please help.

it seems re-routing of pipe is not possible, so I opt to use expansion joint and would try to explain the effect of using it...

i got basic info on the use of expansion joint. however, please throw some light on some of the questions i got.

1. i would like to use a tied expansion joint (lateral movement only)... CAESAR got steps on modelling it... however, i got one procedure which i thought is also applicable. clicking the expansion joint on spreadsheet... input the axial and torsional as totally rigid bellows (1E12), and "assuming 300 N/mm" initially as my transverse stiffness until i got the calculated loads within the allowables... is this one of the "right" approaches?

2. how can i assume the length of the expansion joint (2 flanges+bellows)? i would like to give the manufacturer a "close" length so i can fix my piping layout.

your response would be greatly appreciated...

Hi stressor88

You can not use stiffness as you want rather than you have to choose data from vendor catalogue otherwise it will become very costly specialized design.
However you may got ties universal bellow rather than using tied single below. You may opt for two hingle/gimble bellow to suit your requirement.

If process permit rubber bellow can be use which have much lesser stiffness that metallic bellow.

Regards

Habib

thanks Habib for the reply.

what i'm thinking is after choosing the transverse stiffness i got, the next thing is to look at the vendor catalogue (our material group is currently looking for the manufacturer) and modification shall be done later (with the data from that catalogue)... could the effect be drastic on the piping system? i just hope the length i assume could not be a problem... can you give me a "thumb rule" on choosing lengths?

thanks again. all the best!

Hi stressor88

You may go for Google search to get vendor catalogue.
By the way rubber bellow offer lesser length that metallic bellow.

Regards

Habib

Hello

It would help if you would give some indication what size, pressure and movement you are dealing with. Large size bellows are normally specials. Small sizes you can find from manufacturer's lists.

Thumb rule for the length tends to be that the pipe designers give too short length. In addition too low spring rate is often specified. General rule is that lower the spring rate longer the bellows must be.

Spring rate can be a real problem to a bellows supplier. Specific manufacturers have possibility to make very soft (low spring rate) bellows. If you happen to select from such supplier and the buying will go to an other manufacturer the supplier can be in trouble. It is better when ever possible to specify higher spring rate.

Another useful approach would be to first determine how much movement is required of the XJ and then take that information to the manufacturer's catalog to select the MINIMUM number of convolutions. Then back those stiffnesses into the analysis to see if the resulting loads are low enough; if not then pick more convolutions.

It seems to me that the expansion joint suppliers are providing stiffness only for metallic bellows and not for rubber bellow.

I checked many catalogues and in each case stiffness were provided only for metallic bellows. For Rubber bellows only allowable movements were provided.

Are the stiffness of rubber bellows usually small enough to be neglected during the stress analysis.

This is just one manufacturer and one model that they offer, but Mercer publishes spring rates for their Style 451 joints.

I got the same problem to find rubber bellows stiffnesses.
Is anyone know a supplier who is giving this information?

Ok - in looking at Mercer's site I see that the spring rates are not listed - but if you contact them they will forward you the data.

http://www.mercer-rubber.com/mercer/html/mercerrubber.html

Pinoy,
Since you were decided to use EJ due to some reason, better refer to EJ catalog. Based upon the design temperature and pressure, EJ can be selected and the properties of that bellows are inserted into the piping model. IF the bellows reduces loads and stresses then the range of expansion movements on bellows must be checked. For each bellows, there is a limit to the cumulative axial, bending and lateral movement that can be absorbed by the joint without excessively deforming the convolutions.
After all movements are sets within the required allowable, then you are completed your EJ design.
Additional information, you can refer also to Technical reference manual of CAESAR II. There are some items that should be considered when dealing with EJ design.

As a preliminary option, some of the big boss here in the company where I worked, they recommend to model simple EJ. All the stiffness are set almost to zero. What they were after here is to gather the axial, lateral and bending displacement. based on the displacement, they were refer it to the catalog. As a result, the catalog will provide corresponding axial and lateral stiffnesses and other properties that can be inserted again to the EJ model. Checking also be done regarding the new displacements and load to support and equipment nozzle.

You can find the rubber belows spring rate in the "Fluid Sealing Asociation" Technical Handbook.

As many companies are members of the "Fluid Sealing Asociation" you can be confident that these spring rates are the maximum.



I think that would be a good idea to introduce them in the Caesar database.

Please note that "filled arch" type will have about four times higher the spring rates, while "double arch" type will have half the spring rates.

Regards,