Large welded steel tank foundation. 1

[BRIS]

We are looking to specify 20 million gallon circular welded steel water storage tanks on a reclaimed site with compacted granular soils. We need to make some preliminary estimates before we appoint a speacilist.

Diameter will be about 63 m and height 16 m. I expect the tank will be founded on a crushed rock foundation with an asphalt layer and a ring beam


1) I am looking for typical ring beam details.

2) What range of foundation settlement is normally permitted a) at the ring beam b) at the tank centre.

Any other guidance welcome

 

[JedClampett]

Take a look at AWWA M42. It has a well organized approach to tank design. However, at first glance, I couldn't find a typical ring wall design or settlement guidance. Instead of settlement, it seems as if the soil investigations are more capacity driven.
As far as ring wall illustrations and designs, see Gaylord and Gaylord, "Structural Engineering Handbook." In the first edition it is in chapter 23.

 

[JoeTank]

API-650, Appendix B provides some guidance and design recommendations for flat bottom ASTs. If your soil conditions for the reclaimed site are poor, the 16m liquid depth may need to be reconsidered.

Non-uniform settlement around the circumference of the tank will be a challenge to design for. As a result, there is a measure of atristy in the sizing and reinforcement of the ringwall. Detailing is extremely important as the wall will have a significant hoop tension force.

The bottoms of these tamk are very flexible; however, the roof structure will settle also. Therefore, you may need to consider a roof structure that has sufficent flexibilty built into it to accomodate the uneven settlements.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[BigH]

Do a search of Eng-Tips. We have gone through this several times in the last year. Big tanks with and without concrete ring foundations; offers of an old paper by Boberg; use of granular ring wall, bearing capacity, etc.
This topic needn't invent the wheel!

 

[tincan]

As a rule of thumb, design the ringwall to carry approximately the same total soil pressure as that of the maximum water pressure on the soil. Design the interior column supports for the roof framing at the same pressure.
At 63M x 16M (207' x 52'), the average soil pressure of 3,300 psf is not an excessive pressure.
Best of luck

 

[JoeTank]

tincan,
As I read your post regarding the allowable soil pressure for the column supports, it looks like one would end up with a combined soil bearing under the columns of 2 x 3300 psf = 6600 psf (3300 psf for the product load and 3300 psf for the additional column load). Is that what your are suggeting? That seems a bit high to me.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[BRIS]

BigH - Thanks I have picked up some of the earlier discussions on similar topics.

My field is hydraulic structure design and I have a MSc covering foundation design, my co worker is a structural engineer - we can calculate the settlements etc - it is the tank structural detailing and permissible limits we are looking for guidance on - The effect of settlemernt on comlumns and the typical allowable limits for roof deflection are of interest. I note from a previous thread that tank designers are often considering settlement measured in "feet" rather than inches!


we are in the Middle East and don't have ready access to a technical library but we have no problem getting publications shipped out once we know what to get - so advice on what guide or publication to request is welcome. –

Thanks Brian

 

[tincan]

To SteveBraune
From Tincan

Depending upon the dimensions of the tank and the spacing of the interior joist/roof beam supports and the column spacing, we have normally had a relative small roof loading to be carried to the foundation. For the lighter loads, we use base plates welded to the bottom plate with concrete footings under the bottom plate. For the 63M diameter (206'), you can vary the number of interior column rings to adapt to the soil conditions.

Yes, there is an increase over that of the direct hydraulic loading. We havent found that to be a problem, as of yet. (It helps to have good soils.) On substandard soils the design of the ringwall and the interior column footings would vary.

The settlement will vary with the soil and its condition when the tank is constructed. Testing of the soil is required before any estimate of settlement can be made.

 

[JoeTank]

tincan,
Regardless of whether loads are light or not, the column loads have to be spread out over an area based on the allowable soil bearing. Often the result is a base plate or an H-base assembly. My experience has been to limit the additional bearing pressure to about one-half of the product bearing pressure. I am aware that some folks have allowed the pressure to equal the product pressure, but I've always thought that was a bit too high. Seldom have I seen the use of individual footings under the columns.

