Buoyancy of underground tank

[ars13]

I know buoyancy is crucial item to the design of an underground tank. I've been asked to design a tank that is 20'x40'x10' deep. I've been informed by the civil engineer that the "water table is well below the tank". Is that all the information I may need? Is that information enough? Is there such thing as say I need to be designing for a 100 yr water table? I would assume at some point the water table would be higher than the base of the tank. Thanks

 

[DaveAtkins]

Were borings taken to verify the position of the water table? A Geotechnical Report would relieve you of making this decision.

DaveAtkins

 

[ars13]

Thats the problem. I suggested a geo report but apparantly that is not possible within the time frame the documents need to be completed. I was told to "make some assumptions". So I'm gathering as much information as possible and using that in the design. Making sure I get my dummy notes on my plans specifying what design values where used. And like always, they never get verified.

 

[Ussuri]

12 x 6 x 3m, so not too big a tank. Can you design for the water level at the top of the tank walls?

 

[jike]

Have you contacted the local or state government or USGS to see what data they may have on a 100 year flood elevation?

Do you know what kind of soil profile that you are dealing with? Don't forget to consider the possibility of a "perched water table".

 

[NS4U]

Assume that the water table is at ground level and run the numbers, if the tank doesn't doen't float, you're all set... if it does then you've got some more work to do (a la the previous response)

 

[JAE]

Attached is a document we just used to base our safety factors on the uplift. One concept would be to estimate a "normal", "high" and "extreme" water table condition and check against different safety factors in each (see attached table).

 

[jgailla]

You could request a signed and sealed letter from the civil engineer stating the the high groundwater table is X feet below the bottom of the tank. I suspect they may not want to do this.

If the civil refuses, you could assume the worst, design the tank for a groundwater table at the ground surface, and see where that gets you, as Ussuri says. This will likely lead to an expensive design, and someone will make the decision to either go ahead with the more expensive design or do the right thing and get a geotech out there.

Is this a remote area? Why is it quicker to design a tank than for a geotech to get out there and provide you with some real information? This sounds like baloney. Somebody's making a stupid decision for the wrong reasons.

It's not in your interest to base your design on a vague, oral statement by another engineer. If you have documentation, you'll be much better off when things go wrong. If it's not written down, it didn't happen.

 

[ars13]

Thanks for all your help. Before I get bent out of shape on lack of information I will try and assume the water is level with the top of the tank (also verifying that the 100 yr flood is not above that) and design accordingly.

jike-the soil is mostly dense clay. With that said, I can't believe the water table is always going to be below the base of the tank as the civil engineer suggested.

 

[JAE]

Not too many geotechnical or civil engineers will guarantee or certify anything relative to future water table levels. Much too much variability.

 

[civilperson]

Dense clay is good material for helical anchors to resist the buoyancy. Very cost effective.

 

[jike]

A soil boring won't tell you much other than where the water table is at the time of drilling. Even if you place a permenent pipe in to measure the ground water and then frequently take measurements, you might not have what you need if it is a unusally dry year.

 

[cvg]

suggest a google search on this site for tank bouyancy, this has been discussed quite a bit in the recent past.

google search for tank bouyancy

 

[kslee1000]

Call the local building authority, usually they have all the historical datas for flood, and datas from other constructions under its jurisdiction. If there is potential for flood to occur, you have to check with your client (in writing) to determine the frequency to design for (50 yrs/100 yrs).

 

[wildehond]

I strongly recommend that you get the water table data in writing so that you can base your design off information that is solid. Else make a "water table at the top of the tank" assumption and do the design.