Soil Friction resistance against uplift

[Okiryu]

We are designing a foundation for a water tank. The foundation is a circular continuous shallow foundation on weathered limestone rock. Soils are very competent. The foundation shape is like a rectangular pedestal (it is NOT the typical inverted T-shape shallow foundation shape). We are using the weight of the tank/foundations and soil friction acting at the sides of the foundation to resist the seismic uplift loads.

Do you see any issues for considering the soil friction contribution to resist the seismic uplift forces? Are any structural codes that prohibit its use?

 

[JStephen]

I'm not aware of anything that specifically prohibits this, but I don't think it's normally done, either. In AWWA D100-11, in the section on elevated tank foundations, it says "The weight of the pier plus the weight of the soil directly above the base of the pier...shall be sufficient to resist the maximum net uplift..." and this is the approach that would generally be used with ringwalls, inverted T-shapes, etc., although that requirement is not in the ringwall paragraph. In API-650, E.6.2.3. says product and soil over the footing "may" be used to resist uplift, but doesn't say anything about soil beyond the footing.

 

[JoelTXCive]

This is a good question Okiryu.

In Houston, we regularly use soil friction to resist uplift on wet wells and other partially submerged tanks, but we are not designing for seismic.

The geotechs usually have us disregard the first ~5 ft of soil, and then they will provide a friction coefficient or a direct frictional resistance stress per surface area of tank.

A question I would have regarding seismic is: If you are in the middle of an earthquake that wants to move the tank, then won't the soil be moving too?

 

[HTURKAK]

Dear OKIRYU,

Your post simply implies that the tank is anchored and the foundation needs the wt of foundation+ product and soil friction to resist to overturning..


Just copy and paste of relevant item API app. E. 6.2

(...Foundations and footings for mechanically-anchored flat-bottom tanks shall be proportioned to resist peak anchor
uplift and overturning bearing pressure. Product and soil load directly over the ringwall and footing may be used to
resist the maximum anchor uplift on the foundation, provided the ringwall and footing are designed to carry this
eccentric loading....)

That is, you can only use the ( wt of product above the ringwall + tank shell wt with roof portion supported by the shell + footing wt + soil above the footing) to resist overturning of the tank foundation..so , you cannot use the friction which may develop between ringwall and the surrounding soil. Moreover,the concrete surface protection ( the use bitumen coating , coal tar epoxy paint, membrane tanking etc ) will reduce the friction..

If you can not provide a reasonable FS , the options are; increasing the wt of footing with adding slab under the ring so use inverted tee foundation, anchoring the footing to the ' competent rock 'as per your description etc..

 

[haynewp]

I don't see there being a normal compression component to develop the friction if you are talking about using the face of the rectangular ringwall and the soil that is up against it to resist uplift.

 

[1503-44]

The problem with friction is that it takes movement to mobilize the resistance force itself. If it moves, its too late, since kinetic fricction is greater force than static friction. Its already exceeded the static friction force. In Houston most of the soil is clay, so you probably have more cohesion than friction. It is definitely unconventional to consider either for prevention of uplift. Water tends to lubricate and reduce friction as well. Can you count on dry all the time? Unless you can quantify that, safer to ignore it.

Statements above are the result of works performed solely by my AI providers.
I take no responsibility for any damages or injuries of any kind that may result.

 

[JoelTXCive]

I think you are correct 1503-44 on the Clay cohesion.

The values provided by the geotechs are very low...in the range of 50psf to 100psf resistance. (They still help though).

 

[jayrod12]

Around here, we regularly use 190 PSF of skin friction in uplift. We neglect the top 5 or 10 feet of pile length. This is in about 25-40 feet of reasonable clay. Gravity loads skin friction is normally between 250 and 300 PSF. Rarely higher, and I've only seen lower once and it was in essentially a swamp. The clay was so soft in that instance.

 

[WARose]

It's typically not done for shallow foundations (that I've seen). One big problem is: you typically don't wind up with reliable side contact to mobilize that friction resistance. They remove the forms......and good luck on what you have on the sides.

 

[1503-44]

Agree for piles. Other than point bearing, skin friction is the only action you'll ever get out of them. Even still, neglect the top couple of feet.

The other problem with Gulf Coast gray, slick, oily, clay is, being that it is high plastic, when its dries in the top 3 feet (it happens sometimes), it pulls away and you've got nothing.

Statements above are the result of works performed solely by my AI providers.
I take no responsibility for any damages or injuries of any kind that may result.

 

[haynewp]

I don't understand how you would come up with a justifiable skin friction resistance in psf for this case anyway. We get that from geotechs with unformed deep foundations after a certain depth. Otherwise, we use a friction coefficient to resist sliding of shallow foundations combined with a normal force. Unless you just assume a psf applies based on some experience in the area without a geotech, which seems risky.

 

[Okiryu]

Thanks everyone for the great input. I am a geotechnical engineer so we provide soil parameters for uplift resistance to the structural engineer so I wanted to hear from other SEs the approach using the soil friction contribution to resist uplift forces in shallow foundations. If soils get softening/loosening under seismic loading, I would not recommend to use soil friction but in this case since the soils are competent, I recommended to use it to the SE. Thanks again !

 

[JLSE]

I dont think friction on the side wall is valid. It could as easily lift, right? friction is determined based on the normal force....... and how are you quantifying that? The uplift is on the side that the structure is moving away from the soil, I presume.

 

[haynewp]

Yes JLSE, the soil structure interaction is questionable at the interface under seismic with the push-pull action against the base structure with a different period of response than the soil. That’s Ignoring the vertical seismic movement of the soil at the same time which complicates it more. I would make it work without the friction theatrics.