Behavior of Shear Connected Cavity Walls
K. Papanikolas, M. Hatzinikolas, J. Warwaruk
September 1990, Department of Civil Engineering, University of Alberta
In this Chapter, the results from all four phases of the experimental program (Chapter 4) are summarized and presented in tabular, graphic and photographic form.
The results of the testing of the three identical wall segments are summarized in Table 5.1. This Table includes information relating to the fastening location (hole), the load type and the maximum (failure) load. Analytical "strain gage vs. load" curves for all the tests are given in Appendix: B.
The average capacity to upward vertical load was 4.39 kN with a standard deviation of 0.335 kN and a minimum value of 3.78 kN. For the downward shear force the average capacity was 3.26 kN, with a standard deviation of 0.466 kN and a minimum value of 2.45 kN. Finally, the average capacity of the shear connector in axial compression was 5.8 kN.
The failure mode was consistently identified as the yielding of the metal plate around the hole in which the bent rod is attached. Plate 5.1 shows the mode of failure of a connector after several tests. Since for all the tests no cracks appeared at the mortar joint where the connector is embedded it can be concluded that the shear connector is well fixed within the concrete block.
5.3. 1 General
The objectives of this phase of the experimental program were presented in Section 4.2.1.
The results of the twelve full-scale cavity walls are presented in two types of diagrams. The first type were pressure vs. centerline lateral deflection curves for both the brick and block wythes (see Figure 5.1 for example). The second type were deflected shapes along the wall diagrams at different pressure intervals (see Figure 5.2 for example). Unless referred to in this Chapter, all such diagrams may be found for each wall specimen in Appendix C.
Table 5.2 presents the test results for all the full-scale cavity walls subjected to positive lateral pressure. For the masonry cavity walls two sets of values for pressure and deflections are presented in that Table. The first value corresponds 10 the yield strength (which was defined by a significant change in slope) and the second value corresponds to failure (ultimate strength). The point or ultimate strength is defined as the maximum pressure resisted by a wall specimen. For the "brick -metal stud" cavity walls only the ultimate strength is presented since the end of the elastic range is not well defined.
The method used to reduce the data from the strain gauges located at the shear connectors of the specimens with large cavities (100 mm) is presented in Section 5.4 and it is shown in Figure 5.12. The results of the strain gage readings will be discussed in this and Chapter 6.
5.3.2 Cavity Wall S1W1
This wall system consisted of a 150 mm wide hollow block backup wythe, a cavity of 25 mm, and a 90 mm wide brick veneer wythe. The shear connector arrangement was type A (see Table 4.2 and Figure 4.),
The pressure versus centerline deflection and the deflected shapes at different pressures are presented in Figures C-1 and C-2 of Appendix C.
From the pressure - centerline deflection curves it can be concluded that the wall system behaved elastically up to a pressure of 0.70 kPa. The corresponding minimum deflection at that pressure was recorded at mid-height of the brick wall and was 1.66 mm. At this point the rate of deflection per load increment increased since cracks started to form at horizontal mortar joints of the block wall. The location of the first cracks was around the concrete block that was set on the top angle. This type of failure can be classified as punching shear failure. That was not the kind of failure that was expected but can be explained by the fact that the backup concrete wythe was ungrouted and unreinforced. The final failure occurred at a pressure of 0.92 kPa, when the previous crack deteriorated. The maximum deflection al failure appeared in the block wall at a height of 2200 mm and was 4.9 mm. The pressure vs. centerline deflection curve remained relatively steep throughout the test and the failure was identified by a sudden drop of the pressure. That sudden failure confirmed the brittle performance of unreinforced hollow masonry wall. Up to failure both wythes deflected the same amount at mid-height, thus composite action was present.
In order to avoid this type of failure specimen S1W3 which was also unreinforced had four out of six voids of the top two layers of the block wythe grouted and reinforced.
5.3.3 Cavity Wall S1W2
Specimen S1W2 had a 200 mm standard reinforced concrete block wall, a cavity of 100 mm, and a shear connector arrangement type A (see Table 4.2 and Figure 4.6). The wall was reinforced by placing two 10M bars at the second from the edge voids and then grouting these cores.
