cured in a bag but the NCMA investigation saturated the prism couplets
24 hours after construction and then bagged the specimens. The NCMA prisms
also had tooled joints whereas the current investigation used struck joints.
2. The 48.8 psi average flexural bond measured by the proposed bond wrench
apparatus is closer to the average 46 psi measured for the air cured specimens
during the NCMA tests [Thomas, 1994] and the average 24 psi for the UT
Arlington tests [Matthys, 1990]. This suggests that just bag curing may not be
providing sufficient water to fully hydrate the cement in the mortar. Additional
moisture may need to be provided to ensure full hydration with the large hollow
CMU units. This lack of sufficient moisture may not only be the cause of the
lower the average strengths but also the increase the variability of the tests.
3. The higher strength of the prisms constructed from the smaller units may also, in
part, be a result of less moisture being drawn away from the joints when smaller
units are used.
4. Specimens with more joints appear to have higher bond strengths than those
with fewer joints with the same curing conditions [Hughs and Zsembery, 1980].
This is likely caused by the additional consolidation provided to the central joints
in the taller specimens when the upper units are added and may be one the
causes for the higher test values for the ASTM E 518 four high prism observed
during this investigation. The use of tooled joints for both of these prisms may
reduce some of this difference by consolidating all joints a certain amount.
Further study of this effect is needed.
19
5. The ASTM E 518 procedures only test the central joint mortar joint bond that may
be stronger that the rest of the joints based on the additional consolidation
discussed above.
The proposed bond wrench may be providing lower that expected flexural bond strength
results due to the actions described above. Additional testing to determine whether this
is the case should be conducted. These tests should be done using prism curing that
includes saturated units, joint tooling and flexural bond wrench specimens that are also
four units high.
5.0 Conclusions
Based on the results of this pilot test program the following conclusions can be
made:
1. The flexural tensile bond strengths measured using the bond wrench testing
apparatus are lower than those measured for similar specimens tested using the
ASTM E 518 procedures. The bond wrench test values appear to be about ½
those measured by the E 518 tests.
2. The bond wrench test results have higher coefficients of variation than the ASTM
E 518 test procedures.
3. The 6 in. hollow concrete masonry specimens appear to produce high flexural
strengths and lower coefficients of variation that for those made with 8 in. units.
4. Much of the observed differences between the flexural bond strengths measured
by the propose bond wrench testing apparatus and the using the ASTM E 518
procedures may be due to differences produced by curing procedures and
specimen size. The use of bag curing may not be providing sufficient water to
fully hydrate the cement in the mortar. Additional moisture may need to be
provided to ensure full hydration with the larger hollow CMU units. Further study
of this effect is needed.
5. The use of non tooled joints and the consolidation of lower mortar joints for large
hollow unit prisms may affecting the measured flexural bond and needs further
study.
6.0 References
1. ASTM. Annual Book of Standards, Volume 4.05, the American Society of Testing
and Materials, West Conshohocken, PA, 2004.
2. Robert D. Thomas, “Research Evaluation of the Flexural Tensile Strength of
Concrete Masonry”, Project # 93-172 Masonry Research Report, MR 10, April,
1994, The National Concrete Masonry Association, Herndon, VA.
3. John Matthys, “Concrete Masonry Flexural Bond Strength Prisms versus Wall
Tests”, Proceedings of the Fifth North Masonry Conference, University of Illinois
at Urbana-Champaign, June, 1990.
20
4. S. K. Gosh, “Flexural Bond Strength of Masonry – An Experimental Review”,
Proceedings of the Fifth North Masonry Conference, University of Illinois at
Urbana-Champaign, June, 1990.
5. A. A. Hamid and Z. H. R. Hakam, “Modulus of Rupture of Concrete Masonry
Using Full Scale Wall Tests and Bond Wrench: A Comparison Study. The
Proceedings of the 8th Canadian Masonry Symposium, Jasper, Alberta 1996.
6. D. M. Hughes and S. Zsembery, A Method of Determining the Flexural Bond
Strength of Brickwork at Right Angles to the Bed Joint”, Proceedings of the 2nd
Canadian Masonry Symposium, Carlton University, Ottawa, Canada, 1980.
21