Concrete Ring Wall Foundation for Tank 1

[EireChch]

Hi all,

The project is a retrofit of an existing 12.8m high x 48.9m dia tank supported on a 0.45m wide 2.3m deep ring foundation. Due to the retrofitting there will be an increase in loads. I am doing a check to confirm the existing reinforcement provided is sufficient. This involves revising the existing structural calcs.

Due to the additional line load, the vertical reinforcing req'd is 0.02% which is well below that ACI318 min of 900mm2 (0.2%) that has been provided. No issues there.

However, due to the additional hoop stress, resulting from the increased surcharge acting above the ringwall, we are above the allowable tensile stress of the 35MPa concrete 3.43N/mm[sup]2[/sup]> 3.29N/mm[sup]2[/sup].

I am still satisfying the maximum crack width of 0.3mm from ACI-244

Maybe I am not understanding the structural theory between load transfer between concrete and steel but the calculation approach seems flawed as the entire hoop stress is applied to the concrete alone without considering the reinforcement. Wouldn't there be some load taken by the reinforcement?

In short I am looking for a way to make this work, within the codes and rules of course.

Any help appreciated.

 

[SlideRuleEra]

1) ...the calculation approach seems flawed as the entire hoop stress is applied to the concrete alone without considering the reinforcement. Wouldn't there be some load taken by the reinforcement?

2) In short I am looking for a way to make this work, within the codes and rules of course.

1) Must be an error, reinforcing steel is assumed to carry 100% of the tensile hoop stress.

2) Test to see if the actual concrete strength > 35 MPa. Since the existing tank/foundation are certainly >> 28 days old, concrete strength may exceed nominal.

Also, passive soil pressure outside the ring wall is typically ignored. Do the calcs to see how the amount of passive pressure needed to "pass" compares with total passive pressure available. If the percentage is low enough, you may want to take a closer look at existing conditions to see if reasonable restrictions on operation are practical.

...and there is always the possibility that the ring wall analysis fails. In this case all may not be lost... perhaps a new, concentric ring wall can be constructed outside (and in contact with) the existing ring wall to contain the existing ring wall.

http://www.SlideRuleEra.net

 

[JStephen]

The crack-width analysis above is not commonly used for this application, in my experience.
Also, to get from the vertical load under the tank to the horizontal pressure involves a fudge factor that can be varied somewhat. API-650 recommends 0.5 for normal fill or 0.3 for granular fill, as I recall. But if you have any basis to more accurately determine that pressure, it may be simpler than modifying anything. You may also have some leeway on load factors used.

 

[dhengr]

EireChch:
Consider wrapping the existing ring beam with post-tensioning. This will help control or improve the existing cracking, while adding any needed new tension strength. Then, cast a new outer ring beam, as much as anything, as protection for the post-tensioning. But, some of the final tensioning may take place after this new beam section is cast.

 

[Settingsun]

Why do you need to 'make this work' if you're still within the crack width limit?

Regarding your question about the share of load between steel and concrete, the steel contribution before cracking was probably ignored for convenience which is about 10% conservative for this reinforcement quantity. However there doesn't seem to be any account taken of tensile stresses due to restraint so the cracking stress is overestimated.

It's a little hard to comment without having ACI224 but it looks like a flexural cracking calculation has been applied to axial tension so some loss of accuracy. Is your job just to update the old numbers or should a new crack width calculation be substituted? EC2 perhaps.

 

[fel3]

Take a look at PIP STE 03020, "Guidelines for Tank Foundation Design" (see attached). As others have mentioned, all hoop tension should be carried by the hoop steel.

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[r13]

Eirechch,

The only issue you have with the calculation is the stress in the concrete is 3.43 N/mm[sup]2[/sup], a mere 4.2% above the allowable stress 3.29 N/mm[sup]2[/sup]. Am I right? As SRE pointed out before, after cured in ages, the concrete would have a much higher strength than the originally specified strength - 35 MPa in this case. Knowing the age of the tank, you can estimate present concrete strength, using readily available charts in any of the reinforced concrete design textbooks.

Also you have to keep in mind is the conservatism in the design codes regarding strengths, allowable stresses, amplification of loadings, after adding all up, is the 4.2% that difficult to accept? It's your judgement call.

 

[EireChch]

SRE - I thought it was an error too but the calcs clearly show that one check is the hoop stress is less than the tensile stress of conc. I was leaning down the road of why are we even validating this check when it seems flawed to also saying its less than 5% over so we can consider it passing. Bit of a judgement call as Retired13 points out and if you got a stickler of an approving engineer it could be an issue.

JStephen - If you can provide some references / code clauses that are more appropriate, I am all ears.

Steve49 - Yes our job was to update the existing calcs. Were geotechs so didnt want to venture into new calcs or confirming that the existing checks are ok. The approving organisation requested the update of existing calcs so we didnt argue. If you have

In the end, I discussed the minor mismatch between required and actual tensile stress and they company just reduced the loads to make it pass. Seems that there had been very conservative with the loads they had provided us.

Thanks all.