From Figure 5.1 it can be concluded that the cavity wall behaved elastically up to a pressure of 1.61 kPa with corresponding maximum centerline deflection of 0.82 mm. After that pressure cracks started to develop at the horizontal mortar joints of the block wall at mid height. This specimen was able to resist a pressure of 3.6 kPa with a maximum centerline deflection of 10 mm. Failure occurred in the brick wythe at mid height between the first two layers (from top) of connectors when the tensile stresses (due to secondary moments) exceeded the tensile bond strength of the mortar. Plate 5.2 shows specimen SIW2 after failure.
A ductile performance was observed between the yield strength and failure. The explanation for this is the presence of reinforcement in the concrete block wythe.
The specimen had four shear connectors instrumented with strain gages. By reducing the strain gage readings, assuming linear strains as shown in Figure 5.12, the axial force transferred through the connectors could be found. From that analysis the endmost shear connectors appeared to carry the largest internal forces. The critical shear connectors, therefore, are the ones located close to the supports. At the end of the elastic range (pressure = 1.61 kPa) the maximum axial force was recorded at the lowermost shear connector and was found to be 0.93 kN. At the failure load (3.6 kPa) the axial force at the same critical connector was 1.87 kN. This value is approximately one third of the minimum capacity of the two connectors tested in axial compression (5.78 kN) as was described in Section 5.2.
Since only the uppermost part of the brick veneer was damaged that part was removed and repaired by adding a shear connector at the mortar bed where failure occurred. The specimen now had a connector arrangement Type B (see Figure 4.6) and was re-tested as specimen S2WI.
5.3.4 Cavity Wall S1W3
Wall S1W3 was identical to S1W1 except that 150 mm standard hollow concrete blocks were used. In order to avoid the undesirable shear failure of S1W1 the top two layers of the block wythe were grouted and reinforced.
The pressure vs. centerline deflection diagrams and the deflected shapes are presented in Figures C-5 and C-6 of Appendix C. The yield strength was well defined to be 0.43 kPa with maximum centerline deflection al the block wall of 1.52 mm. After that point, the deflection increased up to 3.00 mm while the pressure dropped slightly to 0.417 kPa. At that lime, tensile cracks in the horizontal mortar joint of the block wall at mid-height opened. After reaching 0.417 kPa and 3.00 mm minimum centerline deflection, the wall system again started to carry increased pressure. The final failure occurred at a pressure of 1.00 kPa when the previous cracks deteriorated. The minimum lateral deflection at failure was recorded at middle - height of the block wythe and was 9.87 mm.
The performance of that specimen can be described as poor and brittle.
5.3.5 Cavity Wall S1W4
Specimen S1W4 was identical to S1W2 except that the backup wythe consisted of 150 mm standard hollow concrete blocks instead of 200 mm.
The end of the elastic range was not well defined but from the pressure vs. lateral deflection curves it can be interpreted to be around 1.0 kPa with a maximum centerline deflection at the block wythe of 2.49 mm. After this load level the deflections started to increase substantially since tensile cracks began to develop along the block wythe. At a pressure of 2.2 kPa a crack in the horizontal joint of the block wythe at a height of 800 mm started to increase. Shortly after, cracks in the brick veneer mortar bed al the same height (800 mm) were detected. Failure, finally, occurred at a pressure of 2.39 kPa when, the previous cracks in the lower part of the block and brick wythes deteriorated. The maximum lateral deflection at failure pressure was 49 mm.
Unexpected high deflections at the base of both wythes were observed throughout the test. That occurred as a result of cracks which formed in the mortar joint at the block / concrete interface. In general, the large deflections (in comparison with specimen S1W2) along the wall can be attributed to the following two reasons: first, the block wall itself was more slender (140 mm wide) than that of S1W2 (190 mm wide), and second, and the most important reason, is that the grout that was used to fill the reinforced cores did not fill the cores completely. As a matter of fact, upon disassembly of the failed specimen, the half lower part of the block wythe was found to be ungrouted. Therefore, half of the block wall acted as if it was unreinforced, resulting in large deflections and premature tensile failure.
From the strain gauges located at the shear connectors along the wall it was found that the critical connector was the bottom one. The internal axial force of that connector at failure was 2.67 kN (2.2 times smaller than the capacity of a connector under axial compression).
5.3.6 Cavity Wall S2W1
As has been mentioned earlier, specimen SIW2 was repaired and strengthened by an additional connector II the failed mortar bed and was re-tested as specimen S2Wl. The two specimens were therefore identical except that S2Wl had a shear connector arrangemenl type B (see Figure 4.6).
