Now-a-days I have seen that few organizations are following a practice which advocates that after normal static analysis is over , same model should be copied and run shall be taken with making friction multiplier as "zero". The logic put forward by these consultants is , once the system has stabilized w.r.t. pressure and temperature , there is no friction and hence the practice. This logic is not digestable , since making the friction zero will not simulate the phenomenon described by the advocates of these method. It will generate a displacement pattern where friction effect is zero, and not the one after the system has experienced friction and stabilized.
Also another logic is , that even if the system has stabilized after stable P & T , friction would still be present. In fact the equilibrium state is achieved because of friction as one factor . ASME code is not addressing this issue. I feel that this practice shall be discontinued as it does not stand the logic.

I see it more as identifying worst case.
You should not take credit for a reduced (equipment) load that is caused by friction as you cannot assure that the friction state can be maintained.
There is also an issue with support design. The piper gives the normal load to the structures people and the structures people would then assure that their support structure can handle the friction load on their hardware. Accounting for friction become much more complicated if the piper is handing over load data that already considers friction.

1) Do not use friction to your design advantage
2) Mu is an easy number to type in but difficult to verify

I agree with both of your observations, but the logic put behind this frictionless run is as mentioned above and repeated below -
"once the system has stabilized w.r.t. pressure and temperature , there is no friction and hence the practice."
They want to simulate the phenomenon which is not what we achieve by frictionless run.

Also it is very difficult to define - "worst case". In the overall piping system where the friction is to advantage and to dis-advantage really can not be generalized. It is like the HP slogan " possibilities unlimited".

Moreover Mu values have been established over number of experiments and not totally without any base.

Expect COADE to take up this matter with ASME so as to have a uniform understanding and implementation in the industry.

If your system drags the support with it via friction, that friction load should remain, even in stabilized conditions, unless something allows the support to move w.r.t. the pipe. Vibrations will do it, and the sources of those vibrations could be from anywhere. Within the pipe, equipment attached to the pipe, vehicles passing nearby, etc.

With that said, I think this is a situation where end users are finding nozzle/piping failures, and this is how they've elected to address the situation.

I agree with Dave that we should not be using (high) friction as a design advantage, but it puts the burden on the analyst to show that it doesn't (or simply do the run without friction to prove it). I think it could be a difficult sell to end-users to add 20-40% more expansion loops or limits with high reaction loads because we ran a case without friction, especially when there are established racks with established piping routes that have been running hotter than "our" pipe for years.

We use an analysis whereby our seismic cases are run both with and without friction. We do this because about the only thing we have any certainty about is that the loadings lie somewhere between an analysis with friction and one without. Vertical acceleration under seismic action will cause both a reduction and increase in friction, meaning it cannot be relied upon. In a seismic event, this will vertical action will vary along a pipeline due to the wave action seen in ground motion.

Dear Guys, I am doing stress analysis for past 6 years in one company. In here when we do analysis we consider Mu value for all supports and we consider the same for nozzle loadings also. We have many problems in restraints loads as we are giving huge loads to civil department. We generally use design temperature and design pressure for analysis and again civil engineers are using factor of safety as 1.5 times as per local standard. When I give friction multiplier as zero I am getting loads in vertical axis only, other axis loads are getting zero in rest supports and I am getting loads only in operating case for guides and directional anchor supports in the system. My doubt is, when the pipe is getting displaced the force will be produced on concern axis due to friction. If we didn’t give that load to civil how they will design support based on friction loads.

In stress analysis whether to consider friction loads are not. If not as you guys mentioned earlier is there any international standard saying piping loads not to be considered by friction only flexibility of the pipe should only be checked with respect to friction.

Re: Ross Sinclair - one might be concerned with the "somewhere in between" is that our analysis turns on friction in all locations, and turns off friction in all location, but the extreme condition could be where some locations have friction while others do not. However, (our) structural engineers also include significant safety factors on (our) loads to offset a multitude of factors, such as this.

Re: Kart89 - if concerned, the pipe stress engineer should have dialogue with his structural engineer to explain what the pipe stress engineer has provided, what is capable of being provided, and what is actually needed. Ultimately it falls upon the responsibility of the structural engineer to understand what s/he has and to communicate what s/he needs to the pipe stress engineer.

Some under-design.Some throw safety factors at it. Some do what's intended by the code.