Allowable stress for Hoop stress 2

[donoharm]

Hello.

As per API620 Annex L, lots of requirements shall refer to API650 Annex E.
Dynamic liquid hoop forces are one of them.

Based on E.6.1.4-6, sigmaT shall be decided by adding sigma h and sigma s.
Which means, total dynamic liquid hoop stress is governed by static liquid pressure and seismic effect.
Problem is API620 tanks have the internal pressure and have another formula for design thicknesses.
The formula to calculate sigma h in E.6.1.4-6 is identical with para.5.6.3.2 (supported by para. EC.9.6.1).

My questions are,

1. Is it OK to apply the formula in API650 E.6.1.4-6 for API620 tanks?
2. If yes, which allowable stress shall be used? API620? or API650?
(As you know, the allowable stresses in API650 are much higher than those of API620)
3. Why para.5.5.6 allows 80% of the specified minimum yield strength for Carbon steel as a guide line for Annex L?
There is no explanation at all in Annex L regarding this value.
Carbon steel material has the lower allowable stress values than API650 and if 80% of yield stress will be used for the seismic design, this is too unsafe.
(it is nonsense to apply 1.33 x design loading condition allowable stress for wind design and 80% of yield stress for seismic design.

Thank you.

 

[JStephen]

I would take sigma-h in E.6.1.4-6 to be hoop stress from hydrostatic head and pressure, not hydrostatic head alone.
I would assume allowable stresses from API-620.
Paragraph 5.5.6 doesn't allow 80% of yield, it allows 133% of allowable but not to exceed 80% of yield, and in most cases, the 133% will control. For example, for A36, allowable is 16,000 psi, 133% of that is 21,280, but 80% of yield is 28,800 psi, so the 133% will limit the design, not the 80% of yield.

 

[donoharm]

I agree with you, JStephen.
Your point is very reasonable and makes sense. But I am wondering which allowable stress is used in the commercial tank calculation programs.
I think you are right because if the allowable stress of API650 will be used then there will be no impact on API620 caused by the dynamic hoop stress.
That is, calculation is meaningless.
But I would like to have some kind of authorities such as commercial programs, any articles, any books and etc.
Without them, even if this approach is very reasonable, tank vendor would keep API650 allowable stress since API650 formula will be used for the calculation.

 

[mariog123]

It is true that, under normal circumstances, the safety coefficients must be uniform thru a Code. From this point of view your last remarks make sense.
However, it is hard to believe that during a seismic event, a tank with the actual sigma-h at a value equal to "the basic allowable membrane in API 650 standard for the shell plate material increased by 33 %" will survive, whereas a nearby tank, API 620 complying, having (by chance) exactly the same sigma-h value in a course of the same material, will not survive because the allowable would be different.

About your remarks, I guess this story will end when LFRD shall be accepted instead ASD in API standards. It is my opinion that E.6.2.4 Hoop Stresses (which requires now an allowable as the lesser of— the basic allowable membrane in this standard for the shell plate material increased by 33 %; or (and?), — 0.9Fy times the joint efficiency where Fy is the lesser of the published minimum yield strength of the shell material or weld material) is rather a confused mix of ASD and LFRD concepts.