Can anyone tell me how to model two 30 inch diameter pipes that are welded together to form a 90 degree bend, as in a mitered joint?A stiffener ring is welded around the joint at a 45 degree angle. This is commonly seen in ducting or large diamter pipe. I tried specifying the node as a bend, indicating 1 mitered cut, but it gives me an error saying it's not long enough to be a bend.

The first thing to do here is see if the bend is long enough without the miter. Make sure the two connecting straight pipes are long enough to contain the bend.

Next, review the article from <font color="#0000ff"><em>Mechanical Engineering News</em></font>, June 2000, page 16, titled <em>"Modeling Widely Spaced and Closely Spaced Miter Bends in CAESAR II"</em>. If you don't have a hard copy of this newsletter, you can find a copy on this web site at: http://www.coade.com/newsletters/menlist.htm .


------------------
Regards,
Richard Ay (COADE, Inc.)

I would like to add an additional thought.

When you do your Caesar II analysis, the stresses that will be calculated will be "beam bending (and torsion) stresses" over the gross section of the (big) "pipe" (a very large calculated section modulus). While useful, these may not be the limiting stresses in the design. The stress intensification factors and flexibility factors prescribed in Appendix D of B31.1 and B31.3 may not be very accurate with this big pipe (remember, A.R.C. Markl did the tests with NPS 4, schedule 40, carbon steel pipe and the resulting data was extrapolated down to NPS 1/2 and up to NPS 72 (quite a stretch) for use in the Codes). With your ring plate reinforcing, all bets are off!

With 30 inch pipe (especially if it has a relitively thin wall - 1/2 inch or less) you should begin thinking about local membrane stresses. You are right at about the limit of the capability of beam theory (beam models). Of course the type of construction that you are describing is very often used in wind tunnel and penstock design. Supporting this pipe also brings some challanges as the pipe will want to "slump" or ovalize. Also, for "branch connections" it may be appropriate to use methods more appropriate than the ASME "area replacement rules".

It may be useful to peruse the AWWA publication "Manual M-11, AWWA Steel Pipe - A Guide for Design and Installation". Also, if you have access to "Steel Plate Engineering Data, Volume 4, Steel Penstocks and Tunnel Liners", American Iron and Steel Institute, you can find some useful information here. These will give you a rational basis for the design of the ring stiffeners. I sort of remember that the Lincoln Arc Welding book by Blodgett (the brown one) also had some design help for this application. With D/t ratios greater than about 100, it may be prudent to perform a modified "Zick anaysis" to get some idea of the circumferential and shear stresses at the supports.

Best regards, John.

[This message has been edited by John Breen (edited January 31, 2001).]

[This message has been edited by John Breen (edited January 31, 2001).]

[This message has been edited by John Breen (edited February 01, 2001).]