I have a couple of questions I’m hoping you can answer in less time it would take me to research them. With Khyle Clute and Brian Smith both recently departing our Methods Section, my time to research issues is very limited. We are developing a detail to show a transition from anchored TBR to unanchored TBR. I have perused TRP-03-180-06 and see a transition was developed for rigid to free-standing TBR (Figure 32 of TRP-03-180-06) based on pinned barrier. Do you know if anything has been (or is being) developed for TBR anchored using tie-down straps? For an anchored installation with crash cushions to protect the ends, what is the minimum number of TBR sections needed to develop the full strength in the system to limit deflection to the 6 inches shown in our BA-401? Thank much for your assistance!
Hi Daniel,
Responses below in red.
I have a couple of questions I’m hoping you can answer in less time it would take me to research them. With Khyle Clute and Brian Smith both recently departing our Methods Section, my time to research issues is very limited.
No transition design has yet been developed with the tie-down straps. The straps have significantly higher deflections that the other two anchorage systems. It could potentially be done, but has not been attempted at this time.
This question is a little more involved. First, the roadside PCB to rigid barrier transition has not been tested to MASH at this time. The median version has, but that uses pins/anchors on both sides of the approach PCB segments.
Second, the asphalt pin tie-down shown in the detail was tested to MASH in November and did not pass due to snag on the PCB joint. MwRSF conducted full-scale crash test no. WITD-2 on the asphalt anchorage of the F-shape PCB. In this test a 2270P vehicle impacted the barrier system installed 6" from the a 3-ft deep vertical trench at 62.0 mph and an angle of 25.1 degrees. During the impact, the vehicle was captured and stably redirected. All of the barrier segments were retained on the asphalt and the maximum dynamic deflection of the barrier was 24.5". However, the left front wheel of the pickup truck snagged on a joint between adjacent PCB segments and was pushed back into the floor pan. This resulted in intrusion and opening of the floor pan in the occupant compartment with a small portion of the wheel extending into the occupant compartment. As such, this test was deemed unsuccessful under the MASH TL-3 impact conditions. Potential system design revisions were noted to the sponsor and a follow-on research project was proposed for the Midwest States Pooled Fund Year 29 research program.
Third, the transition designs we have tested were all developed for connection to rigid concrete barriers. They have never been evaluated with crash cushions or their connection to the PCB. This issue has been brought up previously by several states, but it is difficult to develop a transition to all of the proprietary crash cushion systems. Additionally, transitions between PCBs and crash cushions is allowed by many manufacturers, but I am not aware of them being tested in the configuration. The transitions we have developed may potentially work with crash cushions, but I wanted to let you know that we did not design them for that and they have not been evaluated in that manner. Vehicle snag and the transition in stiffness may not be the same.
With regards to the question of the number of PCB segments needed to generate similar deflections to the original anchors PCB adjacent to a drop-off, that question has several aspects. First, the tested system had deflections higher than the 6” shown in your detail. The 6” gap was sufficient for crash testing to maintain all of the barriers on the pavement without falling into the trench. However, the deflection was higher (approximately 21.8” of dynamic deflection at the top of the barrier and 11.1” of permanent set deflection at the base. These deflections were due to rotation of the top of the barrier and disengagement of the soil/asphalt adjacent to the trench. We would expect lower deflections if the trench was not present, but testing of the bolted tie-down anchorage for the F-shape PCB under MASH resulted in a dynamic deflection o 14.1” at the top of the barrier. This is still higher than the 6” in your detail.
The second aspect is how many barriers do you need to develop the tested system deflections. That question is based on a couple of factors, including the overall system length and the distance to the end. This analysis has not been done for anchored PCBs, but we did look at it for free-standing PCBs.
https://mwrsf.unl.edu/researchhub/files/Report331/TRP-03-337-17.pdf
For free-standing PCBs, we found a minimum system length of nine barriers was recommended with three barriers upstream of the beginning of LON and five barriers downstream of the end of LON. For anchored PCBs we would expect system length to have less of an effect on the overall length and the barriers upstream and downstream of the LON, but this has not been analyzed or evaluated. An anchored PCB system would tend to engage the upstream and downstream PCBs much less than free-standing systems and would not require the upstream and downstream barrier mass and friction to restrain barrier motions. It would seem reasonable that the guidance for free-standing barriers would be adequate for anchored installations. The minimum length free-standing PCB system required additional barriers on the downstream end of the system. For an anchored system, this would not likely be necessary as the anchors provide most of the lateral resistance for the barriers. Thus, a more reasonable recommendation for anchored PCBs would be a minimum system length of seven barriers with three barriers upstream of the beginning of LON and three barriers downstream of the end of LON. Again this has not been formally evaluated, but it seems reasonable based on the data we currently have.
Some parts of this site work best with JavaScript enabled.