Spread footing abutment question

[71corvette]

I've been tasked with designing a pair of stub abutments on spread footings for a highway bridge. I've developed a foundation design based on AASHTO's LRFD criteria that seems to make sense to me, but I'm looking for a quick sanity check.

Whenever I've looked at an abutment standard detail, or even in the AASHTO code, cantilever abutments are always shown with the heel longer than the toe. In the case of tall cantilever abutments this makes sense since the substantial fill over the heel is very effective at resisting overturning forces and helps to keep eccentricity to a minimum (let me know if I go off track here...) However, in the case of this short stub abutment, my analysis is showing that a short heel and long toe is the most efficient design. Does this make sense?

The abutments are supporting a relatively long span of 115' which results in substantial superstructure loads applied to the abutment. In addition, the abutments are fairly short at +/-15' total height with 6' embedment for frost protection. What I'm finding is that, in order to keep eccentrity down (and thereby keep the bearing pressure down), I need to use a 12' wide footing with a longer toe and shorter heel to position the superstructure reactions more toward the center of the footing. The foundation material is compacted sand/gravel with an assumed nominal bearing capacity of 10 ksf.

As I said, I'm second guessing myself since I can't remember ever seeing a detail showing an abutment with a longer toe than heel. Is there any specific reason why a longer heel is requried?

Also, is anyone aware of any available spreadsheets for developing abutment footing designs with LRFD? Whereas this is the first LRFD footing I've done I'd like to do a quick check of my results.

Thanks!

 

[miecz]

Another reason for a long heel is to generate enough dead load to resist sliding. If you have substantial dead load from your superstructure, that may by the reason you don't need the long heel. Do you hav a construction condition, where there's no dead load from the superstructure?

I'm developing a mathcad sheet for the same condition. So I'm curious as to how you are handling the bearing pressures. At this point I'm only checking bearing pressures and the service limit state. Are you checking bearing pressures at the strength limit states?

 

[71corvette]

I have checked a construction loading case - basically all loads except those from the superstructure.

To answer your other question, I've checked bearing, sliding, and eccentricity at the Strength limit state as required by AASHTO. I believe (from memory - my manual is at the office) that service limit state is used only to check settlement and global stability.

 

[BigH]

If the stub foundations are close to the embankment front face slope - don't forget to consider "footings on a slope." If you are assuming 10ksf bearing pressure, you need to ensure you have proper material gradation and crushing of the gravel (like MTC(Ontario) Granular A); and the level of compaction is correctly specified and achieved.

 

[civilperson]

Remember that the resultant of the soil pressure beneath the toe will be 2/3 the distance from wall to toe face. The soil pressure will be triangular, (or trapezoidal), in shape with extreme pressure at the toe face.

 

[miecz]

My manual is at the office as well. I asked about the bearing because I can never get my geotech to give me nominal bearing capacity. That aside, somethings not adding up here. Are you using the passive resistance of the soil in front of the stem to resist the sliding?

 

[71corvette]

Civilperson - The AASHTO LRFD mandates a uniform bearing pressure distribution (over a reduced bearing area based on eccentricity) for foundations on soil. Foundations on rock are designed using a trapezoidal (or triangular) distribution.

Miecz - You've probably already seen this, but AASHTO provides a chart of nominal bearing capacity for a variety of soil types in chapter ten. And to answer your question I'm not using passive resistance to work against sliding. Our state agency doesn't like to do so, and besides, there is some minor potential from scour here that would make using passive pressure unwise. What's causing your suspicion?

 

[miecz]

My gut feeling was that you would have trouble resisting the lateral sliding force from 15 feet of soil with the friction from the weight of the abutment and the soil over a four foot heel. After running some numbers, I see that it's not such a problem.

Like I said earlier, I haven't had a lot of experience calculating bearing pressures with LRFD, but I think the bearing pressures seemed to blow out easily with factored loads.

Also, I seem to remember a discussion on this forum where no one could find an example in the literature calculating bearing pressures under retaining structures using LRFD.

 

[Riggly]

Looking at the forces on the abutment, a longer toe makes sense if the vertical load from the superstructure is very high. Looking at moment about the heel, the soil pressure at the base is what will be providing the stabilizing force. Considering uniform distribution of soil pressure, to provide an effective resisting force the resultant force from the soil has be be closer to the toe. As mentioned in one of the post, the heavy bridge appears to be providing the force to compensate for sliding resistance. Without that heavy load, a longer heel would be required. As long as the longer toe does not interfere with other structure, or cause unwanted obstruction, I would not worry too much.

 

[MIBridgeEng]

Maybe your've already done this, but also the check the case where the wall is built and backfilled without loads from the superstructure. This could be the case during construction (i.e. building the abutment fully prior to setting the beams) or during a future rehabilitation (i.e. superstructure replacement where the beams are removed). In these cases, the dead load of the superstructure and live load on the superstructure do not "help" the stability. Checking these cases could extend the heel longer.