3. Fatigue Test on Straight Composite Girders Subjected to Hogging Moment
3.3 Fatigue Test Results and Discussion
3.3.2 Load-crack width Response
The past research indicates that for composite beam subjected to negative bending, initial cracking would develop in the concrete surrounding the tension elements and reinforcing bars when the flexural tension exceeds the rupture modulus of the concrete (Zia et al., 1976). Tensile stress of the concrete will increase as the applied load increases in the static test, and finally the cracking of the tension elements of concrete would occur when the tensile strength of the concrete for the tension element is exceeded.
In the intermediate static test, crack width was recorded by using 14 π-gauges on the top surface of the concrete slab. Π-gauges were arranged around the span center, and gauge number and location of each π-gauge are illustrated as shown in Fig.3.7. Crack formation and distribution on the surface of the concrete slab was recorded during each intermediate static test, and the details will be described in the following.
Span center
264 265
266 267
268 269
270 271
272 273
274 275
276 277
(a) Π gauges arrangement
(b) Π gauges arrangement in the test
Fig.3. 7 Arrangement of π-gauges (100mm square meshes on the surface of the concrete slab).
Before the fatigue test, a static test (0 cycles) with the limited load of 200kN was performed on FT-1, but no cracks were observed. Observation of the exterior of the specimens in the subsequent test showed that the initial crack was formed in the span center section when Nre equals to 10,000 (as shown in Fig.3.8), and no other cracks were found in other locations.
Theoretically speaking, as 200kN is the theoretically initial cracking load, no crack should initiate in the concrete slab except the span center section under this magnitude of load intensity.
However, after the subsequent fatigue test with different number of cycles (Nre >10,000), traces of small cracks was also found in the other locations, not just the top surface of the concrete slab, but also on the side faces, as shown in Fig.3.8. These evidences clarify that, unlike the static test results, both the amount and length of cracks were increasing with the increase of repeated load cycles.
Furthermore, the crack width recorded by π-gauges during each intermediate static test was given in Fig.3.9. But as have already mentioned above, the π-gauges were only arranged in the span center locations, so only cracks in this location were measured. Besides, as the static load was relatively small (200kN) and limited number of cracks was formed, so some π-gauges may be placed in locations where the concrete does not crack. Therefore, only results of π-gauges that located in the cracking region of the concrete slab were presented herein. The initial crack observed in the test is shown in Fig.3.8, and the gauges numbered 266, 267, 270, 271, 275, and 277 were selected and the measured crack width was presented as shown in Fig.3.9.
As have already emphasized, the residual crack was width eliminated in the new intermediate static test. Obviously, results in Fig.3.9 indicate that (1) in the static test without any fatigue test (Nre
=0), the crack width was found increasing with the load increase and decreasing with the load decrease. “Crack closure” of these cracks was observed during the test and residual crack width is relatively small; (2) when Nre =10000, unrecoverable crack width was observed in gauge 277 location, and the residual crack width was about 0.04mm. No other unrecoverable crack was observed in other locations; (3) when Nre >10000, the residual crack width is almost “0”, but the crack width at the same load levels will increase as the increase of the repeated load cycles, which seems to indicate that the girder stiffness becomes smaller with the repeated load cycle increases. All in all, it seems to indicate that when the repeated load equivalent to the initial cracking load, limited number of unrecoverable cracks might occur with the increase of the repeated load cycles; simultaneously, the cracking stiffness was found to decrease and crack width at the same load level was found to increase.
There is another interesting phenomenon was observed in the test. In general, the crack width (here we may define it as the “mean crack width”) increases with the load increase and decrease with the load decrease, and should never change from “+” to “-”. However, the measured results shown in Fig.3.9 indicate that some crack width changes to “-” when the applied load was removed. This is presumably because the heterogeneous crack width was formed in transverse direction of the concrete slab, which means the same crack has different width in different locations, as shown in Fig.3.10. In this condition, the crack in some locations might still has the “+” crack width even if the same crack in other locations has already closed after the load was removed, then the crack closure tendency of these unclosed crack will cause the compression on the closed crack, shown as the “-” crack width in Fig.3.9.
