Stainless Bellows Expansion Joints in a Process Air Line

[kkcrowe]

Here is my situation, My company has installed about 40 or more of these Stainless Bellows expansion joints in a process air line. The diameters of pipe are ranging from 16" down to 2". The operating pressure of the system is about 9 psi. @ 200 degrees. The expansion joints were shipped minus the "Shipping Bars" that usually hold the joint to a specific "over all length". Each joint is designed to consume 2" of contration and expansion - total 4" movement. Using a tempature delta of 150 deg. and the longest run of pipe involved between expansion joints, I total approx. 1/4" of movement. Now on to the problem...durring installation, some of the EJ's were installed at "plus or minus 1/2" of the original overall length. The manufacture of the joints has given me a letter confirming that this in no way will damage or hinder the operation of the EJ in the system that it is being required to operate within. However, I am looking for any documentation that could show that these joints also have a manufacturing tolerance or an installation tolerance. I need to show that there is no concern for the operation of this system and that I provided a joint with more movement than ever could be required by this system therefore allowing myself some movement within the range of the EJ for installation purposes. Thank You for any help in this matter Chad Rowe Project Superintendent

 

[Scipio]

Installation tolerance should be irrelevant assuming it's within the original design deflection - a joint with a +/- 2" deflection doesn't care whether that deflection is during operation or installation, so long as the cumulative deflection is within it's design. As for manufacturing tolerance, that would vary from vendor to vendor, however the bellows on these expansion joints are usually roll punched, so the actual bellows length should be identical from joint to joint - any discrepency in overall length would be in the flanges, and even then I wouldn't expect it to exceed +/- 0.125". And even if you do manage to exceed the rated deflection, depending on the construction it may not be that big a deal - bellows joints have a finite lifespan, sometimes excess deflection will just mean the bellows has to be replaced in three years instead of five.

I actually had one bellows joint that was installed with an initial radial deflection of about 3/4", when it was only designed for a maximum of 1/4". After checking with the manufacturer, we pulled the joint, re-aligned the piping in question, pressure-tested the expansion joint to make sure we hadn't cracked any welds, and put it back in service with the proper misalignment. It's been in service now for about 5 years with no problems.

 

[StressGuy]

I would tend to agree with Scipio. If the joint is good for +/- 2" of movement, then you should be OK as long as the total movement is less than that.

Although, for future reference, it is very bad practice to use EJ's to make up piping misalignment like that. One of the main reasons joints are shipped with shipping bars is to keep construction guys from going exactly what they did to you. I'm really surprised a vendor would ship joints w/o bars.

I must confess, I've also never heard of using expansion joints in air piping. Wonders never cease

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[unclesyd]

It is not unusual to use expansion joints in air lines. We used them in most lines of 6" and above if any temperature was involved. Most of the larger expansion joints in the higher temperature, > 250°F, systems were of the tied/guided variety to mitigate the possibility of squirm.

 

[SnTMan]

kkcrowe,

All well and good,but I suspect whatever documentation you need is going to have to come from the joint mfg.

Good luck

 

[Scipio]

StressGuy,

Actually I hate using expansion joints unless absolutely necessary just for your reason - if you get a bad fitter or designer they take one look at that joint, which is usually just to compensate for thermal expansion in a single direction, and consider the joint a fudge-factor for putting the system together. "We're out half an inch? Don't worry about it, we'll just pull the expansion joint to fit, no need to waste 10 minutes adjusting pipe supports to eliminate the misalignment properly" The part that always bothers me is that the guys installing the joints rarely, if ever, even know how much deflection the joint was designed for.