Dear All,

Is it possible(practical) to put an axial (limit) stop, on an expansion loop? so as to isolate different parts of the piping (existing/new).

I have tried this in CAESAR II and the load on the limit stop is reasonable. Could you please advise me the practice that is followed.

Attached is a sketch of what I had in mind.

Waqar,
I have seen the application of limit stop for a similar situation in a power plant. there should not be any problem if the pipe stresses are within limits and if the load on the stop is also reasonable. U may be providing this stop to limit the load on some component.
But i also have one doubt that when modeling such stops and anchors whether we need to specify an SIF at that location, because here a flat plate will be welded to the pipe and therefore stress intensification is always a possibility. I do not know the general practise in this case.

Hi Waqar ,

Generally loops are highly stressed regions in a piping system.
So it would be a better idea to include atleast upto the next semianchor in the existing portion of your sketch while doing the analysis.

Happy working!

While non-conventional an axial limit stop at the top of the loop can be used to limit thermal growth in that locale, however be careful as it regards the design of the stop itself. You must address the pipe stresses as well as their interaction with the lug, trunion etc. See WRC 448 for instance....

Hi Waqar ,

Please consider lateral movement at line stop. High lateral movements at line stop, if accompanied by rotations can lock the lateral movement at line stop. According to me you are overstressing existing configuration slightly. Modeling friction in line stop may solve problem of stresses but I'm still not confident about the smooth lateral movement at line stop in this configuration.
Anybody with solution??

Hi Wagar,
Let me add my ideas as follows:
An axial stop on an expansion loop can be a requirement if for your pipe system has to be considered seismic load or other occasional loads like water hammer to avoid large unexpected displacements and stresses in such events. On the other hand the location of a expansion loop including an axial stop should be selected where the thermal axial displacements of the loop are small to avoid additional thermal loads between axial stop and anchors in the main pipe.

Expanding on uhloop's comment:

For dynamic loop design, it is NOT uncommon to have three (3), yes 3, limit stops on a loop. One as shown at the outer "loop width" and one each on the "loop height" legs.

Review the rotations at the support points and provide gaps to avoid locking/binding up the stops.

I also agree with Bob Zimmerman. It is usual to Provide such stops (with small gaps) in case of systems with very long loops & particularly with dynamic analysis when high displacements and very lower frequency modes needs to be arrested.

Can somebody explain how we can make sure (analyze) that the gap which one provides for free rotation of the pipe will not lead not 'slapping'of the pipe on the support under dynamic loading conditions ?

John C thanks for the WRC 448 reference. Is there any similar bulletin/paper for analyzing stresses at anchor plate and pipe junction ? or for say vessel flat end and nozzle junction ?

It appears to me that the current treatment of gapped horizontal supports for seismic analysis, provided the gap is less than 2 inches, is to have the restraint active for the seismic analysis but then double the calculated support load when designing the support. The restraint is active because the probable (?) contact during the event will "kill" that mode of vibration. You double the load to account for impact (dynamic load factor = 2).

Looking sideways.

Whenever i do loops in pipework (steam mains for example) i try to have the loop behave, by which i mean have the bending symetrical about the centre of the loop. Line stops here can be handy.

I am new to pipelines. I have a problem to providing expansion loops on a long pipeline. If I have the following:
1) Material (Hence allowable stress, coefficient of thermal exp), diameter & thickness of the pipe.
2) Temperature of the pipe.
3) Insulation information (Type, density, thickness).
4) the applicable code (B31.4 or B31.8)
How can I get the anchor to anchor length (between which I have to give one loop)using above information? Is there any equation / procedure? Or is it only some rule of thumb? Can anybody help?

I have the same problem. Would be nice to get some hint on that, thanks!

Read this;

http://65.57.255.42/ubbthreads/ubbthreads.php?ubb=showflat&Number=19506#Post19506

Regards,