Expansion Bellow Installation

[Stiffmeister]

Hi Guys,

Recently there was Stress analysis study prepared on the Asphalt loading system I am working on. The analysis considered a specific brand /properties of Single expansion joint in their calculation. However during the procurement, a different brand was selected and thus properties are somewhat different. Now my question is, how can I check if the procured expansion joints are still acceptable?

I'm a process engineer so I am not quite familiar with pipe stress analysis nor selection of expansion joints. Is there someone who could point me in the right direction on how to proceed? I know this work is for pipe stress engineers but I am curious to know what needs to be done on this case if in case a pipe stress analysis software is not available.

 

[RVAmeche]

How different are the properties of the procured joint vs the assumed joint? The by the book answer is to re-run the stress analysis with the actual properties, but realistically if they're close to the assumed properties and the code stress wasn't right on the edge, it's probably fine. Note the definition of "close" will depend on the service conditions and required growth to account for.

 

[Stiffmeister]

RVAmeche,

Thank you for the quick response.

Autopipe results on the report show that the Code allowable values are four (4) to twenty (20) times of the code stresses. Considering these values and that the properties are about 10% different, can we safely say that all fine for the system?

 

[RVAmeche]

I assume you mean 4-20 percent of the code stresses, otherwise there are other issues.

Not sure what the service conditions of your system are, but unless there's significant movement, you're probably okay.

The properties of the bellows are stiffnesses, so if you need to accommodate large movement, a higher stiffness could result in significantly more force. Conversely, a significantly less stiff bellows may expand/contract more in the same application and may end up outside the allowable movement parameters.

 

[Stiffmeister]

RVAmeche,

Yes, sorry for that. It's 4 to 20 of code stresses.

Could you recommend me a good reference for manually estimating bellow stiffness?

 

[KevinNZ]

Bellows stiffness should be much less than the pipe it is attached to. Check the units used in Autopipe. Has the bellows been changed to a different type. e.g. hinged from tie rod type.

If your stress is 20x allowable then the pipe model is way off the piping engineer should be able to work out where the problem is.

 

[Riley15]

If the highest load case is only at 20% of code allowable (ASME B31.3?)and the new expansion joint's spring rate is only 10% different from the one used in the model, I would bet that everything is fine.

 

[jojoyohan]

At this point with a real expansion joint in hand I wouldn't estimate anything. Get the expansion joint data sheet from your supplier. If the spring rate is lower you will be good on the stress front. Check the displacement to verify that the expansion joint was designed/calculated with that or a larger displacement. The higher the displacement, the higher the stress and lower the cycle life of your expansion joint. You should also verify that the B31.3 cycle life of the expansion joint is acceptable with this new joint. The EJMA cycle life will overestimate the cycle life of the joint.