UG-16(c) not good enough for AI? 3

[ConstantEffort]

An AI is threatening to reject vessels made of SA-516 (i.e. subject to SA-20 tolerances) plate which are delivered within tolerance but thinner than the required thickness. I think I'm covered by UG-16(c), but I'm new to Sec VIII. Will someone double check my logic before I make a fool of myself?

E.g. Vessel requires 0.495 after considering pressure and corrosion. I specify SA-516 Gr 70 at 0.500in. Mill delivers 0.01in under, or 0.490in. This is less than the required thickness, but within the tolerance of 0.01in.

The way I read it, UG-16(c) applies and I can use the plate to the full design pressure for the thickness ordered.

SA-20 limits the under tolerance to 0.01in. 0.01in is less than 6% of 0.5in. So the tolerance is not more than that required for UG-16(c) to apply. Ergo, so long as the delivered plate is within spec, I can use it!

[Conversely, if my PO inexplicably loosened the under-tolerance to 0.011in, I would be SOL and the AI would be justified in rejecting the material.]

 

[weldtek]

Technically you're correct, and I'm sure I recall an interpretation stating so but, when a fabricator designs and orders materials as you have here, with basically no room on thickness, you could be setting yourself up for a lot of grief. For instance, what are you going to do when your helper grinds off a temporary attachment and inspection shows the material is thinned out by an additional 20-30 thousands, and in multiple locations on the vessel? I think most experienced fabricators would agree that what I just described is not uncommon.
Of course there's always App 32.

 

[bernoullies123]

ConstantEffort, I would have to concur with your understanding. What has worked for me in the past is to sit down with the AI and show them the clause of the code and ask them for their opinion. We never want to be wrong, eh! I'm sure they will agree and will be happy to sign off on the data form.

As weldtek has mentioned, your spec may be a little close for comfort, but hey maybe your fabrication practices are top notch and you can spec on the limit to win jobs. Well done.

 

[GenB]

Okay. Give it to him. Or her' take away the corrosion and you may be home free...

 

[GenB]

And specify required materials t. 0.490... No AI will refuse it... On the other hand if you had an AI like ours who will definitely will not sign I'd he sees no corrosion allowance to his satisfaction you may be out of luck

 

[ConstantEffort]

Okay, points taken. And taking the discussion one more step...

How is cutting it this close on nominal wall vessels any different than cutting it even closer on mill run minimum wall vessels?

In my short time in vessels, I've seen several such vessels. In my piping experience, min wall pipe was common, even in piping requiring an ASME stamp from an AI.

All this is from the point of view of a specifying engineer. I've not been with the actual fabricator seeking the stamp.

 

[GenB]

Good luck

 

[weldtek]

ConstantEffort,
I guess not much, but over the years I've seen several situations where designing too close resulted in issues for the shop. In one case on a very heavy wall vessel, just rolling it on turning rolls deformed a local section on the shell thinning it just enough to cause a problem. Why go there?
BTW, should you need the info for your A.I., there's at least two interpretations on this subject. VIII-1-86-67 and VIII-1-04-92.

 

[SnTMan]

To echo weldtek, some extra steel is often cheap insurance, both against actual physical problems and paperwork snarls, which can often be more diffcult to correct

Regards,

Mike