Cavity wall S2W1 had a more flexible behavior (larger deflections) in the initial (elastic) range than specimen S1W2. That can be attributed to the fact that specimen S2W1 was re-tested. Since the elastic range was exceeded in the first test (S1W2) it is apparent that minor cracks were formed along the reinforced block wythe. Cracks were, therefore, present from the beginning of test S2Wl, yielding to a reduction in the stiffness. As a result, no elastic range appears in the pressure deflection curves, as can be seen from Figure 5.3 and Figure 5.5.
The wall assembly failed at a pressure of 4.73 kPa with a maximum deflection of 16.8 mm at the middle height of the block wythe. The failure was identified as sliding of the concrete block wythe at the block wall / concrete slab interface. Plates 5.3 and 5.4 show specimen S2W1 after failure. That kind of failure would not have occurred In an actual cavity wall construction since reinforcement from the slab is used to anchor the block wall and to improve the friction and the integrity of the interface. In order to avoid that undesirable failure for the remaining tests, the bottom of the concrete block will was secured by an additional HSS beam member connected with adjustable steel rods to the two W shaped columns of the testing frame.
From the strain gauge readings it was again found that the critical shear connector was the lowermost and just before failure the axial compression force In that connector was 3.86 kN.
Both wythes acted compositely throughout the test and a ductile behavior of the wall assembly was observed. In addition, upon disassembly of the failed specimen, the shear connectors were found to be undamaged and the backup concrete block well grouted along its height.
5.3.7 Cavity Wall S2W2
Masonry cavity wall S2W2 was identical with specimen S1W4 except that it had a cavity of 50 mm instead of 100 mm.
As expected the performance of the wall assembly and the failure mode were similar. Composite action between the two wythes was achieved throughout the test. From the pressure vs. centerline deflection curves given in Figure C-11 it can be seen that the elastic range is not well defined. At a pressure of 2.75 kPa a major tensile crack in the horizontal mortar bed was formed. The corresponding maximum deflection was 26 mm. That crack caused the wall assembly to fail at a pressure of 3.09 kPa with a centerline deflection of 43.7 mm. Plate 5.5 shows specimen S2W2 at the end of the test.
Large deflections at the base of both wythes and along the height of the cavity wall were recorded during the test. These deflections were similar to those observed in specimen S1W4. Explanation for the above were given in Section 5.3.5.
Although special care was taken in reinforcing and grouting the specimen after disassembling the wall system it was found that the five last courses of the concrete block were ungrouted. No damage was detected to the shear connectors.
5.3.8 Cavity Wall S2W3
The special features of specimens S2W3 and S2W4 is that they had both wythes reinforced and a shear connector arrangement type C (see Figure 4.6). Cavity wall S2W3 had a 200 mm standard reinforced concrete block wythe and a cavity of 100 mm (see Table 4.2). Four voids out of the six in the concrete block wythe were reinforced using 10M bars and were then grouted. The sixteen cores of the brick wythe were reinforced using steel reinforcing wire 7.19 mm diameter. Plate 5.6 shows the reinforcement of both wythes.
The pressure vs. centerline deflections and the deflected shapes for both wythes are given in Figure 5.6 and Figure 5.7. From Figure 5.6 it can be seen that the elastic range is not well defined. The first major crack appeared at a pressure of 3.6 kPa at the tensile face of the middle height mortar bed of the block wythe. The corresponding centerline deflection was 4.8 mm. The slope of the pressure deflection curve was then decreased and remained constant up to failure, which occurred at a pressure of 6.2 kPa with a corresponding centerline deflection at the brick wall of 16.3 mm. At that pressure the two uppermost shear connectors buckled and the brick veneer rotated about the mortar joint corresponding to the shear connectors located at 1600 mm from the bottom. This rotation resulted in a sequential buckling of the connectors located at the half upper part of the wall. It was noted that all the failed connectors were buckled in the same side. Plates 5.7 and 5.8 show specimen S2W3 after failure. Composite action between the two wythes was observed throughout the test as we can see from Figure 5.7.
