"Side Friction" to resist overturning in shallow precast foundation

[RDR89]

Hello all, I am reviewing a calculation and had a question regarding the use of "side friction" to resist overturning in a small precast shallow foundation. Not sure if I should post this in the geotech group or structural but since my background is structural, I started here. I have never seen this practice and don't agree with it for a number of reasons. For one, I don't see how you can develop "side friction", which I am assuming is akin to skin friction, particularly in a shallow foundation. It's not like a driven pile for instance where there is force built up in driving the pile in and you have the depth of soil in your favor. Secondly, assuming that there is some force resistance developed along the sides, I do not see how you can use that to resist overturning. The foundation I am looking at is 22"x22"x24" deep, precast and will be lifted in place after excavation and backfilled around with granular backfill.

Am I off-base here or is there something I am perhaps not considering?

 

[driftLimiter]

Its a pretty small footing to resist any kind of overturning load of structural magnitudes.

Small loads for misc equipment I would be more accepting of skin friction especially if its a poured footing. For precast footing embeded 24" or less I dont think its a reliable enough path.

 

[Greenalleycat]

The theory:

A simple thought experiment - with your engineering knowledge, would you rather drag a precast concrete block through an open paddock, or along a narrow path with stones grating against the sides of the block at every step?
I assume the answer is the first one - we expect that stones being there will do -something-

Side friction can be approximated as the interface friction angle mu = tan(phi) times the confining pressure on the sides of the concrete from the stones
That will be standard theory [density * height * Ka) etc
Overturning on the foundation summed around the base of the footing (24" below ground) will be resisted by this interface friction because that is acting parallel but opposite to the overturning force at a height of 8" above the base of the footing
So, it's not going to give you much, but it does exist

My thoughts:

I have used this approach before, but only for assessment, never new design
It feels way too unreliable for new design - what happens if you get slight uplift of the block and your granular fill gets wedged under it, for example?
Sounds like someone is trying way too hard to avoid making the block 1' bigger
I also don't like using this on precast - the interface is so smooth that I can only expect the friction to drop substantially compared to something being poured insitu

 

[TLHS]

Are they using the true front face dimensions for passive pressure? If I'm reaching for capacity on a small item, I'd start by accounting for horizontal spreading of the passive load. If you do that, though, then side friction becomes pretty unjustifiable because it's going to activate the same passive block you've already accounted for.

Realistically, though I can't see how much side friction they can be getting. At best it would be the at-rest pressure pushing at it. That multiplied by a friction factor and a safety factor and then trying to use it for overturning by having a couple force is going to give you basically zero isn't it?

 

[RDR89]

TLHS - it is not a lot. The foundation is supporting small equipment so the side friction is giving them a small "bump" in capacity. They are also not using a safety factor in their determination of the side friction value. They are calculating out the full resistance to overturning using the foundation weight, passive soil resistance in front of the foundation and this "side friction" we are discussing, and comparing that to the full overturning moment due to wind.

 

[RDR89]

GreenAlleyCat you echo my concerns. I just do not see how they develop friction resistance along the sides due to this being precast. If anything goes wrong with the backfill type or installation, my thought is that this calculation is null and void.