Torsion on Pipe

[bangerjoe]

Hi All,

i have a pipe with a eccentric hanger. No point describing it too much, i've got a screenshot!

the plate will be welded circumferentially both sides in a fillet. the pipe will be ridgidly supported (i.e. can withstand the torsion) each end.

just wondering what checks i have to do on the wall thickness of the pipe and on the connection. i note that it is not a full circle weld, only where the steel is touching *like a half moon shape*.

i have run in plate element finite element and it's all OK, but i'd be comfortable looking at some hand calcs.

Thanks, Joe

 

[bangerjoe]

not sure if it uploaded last time!

 

[SAIL3]

load, pipe dia & wall thickeness would be helpful in getting a more accurate feedback?

 

[bangerjoe]

Hi, wanted to keep it high level so theory could be discussed,but t=2 mm l=2m OD=100 mm and plate t=4mm. Suspends load of 40kn at E=100mm

 

[KootK]

Some stuff to check:

1) Combined flexural + St. Venant torsion stresses on tube.
2) D/t ratio to prevent local wall buckling.
3) Deflection and twist if your care.
4) I'd resolve the plate connection as two segments of weld on either side, as short and vertical as possible.
5) If #4 doesn't work out, you might have to look at it more closely as a Geoffrey Packer style CHS connection. See AISC/CIDECT design guides.
6) Use your judgement to check the plate in some fashion. Bending, shear, local buckling.
7) Edge distances to preclude bolt failure.

That should keep you busy and hour to two.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bangerjoe]

Thanks KootK. you're not wrong about keeping busy!

just wondering if you could help with #2? point in a direction to help me consider this?

Thanks again.

 

[KootK]

What country are you practicing in?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bangerjoe]

i'm in tasmania, australia.


I'm using 3990 (ASD) and AS 4100 (Limit States), but more than happy to learn new info!!

 

[dhengr]

Bangerjoe:
You said...“wanted to keep it high level so theory could be discussed.” That shouldn’t mean, make the problem look like every other college text book problem you’ve done over the years; looking to explain a limited number of (or one) concepts, in a section of a chapter of the book. You’ll find, after practicing a while, that it is really helpful to do sketches which are fairly well proportioned, with some dimensions and sizes, and to give enough meaningful engineering info. to really give the problem some context and a real world feel. An experienced engineer will have first impressions of the problem based on this provided info., and these can vary quite a bit based on the info. you provide. Your sketch without any other info. could be your 100mm dia. by 2mm thk. by 2M long pipe with the 40kN load; or it could be a 24" dia. by 12" long pipe with 100kips hanging out there 3'. Same sketch, but two pretty different problems; so which theory do you want us wasting our time on?

I think KootK has covered most of the basic points in his listing above. Add to his #’s 1, 2 & 3; give a little thought to how you support the 4" pipe at the ends, and what portion of the bending and torsional moments go to each support. Are they fixed or simple supports? We all understand the beam and bending moment and shear thing. But, re: the torsion what does twisting compatibility at the load pl. have to do with the problem? RE: Koot’s #’s 4 & 5, I’d add; conservatively, I’d resolve the torsional loading from the pl. (40kN)(100mm) into two lines of weld (reactions) about .75 or 1" long and 100mm apart, as a back span on a cantilever, about what he suggested. And, I would size the weld terminations somewhere near the size of these reactions. Then I’d size the girth weld to take (40kN)(100mm)/(weld arc length)(50mm), a shear flow per mm of weld. I don’t like the weld terminations on your detail, particularly the one nearest the load, acting in very high tension and some prying. There is just no nice way to terminate this weld on your detail. I would likely put a 30-40mm wide flange on the top of the load pl. so I could weld this flg. pl. to the pipe and protect the girth weld termination, which should stop short of the flg./pipe/load pl. corner. I might also put the same flg. on the bottom of the load pl. to take the compressive bending loads, pl. buckling issues. D/t is an important ratio as relates to the pipe buckling, take a look at a good Strength of Materials or Theory of Elasticity text book.

 

[bangerjoe]

Beware the self-righteous fool, eh dhengr?

Thanks a lot for your expansion on KookK's response - I very much like your approach with the weld sizing.

 

[KootK]

I don't have an Aussie steel code I'm afraid. Here's what the US AISC manual has to say:

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bangerjoe]

KootK with the goods! Thank you so much.

Regards, Joe.

 

[SAIL3]

I am not aware of any complete theory to cover the OP's loading case...there may be some empirical/test data available, bu, I doubt it....the eccentric load causes local out-of-plane moments in the pipe @ the conn of the pl to the pipe...in a case like this I would put a complete ring on the pipe @ this location and analyse the ring for these local reactions....