Pipe supports / shoes / clamps - degrees of freedom for stress analysis 3

[TomoB]

Hello,

Since the start of my career as a pipe stress engineer I can't crystalize the matter of whenever does the pipe shoe restricts rotations or not (and which directions).

1) Two major industry companies for which I have worked have guidelines that clearly points out for standard fix points (clamped to structure), for computational elasticity analysis, 3/6 restriction should be used - restricted movements, rotations allowed (does not absorb moments).
This is due to the fact that the clamp system is considered flexible enough to allow slight rotations.

https://files.engineering.com/getfi...-a8a4-4c7f36498734&file=Clamping_system_1.png

2) For the current project I am working on I wanted to use the same assumption, although the clamping system for fixed points is not the same (HILTI MIA).
Since the project is split between two companies, into two parts that are connected, I got a sneak peak into other side's stress model and saw they are modelling fix points as full restriction supports, that is 6/6 restriction.
I have contacted HILTI about this matter but didn't got the answer.

https://files.engineering.com/getfi...-9f25-780967f4a294&file=Clamping_system_2.png

Now I am rather confused as what is the correct assumption in the second case?

3) I guess the rotational rigidity of pipe shoe is also highly dependant on shoe design.
I also noticed that whether someone incorporates it in the calculation is highly dependant on the common practice in that company.

https://files.engineering.com/getfi...43e8-8ce4-d3cd4c48ea89&file=Pipe_shoe_1_2.png

Does someone have an advice when to model supports more in detail?

What are common practice and assumptions in your calculations?

Thank you for your time!

 

[KevinNZ]

If all your pipe shoes are fully clamped to the pipe and support beam (anchored) is there any point running the piping a stress program? If you think you need to allow some movement of these anchored points then the piping is over restrained and needs a rethink.

 

[1503-44]

Great answer KNZ.
I was trying to think how to answer it. You nailed it with one sentence.

 

[TomoB]

I don't know, maybe my question was not clear enough.
I never said all the pipe shoes are anchored. I don't see how that would be possible (T>300°C).

It is not the question of do I need to allow some movement of the anchor point, it is not about movements at all - it's about rotations and moments/torque.
The question is what is the correct approach?
When to model anchors (or better question is maybe which ones) as a fully restrained support, and when as a moments free support?

 

[saplanti]

Some companies use the term of Anchor for restriction in the axial direction. Some engineers invent funny applications sometimes but they cannot convince themselves and ask others to accept or confirm to use. Sometime this may be OK for limited applications. Sometimes a special support is essential for certain application, but this is not the case here I guess.
I suggest you to have a look at the PIP pipe support guidelines or check a company pipe support system to suit for your application.

 

[Joe591]

I'm not a piping stress expert and I do not do stress analysis on piping systems. I'm a construction supervisor. I take drawings, a team and material to a site and then build what needs to be built. I have been involved in the assembly of "data packs" for steam piping and have managed to have a quick look at calculations etc.

Our piping engineer models the stress analysis according to what he specifies. If he indicates a fixed anchor point on the drawing it means that we are to weld a piece of square sheet metal to the pipe and then weld the other side of the square sheet metal to a beam. In other words-if it is an anchor point we have to make sure that that point goes nowhere, we literally weld it to the wall, so to speak. This is to control how the pipe expands and in which directions it is allowed to expand.
If we are to use pipe hangars with threaded rod in certain sections then he will often model the pipe supports as little springs. In other words there is limited fixity.

Pipe shoes he often models as sliding supports in one direction (along the length of the pipe) and completely restrained in the other two axis.

If there is a long length of steam pipe and then an elbow followed by another long length of pipe he will sometimes specify that the first pipe shoe after the elbow should be installed without it's lateral restraints so that the pipe can expand freely. You do lose some support, but that is made of for by other pipe shoe restraints further down the line. The point is that how you model the pipe supports depends on what kind of pipe supports you use. You will have to get an idea of what the company is using and work according to that.

All of those pipe restraints that you are showing in the pictures are based on pipe clamps. Maybe the justification for modeling it as being partially restrained is that some slip is theoretically possible. Perhaps the other company ignores this and just assumes that it will never slip. Once again, I'm not an expert. Just my 2 cents.

Whether a pipe support absorbs moments or not will probably depend on the support as well. the supports that you show that have one pipe clamps will probably not absorb as much rotation as the ones that have two pipe clamps some distance away from each other. It's going to be up to you to determine how much rotation is absorbed though. With the pipes we do we seem to stay away from restraining rotation.

 

[EngrPaper]

So this question as I understand it boils down to what level of detail the support should be modeled in within the pipe stress software. Note that as I understand it in your pictures, the shoe is completely fixed to the base structural steel, so we are really talking about the shoe to pipe connection and what movement that allows.

For item 1, I agree that this likely fixes displacements and allows moderate rotation (neglecting restraint stiffness, etc.). It is essentially a fixed point support (3 displacements fixed, 3 rotations allowed).

For item 2, from the pipes perspective you basically have two of item 1 a short distance apartment from one another. Assuming that the shoe itself is stuff enough (something you would need to determine and feel comfortable assuming), then you have two point supports, so effectively a complete anchor when taken together. The amount of rotational displacement between the two point supports would be negligible enough that we could say it doesn't allow rotation. If I were to model that in a pipe stress software, I would take it as a full anchor. You could get into more detail by fixing the pipe with 2x #1s at their set distances, but I don't think it would buy you much.

 

[MJCronin]

IMHO, This question has been posted in the wrong forum .... It is not a general question about piping and systems design

Because it indirectly deals with the all-important topic of pipe support modelling and stiffness, it belongs in the "Engineering Computer Programs: COADE; CAESAR-II" forum

BTW, I have been involved in management and supervision of piping stress analysis for many years and have NEVER accepted a clamp-on piping shoe as an anchor.

If it is not welded to the pipe wall, it is not an anchor !!!

"In-line" anchors should be rarely, rarely used in piping design and stress analysis work .... They are expensive and may become points of failure.

The only times that they should even be considered is to mitigate stresses and loadings from events such as steam or water hammer or to meet nozzle allowable loadings ....

Additionally, I have seen their use in the rare circumstance where several attempts have ben made to meet vendor nozzle allowable forces and moments for a piece of equipment.

Also see this thread for the only proper attachment of in-line anchors to a piping system:

https://www.eng-tips.com/viewthread.cfm?qid=484314

MJCronin
Sr. Process Engineer

 

[DDDDMScEng]

I would be tempted to make a model of the clamp to find the clamp's flexibility. It doesn't have to be complicated. It wouldn't hurt to do some basic calcs by hand as a cross-check.

The complexities of software can prevent us seeing a simple solution. Once done, you will have a better idea of what's going on "because I've used a simple check-model and some cross-check hand calcs" you can tell anyone who might question your work.

HTH

 

[TomoB]

Thank you all for your answers, they were helpful.

Some companies use the term of Anchor for restriction in the axial direction.

-I agree, so can we say that:
1) anchor is a full restriction support
and
2)fixed point is partially restrained support (moments free)?

If it is not welded to the pipe wall, it is not an anchor !!!

-This is basically the answer I was looking for. Thank you!

it belongs in the "Engineering Computer Programs: COADE; CAESAR-II" forum

-I agree, my mistake.