Ring wall and Tank Center Differential Settlement

[Casanova1]

I am designing a ring wall foundation for a 65 fr dia x 65 ft high tank. The soil is clay and bearing capacity value for flexible foundation is 200 KPA with 150 mm allowable settlement. However for ring wall bearing capacity value is 70 KPA.

My problem is that for the center of tank using the liquid height x by density of water, we are using flexible foundation bearing capacity having pressure close to 200 KPA and for the same having settlement close to 150 mm . The ring wall is sized to have bearing pressure less the 70 KPA allowed, therefore for the same the settlement is around 45 mm.

The differential settlement is around 105 mm which is quite high. The dia of the tank cannot be increased due to space constrains.

This is my first Ring wall design, can any one tell me what i am doing wrong. It may be elementary but there must be something i am doing wrong.

One of my seniors say's that the ring wall will settle the same as the center despite having a large base. Is it the case?

Regards,
Casa

 

[FixedEarth]

You can replace the upper soils below the tank center with compacted road base. Choose the thickness of a base that will give you, say 50 mm total static settlement. Then widen the ring wall footing and also place required road base thickness that will give you say 25 mm total static settlement. This will give you differential settlement of 25 mm.

By the way, 150 mm settlement is too high and may induce tilt. Your geotechnical engineer can run the settlement analysis for you. Ring wall will only settle the same amount as the center if you design it that way and not automatically.

 

[Casanova1]

@FixedEarth.. Thanks for the reply.... how do you suggest to acquire the road base thickness, using CBR value? can you give me an idea..

Regards,
Casa

 

[BigH]

I'm a contrarian - my question is why are you using a ring wall? I've been involved with many tanks in Ontario and elsewhere and none used ring walls and many have been constructed on soft to firm clays. There have been a number of threads in the past discussion tank foundations and the pros and cons of ring walls.

It would be nice if you had given the undrained shear strength; you have indicated a bearing capacity of 200 kPa (note k is not capitalized - capital K means Kelvin degrees).

Your ring wall will undergo more settlement than just the amount you compute for a footing on clay as the soil underneath the ringwall is also being consolidated by the flexible tank loading. What is the tank center settlement? (150 mm?). If so as you have indicated, the edge of the tank, using flexible elastic theory (Milovic) would be very roughly estimated at about 70 mm (about 40% of the center settlement).

Personally, I would put the tank on "tank pad" consisting of a well graded sand and gravel base (crushed road base course) - compacted to 95% modified proctor - say 2 m thick. Extend the material at least 3 m beyond the tank edge. - and do away with the ring wall. For a flexible tank bottom, differential settlement over the radius of say 80 mm/10,000 mm is approximately 0.008 angular distortion - which isn't that "large" for a flexible structure - remember that most angular distortions (differetial settlement) are for rigid or semi-rigif foundations or frame structures. At least this is my opinion.

 

[JedClampett]

Is this differential settlement for a steel tank that big of a deal? The bottom of the steel tank is nothing but a bunch of steel plates lapped and welded. They're relatively thin (5/16"), so it seems like that amount of deflection would be well within their elastic range.
Have you asked a tank supplier what kind of settlement their tanks can tolerate?

 

[BigHarvey]

Totally agree with BigH, 50% of constructed tanks have a ring beam and 50% don't. Boussinesq gives you settlement of the edge = 50% of settlement at center. You can reduce total and differential settlement with rigid inclusion plus didtribution mattress ( and, if necessery, geogrid or galvanized wire mesh in the mattress ).
Steel tanks do have spec for max settlement. A good paper on the subject : "Criteria for settlement of Tanks" by W A Marr, J A Ramos and T W Lambe, ASCE Journal of Geotechnical Engineering, August 1982.

 

[Casanova1]

Ring beam is a provided due to Mechanical Anchors requirement. The tank is in high seismic area, therefore requires anchors.

 

[fattdad]

consider reading this journal article:

http://cedb.asce.org/cgi/

http://WWWdisplay.cgi?41728

+1 on BigH's post too.

f-d

¡papá gordo ain’t no madre flaca!

 

[FixedEarth]

The 65 ft high tank is rather tall and seismic overturning may force you to use drilled pier foundation in conjunction with a ring wall perimeter. A mat foundation may not take the seismic overturning moment. For the interior, you can remove the upper three or four feet until you encounter very stiff soil. Make up that undercut with crushed rock and may be have a 4 inch topping sand. Since the perimeter is on deep foundation, settlement will be less than 0.5 inch and that is why you need few feet of gravel under the interior to limit the tank settlement in the center to 1.0 inch maximum. (therefore 0.5" differential maximum)

To determine your road base or crushed gravel thickness, take your uniform applied stress and apply Boussinesq stress with your 65 ft circle diameter and see for example the stress at -4ft. Then get the consolidation characteristics of the soils and see if the settlement in the center is 1.0 inch. If not, keep increasing gravel thickness until you hit it the 1" mark. If your calculations show unrealistic gravel thickness, trick the soils. Use 2 ft of low strength concrete under the tank topping sand layer and then gravel below that for few feet before the Clay subgrade. This will create a 2 layer system whereby the upper layer is more than 10 times as stiff as the underlying layer & this will drastically reduce your settlement in the center. The low strength concrete layer will in effect bridge it. If you do this 2 layer system, use Westergaard stress distribution and not Boussinesq since the "soils" are no longer of uniform consistency.

You may try

http://www.soilstructure.com/settlementanalysis.html

or ask your soils engineer to assist you in settlement computations.

If the Clay thickness is deeper than 50 ft, then liquefaction and seismic settlement will not be a concern. However, if Clay is only upper 20 ft then check for dynamic or seismic settlement for the soils beneath the Clay. Good luck.

 

[Casanova1]

Thank you all, for your value input.

Regards,
Casa