Expansion Joint Failure 1

[mjpetrag]

I have an expansion joint carrying 250# steam at ~550 deg F that has only been lasting about 6 months to 1 year before it starts leaking. It is an enclosed, pressurized spool piece rated for 300 psi @ 800 deg F and a design cycle life of 1000 cycles.

The problem I am having is what is considered a full cycle? Is there any way to prevent this joint from failing so quickly?

 

[pennpiper]

mjpetrag

As any well trained and experienced piper will tell you, "the best and only way to prevent an expansion joint from failing is to NOT have an expansion joint in the first place."

You need to go beck and look at the configuration of the line along with all aspects of the sorounding area and see if you can do a retro-fit and add an expansion Loop.

You should already have anchors at both ends of the run and you should already have pipe guides on the line. You just need to see if there is room to add a loop that will take the same thermal growth as the joint.

Check it out, it might fit and then you will not have a shut-down every 6 months or so.

 

[gr2vessels]

mjpetrag,
The expansion joint is obviously selected correctly for the pressure and temperature, but the devil is in the details, like off-set expansion due to wrong pre-set, or incorrect assessment of the expansion, etc..Where is it leaking, at the top of a convolution, at the fillet weld to the pipe end, at the flanged end weld, where? Full cycle is the complete travel of the joint from the relaxed position to the rated maximum deflection and back. You have to include the pre-tension in the full cycle. I would re-check the calculations and do some site measurements, at full operating and shut-down condition, comparing them with the original dimensions. Did you check the welding procedure for the installation of the expansion joint?
cheers,
gr2vessels

 

[mjpetrag]

It is leaking at where the pipe end meets the inside flange. So a full cycle would be when the steam is on to when the steam is shut off and back again? The steam line is continuously running. The weld procedure is the same as the expansion joint installed ~20 feet behind this one that failed, and that expansion joint is fine. I was thinking that maybe when they installed it, they took the guide bolts off too soon and it was misaligned.

 

[krb]

See the expansion joints design standards here

http://www.americanboa.com/lite.htm

for an explanation of cycles.

 

[StressGuy]

Guide bolts?

If you're talking about shipping bars, those should be cut off AFTER the joint is installed. Part of the reason for that is to keep construction from trying to use the joint to "correct" for piping fabrication errors. That is not the purpose of an expansion joint.

If you are talking about limit rods or tie rods, those should never be removed as they are integral to the proper function of the joint.

I would definitely recommend having the engineering firm that designed that system for you get back out there to look at the system and determine the answer some questions:

1. Is what was installed what the engineering firm designed?

2. Was the system modeled accurately?

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.