Design of penetration through jackets

[LVSoft]

I have a jacketed vessel with 7 bar overpressure in the jacket and vacuum conditions in the vessel.
The nozzle in the cylindrical part is designed according to ASME VIII-1 Figures 9-5 d1) and 9-6 a).
The nozzle calculations should be performed according to UG37 etc.
The vessel wall is partly loaded by atmospheric pressure and by the overpressure in the jacket (7bar). Calculating the nozzle with 8bar external pressure wil result in quite high wall thickness.

Questions:
1. Is there a way to calculate the nozzle with a reduced pressure?
2. If I increase the diameter of the collar in a way, that it is greater than the effected length around the nozzle, can I then use atmospheric pressure for nozzle calculation?

Comments and answers highly appreciated

Best Regards

Andy

 

[SnTMan]

LVSoft, not sure about some of you terminology, but I am getting that a portion of your nozzle is enclosed by the jacket and is exposed to 8 bar differential pressure.

1) You must design each component for the greatest coincident pressure it is exposed to, per UG-21.
2) Again, not sure of the meaning of the term "collar", but if the jacket does not terminate on the nozzle, such as Fig. 9-6(b) for example, then the nozzle is presumably exposed only to the vessel internal vacuum condition.

Comment: It would seem that the portion of the nozzle exposed to the 8 bar is short, and not of especially large diameter such the 8 bar would not yield an especially thick wall, at least in carbon steel, but impossible to say on the given data.

Regards,

Mike


The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LVSoft]

Mike,

thanks a lot for your fast reply.
To clarify the problem see attachments.
The closure diameter on the nozzle T1 is surely greater than the influence length of the nozzle.
The closure diameter on the manway is within the influence length.

The manway sees partly atmospheric pressure and partly the overpressure of the jacket.

Best rgeards

Andy

 

[SnTMan]

LVSoft, OK, I think I understand your question / problem better now. I infer it is with the nozzle reinforcement, rather than the nozzle wall itself. I expect that the 8 bar external pressure condition is governing the main shell thickness and little excess is available to contribute to nozzle reinforcement.

Referring back to the second question in your original post, I think by "effected length around the nozzle" & "influence length" you are essentially speaking of the limits of reinforcement. If so, then yes, you can place your collar outside the limits, or putting it another way, restrict the limits to exclude the collar.

It seems to me that by doing so a large excess in the main shell thickness (A1 per UG-37) would be available for nozzle reinforcement since the required thickness within the limits is now due to 1 bar external rather than 8 bar.

Hope this is helpful.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LVSoft]

Mike,

thanks a lot. Your comment was very helpful.

Best regards

Andy