Remove 1-part of 3-part pressure vessel - Any special considered needed?

[mp08]

So we have a 'distillation column' that is made up of 3 parts (top, middle, and bottom). The three parts are connected by flanges. All parts are manufactured to ASME VIII and share the same National Board number.

We want to run a test where we remove the bottom part and blank off the end of the middle part so we only have a top and middle.

My question is what additional work / testing would have to be done for the modification? Since all three pieces are under the same NB# and are all ASME vessels, can we do the modification and then do a pressure test (not a true hydrostatic pressure test but just enough to make sure that the flange is holding) and be okay or do we need to get a certified inspector in and have updates made to the NB records? Since it's all flanged and no welding/cutting/grinding type of work is needed, I didn't think much needed to be done, but we want to be careful.

 

[moltenmetal]

If the connection between the parts is by flanges, and the flanges are standard ASME B16.5 or B16.47, then you have to do nothing. That assumes you've followed all applicable codes for the piping changes, HAZOPed the revised equipment to make sure it is properly relieved etc.

If the flanges are not B16 type but are designed flanges, the new "cover flange" becomes a subcomponent of the "new" vessel. I'm not sure how it would be registered, or tested...

 

[GenB]

Same NB no. is one vessel and any modification legally needs to be done by an R stamp holder and follow inspection procedures. Mechanically assembled units without modifications can be assembled and disassembled by anyone without inspection.
Notice that) n most US states the Jurisdiction has to authorize the repairs or alterations of any boiler and p.v.

 

[mp08]

So I looked at the flange and it was a designed flange (no ASME/ANSI stamps on it). So if I'm to remove part of the vessel and put a blank on it, that work needs to be done by someone with an R stamp and then have to figure out what the inspection procedures are?

 

[Duwe6]

". . . put a blank on it, that work needs to be done by someone with an R stamp . . " Yes. And the holder of the R-stamp will perform the necessary inspections and testing. It may end up being an R-2 Alteration and require a rehydro. Usinga stamp-holder gives you "one stop shopping"; they are required to perform all the necessary work, including PWHT, NDE, etc. Only thing left for the owner is touching up the paint.

 

[GHartmann]

You will certainly need some design work done on the "blank". Depending on the diameter of your vessel, a flat plate may require too much thickness and may not be available. The "blank" might then become a flanged and dished head that is essentially a code piece.

 

[sharmass]

It can be done under R stamp using NBIC
It shall be considered as Alteration as the overall length is changed that may affect the loading conditions. Refer 3.4.3(d)
Additional blank must be designed unless standard flange with suitable pressure temperature rating is used.
If the name plate was attached to the bottom part, it shall have to be transferred to another location
Discuss with your AI in details before proceeding.
Original vessel should have suitable markings on each part as required by UG-116(k)

 

[mp08]

Thanks for all the input. I was afraid that it was going to be more work than simply installing a blank flange.

 

[mp08]

What about just putting in a slip blank between the flanges? Would that still need the additional resources?

 

[StoneCold]

mp08
Why don't you tell us how big the whole vessel is and how big the individual parts are. What is the pressure rating of the three part vessel. Many people work on very large equipment where it would not be uncommon to invest thousands to get this done. Tell us what you have. What you are doing, and why you are doing it. What you hope to achieve. We might be able to help you a lot more.

Regards
StoneCold

 

[mp08]

The vessel acts like a distillation column off a reactor for glycol and water. It was designed so that the water evaporates and the glycol goes back to the reactor. We want to get rid of the bottom part (which is setup like a vertical heat exchanger) since the glycol just seems to settle down there and even if we are to empty it, we can't reuse it due to possible contamination. The exit piping to the reactor is above this bottom piece. We can't just repipe since the bottom of the vessel is lower than the reactor nozzle. We try to recover as much glycol as possible for cost savings. If we get rid of this part, more glycol can be returned to the reactor.

The entire vessel is 32in ID (which is why I'm thinking a slip blank may difficult) and about 42ft tall. The bottom section we want to take off is about 8.5ft long. The whole system is rated for 43.5 psig.

I hope this helps.

 

[moltenmetal]

I can almost guarantee you that it will be cheaper to have a short dished head with flange constructed than to try to do this with a blind. Either way the new component will need its own U stamp, which begs the question of how to minimize the cost of testing it. It's unlikely that the shop is going to have a 32" blind just laying around.

 

[StoneCold]

mp08
Indulge me for a moment. At my facility we had a column that was completely below the reactor that fed it. The vapor came off the top of the reactor and ran down the side of the column and entered just below the packing. Liquid was pumped from the bottom of the column back to the reactor. Due to the size we used a positive displacement pump that we adjusted the stroke on to keep the liquid pumped out.
You could do this many other ways, but instead of modifying the column you could just return the liquid to the reactor with a pump around loop.

Regards

StoneCold