Experimental Results

This section presents the results of each of the three phases of the experimental program and discusses the relevant research findings. The three phases of the experimental program are presented separately in the following sections.

Findings of the Overloading Study

The load-deflection relationships of the three beams that were tested in the overloading study are presented in Figure 4. Beam ST-CONT remained unstrengthened to serve as a control beam for the overloading study while the remaining two beams were strengthened with different reinforcement ratios of CFRP. All three beams were unloaded and reloaded at various loading stages to simulate the effect of overloading conditions.

The load-deflection behavior of the three beams was essentially linear up to yielding of the steel. Unloading and reloading was essentially linear and exhibited minimal hysteresis. Prior to yielding of the steel, all three beams exhibited minimal residual deflections upon unloading. However, after yielding of the steel, the unstrengthened beam exhibited a significant increase of the measured residual deflection, as shown in Figure 4(a), while the two strengthened beams continued to exhibit minimal residual deflections up to rupture of the CFRP. After rupture of the CFRP occurred the behavior of the strengthened beams followed a similar trend to that of the unstrengthened beam.

(a) beam ST-CONT (b) beam OVL-1

(c) beam OVL-2

Figure 4: Load-deflection behavior of the three overloading test beams

Table 2: Comparison of the overloading beams

Beam ID

Reinforcement

Ratio

Stiffness

Increase

Yield Load

Rupture

Load

Crushing

Load

ST-CONT

0 %

N/A

137 kN

N/A

222 kN

OVL-1

4.3 %

27 %

181 kN

259 kN

216 kN

OVL-2

8.6 %

46 %

253 kN

358 kN

216 kN

Table 2 presents the elastic stiffness increase, the yield load, the rupture load and the crushing load for the three beams tested in the overloading study.

The ultimate capacity of the unstrengthened beam, ST-CONT, was governed by crushing of the concrete while the ultimate capacity for the two strengthened beams, OVL-1 and OVL-2 was governed by rupture of the CFRP. The elastic stiffness, yield load and ultimate capacity of the beams were increased by 46 percent, 85 percent and 61 percent respectively using the higher reinforcement ratio. Inspection of Table 2 indicates that doubling the reinforcement ratio of the applied CFRP, from

4.3 percent to 8.6 percent, approximately doubled the elastic stiffness increase of the beams. Increasing the reinforcing ratio by two times also approximately tripled the increase of the measured yield load and ultimate capacity of the strengthened beams. This demonstrates that increasing the reinforcement ratio did not reduce the efficiency of utilization of the CFRP material. The improved performance of the strengthened beams indicates that it may be possible to increase the allowable live load level of a steel-concrete composite girder strengthened with HM CFRP materials. A proposed methodology for determining the allowable increase of the live load level is described later in this paper.

As mentioned previously, the presence of the HM CFRP materials helped to reduce the residual deflection of the strengthened beams due to the effect of overloading conditions. The residual deflections of the three beams that were tested in the overloading study were compared to evaluate the effectiveness of the HM CFRP strengthening system. Each of the three beams was unloaded at a load level of approximately 175 kN to simulate a severe overloading condition. The average measured strain at the steel tension flange of beams ST-CONT, OVL-1 and OVL-2 at the 175 kN load level were 2.5 ey, 1.0 ey and 0.6 ey respectively, where ey is the average yield strain of the steel determined from coupon tests. The beams were unloaded to a load level of 45 kN to simulate the sustained load acting on a structure, due to self-weight, after an overloading event. The plastic component of the residual deflection, after subtracting the initial measured elastic displacement of the beams at the 45 kN load level were compared. The measured residual deflection of beams OVL-1 and OVL-2 were respectively 5 times and 6.5 times lower than the measured residual deflection of beam ST-CONT. Consequently, under severe overloading conditions, an unstrengthened bridge girder may require repair or replacement while a strengthened bridge girder may remain serviceable.