Dear all,

I have seen in several support drawings that a typical erection gap of 1 or 2 mm is indicated between the pipe shoe and the lateral restraint in supports that are supposed to be lateral restraints with no gap. Some of my colleages tell me that those gaps must not be considered in the CAESAR calculations since they are negligible and do not exist during real operation (??). Which is your opinion? Those gaps must be considered or on the contrary are supposed to disappear during operation? Which is the reason for those little gaps?

Thank you all.

Hello Carletes,

We often work with construction tolerances of +/- 1/4 inch (6mm). That means that whatever is fabricated may be slightly off the drawing dimensions and still be acceptable to use in the project. This is an insignificant amount of "error" in the construction of piping systems and if all the lateral retraints along a run of piping had absolutely no "gap", there would be problems with alignment. It is a reality of construction that there will be some amount of dimensional variance from the drawing and the purpose of specifying a "construction tolerance" is to place a limit on the amount of variance that will be accepted by the inspectors.

Regards, John.

Carletes,

I disagree with your colleagues' comments. If the pipe support drawing shows a gap, a good field crew will build the support with the gap as close to the dimension indicated as they can get. My approach is to put in a nominal gap as shown on the standard support drawing or whatever gap I think I'll need (usually 3mm on metric projects or 1/8" on U.S. standard jobs). Sometimes I'll try different gaps to see the effect, but generally if a couple of millimeters makes a significant difference a more drastic redesign of the job is probably called for.

As Mr. Breen points out, we know the gap is not going to be exact, but keep in mind that the entire model is a theoretical representation of the final design: The installation temp is not going to be exactly 70 F; the operating temp will not exactly match what is on the line list; the piping and supports will not be built to the tolerances implied in the model; any piping system is going to have some degree of cold spring or other construction related stress; etc.

Regards,

How many angels are there on the head of a pin?

Gaps that are small ~1-2mm or 1/16" are tough to control during construction.

Then assuming you get that gap what happens if the gap closes during heat up to ~ 0.5mm???

Academic types split these hairs all the time... however IMHO unless your model incorporates all facets of the boundary condition (this includes local steel spring rates which may be significant) your fooling yourself by worrying about 1- 2 mm. (How good is your temperature data???) Have you run enough temp cmbinations to grab all of the worst cases??? etc., etc., etc.

The folks doing section III work will get all fired up about this but maybe thats one of the reasons nuke plants cost so much.

So unless you need a major gap... such as used in a box guide or limit stop quit worrying about it.

Horses for courses.

Gaps on guides on long runs of pipework which are keeping thermal growth stable need not be too accurate. However,gaps which are controlling growth such as limit stops need carefull consideration.
Two examples are:
1) Static analysis of Prv pipework with forces vectored into the analysis. The size of gaps here can have a huge effect on stress levels and nozzle loadings.
2)Equipment Calcs, in reducing loadings on pumps/nozzles etc, the gaps on the stops can greatly affect the forces/ moments etc.

So as stated in an earlier responce, i feel it is important to model the supports/pipework/gaps as accuratly as possible. Guides/linestops etc should be individually assesed for the criticallity of the gap.
If indeed the gap is crucial, the state clearly on the iso's and have it checked in the field.
In my opinion, 0(zero) gap supports should not be moddelled in Caesar. if a zero gap support is required for code compliance, then redesign the pipework as zero gap supports are impossible to engineer.

my twopeneth so it is!.

Thank you all.

I don't want to go crazy because a gap of 2 mm when, as you said, there are other many uncertities affecting more to the calculations but my doubt was about if those little gaps that I see lots of times in drawings of supports that must be restraints with no gap, are inevitable because of construction reasons and if in that case they should be considered or not. With your answers I see that I will prepare two models (with zero gaps and with the little gap of 2 mm) and I will take the worst scenario. Thank you.

After running a few jobs both with and without gaps, you will soon establish your own modeling criteria - where gaps are significant and where they are not.