By reducing the strain gauge readings it was found that the buckling compressive force of the top connectors was 5.2 kN. The theoretical buckling load for that connector was calculated using the interaction equation (Section 13.8.3) given by the CAN 3-S16.I-M84 "Steel Structures for Buildings" (Ref.8) and was found to be 2.67 kN. It is known that the above theoretical approach is conservative, and can therefore be used safely for design.
5.3.9 Cavity Wall S2W4
Specimen S2W4 was identical to S2W2 except that the backup wythe consisted of 150 mm standard hollow concrete blocks instead of 200 mm.
The pressure vs. centerline deflection and the deflected shape curves are given in Figure C-15 and Figure C-16 of Appendix C. Since both wythes were reinforced the wall assembly had a ductile performance and therefore the end of the elastic range was not well defined. At a pressure of approximately 1.5 kPa a crack appeared in the horizontal joint of the block wythe at a height of 800 mm. The corresponding centerline deflection was 3.81 mm. After this pressure the deflections' started to increase substantially since tensile cracks began to develop along the block wythe. Failure finally occurred at a pressure of 3.16 kPa when the previous cracks deteriorated. The maximum lateral deflection at failure pressure was 18.6 mm. After that peak point the pressure dropped at 2.65 kPa and the wall assembly experienced a ductile behavior with constant pressure up to a deflection of 58 mm (recorded at the block wythe at 800 mm height). At that point the air bag pressure started to drop and the specimen was unable to carry more load.
Since both wythes were reinforced it was expected that the connectors would buckle before the masonry components failed. The premature cracking of the concrete block wythe can be explained by the fact that upon disassembly of the specimen, the half lower part of the block wythe was found to be ungrouted (see Plate 5.9). Therefore, half of the block wall acted as if it was unreinforced, resulting in large deflections and premature failure.
5.3.10 Cavity Wall S3Wl
All specimens of the third series consisted of a metal stud backup system with a brick veneer facing wythe. Specimen S3W1 had 100 mm cavity and shear connector type A (see Figure 4.6 b). The pressure vs. centerline deflections and the deflected shapes of the two wythes are shown in Figure 5.8 and Figure 5.9 respectively.
As can be seen from these Figures, the two wythes did not act compositely since the brick veneer experienced larger deflections. At the location of the connectors the wythes underwent the same deflections, but between the connectors they had different deflections. An explanation for this is that the stiffnesses of the two wythes are different and under the same pressure they undergo different deformations. By placing connectors in closer proximity the specimens will be forced to undergo the same deflections.
At a pressure of 4.2 kPa the two uppermost shear connectors buckled and the brick veneer rotated about the mortar joint located at 600 mm from the bottom. This rotation resulted in a sequential buckling of the rest of the connectors. The maximum deflections recorded just before failure at the center of the backup and brick wythe were 9.87 and 3.96 mm respectively. Plate 5.10 shows specimen S3W1 after failure.
Upon unloading the specimen it was found that the metal stud backup system returned to its undeformed configuration proving an elastic behavior.
5.3.11 Cavity Wall S3W2
This specimen was identical with specimen S3W1 except that S3W2 had a shear connector arrangement type C instead of type A. The pressure vs. centerline deflections and the deflected shapes are given in Figures C-19 and C-20 (Appendix C) respectively.
From these Figures it can be seen that both wythes acted compositely throughout the test. The specimen seemed to have behaved elastically up to failure, which occurred suddenly and in a similar manner to that of specimen S3W1. At a pressure of 6.48 kPa the top shear connector buckled and the brick veneer rotated about the bottom mortar joint towards the backup, forcing all the other connectors to buckle. The corresponding maximum mid-height deflection at failure was 14.34 mm.
Large deflections were recorded at the base of both wythes at the end of the test. That probably occurred when the three expansion anchors, used to connect the backup system with the bottom concrete slab exceeded their slip resistance capacity. At the same time cracks were formed in the mortar joint of the brick / shelf angle interface.
5.3.12 Cavity Wall S3W3
This specimen was identical with S3W2 except that the cavity was 50 mm instead of 100 mm. The pressure vs. centerline deflections and the deflected shapes of both wythes are shown in Figures 5.10 and 5.11 respectively.
From the two previous tests it was found that the critical limit state for this series was the buckling of the connectors and not the failure of the wythes. Therefore as the cavity decreases the effective length of the connector decreases and its buckling load increases. A higher ultimate capacity was therefore expected for the connector components of the assembly.
