If adding a flange on pipe nozzle, R stamp required?

[wangzh2k]

Hi all,

A existing vessel has a pipe nozzle connection. Now, we want it to be changed to flange connection, cutting the pipe and adding a couple of flanges.

For this change, do I need a R stamp for the existing vessel and re-submit calculation and vessel drawing with added flange to NB?


Thanks in advance.

 

[vesselfab]

you are "altering" the vessel, not repairing.
you are modifying the pressure containment material.

you should:
notify your AI
peform calcs/sketch
assign weld procedure
do the work
test as prescribed by NB/AI
affix "R" symbol/plate
send in "R" form to NB if it is a NB vessel

 

[wangzh2k]

When I send "R" form, do I need to send the calculation with sketch to NB, or only "R" form?

Thanks again.

 

[jte]

wang-

My initial impression is that no, this is not vessel work, thus no R stamp.

Now, just to be clear: If I understand your situation correctly, you have an existing vessel with an existing nozzle which left the fab shop without a flange. This nozzle neck is welded directly to the piping. Thus, the scope of VIII ends at the weld prep of the nozzle neck where it connects to the piping (presumably B31.x). The weld metal itself is out of VIII scope. If you cut the piping at the original seam and replace it with a weld neck flange, the butt weld remains a piping weld, subject to B31.x requirements, not VIII or NB.

On the other hand, if you are cutting and welding to anything at the shell, you are in either a repair or alteration mode per NB.

jt

 

[wangzh2k]

jt,

After replacing the pipe connection by a couple of flange connection, one welded flange and welding will be within pressure vessel boundary per ASME VIII-1 definition.

The inserted two 150# flanges are bolted. One is welded in piping line. The other is welded on the weld prep of the vessel nozzle neck.

I still do not understand the added flanges to be within the scope of B31.X. Could you please give further instruction?

Thank you so much for your response.

 

[vesselfab]

jte

I read the orignal question wrong
I thought he was adding a small nozzle in an existing larger neck.

sorry

 

[wangzh2k]

jte,

Thanks for your post. Sorry for my poor english writing.

 

[dcasto]

We just finished doing this ourselves. The latest version of VIII really puts a dampener on this type or change. Our AI insited that we had to do an R stamp. The vessel is a distillation tower with the overhead lines and feed lines welded to the vessel and have bolted stand offs and they were flanged near ground level. The vessel is circa 1981.

In any case we had the code shop remove the slip on flanges and weld on RFWN and the R stamp cost $500. Just 3 years ago I did a similar thing as per jte.

 

[jte]

dcasto-

Can you be more specific as to which change in the 2007 ed you are referring to? I checked the "summary of changes" document and found only two changes to U-1, one fixing an incorrect reference while the other changed the hot water tank part.

wang-

Look up interpretation VIII-1-86-193 in Interpretations Vol. 23, page 319. Alternatively, -83-329 in Vol 17. If your AI is fussing about welding a WN flange to the nozzle neck, I'd suggest adding a one pipe diameter 6" max long pup piece to the WN flange, then welding the pup piece to the nozzle neck. This accomplishes your goal while being very clearly in compliance with U-1(e)(1)(a). There is no question that the weld metal is out of VIII, and it is recognized that some of the vessel material at the weld will be melted and refrozen in the welding process.

jt

 

[wangzh2k]

jt,

Thanks for your message. I just checked code and found the limit of pressure vessel boundary when external piping or other mechanical equipment or pressure equipment is to be attached to the vessel:

(1) for welded connections, the first circumferential joint
(2) for screwed connections, the first threaded joint
(3) for bolted flanged connections, the face of the first flange.
(4) for proprietary fittings, the first sealing surface

For my case, it sounds I cannot get rid of (3). The welded flange always include in the limit of pressure vessel boundary.

Thanks for your further posts.

 

[wangzh2k]

jt,

I understood your idea. I hope I can convince AI for this change. After I add a long pup, it will become case (1).


Thanks,

 

[wangzh2k]

jt,

Why the pipe diameter is 6" maximum? If larger than 6", is it acceptable?

Thanks in advance.

 

[jte]

wangzh2k-

For most situations, a 6" separation between circumferential welds should be enough... Just a rule of thumb, no specific technical justification.

As for your cases: If the vessel paperwork (U1 form) includes a flange on the nozzle neck, then, yes, you are in case (3). If the vessel came from the shop without a flange attached, then you are in case (1) even if you later attach a piping component commonly known as a "flange" to it. The main issue here is "What was the code boundary when the U stamp was put on the vessel?" If folks get uneasy with a directly attached WN flange, the pup piece should remove all doubt.

jt

 

[wangzh2k]

jte,

If an existing vessel without a flange attached (nozzle pipe only) even if I later welded a flange on the nozzle neck, the limit of pressure vessel boundary is still in the first circumferential joint, saying case 1?

Not the face of the welded flange?

I totally lost.


Thanks for your post.

 

[gr2vessels]

wangzs2k,
The limit (1) is the first circular weld of the nozzle neck (pipe) apart of the welded connection to the vessel shell or head, if you use a long nozzle neck made-up of several pieces of pipe, bends, elbows, etc. If using one single piece of pipe for your nozzle neck, then the bolting flange attached/welded to the nozzle neck becomes part of the vessels and the limit becomes (3). Welding a bolting flange to the vessel/nozzle neck, you are altering the vessel and you should follow vesselfab instructions.
gr2vessels

 

[jte]

gr2vessels-

Clearly, this is not a black and white issue. I read the (a) thru (d) of U-1(e)(1) as "or" statements while you are interpreting them as "and" statements. In nearly all cases with new construction, the WN flange is shop installed and thus it is easy to include the flange in VIII-1 scope in accordance with U-1(c)(2). However, I'd say that it is not mandatory to include the flange in the scope.

I would think that you'd agree that my pup piece recommendation would make the neck to pup piece weld "piping" since this is in direct agreement with Interpretation VIII-1-86-193 which I referred to above. Now, would you argue that the additional weld of the pup piece to WN flange makes the system as a whole more or less robust?

jt

 

[gr2vessels]

Personally, I tend to be conservative with the interpretation of the first circular weld;- any long drain nozzle of a vertical vessel/column, sticking out through the skirt IS a nozzle upto the face of bolting flange, case(3) above. The vessel hydrotest is also the hydrotest for the whole nozzle. I have had cases however, with the Client requesting to weld a cap to the nozzle neck at the first weld, to enable the column hydrotest and perform B31.3 testing of the remaining nozzle neck piping. Weird and narrow code interpretation. Further, this interpretation opens the way to weld the remaining nozzle neck on site, after the vessel erection, when full hydrotest is out of question, because it only counts as field pipe weld...

jt, the pup solution is correct. Again, the pup and the flange will have to be tested separately, as per piping code, but the vessel with the nozzle neck cut short will remain untested as it is considered unchanged. In reality, with or withot pup you are adding a new feature to the vessel, the nozzle flange, which you should test to prove the whole vessel integrity. That makes it vessel modification which require full documentation to prove that the vessel is safe for use.
This might raise a few eyebrows, but it's still better engineering than no engineering.
gr2vessels