ACI 350.3 Multicell Tank Question 1

[EngInternational]

In designing a multi-cell concrete waste treatment tank (approx. 9.5m x 4m) with five compartments in series (1m-1m-2m-1m-3m), the following came up:

In ACI 350.3, the definition of "L" (for use in L/H_L) states that L is the inside length of a rectangular tank parallel to the EQ direction. If a tank has multiple cells of varying size, does one use L for each compartment and, further, calculate Wi/Wl and Wc/Wl for each compartment separately.

Additionally, what are the realistic load cases that should be considered, taking into account construction works such as leak testing, backfill, with and without roof, etc.?

Thanks in advance for any replies.

Cheers.

 

[JedClampett]

First of all, there's a lot of detailing and design customs for concrete tanks which are better transferred through your supervisor or someone who's done tanks before. For instance, there's load combinations from ACI, and there's load combinations which experienced tank designers use. And realistic doesn't necessarily play into it.
As far as the seismic mumbo jumbo, it very seldom controls for me, so for the multicell case, I would just use whatever gives me the greatest load. For the other load combinations, I design for the tank full (not operating level) to either the passive overflow or the top of walls. The tank should be designed for being full with no backfill and with backfill and no liquid. As far as the roof, if they're going to leak test without it, you'll need to check that case too.

 

[Gumpmaster]

L is for each compartment I think.

Jed, seismic often controls in determining the freeboard. Most times you can't slosh out of the tank, so you need to have sufficient freeboard. Sloshing can often be 4 feet or more, so you get some pretty significant freeboards.

 

[JedClampett]

And I would say unless you have a passive overflow once you add that freeboard, you need to design for the water at that height (top of wall). And that's likely to control.

 

[EngInternational]

Thank you for the replies.

Mr. Clampett, I am curious why you say that the seismic "mumbo jumbo" seldom controls. (by the way, seismic "mumbo jumbo" is a newly found technical term that I will sprinkle into my vocabulary in the future.) What typically controls?

Can anyone confirm if Gumpmaster is correct that L is for each compartment. I think he is, but I'd like to be able to say that with confidence.

Cheers

 

[paddingtongreen]

You would not be wrong either way, so why complicate it. I go along with Jeb, fill the tank to the top for design, and with multi-cells "one empty, one full", sometimes openings in dividing walls are closed for repair work.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin