PCA Circular Concrete Tanks 1

[Dave7705]

Help! I'm analysing an existing circular tank, open top with base fixed to the bottom slab. In "Circular Tanks without Prestressing" by the PCA, it is recommended (section 13) to combine and distribute induced moment from the base slab with moment in the wall in order to accurately predict maximum ring tension. However, for the life of me, I cannot visualize how moment in the base slab could exist simultaneously with moment in the wall - if the tank is full, the wall has moment and ring tension but the base slab is everywhere loaded in opposition to the soil reaction; and if the tank is empty the base slab will have moment but the wall will not be loaded. The PCA example is short on delivery of how the base slab in the example has moment...

I know you guys who do tanks everyday know the answer to this one - please share.

Thanks!

 

[JedClampett]

Think of the wall slab interface the same as you would for a cantilevered retaining wall. The soil reaction is reduced for some distance until the moment equlibrium is satisfied.

 

[Dave7705]

Yes, but.. I get it if the moment from water or earth pressure causes moment at the base of the wall. However, the PCA book indicates the base slab is under a separate moment which is additive to the moment in the wall. I can see a foot wide strip of base slab modeled similar to a strap footing, with negative moment at the center and positive moment at the center of the tank and at the base of the wall, but I can't see a loading condition where base slab and wall moments exist simultaneously - or am I just being dense here?

Thanks!

 

[JedClampett]

PCA publications are notoriously poorly written. I'll check when I get in the office tomorrow.

 

[Ussuri]

Possibly because your walls are heavier than the water? You have a UDL across the base slab due to the weight of water (1000 kg/m3) and a load around the perimeter due to the weight of the walls (2400kg/m3).

 

[Dave7705]

Thanks Ussuri - In fact, the tank full imposes a load that would overcome the wall weights. I think Jed's on the right track.

Jed, thanks for looking into this for me. I'm trying to get a copy of Batty's tank book, but my deadline for this study phase will overtake bthat effort. Any insight you can lend would be greatly appreciated.

Thanks again...

 

[JedClampett]

I've never used a moment from the slab like the example in Section 13 does. The only time I could think of a moment from the slab is when the tank is supported by piers or piles and there is a moment from the tank slab spanning between them. If you look at the example, they just pick a moment and never calculate how they get it.
We design for the moment from Table A-2 and that's pretty much it.

 

[Dave7705]

Yep - OK, thanks Jed. Confirms our thinking, but we were concerned we were missing something.

Thanks again...

 

[letrab]

Hi Dave,
The way I understand it you have three options at the wall base connection Sliding, all in hoop tension in the wall
Pinned base, water resisted by wall tension and vertical moment in the wall. The moment is obviously zero at the bas and consequently no moment in the base. The last option is a fixed base and the one you are concerned about. The wall moment is reduced as it is fixed at the base, this moment must go into the base. The base must therefore be big enough to develop that moment otherwis it will rotate a little bit and act as a pin.
The easiest way to analyse the wall would be to model it as a beam on elastic foundation with the base stiffness equivalent to the wall stiffness. You can then play with the support fixity and see how the moment and hoop force vary