Vaccum stiffening rings that interfere with nozzles

[gsail]

I just started the design work for a replacement tray tower for a local refinery. They want it built pretty much as is but to meet their current specs and reuse existing ladders & platforms. This vessel has 7 vacuum stiffening rings for external pressure. Two of the rings pass through nozzles and one of the two actually passed through 3 nozzles. I have never designed a vessel that this was done and always considered it bad practice. Compress shows it as a deficiency in the report, but not as a failure. The customer specs say nothing about the practice, the vessel was built prior to the current company's ownership of the plant. Is this a viable method of placement of the rings? I am looking a moving the rings as an alternative but there are plenty of other interferences that will make that difficult. I have put a call out for our AI to get his views. Any of you out there dealt with something similar?

 

[MrTank]

Just a curious question, have you seen or verify with the customer that the actual vessel is interfering with the stiffeners or are you just looking at old drawings?

-MrTank

“I have not failed. I've just found 10,000 ways that won't work.”
-Thomas A. Edison

 

[gsail]

MrTank; You are right the drawings are from 1989 and hand drawn and questionable, but because the client wanted to re-use the ladders and platforms they had a third party engineering firm do a laser scan of the tower to verify clip locations. The resulting information we received confirmed the nozzle/ring interferences. Since I was not comfortable with the previous design, we have since decided to rearrange the rings and avoid the problem. It was a real headache to do because of all the other clips and supports that were also issues, but I will sleep better knowing we gave them a much better product.

 

[MrTank]

Nice job gsail. I would have done the same. It's more work but it is the right kind of work. How close were the nozzles and rings to each other? The third party firm might have produced questionable results too. I am not saying that they are wrong but until you visually inspect yourself that the ring and nozzle interfere with each other, then you can't satisfy yourself. Have you seen pictures that confirm the interference or was it just a report? But either way, good job with the better design.

-MrTank

“I have not failed. I've just found 10,000 ways that won't work.”
-Thomas A. Edison

 

[jtseng123]

Is the ring welded to the nozzle ?

 

[gsail]

Yes the rings were welded to the nozzles so I guess you could say the nozzle neck became part of the ring and it was continuous, I am not sure I would buy that. UG-29b says the ring must extend completely around the vessel except as in UG-29c which deals with internal rings. I talked with one of the ex-employees of the company that built the original vessel and he said that practice of letting a nozzle interfere with a stiffener ring was common in their shop.

 

[Karloss12]

I think the second paragraph of UG-29c is for rings in general (The first paragraph applies to internal rings only). I think Compress brings it up as a deficiency as it is unable to carry out the checks specified in the second paragraph of UG-29c. You have to carry out these checks manually. That is interpretation. Anyone got any improvements on that?

Some years ago I had to design vessel with rings interfering with nozzles.

I was a trainee at the time. My senior engineer instructed me that stiffening rings shall go through the centre of the nozzle or be completely clear of the nozzle. I don't know where this rule is stated. It may not be a rule and is just good engineering practice.

 

[innovation2]

Check the nozzle for interia values and verify that is geometery produces values larger than the ring materials. I have put rings through nozzles before and generally you will find that the nozzle is last thing to buckle. Otherwise you could run an FEA and verify.

Good Luck