ASME SECT VIII Hillside hole in eliptical head

[M1Can]

I am reviewing a pressure vessal calculation for a customer that has a connection into an elliptical head. Rather then the nozzle coming in perpendicular to the crown of the head it is coming in at 90 degree. Example the vessal is vertical and the connection is horizontal, perpendicular to the shell of the vessal.

The connection is like a hillside connection that passes through the knuckle of the head and ends tangent with the bottom of the crown radius. I have never seen a connection like this and I am not sure how to approach the calculation. The customer has submitted a standard reinforcement calculation but my feeling is that this is not applicable.

Any advice will be appreciated. I tried to attach a sketch to the post but I am not sure how.

Thanks in Advance



M1 Engineering

http://www.m1engineering.ca

 

[deanc]

See Appendix L L-7.7

 

[M1Can]

Similar to Appendix L 7.7 , except 7.7 shows a hillside connection into a shell. This connection is into an eliptical head and it is not entirely in the crown radius section of the head it runs through the knuckle radius , including a little of the straight ( that fits to the cylindrical shell) and extends throgh the crown radius to about the center of the inside of the crown radius.

I dont think example 7.7 is the same

M1 Engineering

http://www.m1engineering.ca

 

[chaulklate]

M1Can,

Best I can tell, it's a regular calculation as your customer has supplied. I just ran a similar scenario using Compress software [would be great to hear from Tom Barsh right about now ] and I got a favorable result. Without knowing your exact design paramaters, there doesn't seem to be anything to be concerned about. As the nozzle is in the knuckle, check to make sure they are only using minimum head thickness and not the nominal value in their calculations. My scenario is a 2" nozzle in a 48" head. If you are trying to put a 10" nozzle in a 24" head...you might have issues.

 

[19750412]

Hello M1Can
Please correct me, but I understand that the nozzle is located in such a way that its opening is exactly through knuckle area. If it is, you can not made standard reinforcement calculation, because nozzle is located in discontinuity of the head and therefore it is necessary to reduce reinforcement area. You can refer to EN 13445 point 9.7.

ARTI

 

[M1Can]

If anyone can give me some advice on how I can add a sketch into the forum it would be great, a picture explains it so much better.

This case the head is for an 8" pipe and the nozzle is 2" pipe.

Thanks

M1 Engineering

http://www.m1engineering.ca

 

[TomBarsh]

I can picture in general what you describe (the nozzle axis is perpendicular to the vessel/head axis). But where exactly is the nozzle located? "...ends tangent with the bottom of the crown radius", so it is like a tangential nozzle on a cylinder? ie: its bore is tangent to the crown of the head? Better yet, just specify the distance of the nozzle axis above the tangent line of the 8" head.

In any case, this should in general be reinforced as per UG-37 if an ASME vessel. ASME Section VIII Div 1 doesn't restrict nozzles from being located in the knuckle of the head (there are some previous threads here about this issue).

But like others have pointed out, the minimum Code rules may not provide all that's needed for an adequate design. To wit, UG-36(a)(1) indicates that the Code rules generally apply to openings only with a ratio of long side/short side less than 2.0. Your opening may exceed this ratio and need additional analysis. Also, do the limits of reinforcement extend beyond the tangent line? If so, how will you handle the reinforcing calculation in this case?

 

[chaulklate]

I could be wrong (it's happend before...here actually) but being a 2" nozzle, there is a good chance area replacement is exempt per UG-36, so having to reduce the reinforcement area as a result of a discontinuity would be negligible...wouldn't it?

 

[M1Can]

x x
x xyyyyyyyy
x x
x x
x x x x x yyyyyyyyyyyyyyy


My best shot at a sketch , the x are the head ( although it is actually eliptical) and the y are the nozzle

Thanks


M1 Engineering

http://www.m1engineering.ca

 

[deanc]

http://www.mech.uwa.edu.au/DANotes/pressVessels/shells/shells.html#top

http://www.pveng.com/Soft/Heads/Heads.htm

http://www.eng-tips.com/viewthread.cfm?qid=162943&page=5

http://www.cascasia.com/PDF/NP06-2004.pdf

See if one of these may help explain.

 

[doct9960]

M1Can,

Out of curiosity, why is the 2" nozzle connected to the bottom head like that? Is there a space restriction? Why not have an elbow connection from the center of the head like this
x x
x x
x x
x x
x x x x x
y y
y y
y yyyy
y
y yyyy

 

[M1Can]

doct9960,

Not sure how my customer ended up with this arrangement, we asked the same question but didn't get a good answer. Only that manufacturing on the tank has started. Apparently their competitor uses this arrangement and they followed suit. Registration in Ontario Canada requires PE review and stamp, and I want to fully understand and accept design prior to approving. But apparently my customer has built these for other provinces (that don't require local PE approval) and the local authority has approved design with no objection



M1 Engineering

http://www.m1engineering.ca

 

[deanc]

http://www.bonneyforge.com/

Have a look at "Elbolets" and "Latrolets"

May save a lot of work. You can use U-2(g)to calc.

 

[vpl]

Guys, though you appeared to solve the "adding a picture" problem, if you click on the "Process TGML" link in the Step 2 Options when adding a message, you will find instructions in how to add a picture.

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[kdale]

I've done a number of similar concections, including manways at the knuckle. I believe it can be calculated as if one the smaller pipe is connecting to the larger size at the outside edge.

The only reasom my company always avoided this type connection was the extra time and diffiucltn of the weld in this area. Also forming an accurate repad (when required) is very difficult.