fuel tank foundation design 3

[jrj]

My company is working on a military project with a 45' diameter fuel storage tank as part of the design. This is the first time we have encountered a fuel storage tank foundation design. The tank will bear on a concrete foundation around the perimeter of the tank and on grade within the ring foundation. Can someone lead me in the right direction as far as designing the perimeter concrete tank ring foundation, checking for overturning during a seismic event, and whether anchorage is required. The only material that has been mentioned so far is the API-650 and our office does not have a copy.

 

[Focht3]

Your foundation is called a ring beam. This type of foundation is quite common in the oil business, and the design is well established. It's normally a shallow foundation. The ring beam has to be designed for the vertical load due to the weight of the tank shell plus the fluid weight (of the portion directly over the ring beam, of course.) Wind and seismic must also be addressed. The ring must also be designed to resist the hoop stress caused by the horizontal loads induced by the fluid weight.

As I recall, the design is fairly straightforward. But I haven't done one in over a decade. There's a design guide you need to find and review; perhaps someone will provide the citation for you...



Please see FAQ731-376 by [blue]VPL[/blue] for tips on how to make the best use of Eng-Tips Fora.

 

[tincan]

AWWA D100-96 "AWWA STANDARD FOR WELDED STEEL TANKS FOR WATER STORAGE" has a section on seismic design.
ASCE " Wind Loads and Anchor Bolt design for Petrochemical facilities" would be two places to start. The 45' diameter will probably prclude overturning. Design the width of the ringwall/ring beam to have approximately the same pressure as the soil loading at the center of the tank. Backfill the interior of the ringwill with a granular material. Design the hoop stress of the ringwall for the surcharge loading of the fuel filled tank.
Check with some of your peers, see if any of them does standpipe design. If he does, he probably has a spread sheet for the seismic and anchor bolt design,
best, tincan

 

[johnd43]

Look in the discussion elsewhere in this Forum: "Concrete pad for tanks" or something like that. You can recognise it because it has 63 replies - all asking for a copy of the reference that I guess will answer your question. there is mention of seismic design too.

Good luck.

 

[tincan]

As an afterthought, when you backfill and compact the granular fill within the ringwall, mound the fill for a 6" to 8" rise at the center of the tank. Long term settlement is greater at the center than at the perimeter.

 

[1969grad]

Based on your stated company's experience and the way you phased your question it sounds like you "do not know what you don't know." It is a formula for an expensive disaster.

Suggest you hire a consultant civil engineer who is knowlegable in soils engineering and concrete foundation design. This is not "cook book" engineering.

 

[BigH]

Why are concrete ringwalls appearing to be so popular? I've worked on many tanks in refineries and storage yards up to 150ft diameter and none were ever on a concrete ringwall. We sometimes used a crushed stone ringwall for founding the tank rim. Praytell??

 

[Focht3]

They have been around for a long time - since the 1950's, perhaps before. Their use predates me -


I don't think the use of ringwalls has increased so much as many new tanks are being built. I suspect it's a combination of the life cycle and (hopefully) reviving economic conditions. And we should remember that many of the engineers who used to perform these designs are probably retired, so the cumulative knowledge about this kind of design is severely reduced in comparison to the design environment of the 1970's and 1980's.

Crushed stone ringwalls can also work, but can have problems with lateral spread - which can result in damage to the floor of the tank. And the viability of a crushed stone pad diminishes as the height of the tank increases. Imagine a 40 ft (12.2m) high tank filled with a fluid with G_S of 2.2 ...



Please see FAQ731-376 by [blue]VPL[/blue] for tips on how to make the best use of Eng-Tips Fora.

