Crack Control for Rectangular Concrete Tanks

[cuels]

I have been using the PCA Rectangular Concrete Tanks book to design some WWTP tanks, as well as, other tanks. I have developed some questions based on the procedure. I am looking for others experience to get a feel for the final design.

My concern is with the Crack Control req'd steel. I am required to maintain 3 5/8" cover on the steel when I compute the stress in the steel. This is due to the DeNeef Swellseal bead to protect the steel at the cold joint and make the tank water tight. The bead requires 3" of cover and the bead is to be 5/8" in order to maintain a waterproof joint. Because of this, my max spacing for the rebar to control cracking ends up being less than 4" (unreasonable). The only way I can see to fix this problem is to increase the size of my wall or reduce the cover on the steel. My tank is a small 12'L x 8'W x 12'H tank. I have been trying to maintain a 12" wall. It just seems excessive to increase the wall thickness to 1.5' just to increase the steel spacing.

The other problem I have is the buoyant force. The WWTP is located on the side of a hill, so I don't see how the soil could become saturated long enough to lift the tank, but the calculation shows that my FS is less than 1 and I have to increase the walls and slab to be huge to get the F.S. to even be 1. I don't see any other way to make the tank heavy enough to resist the buoyant force.

So, I guess my question, on this small of tank, do you go to the extent of installing helical piers or is there other justification for resisting the buoyant force? Also, is my crack control spacing out of control? I just need a feel for good practice to know whether I am not trying to make the tank too small, and should be looking at thicker walls, slabs, etc. Mass concrete or economy?

Thanks for your comment/sorry for the long explanation!

 

[kslee1000]

If you have more than 2" clear cover, use 2".

 

[kslee1000]

Use 2" in crack control cal. (if I remember correctly, the crack control criterion was developed based on it, the excessive cover has no bearing on this regard).

 

[civilperson]

Put the bead where it is required to be and the steel at 2" cover. Why put the bead outside the steel?

 

[kslee1000]

for uplift, I would try to make the base slab larger than the perimeter of the wall. Draw a line (35 deg from vertical - consult w/geotech at final design stage) and include the enclosed soil weight. Automated drainage system (sump pump & drainage fabric) may be helpful, but failure-prone (fabric could be clogged by fines that restrict flow, power outage..etc).

 

[LobstaEata]

Perhaps helical anchors could be used to resist buoyant forces? Cost is ~$1500 each.

Additional weight of concrete might end up being more economical though.

 

[cuels]

So, kslee 1000, I can still use 2" in the calcs even though it is more like 3 5/8"?

For civilperson, I would like to put the swellseal bead in front of the steel to protect the steel at the cold joint at the bottom of the wall. However, maybe I am being over conservative. I am assuming that the wastewater from the tank would leak into the cold joint and corrode the steel away. Am I correct in my thinking?

 

[kslee1000]

cuels:

Yes, that was our practice (for water retention projects). I remember there was a paper discuss that, unfortunately, I don't have a copy on hand.
But one argument I remember well is that with all steel at the same stress (tension) level, so the concrete around the steel, then how could the extreme fiber of a concrete section with more than 2" clear cover experiences same level of, or higher stress (that causing concrete to crack) than a section with 2" cover only (note both cases have the same lever arm "d", the same load capacity, but different depth "h")? I hope you can see clearly the logic behind this argument.

 

[kslee1000]

The argument above is similar to the ACI consideration of "Architectural Columns" vs "Structural Columns". The excessive area serves a function but structural. Hope this helps.

 

[civilperson]

Concrete is NOT water proof, even with NO cracks. The normal cover and density has a permeability to moisture that allows small amounts of water to transverse the thickness of the wall. This amount is not determental to the steel reinforcement since the initial coating of oxide prevents further corrosion.

 

[kslee1000]

I have no intention to debating reinforced concrete behaviors, there are many knoweledgeable persons handle that. Just want to point out the fact that water tight (actually dense - low permeability) concrete has been widely used in water front projects for many years. To counter buoyancy, and for better protection of reinforcing steel, the structures are usually massive with clear cover arranged 4"-6". That's the reason we investigated, and determined to use 2" for crack control calculations. We provided concrete keys/grooves and continuous water stops along all construction joints to lengthen the path of seepage. Some of projects (submerged powerhouse) are more than 30 years old, I haven't heard any structural problems yet. Please note the reinforcement were plain grade 60 steel, no coating, and no water proofing was applied inside or outside.

 

[oneintheeye]

dont get this, the calculated crack is controlled by the reinforcing and is a function of the position of potential crack with worst case being furthest from a bar. The calculated strain in the face of concrete will depend on the thickness and reinforcement amounts. The ability of the reinforcement (and hence amount required) to prevent cracking depends on how close it is to a potential crack, hence greater cover means its more susceptable and should be allowed for in calculations.

 

[kslee1000]

Another way for thinking:

For a section with constant d, As, constant bending moment, in the elastic range, which one would crack first - with 2" cover or 6"?

 

[oneintheeye]

possibly, but I would certainly use the actual cover and prove it. The strain would be less using full depth obviously, dont think that will automatically mean less chance of cracking. Not in terms of the rigourous requirements of water retaining sections. I'm UK based so not sure how code equations etc work states side.

 

[kslee1000]

I have no doubt UK handles crack control in another manner than the USS, which requires more steel for section with thicker cover while all else being constant as stated above. You may want to give USS method a try, then you know the confusion and delimma of designers here.

 

[oneintheeye]

its not just over there where engineers are confused.

thicker sections requires greater shrinkage steel. But shrinkage crack control is so dependent on site control designing to the required accuracy for 0.2mm crack widths seems a bit bizarre, to me anyway.

 

[kslee1000]

In the US, the crack control focusing on cracks initiated around reinforcing steel due to flexural tensile stress. On the other hand, T&S steel is a "service" concern. The two concerns are treated seperately in the ACI code.