UHX: Opening very close to Tubesheet (with integral Shell/Channel). 2

[Ganfoss2]

Dear all,

I need to clarify the following issue: if a "discontinuity" (like a saddle, a "small" opening to drain/vent the shell/channel chamber, a "large" opening for the INLET/OUTLET of the Process Fluid, ...) may be "close" to an INTEGRAL tubesheet ["close" mean distance less than: 1.8 x RADQ (D x t)] when the shell/channel thickness is used as "reinforcement" to reduce the tubesheet thickness ?

(See UHX-12.5.10 and UHX-13.6 for references)

I was not able to find any of such constraints in ASME VIII-1: the only available "criteria" is an extrapolation of the MIN. Distance between two LOCAL AREA of DISCONTINUTY (where S >= 1.10 x Sm) stated in the ASME VIII Div. 2 Code (Ex. Appendix 4).

Does someone has a "criteria" to suggest? I see very often request of quotation of Exchanger with "large nozzles" or "Supports" very close to an Integral Tubesheet with reduced thickness ("clamped" at shell/channel shell barrels ends).

NOTE: The New European Codes (EN-13445/3) and other Old EU National Codes (Ex. BS-5500, VSR) state a MIN distance respect to a "discontinuity" (Flange, Cone-to-Cylinder attachment, .... )!

 

[DonKurle]

Ganfoss2,

Currently UHX does not have any rules that limit the size or location of a nozzle in an integral shell or channel relative to the tubesheet. We (SG-HTE) have an open item to investigate this issue but at this time we do not have a proposal for consideration.

When a shell or channel is integral to a tubesheet, UHX considers it to support the tubesheet regardless if the shell is thickened or remains constant thickness the entire length. Therefore, an integral shell or channel will always provide support to the tubesheet.

The criteria I used is to keep the distance from the back of the tubesheet to the OD of the nozzle 1.5 x sqrt(Rt,where R is the shell radius and t is the shell thickness, or the distance from the nozzle centerline to the back of the tubesheet d, where d is the nozzle ID, whichever distance is greater. As a designer, you must use engineering judgement to decide if the nozzle is large enough to be of concern. For example, a NPS-2 nozzle in a 48 inch shell would not be of much concern but if the nozzle were NPS-12, it may be of concern.

 

[Ganfoss2]

Ok! I agree with your criteria!

Thanks

 

[SnTMan]

Ganfoss2, my reading of UHX 12.5.10 & UHX 13.5.10 is that an opening in an intergral shell or channel would not be permitted within 1.8*SQRT(Ds*ts) of the tubesheet. I think that a support located within this distance would not affect the stiffening effect of the cylinder on the tubesheet where a "large" opening could.

Its' true that the people that prepare the RFQ's don't always allow for some of the Code requirements, they have other fish to fry, like getting the new exchanger to mate with existing piping and foundations.

You could, if in doubt, treat the tubesheet as simply supported and avoid the issue at the cost of a thicker tubesheet.

BTW, I am not a big fan of Part UHX. I miss the old days

Regards,

Mike