Dynamic Impact Load Tests of a Bridge Guardrail System
< Prev | TOC

6. Observations

The outcomes of the prior laboratory testing at CSU of a then conventional all timber bridge guardrail were reported in past publications (Pellicane et al. 1990, Gutkowski et al. 1994). The alternative bridge guardrail system is a definite improvement in strength. During the ramp load test, the alternative guardrail specimen was loaded up to about 32 kip (142 kN) of transverse load at the center post. Thus it was subjected implies a safety factor of 3.2 compared to the 10 kip (44.5 kN) static design requirement at that time (AASHTO 1989 etc.). This load is 33% higher than the 24 kip (107 kN) load for which the conventional guardrail system completely failed. No visible damage to the specimen occurred. Indeed, since the time of this study, changes and improvements evidently were made independently by others to strengthen the conventional guardrail system (Ritter et al. 1998). Subsequently, the modified conventional guardrail system was successfully subjected to a crash test within the requirements for a then federal Performance Level 1 (PL1) test (Ritter et al. 1998). This suggests the alternative guardrail system might perform successfully if subjected to such a test.

In the ramp load testing, none of the transverse members in the test specimen carried more than 35% of the total axial load transferred to all of them collectively. However, the load distribution observed in the transverse members of the specimen is likely not the same distribution that will develop in the real bridge under similar loading conditions. The laboratory set-up only simulated a portion of the real bridge located well into the middle of the span. Although intended to allow the specimen to move horizontally, the supports and sleeves introduced unintended friction forces in the direction opposite to that motion. On the other hand, the support conditions of a real bridge surely provide even higher countering forces. Thus, the 35% maximum load share is believed to be a high estimate of the maximum load share transferred to a single transverse member, i.e. to a single hanger, and is very conservative.

The exploratory pendulum test with the massive concrete block, suggests good performance of the specimen The impact severity was about 12.8 kip-ft. (17.4 kN-m). This is comparable to a test Level 2 requirement (13.3 kip-ft. [18.0 kN-m]) if conducted with a small sedan 1807 (820 kg) based on federal recommended test procedures (Transportation Research Board 1993). Impact severities required for a 4400-lb. (2000 kg) single unit truck for Test Levels 1, 2, and 3 are 25.4 kip-ft. (34.5 kN-m), 49.9 kip-ft. (67.6 kN-m), and 102 kip-ft. (138 kN-m) respectively.

7. Conclusions

The overall ramp and dynamic load testing showed encouraging performance of the hanger detail and the special steel connector attaching the through girders to the floor transverse beams. No significant distress in either the steel connector itself or in the nails between the connector and the timber members was observed. The main girder had no visible damage. Loads applied significantly exceeded those that had caused major failure in the conventional guardrail systems when subjected to the same test.

The performance of the test specimens under the concrete block pendulum test is inconclusive. Impact severity achieved was very low, comparable to a small sedan in an NCHRP Test Level 2 setting. Impact severity levels required for a single unit truck are substantially higher. However, the performance in a laboratory test under ramp loading substantially exceeded that of the conventional guardrail system also tested in that manner. When modified, the latter survived a federal PL-1 test concluded by other researchers. These events suggest the test specimen likely had damage to the main girder during its repositioning, thus compromising its capacity. The main girder had some horizontal cracking occur, but extreme handling procedures used to put the specimen in place may have contributed to that. In any event, catastrophic failure events did not occur.

8. References


< Prev | TOC