Invalid Intersection at a Bend

Posted by: Jack Kowalewski

Invalid Intersection at a Bend - 05/06/05 08:34 AM

I would like to model a pipe that turns in the horizontal(x,z)plane after a 90° bend at the end of a rise in the +Y direction. If the turn in the x,z plane(about the Y axis) is a multiple of 90°, the bend is formed correctly. If the the turn is a 45° or 30° (or any other I would think...) the result is a gap between the "out" side of the elbow and the "in" side of the following pipe segment. This gap results in a fatal error when I try to run the analysis.

Any help on how to model this situation correctly would be appreciated. Thanks in advance.
Posted by: Richard Ay

Re: Invalid Intersection at a Bend - 05/06/05 10:57 AM

Your problem is that you need sufficient length to contain the straight lengths from both bends. So if we assume 6" pipe, with a 90 deg bend followed by a 45 deg bend, the distance between the two TIPTs is:

delta = L1 + L2
L1 = bend radius = 6 * 1.5 = 9

L2 = R * tan (angle/2)
= 9 * tan (45/2)
= 3.73

So you delta between TIPTs is 9 + 3.73 = 12.73.

Try this:

Element 10-20, dia=6, wt=S, DY=5ft, bend=Yes
Element 20-30, DX=12.73in, bend=Yes
Element 30-40, DX=20, DY=20

Note, with this set of numbers, you can get the plot to look right with a distance of 10in between TIPTs, but then the Error Checker will tell you the element isn't long enough.