I couldn’t find any topics in CAESAR II forum archive related to applying friction co-efficient on guide supports. So I am posting this new topic.

For Y restraint + guide supports, normally I will apply the friction co-efficient on Y restraints only and I don’t apply it on guide supports. But some of my colleagues are in an opinion of applying friction co-efficient on guide support in addition to the Y restraints. I have done some trail run about applying friction co-efficient on guide supports.

In my trial model CAESAR output restraint report, the model without the guide friction co-efficient gives an axial force equal to Y restraint force multiplied by friction co-efficient. The model with the guide friction co-efficient gives an axial force equal to Y restraint force multiplied by friction co-efficient + guide lateral load multiplied by friction co-efficient. This is going to increase the axial force acting on the whole system.

It will be good to get some clarity on this subject.

I agree
In my opinion, if pipe movement to axial direction is excessive(such as 100mm) we should take a look at the load form guide friction(Or omit the guide at that point).
But off the top of my head, the load created by friction is important when it nears to the equipment nozzle or so. It you have excessive displacement near equipment nozzle, you need to think about it again.
Might it be the reason we don't care friction of guides? Let me think...

Hi MK,

Thanks for your response. We can't avoid the guides in our piping systems. But if we consider the friction co-efficient on guides, it is increasing the axial force drastically based on the loads acting on the guides.

Best regards

James

I am a bit conflicted about friction on guides. Without any deadweight, these restraints build normal load only as the pipe heats up. The friction algorithm uses the "final" normal load to set the magnitude of the friction force and there may be some axial movement before the builds to the point where the pipe "sticks". It's more pronounced where gaps exist as there could be substantial (axial) movement before the pipe "sticks" on that guide - axial movement not considered by CAESAR II.
Run it both ways - with and without guide friction - and figure the real response to be somewhere between.
In my (current) opinion, if you are using friction, use it on weight-bearing supports.

Usually we set clearance gap for the guide such as 2mm. So as Dave says, untill pipe sticks to the guide, load by friction won't occur even the axial movement is excessive. I do compare with 2mm gap or 0 gap for sure.

In one of my project, high temperature(560º celcius) riser pipe stuck to the guides and buckled due to lack of the guide gap. It won't move to axial direction because of the friction force was too high.

But, I've seen one of the HRSG vendor uses friction on guides so if the friction forces are excessive they are using teflon or graphite sheets. I think it's too much, but they may have an experience or reason for this....?

Hi,
setting 2 mm gap is too close considering practical applicability.
Mathematically it will be OK, but on site may be it is a problem.

However, whatever may the gap size is, we must provide a friction at the guide gap to have the result if the guide gap is closed.
If it is not closing (mathematically), giving rise to non convergence; then one can delete friction with a caution that the support is not in a close proximity of the equipment nozzle.
Simply not specifying friction at the guide gaps is not close to actual simulation. Try to provide boundary conditions as close as possible to actual situation.

Another thought behind this is, one may say that, after the line is fully hot or stabilized when in full operation, there is no friction. But there is a situation, when the line is being heated up progressively, giving rise to frictional forces; which will be passed onto the nozzle. & nozzle must sustain these too.

Search this Forum for the word "convergence", there are a number of discussions and ideas.