When I input the effective area in the Untied Bellows and place limit stop with guide minimum to the bellows, the limit stop is not absorbing the pressure thrust. I want to place untied below to absorb axial thermal expansion in which placing a tie-rod will not serve the purpose.

Please share your experience with this type of problem.

Cheers!!!

Double check the effective area value, axial stiffness and pressure.

Hi Dan. Thanks for the reply. All the 3 variables have been considered in the input.

Any other opinion is greatly appreciated.

Cheers!!

Maybe is the case to ask Coade support. Who knows ...
Regards,

Caesar II wants the Deff, i.e., the effective DIAMETER. This has always worked to the horror of many first time modelers of EJ trying to use untied EJ in systems with a decent amount of pressure.

Double check your units and some basic inputs.

Thanks Bob & Dan for sharing.

Any other experience with the same is highly appreciated.



Cheers!!!

CAESAR II will give correct result but the modelling has to be right. The problem is that to have axial movement requires anchors on both sides of the expansion joint. Using limit stops is just an emergency safety mechanism if one of the anchors fail. Typically such is done with rods and nuts with gaps or slotted hinges.

It is impossible to design such design to work during the operation without having the pressure thrust on the anchors. Anchors can be steel structures, pumps, vessels etc. but most of the time the forces are too high. I have a case where partially buried concrete block about 1 m^3 moved on NB600/6 bar line.

Unrestrained axial expansion joint is a dangerous item. Nearly every time where such is used the pipe design is poor or bad or even dangerous.

Thanks Jouko. That's what I have done. To put rods and nuts with gaps to control the axial movement.

Any other experience is greatly appreciated.

Warm Regards,

Nuts with gaps do not control the axial movement. They only limit it to specific amount provided the unit is installed correctly. Unfortunately that doesn't happen always. Computer gives an answer but the reality is different. Even the heat expansion is unknow. Pipe metal temperature used in the CAESAR II analysis is not what it is during the operation. It is unknown.

I design expansion joints since over 20 years. I also give training and it includes most common mistakes made by the pipe designers

Very common mistake: Pipe designer specifies expansion joint axial movement to be -25/25 mm (-1/1") on 450 C (840F) on a hot line. What is the (bad) mistake?

Thanks for sharing your experience Jouku. I'll follow your advice. So far I haven't design up to the creep temperature range for expansion joint. As much as possible I don't want to use it as it is a weak spot in the system if improperly designed. It's still better to design using the natural flexibility of the piping system if space permits. I just happen to use it due to very low allowable nozzle load at the equipment connection but temperature is low.

Any other experience is greatly appreciated.

Cheers!!

450C is not really in creep range. Process plants have quite regularly expansion joints in those temperatures. Temperature is not the problem. Movement specification is wrong. There is zero change that expansion joint will expand as much as it compresses. Giving such specification on an expansion joint is making the line far more unsafe than giving correct spec.

Only place with very rare exception where pipe engineer must specify equal 2 way movement is movement caused by mechanical movement or for instance wind.

I use only Celsius for the calc. Expansion joints are delivered for ambient temperature, e.g. about 20 C. Temperature increase is 450 - 20 = 430. For the expansion joint to expand the same as it compresses pipe has to cool down to -410 C. In our dimension there is no such temperature. Designing expansion joint for equal movement makes it longer and prone to instability.

Pipe designers spend a lot of time looking into spring rates (and sometimes forgetting pressure thrust). My problem is that most of that time is wasted. High number of expansion joints pass a yield point during their movement. That means that when the pipe cools down the expansion joint has to be pulled to its original length. I do not know any pipe designer who considers that force.

Old times pipes failed because they were too rigid. Today pipes fail because they are too flexible and often also have insufficient axial stops and guides. Stress analysis may say all is perfect and line fails during operation, sometimes during the pressure test.

Thanks for sharing your experience Jouku. I will consider this in case I will face such design in the future.

Any other experience is greatly appreciated.

Cheers!!