I am trying to model an existing slip joint in an 8" 15 psig steam line. It is over 40 years old and the manufacturer is defunct. A similar slip joint failed during a system startup after pipe modifications (designed by others), possibly due to a water slug resulting from insufficient trapping. Now the joints have become a concern.

I looked at the on-line help and modeled the joint accordingly, using slip joint data from a catalog, but I don't have a feel for if the output is correct. The data on-line did not show input related to pressure thrust. Shouldn't this be included?

Catalog data for calculating anchor loads:
Fc = 8000 lb packing friction force
Fp = 876 lb pressure thrust for 15 psig steam
Fs = 1275 lb pipe support friction per 100 ft pipe

I modeled:
K1 = blank (rigid)
K2 = 100 x N / a = 100 x 8 / 1.9 = 421
Fy = 8000, based on packing friction force per catalog
No thrust load.

Does this sound correct? I'm not sure if I used the correct values. Also, it seems thrust load should be included.

How does this model respond? My guess is you will see some thermal movement of the joint but nothing else.
If you wish to include pressure thrust, just add a force set. This force will have two components, one upstream and one downstream equal to P times inside area. I would place these forces at the upstream and downstream surfaces perpendicular to the axis of joint - typically at the elbows boounding this joint but a reducer before an elbow would share some of the load.
Based on your pipe size and pressure that load of 876 lbf will not bust the friction at the joint. And the support friction will also hold the joint. However, I do not think it good design to rely solely on friction to hold a force-based load such as pressure thrust.
Your value for K2 comes from our Help text. I doubt that has any impact on this analysis. If it does, you may want to see what happens when you increase and decrease that (uncertain) value.

Michele,

The slip joint would have different modeling factors compared to a more common convoluted bellows expansion joint that most designers might be familiar to analyze. It takes some thought about the degrees of freedom and packing friction behavior.

First, the easy point about pressure thrust. The Fp = 876 lbs would appear to be approximately correct for 15 psig steam. The slip joint could possibly be modeled as a guide restraint with a CNode connection between the inner/ outer pipes. A more exact model could place guide and CNode at each end of the inner / outer pipes of the slip joint plus small gaps for the guides to allow for angular deflection in the slip joint. There is axial rotation possible in a slip joint - something a bellows expansion joint would be essentially rigid in that axis.

The friction of a slip joint would be the most significant of the factors given. The 8,000 lbs of friction would act in opposition to the travel direction. During heatup the friction force is opposite to expansion, while during cooldown the friction opposes the contraction of piping. The bellows expansion joint would act more like a linear spring.

One approach to modeling the pressure thrust would place the 876 lbs at end elbows of the straight run of the pipe with slip joint. The friction forces might be placed at the inner and outer pipe ends of the slip joint. The direction of the friction loads would be opposite for the outer pipe from the inner pipe. If the magnitude of the anchor support loads is used and understood to be reversing in direction, then it might be an accurate result. When the friction load and pressure thrust load are acting in the same direction, i.e. the heatup condition, then that would be the larger load case of 876 + 8,000 lbs + other supports friction load (1275.) to act upon the anchor supports at opposite ends of pipe run.

Thanks for your reply.

The model without pressure thrust responds exactly the same as when I put a spool piece in place of the slip joint. I don't see any thermal movement at the joint due to the flexible pipe configuration. It is not installed per manufacturer's recommendations. There is nothing to force the growth into the slip joint. This system has no guides. It is supported on yoke rollers. On one side of the slip joint, it is a straight run to the anchor. On the other side, there are 3 elbows between the joint and the anchor. Stresses are very low.

If modeled with the pressure thrust, I think it would respond the same way because the thrust is so much less than the friction at the joint. I, too, do not think it is a good design to rely solely on friction to hold the load.

One question: I think the thrust forces would be parallel to the axis of the joint. You said perpendicular. Which is correct?

The pressure force is in line with the slip joint. But this force originates on those surfaces which are perpendicular to the joint centerline.
The pressure surface is perpendicular; the force is parallel.

Richard Yee,

Thanks. Your explanation is clear and points out some of the things I have been thinking about expansion joints vs slip joints. My model does include a pair of guides at the slip joint (one at each end of the overlapping pipes), but without any gap. Rotation is not restricted. I will add pressure thrust, but do not expect it to change any results, since the slip joint friction prevents the pipe from joint from displacing.

Thanks for the input.

Michele

Dave,

Thanks for the clarification.

Michele

The value of K2 is 421, then what is it in Newton-meter/degree?