Custom flat-faced flange with full-face gasket -- does ASME viii cover this?

[chrisstandring]

I am making some design changes to an existing air dryer that was designed to ASME viii Division 1 back in 1984. Part of the modifications consist of adding a short shell section to the existing shell by bolting it to the end flange. The end flange is a flat-face slip-on type and uses a full-faced gasket. The dryer shell diameter is such that a custom flange had to be used as the diameter is different to any standard pipe. The new shell section therefore also has to use that same custom flange. I cannot use an off-the-shelf flange. Right now I am putting together the calculations to support the modifications. Unfortunately I do not have the original calculations for the dryer. The problem is that I cannot find anything in ASME viii that covers the design of custom flat-faced flanges with full-faced gaskets. Flanges with gaskets entirely within the bolt circle are covered but that is of no use to me. Is it even possible to design a flat-faced flange with full-faced gasket to comply with ASME viii? Can anybody help or advise?

Cheers,

Chris

 

[C2it]

Brits to the rescue !

Correct, ASME VIII only deals with metal to metal contact outside the bolt circle.

Check out PD 5500 (2012 or later) section 3.8.4 and working form 13.

 

[chrisstandring]

Excellent! Thanks for that. That confirms my fears. I saw the bit about the flat-faced flanges as well and thought "yippee!" -- until I read about the metal to metal contact outside the bolt circle -- bugger!

Never mind I will check out PD5500 next week. I've got a copy on my desk but I suspect it is older than 2012. I might have to get our standards guy to get a copy in.

 

[SnTMan]

chrisstandring, Sec VIII, Div 1 does not cover full face gasket flange design, but Taylor Forge Bulletin 45 has been used for years and years under U-2(g). Some software implements it.

Regards,

Mike

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

 

[C2it]

I suspect that you will find many similarities between PD 5500 and the TF code recommended by SnTman.

I don't think there has been any change to the method in PD 5500 for many years, so a new purchase may not be necessary.

 

[chrisstandring]

Thanks guys for the help so far. It turns out that there have been a few changes in BS 5500 / PD 5500 since 1996!

Looking at PD 5500:2015 section 3.8.4 there appears to be some discrepancies in a couple of formulas.

In Suggested Working Form 13 the formula for hR is

hR = (G0 - C - dh)/4 + d/2

Whereas in 3.8.4-1 the formula for hR is

hR = (G0 - C + dh)/4

Now d = U/VL.h0.g0^2 (3.8.2) which suggests to me that the units for d is mm^3 and that conflicts with the units for hR which is in mm (assuming the factor VL is dimensionless -- I could be wrong but that is what it looks like). Further, for the particular application I am looking at, d computes to a fairly large number which makes hR as computed with the formula in Suggested Working Form 13 much larger than the physical size of the flange (about five times larger than the flange OD). That makes no sense to me and I believe that formula is incorrect.

If I use the formula for hR in 3.8.4-1 then I get a number that agrees with the diagram.

So which formula for hR is the correct one? Looking at the diagram on Suggested Working Form 13 it looks like the formula for hR on that form has a typo and d should actually read as dh, as follows

hR = (G0 - C - dh)/4 + dh/2

= (G0 - C + dh)/4

which agrees with the formula in 3.8.4-1. So I am sure that the formula for hR on Suggested Working Form 13 has a typo.

Can anybody clarify this?



Cheers,

Chris