The mode of failure was similar with that of the previously described test of this series. The only difference is that the failure of S3W3 occurred at a pressure of 8.34 kPa. The corresponding maximum deflection was recorded by the centerline LVDT and was 20.51 mm. From Figures 5.10 and 5.11 it can be seen that both wythes acted compositely throughout the test.
Upon disassembly of the failed specimen, as was the case for all specimens of series 3, the metal stud backup wythe was found to be undamaged. Plate 5.11 shows specimen S3W3 after failure.
5.3.13 Cavity Wall S3W4
Specimen S3W4 was identical with S3W1 except that the cavity was 50 mm instead of 100 mm. Figures C-23 and C-24 (Appendix C) show the load v s centerline deflection and deflected shapes of both wythes.
As was the case for S3W1 these Figures show that the two wythes did not act compositely. The deflections of the brick wythe were at least twice those of the backup system. Failure was similar to the one described in section 5.3.10, and occurred at a pressure of 3.9 kP, with maximum centerline deflections of 13.47 mm and 5.25 mm at the brick wythe and backup respectively. After failure the backup system was found to be undamaged. The specimen after failure is shown in Plate 5.12.
The deformations induced to the masonry cavity wall exposed to the climatic conditions were due to the:
- Shrinkage of concrete block wythe.
- Expansion of brick veneer.
- Thermal deformations.
Therefore relative deformations between the two wythes were present.
From the strain gauge readings which were recorded during the testing period of time (10 months), it can be concluded that the shear connectors partially restrained these opposite vertical movements and induced forces in the masonry components.
The environmental humidity, external temperature, internal temperature, and material properties of the masonry components and shear connector plates are some of the parameters that affect the amount of the internal forces induced in the masonry components. It was therefore difficult to isolate one parameter and investigate its effect on the internal forces of the shear connected cavity wall. After the data were reduced to allow for the thermally induced strains on the connectors, a linear distribution was assumed for the deformations or the connectors. By using linear regression a linear strain distribution that best fits the three strain gauge readings was found. Based on this strain distribution and assuming an elastic modulus of 200 GPa for the connector material (mill galvanized thick plate) a linear stress distribution was found. Finally, by imposing equilibrium relations at the cross-section with the strain gauges the shear and axial forces at the connection of the tie with the connector can be calculated. Figure 5.12 shows the method used to reduce the data. Although the calculated internal forces were not very accurate, since the real strain distribution is not exactly linear, an estimation of their magnitude was obtained.
The maximum internal forces were recorded at the top connector. Figure 5.13 shows a typical diagram of the difference of temperature (internal minus external) versus the shear forces at the critical top connector. The maximum calculated shear force was 3.50 kN and corresponded to a temperature difference of 40°C. Although, other parameters (such as humidity and material properties) affected these values, Figure 5.13 shows that as the temperature difference increases the shear force also increases. This shear force induced on the connector relates to the axial force transferred to the masonry components. It was also found from the results that the brick wall was subjected to compressive forces, while tensile forces were generated in the backup concrete block wall. Therefore, such a system has the advantage of generating tensile forces in the block wall, which can be taken care of with suitable reinforcement.
In addition to that, no damage was observed to the shear connectors or wall components of the tested specimen.
These test were conducted to estimate the tensile strength of the Simplified Shear Connector. The test results for the four different types of shear connectors and the two different types of loads described in section 4.3.4 are summarized in Table 5.3.
As it can be seen from the average values and the standard deviations the best and most consistent performance was observed for type D. In addition to that, by comparing shear connectors type B with type C it can be concluded that the corrugation seems to be more effective than the holes in improving the tension capacity of the connection.
The failure mode for specimens with a shear connector type B. C or D and the load applied at location indicated by (a) (Table 5.3) was identified as cracking of the mortar joint at the junction of the connector with the block wythe. For specimens with connector type A and the same load condition, the failure occurred when the connector pulls out of the joint without any damage to the masonry wall segments. When the load was applied at location (b) indicated on Table 5.3 and for all types of connectors the steel plate around the hole, used to connect the tie to the shear plate, failed by yielding. The connection of the plate with the block wall did not fail. Therefore the capacity of the system can be increased by increasing the thickness of the plate or the metal between holes.
« TOP l CHAPTER SIX »