Out of Roundness for pressure vessels

[izamil]

Thank yoy all for helping me in the past. I have yet another request for which I seek your help on.

I know of someone who came to asking the following Quote:

We have fabricated a vessel 40 meters long X 4.2 meters inside diameter with 16mm wall thickness. The vessel required PWHT due to service requirements.

The vessel has been designed to ASME section VIII DIV 1:1998 with 2000 addenda. Per UG 80 (a) (1) (2) permissible out of roundness allows 1% of the nominal diameter for an internal pressured vessel.

The maximum out of roundness is 102mm adjacent to a 36” nozzle. The above vessel is stated to be not acceptable per code, and we intend to contend this client rejection because PWHT is not a code requirement and we are stating that because he insisted on PWHT for service reasons, these are the tolerances that we have had to work to.

For your information, the vessel has many openings and stiffeners so the 2% addition of the inside diameter of the opening is a factor addition to the 1% out of roundness tolerance of the vessel. What this means is we have 42mm for the inside diameter of the vessel plus 24mm of the inside diameter of the 36” opening making a total of 66mm permissible versus actual being 102mm.

We require your assistance in providing as many possible solutions to satisfy the code.
Is there any way this vessel can be acceptable by ASME code referenced above?

Moreover, the client measured the oval dimension using a tape measure (top to bottom/side to side) and the 102mm was found in this way. We measured it radial (sweep board) and found a maximum of 52mm out of roundness.

We need all the help we can get.

End quote.

I would appreciate if you could help my friend by answering back to this.

 

[MarcoE]

Izamil,

In your email it is unclear on exaclty what you have there.
The ID of the nozzle is 36", what is the ID of the shell?

Marco

 

[spector]

Marco,
Please ignore Dikkala's response and attempt to answer the question. I noticed Dikkala didn't venture any suggestions, but instead chose to berate you.

It's apparent these folks have invested a lot of money and time constructing a vessel that may be rejected when in fact, it just might be servicable. It makes me wonder if they still want this device or if they are looking for a way out. In fact, the correct answer might be worth something to the builder. I would hate to have to scrap such an effort, and would consider rewarding anyone who might save my butt.

I suggest they make an effort to exactly measure and precisely document their findings and the methods used to determine them. If and when this goes to court, the guy with the best, most meticulous paperwork will win. Just make sure to include the part about the tape measure!

In cases like this, it's ok to cheat....I doubt this vessel is standing upright as it is 40 meters long. So roll it over 90 degrees and measure again. Enlist the forces of gravity as required to help you pass final inspection.

And good luck!

 

[prex]

It is not clear to me what relationship you intend to make between the dimensional deviation and the PWHT. Of course, if you could provide some data or evidence that the deviation was within the limits before the heat treatment, then I think you win.
Also I can't see how you can obtain half the deviation by a different measurement: the code is quite clear on how to determine the out of roundness, it is a difference of max and min diameters in the same vessel cross section. Can you better explain this point and also, qualitatively, how that section is shaped?
I join also the suggestion of spector: by turning and correctly supporting the vessel you could force the section to a better result... prex

http://www.xcalcs.com

Online tools for structural design

 

[MarcoE]

Izamil,


The numbers I'm getting are a little different.
(4200*.01)+(36*25.4*.02)= 60.28 mm permissible.

from what you describe, you would still be under the limit. This really sounds like the purchaser doesn't have the jack to complete the purchase.
I agree with Spector, roll it and try to make it work out.

Thanks Spector for the kind words, but it's prolly what I deserve for trying to help someone before totally waking up in the morning.