ASME Expansion Joint Calculations

[Lukasz1]

Hello,

We are in the process of designing small shell&tube heat exchanger with two fixed tubesheets (shell diameter is 5") acc. to ASME Sec. VIII, Div.1 requirements. Whole construction is made entirely from copper grade C11000 and C12200. It looks that we would need to implement expansion joint for this design.

I have never calculated such element acc. to ASME requirements and would like to ask for some guidelines. I did my research and from what i understand there are two ways to do it:

1)Calculate acc. to App. 26 - i tried to start calculations for single convolution and single plie but from 26-2(f) i read that fatique equations are only valid for austenitic chromium nickel stainless steels, UNS066XX, UNSN04400. Does it mean that other materials are not allowed? Can 26-4.1(f) be applied in this situation?

2)Calculate acc. to App. 5 - i cant find any reference to calculation formulas. Does it mean that we need to use other standards like TEMA, EJMR or FEA to perform such calculations?

If anyone more experienced in this subject could give me any guidelines i would be very appreciated.

Regards

 

[SnTMan]

Lukasz1, 1) see 26-2(f) and 26-4.2(b). Another option for you might be to let an expansion joint vendor supply the design, to your stated conditions. Your problem then becomes one of managing the vendor.

2) Yes, however, for F&F joints the design is generally carried out by the vessel designer and the components made by vendor who has no responsiblity for the design whatsoever. I doubt it will be within the scope of an EJMA manufacturer.

My advice? Find some way to eliminate the joint.

Regards,

Mike

 

[fegenbush]

To elaborate on what SnTMan said, here are a couple comments:

1. Buy the expansion joint: for a custom design that you are unlikely to encounter again, this is a popular option. However, copper is a specialty material (even moreso than the nickel alloys) and you may have difficulty finding a vendor to perform the work.
2. 5-3(b)states that the calculation of stresses is up to the designer. I find that Roark can be a good resource for this (Table 11.2 Case 1b). UHX-17(a)(2) excludes the inner and outer torus rotational stiffness, so use that as you will.

 

[tigny]

Hi,

you may want to have a look at TEMA (eight edition) code, RCB-8 Flexible shell elements. It seems limited to Poisson's ratio = 0.3

yours,

 

[Lukasz1]

Thank you all for your comments.

We decided to try TEMA 9th Edition FEM approach.