Heads out of tolerance

[mengnr]

I procured a number of smaller pressure vessels with torispherical heads. On QC, some of the heads are out of tolerance. The inside radius is nominally 24 in, allowed to be up to 24.3 per Sec 8 Div 1 UG-81, and actually measures up to 24.5. I would rather just use these if there's a way to show they're safe, rather than returning them and having to wait for new ones to be manufactured. I think they are a little over the required thickness, because I rounded up to a stock size and also to allow for any thinning during forming. I'm trying to avoid having to move into Sec 8 Div 2 because it's been a long time since I used Ansys... Any suggestions?

 

[r6155]

Tolerance is with the inside diameter D, not inside radius.

Regards

 

[mengnr]

Excellent point. I was unclear and maybe wrong in my description of the problem.

D = nominal inside diameter of the vessel shell at point of attachment = 23.5"
So the inner surface of the head can deviate + 0.294 / - 0.147.

Nominal *radius* of inside surface of torispherical head is 24". I had been thinking this meant the allowable radius was 24 + 0.294/-0.147.

Should the tolerance be applied to the diameter of the inside surface (48 + 0.294 -0.147)? I guess I was confused because in the next sentence, it refers to the knuckle radius...

 

[r6155]

Not use 48". Please read UG-81 again (slowly).
Use a template with theoretical dimensions and measure the head.

Regards