 

[BigH]

You might be right on some lateral spread if the clay is soft - if that is the case, the whole foundation solution would be more geared towards ground improvement. We've put 40 ft high tanks on Su=800psf clay before without even crushed stone ringwalls. I guess that in my experience in some big Ontario refineries, ringwalls have not been popular. As for the high Gs liquid - well, that is part of the design and such ringwall foundations might be more appropriate - but I would suggest that most storage tanks house either refinery products, water or other liquids of similar Gs. So . . . such ringwall use would likely be for specific design conditions and not "general".
Best regards and

 

[Bbird]

BigH,

My experience goes with Focht3. There are two methods; one with a concrete ring beam and the other is simply a thick layer of crushed stone overtopped by a layer of bitumen sand mix.

I don't think there is any particular reason why one is better than the other but the footprint of the former should be slightly smaller than the latter. Without a ring beam the crushed stone has to be extended wider than the tank in order to spread the load. What most people do not realise is that large settlments of a fuel tank is tolerated and the magnitude could cause a heart attack to an engineer who only designs buildings (in feet instead of inches). The materials used in the two designs are similar but their quantities are not and so a direct comparison is not on a like to like basis.

Any structural engineer can confirm that a RC ring beam can provide a stronger and more uniform support for the vertical wall of the tank.

An important consideration is to prevent contamination of the soil by the fuel in case of a leak and it is sometimes necessary to cast the tank farm area with a concrete slab to form an imprevious layer with a high bund wall along the perimeter. Thus the ring wall beam solution fit neatly into this application.

 

[tincan]

When the direct loading of the liquid(height x 62.4G) exceeds the allowable soil pressure, a solid footing is required to distribute the loading (liquid + wind + seismic). When the direct loading of the liquid is less than the allowable soil pressure the tank may be placed on a ringwall or a crushed rock foundation. The ringwall is usually required when the wind and/or seismic loading results in uplift of the tank. The tank is then secured to the ringwall with anchor bolts to compensate for the uplift.

 

[cvg]

tincan is right on. Water tanks in southern California are typically constructed with the reinforced concrete ring wall foundation and seismic loading is a real concern. Couple that with the requirement / desire to maintain adequate supplies of potable water after a major earthquake and the water tanks need special consideration. Our designs never considered anything but RC ring walls.

 

[BigH]

Wow - I guess that we've been out of in in Ontario for too many years, eh? All our Sarnia Refineries and Lake Ontario refineries . . . So the 150 ft dia tanks with no ring walls under the tank edge that have behaved fine for over 20 years now are what?? passe??
There is no question about the bunds being needed in tank farms for the spill protection - but to say that RC ringwall is a positive measure in case of rupture when the area is underlain by clay . . .
I still say that your choice of a concrete ring wall or crushed stone ring wall or no ring wall under the tank edges is dependent on the site conditions and the tank under consideration. I wouldn't put in the ring walls unless it was absolutely necessary.

 

[tincan]

The weight of the wall + that portion of the roof carried by the wall is considered to bear on an assumed annular ring portion of the floor plate. When that loading exceeds that of the allowable soil pressure, or when that loading exceeds that of the hydrostatic loading of the tank (height x liquid loading), a ringwall is normally used to distribute, or equalize the load between the soil pressure due to the direct liquid loading and the soil pressure carried by the annunlar ring. This tends to reduce the differential settlement and allows for a more equal settlement of the tank. Sometimes, you will see a stiffener angle or plate at the base of the tank wall to distribute the wall loads.
A large majority of petroleum tanks where the diameter >> height have no footings, the overturning due to wind and or seismic is << that the dead load resisting moment and require no tie-downs.

 

[BigH]

Thanks - but I don't ever remember this concept of edge loading being taken into account the way you indicate - the edge is &quot;tied&quot; into the tank - to use a very thin edge load holding up the tank wall when you have distribution and weight of the liquid - seems overly restrictive. Maybe for tension capacity of the plates, but not for bearing pressures.

I bring to our attention paper by Bjerrum on Edge Failure of large oil storage tank - International Conference on Soil Mechanics and Foundation Engineers, London, 1956. It's an oldie but a classic.