Pipe Support / Stress Question 2

[mylexicon]

I guess I should have directed this question to those who own a copy of “CASTI Guidebook to ASME B31.3”.
Anyhow, here is my question: Trunnions or dummy elements welded to a straight run pipe and serving as pipe supports will develop certain moments and respective stresses(longitudinal & circumferential) at the connecting point to the pipe wall. I am basically talking about the moments induced by the restraining loads at the other end of the trunnion (resting on or pinned to a structure)which translate to moments at dummy/pipe connection. The formulas, for calculating these local stresses and moments at dummy to pipe connection, extracted from the well known book “Design of Piping Systems” and simplified and then printed in P. 136-138 of “CASTI Guidebook to ASME B31.3” are often used for checking the stresses for these types of pipe supports.
I’d like to know whether these relations are also applicable/credible at Trunnion to Elbow connections? I acknowledge that there is a SIF value associated with trunnion to elbow connections. Does inclusion of the proper SIF value in the abovementioned formulas make their application to the latter case acceptable?
Thanking you in advance
drno

p.s. following are the relations for "Longitudinal Bending" ONLY:

f = M' / (pi *r^2) ; where M' = FL * trunnion length

S = [1.17 f * sqrt(Rt)] /(t^2)]+ (PR/2t); where sqrt(Rt) ? 2in (i.e. ; 2in minimum)

M = [2.685 * S r^2 * t ^(1.5) / sqrt (R)] – (PR / 2t)

Nomenclature:
FL = Long. force (parallel to pipe axis)
M = moment, in-lb
r = outside radius of trunnion, in.
R = outside radius of pipe, in.
P = internal pressure, psig
t = corroded thickness including reinf. Pad, in.
S = bending stress, psi

 

[DSB123]

There is a British Standard which covers this "Kellog" approach. I do not have it with me at the moment but from what I remember it covers the scenario you talk about. I'll see if I can dig it out to give you a reference.

 

[Frank1344]

HELLO,
i have a copy of the CASTI latest edition in my hand.
it does not limit the scope of using trunnions in only specific locations.
to me all the equations can be applied to the elbow support as well.

what i do is usually i model the dummy leg as well and i had cases which the dummy failed due to sustained and thermal load and i chose a dummy either bigger size or heavier in wall thickness.

hope it helps.

Frank.
Calgary

 

[NozzleTwister]

mylexicon,

Find below a link to another thread on Coade's website forum on the same subject for your reference.

http://www.coade.com/cgi-local/ultimatebb.cgi?ubb=get_topic;f=1;t=000928

Good luck,


NozzleTwister
Houston, Texas

 

[mylexicon]

Thank you all. I have also tried to contact the author of the book at casti@casti.ca and am waiting for his reply. I'll let you guys know as soon as I hear back from him.

all the best,
drno

 

[mylexicon]

NozzleTwister

Thanks for the link. I just have to say Wow.... This was like opening the Pandora's Box. I had no idea so many Pipe Stress specialists out there were trapped in a limbo on this issue. It made me feel better. Anyhow, it was highly educational. On the Coade forum, a gentleman named Aaron had kindly offered to distribute copies of the MW Kellogg papers on trunnion design by email; did you ever get a hold of such paper? If so, did it erase the ambiguities on trunnion design and credibility of the sources?

many thanks,

drno

 

[mylexicon]

hello again,
it took a while but finally the author of "CASTI Guidebook Series -Volume 3,ASME B31.3 Process Piping, 3rd edition" replied as following:

"It is true the MWK presentation is for trunnions on straight pipe.
However, I do feel the same equations and approach would be
applicable for trunnions on elbows. I believe this approach
for elbow trunnion installation would be conservative over
straight pipe in the sense that the trunnion, fitting on a curves
elbow surface, would offer a greater junction contact area which would
result in a greater area to distribute the stress. This would result
in a lower stress.

A parallel discussion could be said when comparing a 90 degree sif to
that of a 45 degree intersection sif. Because of the greater
area distribution offered by the 45 intersection at the run pipe
intersection,
the 45 degree intersection sif would be lower that the corresponding
size 90 degree intersection.

Glynn E. Woods, P.E."

The last couple of days, I spent a few hours reading older threads on this topic but didn't reach a definitive and full-proof conclusion. Evidently, stopping short of FEA methods, these formulas are the most reliable/practical if not the most accurate method of calculating local stresses at trunnion to pipe connections.

regards,

drno

 

[MechEIT]

I've been reading through MWK and Roarke and a host of other pipe stress literature (new to the field). What I'm wondering is, is there a significant reduction in elbow flexibility as a result of integral attachments to the fitting? Could this be handled by an altered flexibility factor for the elbow?