Tank ring wall foundation question 1

[suocsiv]

Which is the best method for a tank foundation(D=11,46m/H=12m) made with a concrete ring wall:

A)Design the concrete ring wall, as described in API 650 Apendix B.4.2,where the centerline diameter of the ring wall is equal to the nominal diameter of the tank??

Or

B)Design the concrete ring wall with a bigger ID so that the tank can settle on a 95%compacted fill in the ring wall??

In other words in the first case, tank shell/bottom plates joint, is resting on the concrete ring wall(bottom is resting on the compacted fill) and in the second tank shell joint and bottom is all resting on the compacted fill.

Anchor bolts are also going to be used.
No annular plate / bottom plates will exceed 50 mm out of nom diam. of tank.

Any advice or recommentation would be appreciated.
Thank you.

 

[BigInch]

I agree. Since the low oil prices in the mid 80's, and early 90's there has been a general exit of experience from the petroleum industry.

Although I would be interested to know how you avoid edge settlements greater than that at interior points, as what you say pretty much contradicts what every geotechnical engineer I've ever talked to has said.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[BigH]

Almost all the tanks we have done were in refineries or storage areas. Our clients always were good but were mainly big players.
Good point though, SRE, about using the ring wall due to other considerations - most definitely not, though, for tank support. It is more of a safeguard then than really for support. As we knew settlements would be more than rigid connections could take, our recommendations were always for flexible connections between the tanks and in/out piping. For large tanks the bulk of the lateral spread would likely be below any shallow ringwall in any event. Look at the horizontal pressures at the edge, at the mipoint to the centre and at the centre.
We have supported tanks on a 5 ft. granular pad takeng some 10 ft beyond the edge of the tank over some quite soft soils. I remember once where Dames and Moore got sued in that their tank settled some 40 inches - doubt it had a ring wall but although the tank was still serviceable, the owner removed it and installed a lot of piles and then sued. I agree that the ring walls seem to be one of those "we have always done it this way" or some are not used to founding without a footing. A granular ring wall would probably work just as good as a concrete one in many cases.
Interesting that some think that the edge of the tank will settle more than the centre. A tank base is not rigid - the base of the tank (large ones) will be such that it will act as a flexible foundation (whether with or without ringwall). Flexible "footings" settle more in the interior than at the exterior. See any text of elastic theory (Polous and Davis (1975)). The ring wall can help minimize differential settlements of the tank edge from one point to another along the edge and may allow it to be not as rigid as if not on a ring wall, but it won't help much for the total edge settlements, I would surmise as these are based on the loading of the stored liquid/silage.
A lot of good points in the thread - thanks.

 

[BigInch]

Easy... theoretically, there is more load at the edge. I think how much depends on if the tank has a fixed or floating roof. Fixed roofs must transfer the roof weight to the shell, then down to the soil, whereas floating roofs do have an inherently more uniform load, with only the additional shell weight acting at the edge, which is probably not going to be all that much pct-wise greater in most cases, but non-the-less is heavier at the edge.

As for regidity, the ring addresses that point quite well, taking into consideration that it is there only to equalize the soil load as much as possible, it recognizes that the system is definately NOT rigid.

Total edge settlements are not of much concern and can be allowed for by pipe flexibility at the connections and by preloading the tank farm soils before construction, remembering that it is always the differential settlements that cause the most concern.

A granular ring wall would probably work just as well as a concrete ring in many cases, as long as the granular material equalizes the underlying soil loading from the shell with the interior tank load going to the underlying soils, given there is no need to elevate the tank from the surroundings and interior lateral load can be considererd nil, or the padding can extend past the edge to the point where there is no effective lateral spreading and a resultant greater settlement of the shell.

Theoretically, I don't think anyone can argue that a uniform load is not be the most desireable condition for rigid or flexible foundations of any kind.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[emmgjld]

Most of my tank experience is with water tanks and small to medium sized petroleum tanks, fixed covers. I tend to use structural fill beneath and beyond the tanks. I also do a lot of overexcavation and replacement, many times with native soils, even silts and clays. The use of 1 or 2 layers of geotextile or geogrid is commen. With silts and clays the geo-products don't develope very much of their potential strength but, the improvement to the native soils is impressive. Many of my sites are 20 to 50 miles from concrete batch plants, over poor roads. The use of native soils, if possible, is desired.

The problems of settlement vary across different climates and the reasons for settlements. If normal soil consolidation is the issue, with very moist to wet soils, The center portion of the structure will 'crater', after everything has stabilized.

If the subsoils are damp to dry, the settlement will begin with whatever portion is getting wetted. Most tank sites have horrible surface drainage. The edges get wet first, so the exterior settles first. If the situation is allowed to stabilize, the center will still 'crater', probably. If the distress is caught early, edges are down and the center is up or 'hogged'. Usually repairs begin at this point and the ultimate soil settlement may not be seen.

In my case, I deal at lot with collapsible soils, thankfully less than 20 feet thick, except in the Rangely Area, where 70 to 100' is commmen. Rangely is the area where I have had the worse experience with poor initital construction and have earned a lot of fees doing repair. Collapsible soils can have erratic stratigraphy and are often wetted unevenly, so settlements are unpredictable.

 

[BigInch]

Most of my experience is the opposite, with expansive clays where settlement occurs first with drying, not with wetting and dishing occurs in the center of the tank as those soils tend to dry as they squeeze moisture down and out. For expansive clays, it seems like a ring wall might inhibit water entry into undertank edge areas, especially if treated and backed up with a vapor barrier.

You don't mention stripping topsoils and preloading to eliminate most of the initial consolidations before construction of the tanks ever begin. Is that something you do?

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[emmgjld]

I see very little Preloading as the projects are usually moving along very quickly. Overexcavation and replacement, often with native soils for a portion of or all of the fill, is more commen. Everyone is in a hurry. It is notable that the vast majority of the foundation plans are done at the time the Geotechnical exploration is being accomplished. Then changes are rapidly produced. No wonder problems of design and construction occur.

Your idea of the ringwall acting as a partial barrier is correct, in my experience. My usual problem with using the foundation as a barrier is the foundation/barrier is usually not deep enough and the moisture loss/gain by the environment is still exerting as large effect on the foundation soils.
When I construct moisture barriers, I try to get them 4 to 10 feet away from the foundation and at least 5 feet deep. For my climate, that seems to work. I am assuming you are doing what seems to work in your area.