DESIGN EXPANSION JOINT - DOUBT

Hello my friend,

I attached a drawing about an expansiont joint that I will use in my project.

So, I would like to know that I modeled it correctly at CAESAR II.

So, follow step-by-step about what I did:

- 10 to 20 - rigid element DY=150mm (weight = 90kgf)
- 20 to 30 - expansion joint DY=700mm (axial stif = 10000000 lb./in. ; trans stif = 952.008 lb./in. ; torsion stif = 10000000 lb./in.)
- 30 to 40 - rigid element DY=150mm (weight = 90kgf)
- 10 to 41 - rigid element without weight DY=1000mm (Restraint Node 41 Cnode 40 Type Y ; Restraint Node 41 Cnode 40 Type RX ; Restraint Node 41 Cnode 40 Type RZ)

So, Thank you very much and I will be waiting for your answer about that.

Best Regards,

Saulo Cunha

Saulo,

I think the way that you do it's wrong. You can insert this kind of joint pre-modeled by CAESAR. Use U-type in CAESAR's EJ module. An change the lenghts and stiff that you need.

Regards,

Miyamoto

Hi Miyamoto,

ok, I did it before, but if you see the vendor just sent me the lateral stiff. So, I took the CAESAR manual and he teach how can we model when vendor send us lateral or bending stiff.

Look at the picture attached.

So, I have an universal EJ and just lateral sitff. I think that if I use the modeler EJ at CAESAR, considering an Universal EJ I will take two bellows and one piece of pipe between of then, and I don't know if I can put the lateral stiff both of then too.

If I understood I just do it if the vendor send me the bending stiff.

Please, let me know if I'm correct...let discuss abou that...what do you think ? I need it urgently because I will buy this EJ and I need to evaluate that in my model.

Thank you very much Miyamoto.

Best Regards,

Saulo Cunha

Saulo,

The type of joint you are using is a “Tied Universal” joint which resists the longitudinal pressure thrust and will accept only lateral movement. So if X is axial, Y is vertical and Z is out plain, then your restraints at the Cnode would be X, Ry & Rz.

In regard to the Spring Rate Values, you can contact the manufacturer to request a single bellows element bellows design analysis (BDA) which give you the actual values to input.

If you’re in a hurry, you can trick C2 into doing the right thing by putting a zero length bellows in the middle of the space. Enter the actual lateral spring rate provided by the vendor, but then enter rigid, ie 10E6, for the axial, bending and torsional spring rates. Also enter 0 for the Effective ID. This method will be accurate for the model but not for the EJ movements. You’ll have to do a hand calc for the axial movement between the tie rods.

Yeap Dan, I agree with you, but what do you think about that:

"Manufacturers publish a wide variety of data for universal expansion joints. In most cases the published spring rates are for the universal joint as a whole assembly. When the lateral stiffness is given for the whole assembly the simple or complex models of single bellows can be used. In this case the manufacturer must also provide a cumulative assembly displacement limit so that the piping designer can verify that neither of the bellows are over-extended".

It's my doubt, because I read it on CAESAR manual...

Could somebody explain me it ?

Thank you very much again.

Best Regards,

Saulo Cunha