I want to split pure thermal force and friction force by following load cases:

L1 = W+T1+P1 (OPE) : OPE_FRIC
L2 = W+P1 (SUS) : SUS_FRIC
L3 = W+T1+P1 (OPE) : OPE_NO_FRIC
L4 = L1-L2 (EXP) : T_PURE + FF = OPE_FRIC - SUS_FRIC
L5 = L1-L3 (OPE) : FF = OPE_FRIC - OPE_NO_FRIC
L6 = L4-L5 (OPE) : T_PURE = (T_PURE + FF) - FF

With:

T_PURE: pure thermal force
FF: friction force

Any comment?

My opinion is that this will provide numbers, but they'll probably be non-sensical except perhaps for simple systems without non-linearities or sign changes with/without friction.

If axial friction at node 1 prevents lateral load at node 2, whereas frictionlessness would produce a higher lateral load at node 2, it's not necessarily accurate to conclude that the F2(frictionless)-F2(friction) is necessarily due to friction at node 2, but rather friction interplay between nodes 1 and 2.

Expand onto this that guides may have gaps, +y supports may be lifted off from, secondary effects of lines moving in the opposite direction, etc.

You could end up with Fx at node 101 = -1000 units with friction and +1000 units without friction - thus friction must be doing 2000 units of force, right?

Well, no, the support isn't actually throwing 2000 units of force - that's just everything working in tandem.

What I can imagine from your reply is as follow:

Considering Stopper/Guide force at a support.

The case F2(frictionless) consists of the thermal force "T1" totally coming from the thermal expansion of the system. No the influence of friction


The case F2(friction) consists of thermal force "T2" and friction force "FF". The "T2" is not the same as the "T1" because it has friction influence which makes the thermal expansion of the system in this case different from the case of frictionless

That's why F2(friction)-F2(frictionless) doesn't give the correct FF value

Correct me if any misunderstanding

My nomenclature was confusing. Let's simplify further.

You have a system that consists of nodes 1010 thru 1040.

1020 is a bend in the horizontal.

nodes 1010, 1030, and 1040 are guides with vertical supports, and let's say guides with gaps, because it's easier to visualize.

It also consists of unknown piping from nodes 10 through 1000.

There's a pipe routing between 10 through 1000 that, with friction, you do not see contact the guide at node 1040, but you have friction loads in the positive direction.

That same pipe routing, same temperature, without friction says that at node 1030 you contact the guide and have moment reversal, and then at node 1040, you see the pipe move in the opposite lateral direction and contact the guide and have contact loads in the negative direction.

+500 vs -1000. The force due to friction at node 1040 is not (+)500 - (-)1000 = 1500. No - the force due to friction is clearly simply 500.

How about if the guide has no gap in the model?

Can I say 1500 is due to friction? Or still, be friction interplay between supports?

I think I know what I want now.

As per your comment: "it's not necessarily accurate to conclude that the F2(frictionless)-F2(friction) is necessarily due to friction at node 2, but rather friction interplay between nodes 1 and 2"

What I want to see is the friction interplay from other supports on the considering support. This helps to identify whether friction or thermal is the dominant effect when looking at a guide/stopper force

The title of this post should be: Splitting the pure thermal force and Friction interplay

Let me know if any comments/misunderstanding

I think we're on the same page, then.

Friction causes a difference in system behavior, and that system behavior dictates the loads on restraints, but it's not fully accurate to state that the loads on restraints are necessarily exclusively to friction.

Friction will typically cause the centroid of thermal expansion to move in a system with sufficient degrees of freedom.

If you remove sufficient degrees of freedom, you will get the correct answer that LX-LY is exclusively friction loads.

Thanks for the comments.