Analysis of Tubesheet when welded directly to hemi-head without skirt

[PVGuy]

ASME Sect. VIII UHX provides tubesheet analysis when a cylinder is attached to the tubesheet but not when a hemispherical head is attached to the tubesheet. I would appreciate any leads on some methodolgies already published on this subject.

 

[GenB]

I do not see a difference cylinder or hemi, your analysis
is on the cylindrical part "skirt", and is taken as a cylindrical part
The hemi is calculated separately for t or p,
ER

 

[PVGuy]

To clarify, I am talking about the absence of a skirt or straight flange.

 

[rmw]

Hemispherical heads used on high pressure Hx's are often attached to a "lip" forged or machined to the tubesheet. This is for, among other things, so that the head attachment weld and any necessary PWHT, since it is done after the tube expansion and/or welding is finished, will not sensitize the outer rows of tube joints.

It is also to properly transmit the loads of the head at pressure to the tubesheet. There is HEI and EPRI information about the radius of the transition of the lip to the tubesheet itself.

Quality manufacturers of high pressure Hx's use from 4-1/2 to 9 inches for this lip dimension. (And yes, I have seen HP heaters without any lip, fact stranger than fiction. Poor outer rows of tubes.)

So, wouldn't the "lip" mentioned above qualify for the cylindrical attachment that you are questioning regarding the tubesheet analysis, assuming we are not talking about the skirt end.

I don't know about this regarding LP Hx's using hemi head channels.

rmw

 

[deanc]

If you do not find a reference in the section,it falls under
U-2(g). You can still use UHX for the rest of the design if you choose and your AIA agrees.

rmw does have a good point. How do you plan to attach?

See Code Case 2429. There are updates for UHX coming out next year.

 

[rmw]

I thought they were coming out mid this year. What is coming out in July. A utilitiy consultant just involked something in UHX that is coming out this year that resulted in a slightly thicker tube sheet for my client.

rmw

 

[GenB]

It does not have a skirt, it is still cylindrical for calcs purposes.

 

[PVGuy]

I can assure from interaction with the code committee that the math assumes cylindrical attachment to the tube sheet by either bolting or welding. I am looking to a have a tubesheet integral with a hemi-head channel that does not have a straight flange, the curvature starts immediately. It may be conservative to approximate it as a cylinder with the diameter of the head; but I am not sure of this at this point in time.

 

[unclesyd]

I just conferred with the engineer that carried out most of our HX calculations. He said that he normally treated it as a very short cylindrical skirt, the same as an external floating head design. This was using a rabbit on the tubesheet to where a full thickness weld was possible.
We also had some hemispherical heads that had a short straight flange and attached to the tubesheet the same way.
We had two from a well known manufacturer that had the tubesheet inserted in hemispherical head a.k.a socket weld.