View Q&A

I have some thoughts on the tubular rail splice shown.

We know from previous crash testing that exposed edges as small as 3/8” thick have snagged vehicle components and either disengaged sheet metal or caught vehicle rims and instigated vehicle instabilities. Thus we have recommended l1/4” or less. The NCHRP 554 study performed at TTI regarding aesthetic bridge rail features came to similar conclusions and found that vertical asperities should be less than 1/4” deep. Your best source for exposed edge information would be in NCHRP 554. We have also addressed similar issue in the Pooled Fund Q&A.

https://mwrsf-qa.unl.edu/view.php?id=1208

https://mwrsf-qa.unl.edu/view.php?id=1118

In your case, the tube rail shown may exceed those limits for exposed edges, however it is also in a location that is less likely to grab the vehicle wheel. I don’t have the heights of the tube rail and parapet or the shape of the parapet, so that may affect things to some degree. However, I would assume that the majority of the wheel interaction would be with what appears to be a vertical parapet and not with the higher tube rail.  Your rail is also a round tube so there is not a continuous vertical edge to grab the vehicle as well. This lessens the potential snag significantly.

There may be some snag on the fender of the vehicle due to the exposed edge. However, we have seen previous TL-3 tests that have had significant fender snag and not had a serious issue.

Thus, we can’t say definitively that the splice shown is not an issue based on existing testing and data. However, we based on the round tube shape limiting the exposure and the location of the splice, we believe the effects would not be as severe as previous guidance would suggest.

One item to note is that there may be a concern with the  vertical plates supporting the tube and termination of the tube posing a snag hazard as well. Again, I don’t have dimensions for the parapet or the offset of these structures from the parapet face, but we have seen that fenders can get above the top of the parapet and snag on these types of surfaces as well.

Thanks

We are still sorting out what to do about the Structural Steel Railing that was discussed below.  As time has gone on we have gathered more information about rail and found research completed on a similar rail.  I am going to try and summarize the problem as we see it and all the things we have reviewed related to the problem and our currently planned response to the problem.  We were wondering if you could review our information gathering and response to see if you agree with our conclusion.

·        Problem      

o   Gaps exceeding the tolerance indicated in the plans at the splices of the elliptical tube rail.  (See Figure 1 Below)

o   Misalignment of the tubular rail sections.  (See Figure 2 Below)

o   There appears to only be one manufacture of the elliptical tube used in the rail and they did not feel they could improve the tolerances we are currently seeing.  The issue was explained to happen when the galvanizing was applied to the fabricated rail sections.  The galvanizing process requires the tube to be heated and that heating is resulting in a warping of the elliptical tubes.

o   There was concern that the beyond specification gaps and misalignment of the rail would result in a railing that would not perform as designed in crash.  

·        Information Gathered

o   DOT field staff gathered several pieces of information for us        

§  How common were the misaligned and over specification gaps in the rail to this point

·         They were very common and the reason this deeper investigation.

§  What were some of the worst gaps in the rail splices (See figure 1)

·         7/16” total vertical gap on the top or bottom of the tube.

§  Where are the worst gaps generally located (See figure 1&3)

·         They are generally at the top and bottom with the gaps on the side being tight.

§  Are there other suppliers that are capable of more closely meeting the plan tolerances

·         There is only one supplier the contractor and inspection team are aware of

§  How many splices will there be in this project

·         There will be approximately 640 splices on the overall project for the bridges.  That is for 20,494 feet of rail. 

o   DOT Design Methods group gathered existing research and reached out to Midwest Roadside Safety, Texas A&M Transportation Institute (TTI) and few state that have installed the same or similar rail. 

§  With the information provided at the time Midwest Roadside Safety responded with and email.

·         Midwest indicated the location of splice made it less likely to snag a vehicle wheel.  They indicated there was the potential for a fender snag on a TL-3 vehicles, but they have seen significant fender snags that did not cause serious issues.  They did note concern with the support post but did not have details of the posts available at that time. 

§  TTI responded with the linked research (http://tti.tamu.edu/documents/0-6946-R2.pdf.)

·         The rail was somewhat similar and used the same elliptical rail section we are using.  The splices had what seemed very similar if not worse gaps and misalignment of the top tube (See figure 4).  The rail was tested at the splice location and passed the tests that were part of the research. 

§  Texas responded with an email

·         They stated that what we were seeing was not uncommon with this rail system.  They felt the splice was high enough that it was not a strength or snag concern

§  Pennsylvania responded with an email

·         Indicated there were no problems observer so far with their installations of this system.

·        Response to Problem

o   The splice gaps beyond what are specified in the plan are undesirable but based on the information available should not significantly increase the risk of poor crash performance.

o   There does not appear to be a good way to reduce the gaps with the materials supplied.  We recommend shimming the rail to improve alignment and try and keep the gap even around the splice as possible.   The shim material and thickness should be reviewed by the construction and materials staff.  The shims should be match the size and shape of the post base

I would agree based on the TTI report that the concern for snagging at the splice is limited based on the gap size, tube thickness, tube shape, and the similarity to the system evaluated at TTI.

That said, there may still be some concern for passenger vehicle snag on the vertical posts of the system. The TTI report your shared had quite a bit of vehicle fender and hood snag, and that occurred with a lower horizontal rail element to help mitigate post contact. Your railing has a taller parapet and a different shape, so the vehicle interaction with the vertical posts may vary, but I believe their still may be a concern there.

A second concern may exist with respect to the horizontal tube location relative to the face of the lower parapet. MASH considers fracture of the side windows due to contact with the test article a failure criteria. As the top tube in this system extends past the top corner of the barrier approximately 3.75”, there is concern that this tube may contact the side windows. We have recently been setting these tubes back from the face of the barrier in our designs. The safety shape slope of the barrier may mitigate this somewhat if the vehicle climbs the toe and rolls away from the barrier, but it is still something I wanted to note.

Thanks