Regarding the base plate or H-base assemblies, API-650 prohibits the welding of these items to the tank bottom. The normal practice is to leave them loose, but provide vertical guides. This allows the bottom to settle without inducing stresses into the bottom plate.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[BRIS]

Sorry to ask basic questions - basically we are going to put these tanks (total storage volume 100 million gallons) out for bidding under EPC contract. We are expecting that we will improve the foundation with vibro replacement piles (stone columns) but we are still waiting results from the SI.

1) I have now done some rough and ready cost calculations and I don't see a significant cost saving from using 1 x 20 million gallon tank rather than 2 by 10 Mg tank. I guess in the industry you have some rule of thumb on optimum tank size?.

2) I am interested in the discussion on column loads - again do you have a rule of thumb on optimum column spacing - I assume that optimum economic spacing?

Thanks for the contribution so far - Brian

 

[JedClampett]

I'm not sure exactly why, but most tanks I have specified have liquid levels around 30 ft. This is nice, because it makes the soil pressure a little less than 2000 psf, which is an easy allowable to achieve. Also, if the steel plates are in 8 ft. wide sections, you can use an even number (assuming a 2 ft. freeboard) and not do a lot of trimming.

 

[JoeTank]

Bris,
Don't worry about the number and spacing of columns. That's the responsibility of the tank contractor. If you set those parameters you will increase the tank costs as the tank contractor then has to move away from the economical details he uses as a standard. Your best place to find such info is to go to the experts... the tank contractors.

Regarding tank sizing... the owner will come to dispise you if you leave him with one big tank rather than two smaller ons. Two tanks gives the owner an option to maintain one of them while keeping the other on service.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[tincan]

SteveBraune
The loading under the interior columns and the exterior wall at the ringwall will vary from the direct soil loading as the liquid level of the tank varies. The pressures will very rarely be the same. We try to match the maximum pressures as best as possible to the allowable soil pressure and to each other. Have also used guides and welded base plates. We normally slope the soil/granular base of the tank 1 inch per foot from the ringwall to the center support column for the differential settlement unless conditions dictate otherwise. I'm allways amazed at the dished shaped condition of the base material when we salvage/remove a tank.

Best, Tincan.

 

[BRIS]

Steve

It will be at least 2 big tanks or 4 smaller ones.

 

[jheidt2543]

Here are some reading suggestions you might want to consider. The construction details and soils might be different, but the discussions and monitoring setups could give you some ideas for your project. All are available from ASCE.

"Steel Tanks Performance Founded on Very Compressible Soils During Water Tests and Service Period" by Atucha, Rivas and Velasco, from "Measured Performance of Shallow Foundations", ASCE Geotecnical Special Publication No. 15. This paper discusses the settlement monitoring of a large steel tank.

"Investigation of Failures" by Gerald A. Leonards, ASCE, Vol. 108, No. GT2, February, 1982 This paper has a discussion of a failure of a large steel tank founded on piles.

"Evaluating Calculated Risk in Geotechnical Engineering" by Robert V. Whitman, ASCE Journal of Geotechnical Engineering, Vol. 110, No.2, February, 1984. This paper discusses probability in geotechnical engineering, with an example of the evaluation of reliability of earth berms around a tank farm.

The last two papers are also found in "Terzaghi Lectures 1974 - 1982", ASCE Geotechnical Special Publication No. 1.

Just some food for thought!

 

[BigH]

You also might want to check out Bjerrum's article at the London (1957) ISSMFE Conf. Talks about edge failures.

 

[BRIS]

Thanks for the responses.

Steel tank design is obviously an interesting field and not one that I have given much thought to before now - As a mater of curiosity how do you deal with roof expansion? - Here in the Middle East ambient shade temperatures range from less than 6 degree C to over 50 degree C. I would expect the exposed roof will exceed over 100 C when exposed to summer sun. The stored water temp will be 20 to 35 C. The water is potable drinking water so the roof has to be weather tight. The walls won't push out so I guess the roof has to lift ??

I am visiting the US next week and hope to pick up some of the above references.

Thanks

 

[JoeTank]

Bris,
The roof plate will heat up, expand and buckle upward. This is common occurance and is not a problem.


Steve Braune
Tank Industry Consultants

http://www.tankindustry.com