Multiple holes in blind flange - alternative to calculation? 1

[timsch]

My company utilizes blind flanges frequently in multipoint thermocouple assemblies designed to ASME BPVC SECTION VIII, DIV.1. Frequently these blind flanges have multiple holes drilled through them in circular patterns around the flange center. Calculations are typically done per UG-39 to determine flange thickness.

ASME B16.5, Table 6 shows single, centered hole size limitations in blind flanges without requiring a hubbed flange. Hubbed flanges in the case of a slip-on flange have a hole through the center the size of the pipe OD that the flange is designated for.

Rather than doing UG-39 calcs, is it allowable to determine whether the holes drilled in our flanges are fully circumscribed by either this minimum allowable diameter hole for the standard blind flange or by the pipe OD of the hubbed flange, and if so, to use the flange as directed per ASME B16.5?

I have seen this proposed as a practical approach to flange analysis by a well known member of this forum in another thread, but I have not seen any indication in the code or related documents that explicitly allow this.

Thanks,
Tim

 

[XL83NL]

Yes, provided there's only one hole in the centre. That's the B16.5 requirement, but based on your post, I guess that's not what you're after.

 

[KevinNZ]

A ring of holes with cause large stresses between the holes. The pressure on the plate inside the holes will be pushing plate out.

I would

1. regard the outside of the ring of holes as the diameter of a large hole in the flange and use this see whether you can use a bind flange or a full hub per B16.5. This will mean the flange is thick enough for bolting stresses.

2, use FEA or other method to calculate the stresses between the holes. Design to ASME VII div 2.

 

[timsch]

The main concern with flanges is whether they will seal effectively, so deformation at the seal face must be minimized. I understand that there will be increased stresses in the material between the holes in a configuration as I described. The question I have is whether a flange such as a slip-on with a full pipe diameter bore in it will deform at the sealing face any less than a hubbed blind flange with identical thickness that has a hole pattern which is fully circumscribed by this full pipe diameter bore.

I see no way that additional material within what would otherwise be a void in the center of the flange would be detrimental to the flange seal.

 

[XL83NL]

Interesting question. My first thought would also be that 'in no way any additional material, within what would otherwise be a void in the center of the flange, would be detrimental to the flange seal'.

 

[SnTMan]

timsch, if your hole pattern can be enclosed in the Table 6 diameter I'd consider the design adequate.

You can of course run the UG-39 calculations.

Another way to do this is to consider it as a loose ring flange as per Appendix 2, the ID being equal to a circle circumscribing your hole pattern. B16.5 thickness may or may not be adequate, as Apx 2 calculations of B16.5 flanges are known to be troublesome.

None of these methods will consider detailed analysis of the stresses between the holes, etc.

Regards,

Mike



The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[timsch]

We currently run the UG-39 calculations for this scenario, which occasionally do result in us needing a non-standard thickness custom flange.

I also would consider the design as described above using the hubbed flange adequate. I frequently have to submit my calculations to a third party for review if it is a PED or a U-stamp job, so I need to be able to convince them and the AI that my use of a hubbed blind flange for this application is acceptable per ASME BPVC Section VIII, Div.1, hence the reason for my post. It would help us in sourcing materials if I could specify a hubbed blind rather than a custom blind flange.

 

[moltenmetal]

timsch: we would typically agree with your assessment. A blind flange can be drilled with a single hole per the B16.5 table, but if multiple holes or holes larger than those permitted in B16.5 are required, a hubbed blind of dimensions equal to the blank hubbed blind typically used to make a reducing threaded or socket-welding flange, can be used for any practical pattern of holes which are circumscribed within the flange's NPS OD. By "practical spacing" I mean a spacing which is practical for NPT threading or socket welding.

Note that we have previously confirmed that reducing socket-welding and threaded flanges are indeed machined from hubbed blinds, rather than being pierced during the forging process.

 

[XL83NL]

Im currently purchasing some flanges for the same purpose (i.e. hole is bigger than that permitted per table 6 of B16.5), and I specify the flange to our supplier 1" MASSIVE SLIP ON (INITIALLY NO HOLE), WITH 3/4" NPT THREAD TAPPED INTO CENTRE OF FLANGE, or simply 1" HIGH HUB FLANGE WITH 3/4" NPT HOLE. Delivery time is slightly more than 'stock', but two weeks usually does it. Prices are also higher, but in an overall project cost insignificant.
We design most our jobs in compliance with PED. Typically the blind flanges dont fall into a category, and therefore using tbale 6 is sufficient.

 

[timsch]

Thank you all for your responses. They are very much appreciated.

moltenmetal, when you use the term "we", you wouldn't be referring to the ASME BPVC code committee, would you?

 

[moltenmetal]

No, by "we" I meant "my company" which shall remain nameless, for the timebeing. They shouldn't be blamed for anything I say on Eng-Tips...!

 

[timsch]

Understood. Thanks again moltenmetal.