Bulge in vertical storage tank

[witchdoc]

We have a storage tank (600000 gal capacity) that has been overstressed due to freezing and expansion of the contents. The original diameter is 60'. The lower shell course has swelled to 60'-10" diameter. The tank was built in the 50's, and drawings indicate it was designed and constructed to API 650. The drawings indicate the materials of construction are a A285 grade c material that is clad with a 1/16 inch thick layer of 347 stainless steel.

How do we determine if this tank is safe for continued service? Do either API 653 or 579 offer guidance on when to repair and how to repair?

 

[ColonelSanders83]

If the tank has indeed swelled in size this indicates the lower portion of the tank has undergone gross plastic deformation (the material has exceeded its yield strength). I have never come across any code allowing the continued use of components that have plastically deformed although my experience does not involve API storage tanks.

In this case I would read the original edition of the API 650 design and construction code to determine how to proceed with the repair. I would also disallow service (if it is still being used) until the issue is resolved.

 

[jmiles]

One thing, I would use the msot current code to deal with repairs, typically once you start to modify a tank or vessel you lose the grandfather clause of an older tank or vessel and you have to bring it up to the new standrds.

 

[JohnGP]

witchdoc,

API STD 653 provides criteria for evaluating pitting and general corrosion of shell plates, and repair techniques, but I think you will need to go to API 579 to evaluate the situation you describe. As its name suggests API 579 provides techniques for evaluating the fitness for service of pressure equipment that has some flaw or suffered damage.

I haven't had to use it, so cannot vouch for its usefulness in your situation, but it is a thickish volume, and for tanks certainly complements API STD 653 in a big way.

Cheers,
John

 

[786392]

If not incorrect

API 650 requires 1/200 limit ensured for

1) Plumpness as well as
2) Shell's Roundness

Probably these referenced Limit values get varied to 1/100 in case of API 653.

Hope this helps the way forward!

Best Regards
Qalander(Chem)

 

[JohnGP]

API 650 is for new tank construction though (but is also referenced by "repair" standards). The 1/100 plumbness limit in API 653 is for a reconstructed tank. To evaluate whether the tank can still be used as is, or whether modified operating conditions are required, due to the introduction of the general distortion described above, will require something like API 579 to be used.

That's just my opinion though.

Regards,
John

 

[proinwv2]

I am not familiar with these codes so please bear that in mind. What I am is curious about this situation. Considering the value of such a large tank, would the following offer a sensible way to evaluate the tank condition, if it is not contrary to a code?

Using NDT evaluate the shell at the largest diameter, as well as where it starts to deviate from its original shape (the change in slope), above and below the distortion. This will be three circumferential bands. The evaluation should look for the amount of reduction in thickness as well as structural damage in the form of cracks or fissures.

Is this sensible or practical, or is it merely late on a Friday evening after a difficult week? (-:

Paul

http://www.ostand.com

 

[witchdoc]

Thanks to all for the responses. I purchased a copy of
API 579 -- chapter 8 offers a complex maze of analysis methods that appear to lead to a non-linear elastic/plastic numerical model as the only method for analyzing this situation. Does anyone know of a good consultant who is familiar with 579 and that is familiar with this technique?

 

[JStephen]

The 1/200 limit mentioned above is a construction tolerance, and not intended as a chech for continued service of a deformed tank.

 

[witchdoc]

HOwever, api 579 fitness for service lists tolerances for atmospheric tanks in table 8-5. height/200 is the limit for plumb.

 

[786392]

If understood/derived accurately from OP's it is a matter of high distortion shaping into bulges of sever intensity and reference code was indicated API 650.

Accordingly both basic references were indicated for reviewing bulges coming in what range(s).

Now as regards API 579 the OP had reservations.

Moreover when we go into 'Fitness for Service' concept to setup the criteria;

then thickness ckecks and many other metal properties checks are also to be considered.Which may not boil down simply into Height/200 idea,I believe.

Other forum colleuges will improvise on the issue!

Best Regards
Qalander(Chem)

 

[jte]

If you're looking for consultants with experience - these folks would qualify. No, I don't work for 'em.

http://www.becht.com/

jt

 

[unclesyd]

I would start getting some budget numbers for replacing the bottom ring.

Is the bulge all around or just in some sections?

Where does the tank regain normal diameter?

 

[dcfrush]

Witchdoc,

Does the tank have a U-1, and if so was it built to ASME code? You'll need the joint efficiency value E from the U-1 (0.7 if unknown), and the maximum allowable hydrostatic test stress in lbf/in^2; use the smaller of 0.88Y or 0.472T for bottom and second courses, use the smaller of 0.9Y or 0.519T for all other courses. To calculate the minimum acceptable thickness for all courses use (2.6*(Height-1)*Diameter*Specific Gravity)/(Allowable Stress*Joint Efficiency).

579 has 13 sections: Sections 1 and 2 are introductory and Sections 3 through 11 are assessments of the following: equipment for brittle fracture; general metal loss; localized metal loss; pitting corrosion; blisters and laminations; weld misalignments and shell distortions; crack-like flaws; equipment operating in the creep regime; and fire damage. I've had the ASME 579 course but we only briefly discussed the shell distortion section and I haven't had to actually do an FFS for this kind of damage. However, if you want I could look into my copy of 579 I have at my office for more information.

If you want to go the 579 route you'll need some accurate measurement of the damaged area and some thickness readings. You can't take credit for the 347 liner for vessel thickness, since it's only there for corrosion protection, not for containing pressure, but you'll probably want to examine the liner for damage as well. If the integrity of the liner is compromised it will need to be repaired, unless the tank is in a different service now and no longer needs the corrosion protection.

If you farm this work out to someone they'll need this information anyway, and if you're having trouble finding someone I could give you the names of some of the companies we've used.

Good luck.

 

[witchdoc]

Thanks for your reply. After studying 579, we have commisioned a consultant to do a non-linear FEA per Chapter 8 and B1. The results of this analysis should establish safe working levels.

What sections of the code address the exclusion of the cladding from the structural analysis?

 

[dcfrush]

I don't know off the top of my head, but typically the liners aren't installed for pressure bearing capability and the plates are attached by fillet welds. If overlay was used you could take credit since the corrosion resistant material would then be integral with the tank wall.

Good luck with your repair.

 

[ColonelSanders83]

When I worked on ships we had two levels of noncorrosive materieal weld protection.

"Cladding" was generaly 3/8" thick and no credit could be taken for strength, the theory was the reduction of strength in the HAZ of the cladding balanced out the strength of the cladding itself. The weld procedure and NDE was also less stringent.

"Buttering" was usually greater then 1/2" thick and credit could be taken for its strenght as a ratio of its yield (create an equivalent thickness of the base material). The NDE requirements were quite strict and it was rarley used.

Don't know if this helps but the reasoning is sound.