As it can be find that some reinforcements were still in tension after the removal of applied load, concrete in other locations may in compression in order to keep the balance of the section forces.
-500 0 -1000
-1500
-2000 500 1000 1500 2000
Fig.3. 8 Crack formation and distribution on the concrete slab of FT-1 during the fatigue test
-0.02 0.00 0.02 0.04 0.06 0.08 0.10
0 50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
-0.02 0.00 0.02 0.04 0.06 0.08 0.10 0
50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(a) Gauge 266 (b) Gauge 267 10,000, initial crack 50,000
100,000
1,500,000 500,000
-0.02 0.00 0.02 0.04 0.06 0.08 0.10 0
50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm) -0.02 0.00 0.02 0.04 0.06 0.08 0.10
0 50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(c) Gauge 270 (d) Gauge 271
-0.02 0.00 0.02 0.04 0.06 0.08 0.10 0
50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
-0.02 0.00 0.02 0.04 0.06 0.08 0.10 0
50 100 150 200 250
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(e) Gauge 275 (f) Gauge 277
Fig.3. 9 Load-crack width relationship of FT-1 after different number of cycles
Fig.3. 10 Formation of “-” crack width
Crack closure zone
Unrecoverable crack Re-bar
Re-bar
-500 0 -1000
-1500
-2000 500 1000 1500 2000
Fig.3. 11 Crack formation and distribution on the concrete slab of FT-2 during the fatigue test
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
Applied load (kN)
Crack width (mm)
0 1×104 5×104 1×105 5×105 1×106 2×106
(a) Gauge 264 (b) Gauge 265
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
Applied load (kN)
Crack width (mm)
0 1×104 5×104 1×105 5×105 1×106 2×106
-0.05 0.00 0.05 0.10 0.15 0.20 0.25
0 100 200 300 400 500 600 700
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(c) Gauge 267 (d) Gauge 270 Initial crack
10,000 100,000
10,000 50,000 50,000
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
0 1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(e) Gauge 271 (f) Gauge 275
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600 700
Applied load (kN)
Crack width (mm)
0 1×104 5×104 1×105 5×105 1×106 2×106
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 0
100 200 300 400 500 600
700 0
1×104 5×104 1×105 5×105 1×106 2×106
Applied load (kN)
Crack width (mm)
(g) Gauge 276 (h) Gauge 277
Fig.3. 12 Load-crack width relationship of FT-2 after different number of cycles
Similar to FT-1, crack formation, distribution and crack width development process on the concrete slab of FT-2 were also recorded during the tests. As expected, the initial crack was observed in the span center section during the initial static test (0 cycles) and more and more cracks were formed with the increase of the load cycles, as shown in Fig.3.11. The cracks propagated and distributed on the surface of the concrete slab. Besides, the cracks mainly distributed in a direction perpendicular to the bridge axis, which means the cracks were mainly caused by the applied negative bending moment.
For FT-2, the gauges numbered 264, 265, 267, 270, 271, 275, 276, and 277 were selected and the measured crack width was presented as shown in Fig.3.12. In the initial static test, the crack was formed in the span center and the crack width was found suddenly increased (often referred to as crack jump) when the applied load is about 230kN, which should be formation of
the initial crack. The results clearly indicate that after the initial crack, the crack width increased very quickly, and large residual crack width was also observed as can be indicated in gauges 265, 267, 271, 275 and 277. The maximum residual crack width in the initial static test is about 0.076mm. In all subsequent tests, the crack width was increasing with the load increase and decreasing with the load decrease. However, the maximum crack width did not increase and the residual crack width does not change with the increase of number of load cycles, which showed that the cracks did not change their sizes and the residual crack width is almost the same as that in the initial static test.
It seems to indicate that when the repeated load equivalent to the stabilized cracking load, some unrecoverable cracks would occur in the static test, but no other unrecoverable cracks were observed in the span center section with the increase of the number of load cycles;
simultaneously, the cracking stiffness and the residual crack width was found keeping as constant and did not change with the increase of load cycles.
Furthermore, “-” crack width was also observed in gauges numbered 270 and 276. But for the gauges in other locations, “-” crack width was not observed in the initial stastic test or subsequent intermediate static tests.