Buoyancy Resistance of Underground Tank 1

[JAE]

I've got an underground tank (reinf. concrete) that we've initially designed. The design water table is assumed at grade level. There is about 3 feet of earth over the tank lid.

The base slab thickness was calculated to provide enough weight to counter buoyancy uplift of the overall tank as there will be times when the tank is empty.

We included the weight of the tank walls and lid as well. We wanted to use a 1.5 safety factor against uplift. We also included the weight of the earth on top of the lid.

The base slab extends out from the perimeter walls a small amount but we did not include any weight of soil directly above these lips.

The question is, if we extend the base slab further out to create larger lips, can we include any weight of soil above these lips to help resist uplift forces from buoyancy?

My initial thought is no - since the soil would be fully saturated with the water table at ground level and the soil would have a highly unknown shear capacity - much like a tank submerged in quicksand. However, I was curious what you all thought about counting on some amount of "help". This would allow a bit smaller, more economical base slab thickness.

The soil is comprised of various strata of lean clays, sandy silts, some trace gravel, N counts about 10 blows/ft.

 

[fattdad]

you can use the effective unit weight of the soil above the concrete mat, where it projects beyond the tank footprint. Some folks also consider a wedge that actually envelopes a greater prism of earty. Conversely you can consider the shearing that may occur along the vertical projection.

hope this helps.

f-d

¡papá gordo ain’t no madre flaca!

 

[cvg]

it seems like you are using fairly conservative parameters with the water table assumed at the surface and 1.5 FS. What is the probability that water will ever reach the ground surface?

I might be a bit concerned with the fact that you include the roof and the cover. How will you keep the tank from floating during construction (of the roof), especially if your dewatering pumps fail? Do you have at least 1.1 FS for this case?

Can you include soil weight that is saturated? Sure, it still has some weight, even when fully saturated and submerged. But you shouldn't be trying to calculate shear capacity to keep the tank from floating.

 

[civilperson]

I would not use the soil over the lid, but I would use the weight of tank height of soil over the exterior lip. 1.5 is quite conservative SF. Try 1.1 or even 1.0.

 

[VoyageofDiscovery]

Check with your geotech, I usually use ~30 deg from vertical based at the edge of the lip.

HTH

 

[SlideRuleEra]

JAE - As you know, I'm not a geotech, but am quite familiar with underground structures in soil with a true high water table (really at ground level). IMHO, use your conservative assumption of only the soil directly above the slab.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[VoyageofDiscovery]

JAE,

I guess this is where your judgement comes in

VOD

 

[hokie66]

Just another opinion, I would use all the mass inside the edge of your base slab extension, with no wedge outside this. My tendency is to use about 1.2 as a safety factor against the most severe buouyant condition.

 

[JAE]

Thanks all.

I guess there is a diversity of opinions on this.

I don't see how one would say to NOT use the soil over the lid, yet DO use soil over the base lips. That seems quite logically inconsistent.

With the soil over the lip, (I'm not a geotech) it seems that you MUST have some sort of shear capacity in the soil to count on it. If the tank was in pure water, the water above the lid certainly doesn't help. If you then add soil progressively to that water, at some point you would MAYBE generate some hold down mechanism....I'm thinking of my "tank-in-quicksand" analogy. Where does the soil on the lip begin to help hold the tank down? How do you know, with fully saturated soil, that the soil had any ability to do that?

The soil over the lid is "there". It seems I could use it but again, if it was quicksand, I don't see how it would be helpful.

But with almost any amount of un-viscousness (is that a word?) I also see that the soil around the tank would hold the soil over the lip in place and certainly be a weight to be displaced should the tank want to rise.

help!

 

[Dirtguy4587]

JAE,

This is why, when considering the resistance provided by the soil above the slab extensions, you must use the effective weight of the soil (i.e. the total weight of soil minus the weight of the water). Use of effective stress is fundamental in geotechnical engineering.

W.r.t. the shear resistance of the soil, fundamentally there should be some shear resistance. However, if you do not have soil testing to rely on to calculate shearing resistance, I would suggest that using the effective weight of the soil block directly above the slab extensions would be conservative.

DG...

 

[SlideRuleEra]

I like to go back to Archimedes Principle and solve this type of problem the easy way; please see the attached file (assume that it is a three dimensional sketch).

1. Calculate the dry weight of everything inside the volume described by the dashed line (including tank contents, if any).

2. Calculate the weight of water needed to fill the volume described by the dashed line (ignore the volume of objects/materials that are inside the dashed line).

3. The difference is the submerged weight (or buoyancy - depending on the sign) of the volume of interest.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[oneintheeye]

use the soil over the toe extensions
Dont use soil over lid as this could be removed, along with the lid when tank is emptied for maintenance cleaning etc.

FOS 1.1 particarly if your assuming water at ground level as it cant get any higher!

 

[JAE]

Water can't get higher than ground level? Tell that to New Orleans.

 

[civilperson]

The height of the water is not a factor in buoyancy once you are submerged. One inch of water cover produces the same lift as one hundred feet of water over the top. Herewegothen has the same reasoning as myself for not using soil over lid, (it is not there during installation and maintenance.)

 

[hokie66]

SlideRule is correct, but I think you knew that. You just have to decide on a safety factor to use. I think herewegothen and civilperson are too conservative about not using the soil over the lid. If the water level is that high, it would be evident and would tend to fill the tank as the lid is lifted. Even supporting a crane to lift the lid would be problematic.

 

[nackra]

I concur w/ hokie66 and sliderule

this link will give you basics:

http://www.highlandtank.com/LIT/HT_Concrete_Deadman.pdf

 

[oneintheeye]

hokie i agree to a point but if the water table is high and maintenence is planned a safe system of work may be adopted that will exclude water from entering the tank, sheet piling maybe or pumping. Therefore the ground water level could be at ground level without water in tank.

 

[Ussuri]

We used to take account the weight of the soil resting on the lip (skirt) directly above. We consider it as pure static mass using the submerged unit weight of the soil, thus for the tank to lift under a buoyant load it must also lift the soil above the lip.

I think in reality it is probably more complicated and you could look at different failure planes, shear resistance of the soil, skin friction between the tank and the soil.

 

[Ussuri]

This would assume that the lip is structurally strong enough not to snap off under the weight of the soil above.

 

[escrowe]

I would use the effective weight of the backfill above the 'lips,'
assuming that I could also recommend a granular, compacted backfill (or stone)
to reduce lateral stress on the tank.
If the owner then chooses to backfill with muck, uncompacted silt, soft clay, etc..
explain that a thicker base would then be required.
(money talks, BS walks)