Circular Plate Bending Between Two Pipes of Different Diameter w/ non-Axisymmetric Loading due to M

[maest]

I've been coming across eng-tips threads for years while searching for discussion of various problems at work. Finally got around to registering and posting my own and I'm hoping to get involved.

I know its best to include sketches but I'm hoping the question is straight-forward enough for me to verbally describe. An 8" pipe is connected to a 12" pipe with a circular splice plate of to-be-determined thickness between them. The 12" pipe can be considered fixed away from the splice plate and the pipes are simply fillet welded to the flat side of the plate. I'm primarily concerned with bending of the plate between the two pipes if the 8" pipe is subjected to bending moment in additional to an axial moment.

If it was just an axial load I would simply turn to Roark as a line load around the 8" diameter with a simple support along the 12" diameter. However, the moment complicates things and makes for a non-axisymmetric loading. I considered a conservative approach of setting the distributed circumferential line load, w in Roark as P/circumference + M/S_weld but for a high moment case it proved over conservative.

Just wondering if anyone else has any idea.

Thanks,

 

[JStephen]

I would use the approach you mention in that last paragraph. You could treat it in vast detail, and if it's a moon rocket or you're building a million of these, it'll be worth it, otherwise, you'll spend more time thinking about it than what a thick plate would cost.

Consider a plate with gussets in there as another option.

I can't tell from your description, but if the application permits, I think the plate would be stiffer without a hole than with a hole...ie, solid circular plate instead of an annulus.

 

[maest]

JStephen-

The plate is to be a solid circular plate and not an annulus. (Screen clip of the Roark case I was looking at attached.)

I may, of course end up going with a conservative very thick plate, but I was just seeing if anyone has come across a similar example and if they did not resort to FEM or a very thick plate what their solution was.

Thanks,

 

[prex]

Assuming from your screenshot that you have Roark's 5th ed., you can use the trunnion loading case 20. Unfortunately it doesn't tell you where the maximum stress is, I expect it to be close to r=b.

prex
[URL unfurl="true"]http://www.xcalcs.com[/url] : Online engineering calculations
[URL unfurl="true"]http://www.megamag.it[/url] : Magnetic brakes and launchers for fun rides
[URL unfurl="true"]http://www.levitans.com[/url] : Air bearing pads

 

[SAIL3]

Doubt if one can get a very accurate result to this problem and if deflection is an issue. So with that in mind I would tend to go with something that I have a better handle on. I would tend to add a ring on the 8" pipe about 4 to 6" from splice and add some stiffeners between it and the splice pl.

 

[rb1957]

how good is the assumption that the outer edge is fixed ? is the 12" pipe much thicker than the circular plate ?

can you run a couple of stiffeners across the plate, to carry the moment into the 12" pipe ?

how sure are you about the moment ? if it is "just" a fixed support in a model, is that real ?

do the two pipes over-lap ? so you could join with two plates (an annular one and the full plate) ?

is the solid plate superior to an annular one ? or is there "stuff" in the 8" pipe ?


Quando Omni Flunkus Moritati