Large tanks on sand

[tj83]

I do structural consulting work for chemical company. A couple of their large storage tanks have double bottoms separated by a 4" void filled with sand. The reason there are double bottoms is the original floors were found to be corroding and now probably even has holes in them. Leak detection pipes are embedded in the void space and have recently began draining pure water. They have asked me to determine the source of water. The site is flat, approximately 1000 feet from a large river. Generally we find about 5 to 8 feet of clay overlying deep sand & gravel. The foundation consist of a concrete ring and sand fill beneath the tank. The tanks are about 70 diameter x 35 feet tall (million gal). The fluid is probably .95 specific gravity. My question is if they removed the clay beneath the tank when they were originally constructed them (1950's) and backfilled with clean sand creating a less restricted vertical path for water movement, would the fluid pressure or loading/emptying cylcle have an effect such that water is being forced to the surface. The elevation of the tank bottom is right at grade level and possibly a few inches lower. There does not seem to be a simple obvious reason how water is getting into the void space.

 

[fattdad]

Does the flow rate increase when the tanks are being filled? How recent is the installation of the leak detection piping?

It seems from your discussion that the tanks have been in service for a while. It's possible that this vertical flow gradient has existed for some time, but it's more apparent as focused flow now that there is an opportunity for the flow to find the leak detection system. That said, it would seem that there is some natural process to "drive" the vertical flow gradient (i.e., a portion of the tank is in a cut) or that this is related to the filling of the tank (i.e., the vertical flow is related to compression of saturated soils).

Interesting problem. . . .

f-d

¡papá gordo ain’t no madre flaca!

 

[tj83]

Thanks for the reply. How would a portion being in a cut affect the situation?

 

[fattdad]

If the cut were to transect the high ground water table (i.e., during this time of year), the water table would then readily migrate through the sand subbase of the tank. Having the leak detection pipes would then be a conduit for draining this water from the subbase and to "daylight".

Just random thoughts. . . .

f-d

¡papá gordo ain’t no madre flaca!

 

[BigH]

Can you install a standpipe adjacent to the tank - to see what the groundwater level is? Even if they are draining water, I don't see what the concern is - I don't see the tank "floating" - . . . or maybe it's too late at night for me.

 

[tj83]

We may try the standpipe. The pipes are for leak detection.

 

[oldestguy]

What about temperature gradients and water migtating up as vapor, going thru corroded bottom plates into the cooler upper plate area. Is your tank containment colder than soil below the tank at times? It it ever is, then look to vapor transfer in cold weather. It does amazing things at times.

Another mode for water movement is from non-salty zones to salty stuff, again related to vapor pressure. With the right medium to go thru, osmotic pressures can be many feet of head.

Is there any opening into the sides of the sand bed so that rain water coming down the tank sides can enter? Wind driven rain also can be a problem

 

[msucog]

Here's a random thought at 3 something in the morning...What sort of rate of flow are you talking about? Any chance that the rate of flow could pipe the sand layer or foundation soils out?
Also, (if i understand your scenario) is it general practice to construct chemical storage tanks (steel? since you noted corrosion) directly on the soil foundation? It sounds like the concrete ring is for "containment" purposes and not supporting the structure. Perhaps I misunderstood your description...

 

[tj83]

Oldestguy, I will check on the temperature difference.
There are no salty soils or substances involved but interesting information. Rain coming down the sides was one of my first thoughts before we went and looked. However, the bottom 8" of the perimeter is sealed with an industrial coating still in tact. Closer inspection may be warranted though.

msucog, Yes, the tank is steel and the concrete ring is for containment only. The flow rate was on the order of a leaking faucet the day we observed. I don't think it is enough to pipe the sand out but I will check again after we get some heavy spring showers.

Thanks for all input - good thoughts.

 

[oldestguy]

On the temperature thing, I suspect your fluid expands when warm and shrinks when cool, meaning that cold sides will cause a movement of cold stuff to the bottom of the whole tank. Ground under the middle is likely not to be cooled that well, resulting in that temperature difference, cooler fluid above warmer soil below.