Vapor Belt Distributor & Mandatory Appendix 14

[Eddy1988]

All,

I have been asked to evaluate an existing heat exchanger (condenser) with Vapor Belt Distributor. I have used the COMPRESS software's calculations for Integral Flat Heads with a Large, single, circular, centrally located opening (Mandatory Appendix 14) capability. Due to the method of fabrication (full penetration corner welds), Mandatory Appendix 5 and Mandatory Appendix 9 are NOT applicable.

The results are very poor. The vessel would need to be down rated according to the results. Has anyone come across this issue in the past?

Thanks,

Eddy

 

[marty007]

The code has some specific requirements on how to handle vapor belts, see paragraph UHX-4(c). How you handle it depends on whether any of the internal shell remains in the area of the vapor belt.

Have you considered this paragraph already?

 

[Eddy1988]

Yes. I have considered this paragraph. It led me astray. I don't think that I could use the guidance provided since it assumes vapor belt or flanged and flued expansion joint connections. These appendices do not apply to the current method of fabrication (full penetration corner welds).

 

[Duwe6]

The results are unsurprising. Square corners calculate to a low MAWP. Without some shell remaining under the belt, that MAWP will be tiny. Eddy1998 is telling you to investigate to see if there is some shell that is under the bolt. You need to stuff someone inside that shell to see if the shell continues on through the vapor belt, and how many and how big the 'windows' are that were cut in the shell to allow the vapor to enter the shell.

 

[marty007]

If you do not have any remaining shell under the vapor belt, and you have a square built-up vapor belt, then as you say the code doesn't directly cover this geometry.

You could invoke good-ole U-2(g) in this case and refer to TEMA for a design basis. TEMA paragraph RCB-8 (Flexible Shell Elements) does cover the geometry of square corners, and gives a full design basis. Note that the latest edition of TEMA (Ninth) has switched to an FEA based calculation method.