Bearing Resistance of Culvert Foundations

[yakpol]

I am designing the arch culverts supported on the spread footing. My question is: can I use larger bearing resistance when maximum stress occurs under the heel of the footing?

[li]Tradionally, the bearing resistance is calculated using footing embedment depth (Df per AASHTO) inside the culvert (at the toe), which is a fairly small due to the scour. The footing embedment at the heel, however, is quite large, equal to the depth from the roadway level. Based on this depth the resistance values are several times greater than using embedment at the toe.[/li]
[li]The maximum bearing stress at the toe occur when lateral loads are maximized. In contrast the maximum stresses at the heel occur when lateral loads are minimized and vertical are maximized. Considering weight of soil above the heel, it's very hard to achieve reasonable footing size for bearing resistance based on 34 degrees friction angle.[/li]
[li]Wondering if I can use higher resistance on the heel side? It's understandable that full advantage of large embedment cannot be taken due to the proximity of the lower ground, like inside and at the ends of the culvert. Still using lower resistance seems overconservative, preventing rational designs.

Appreciate your opinions on this subject. Please see the attachment for a visualization. [/li]


Thank you!

 

[BNGeo]

I expect if you do a refined analysis, the initial stress condition would be somewhere between the toe and the heel. Especially for small structure widths where the walls are not all that far apart.

In Massachusetts, MassDOT recently specifically directed on a project that only the overburden on the toe be used (lesser overburden). I believe there is also a similar reference in AASHTO.

 

[geotechguy1]

Interestingly I think this is actually a question that has never been answered decisively in the literature. Someone should do a PHD on it. There are a few lengthy threads on the same issue as it pertains to footings sitting at the base of a basement excavation which is a common case, where one side of the footing is 'embedded' 2-3m+ and the other side is only 300mm. There is no agreement; I've also asked numerous people over the course of my career and have never found any consensus.

None of the conventional bearing capacity equations capture this case.

eg:

https://www.eng-tips.com/viewthread.cfm?qid=350831

 

[yakpol]

Thank you for responses. I was surprised how gray this subject is. We proposed to use bearing capacity based on the larger embedment if eccentricity (M/P) exceeds 0.1 of footing width, which ensures that linear bearing stress on this side is at least 4 times larger than on the side of shallow embedment. It sounds rational, but has no research behind.