Logged in as: Public User

Guardrail over shallow culvert

State ME
Description Text I was recently introduced to your joint report – Paper Reference NO. 0824-000067, A Synthesis of MASH Tested 31-in. Tall, Non-Proprietary, W-Beam Guardrail Systems, and found it to be extremely helpful. Thanks for your efforts to assemble everything in one place and clarify things.

A couple questions:

It was good to learn that the omission of a single post without any further action (no double nesting rails, etc.) actually passed MASH Test 3-11. My question relates to leave out multiple posts in an area with a shallow culvert. We have tried to avoid wood posts for long-term maintenance reasons, so we would prefer not to use the MGS long-span guardrail system, unless a steel post option is available (any thoughts on that?). Attaching guardrail to a culvert looks like a good alternative. The top mounted system (Figure 8b)requires a minimum of 9” of soil fill. Is there an upper limit to the amount of soil fill allowed? If there were 3 feet of fill, would it still be acceptable to use the top mounted system? In the past, we have embedded cut-off posts in concrete. Has anything like that ever been tested?

Thanks for any help you can give!

  • Guardrail
Other Keywords Culvert
Date September 7, 2018


MwRSF tried to incorporate steel posts within an elongated version of the MGS long-span system a few years ago. In this configuration, the wooden CRT posts were replaced with Universal Steel Breakaway Posts, and the unsupported span length was increased from 25 ft to 31.25 ft. Unfortunately, the system did not pass MASH TL-3.  This testing with an elongated span was the only attempt that I am aware of which tried to incorporate steel posts within the MGS long span.  As such, we currently do not have a steel post long span system.  I have provided a couple links to reports dealing with this project





With the top-mounted posts on low-fill culverts, thre currently is not a maximum embedded distance for the posts.  The top mounted posts within shallow fill depths should be stiffer/stronger that standard posts embedded 40-in. in to the soil.  However, as the embedment depth is increased, the post stiffness becomes closer and closer to the stiffness of the standard post in soil.  We currently don’t have the testing necessary to identify embedment depths less than the nominal 40 in. that would result in crashworthy system.  Thus, our current recommendation has been to use top mounted posts until an embedment depth of 40 in. can be achieved.


Posts embedded directly into concrete have created problems in the past.  The concrete does not allow the post to rotate back through the soil and results in a very stiff system.  Crash tests into such guardrail configurations have resulted in failures due to rail tearing, loss of containment, and excessive vehicle decelerations.  Thus, embedding posts directly into concrete is not recommended.  To place posts within an concrete or asphalt pavement, leave-outs should be utilized around the post to allow post movement.  The currently approve leave-outs measure 18-in. x 18-in. and are filled with a very low strength grout (200-300 psi) after the post is installed.


Let me know if you have any further questions.

Date September 7, 2018


Thanks for the clarifications.  As a follow up I thought I would send a detail sheet showing a typical situation we run in to.  In these situations we are not dealing with a headwall and a vertical drop, but rather a 2:1 slope down to the top surface of a much longer box culvert.  Now that we are typically installing MGS rail at 31” height, we want to make sure we handle these situations correctly.  Top mounted rail seems to be a good fit.


I’ve also attached a detail that we used before switching over to MGS.  This system doesn’t seem a lot different than the top mounted system.  When the top mounted system is impacted, how do the posts react – do they break away or do they deform?  Also, wondering if you observed any damage to the concrete structure when the rail was impacted during the testing?


In case you are interested, we typically install guardrail at the top of 2:1 slopes.  Our preferred cross section would include 18” of level terrain behind the 7 foot posts.  Alternatively we will often use 6” of level terrain behind 8 foot posts.  The longer posts would account for variable soil types with less than 2 feet of level terrain behind the posts.  This seems to fit with what you are recommending.


Thanks again for your help.  I really appreciate you taking the time to respond!



Date September 8, 2018
Attachment 20180910084908762.pdf
Attachment 20180910090429408.pdf


The top-mounted posts for MGS that we have been discussing will not breakaway. The posts will plastically deform near the baseplate as the post is deflected backward.  Occasionally, a post flange or 2 will tear, but the post has remained attached to the baseplate.


We have not observed significant damage to the concrete slabs during our testing of guardrail with top-mounted posts. We have evaluated both through bolt and epoxy anchored bolts.  Neither pulled out during testing.  Note, the concrete slabs were all reinforced and at least 8” thick.  Thinner slabs or unreinforced slabs may be subjected to concrete damage.


We have successfully tested a standard MGS at the slope breakpoint of a 2:1 slope to MASH TL-3 criteria.  See the following link for the report: https://mwrsf.unl.edu/researchhub/files/Report325/TRP-03-320-16.pdf

That said, we have never evaluated top mounted W6x9 posts adjacent to a 2:1 slope.  The top mounted posts should provide a little more stiffness than a standard 6-ft post (as discussed previously), so the deflections should be held to within the tested bounds of the MGS.  Thus, I think this configuration should perform as intended.  However, I have no support to prove that it would be crashworthy.


I hope this helps.


Date September 10, 2018

Contact Us:
130 Whittier Research Center
2200 Vine Street
Lincoln, NE 68583-0853
(402) 472-0965
Email: mwrsf@unl.edu
The information contained on the Midwest Roadside Safety Facility (MwRSF) website is subject to change without prior notice. The University of Nebraska and the MwRSF is not responsible or liable, directly or indirectly, for any damage or loss caused or alleged to be caused by or in connection with the use or misuse of or reliance upon any such content, goods, or services available on this site.