Loose Flange Made from Two Half 2

[McDermott1711]

Hi there,
Is it acceptable to have a 28" series B loose flange being made by two half (and bolted together) according to ASME B16.47? I'm affraid making a flange by two half has risk of not having a flat contact surface due to looseness of bolting between two half.


Nothing is more fairly distributed than common sense: no one thinks he needs more of it than he already has. Rene Descartes

 

[LittleInch]

Try re-attaching the attachment. Just click on the bit wher eit say go back to website and don't copy it. your link doesn't work.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[McDermott1711]

Reloaded.

Nothing is more fairly distributed than common sense: no one thinks he needs more of it than he already has. Rene Descartes

 

[TGS4]

Unless you have analyzed it for that configuration, then I would say no. I have evaluated theses routes if flanges with FEA, and I can share with you that they behave VERY differently from a flange that is one piece.

The reason it behaves differently is that a large portion of the load is carried in the flange through bending in the circumferential direction, or you can think of it as a torsion around the flange circumference. Breaking the load path by introducing such a division renders carrying that load much more difficult.

So, using a standard flange at the rated pressures: no.

 

[donf]

I agree with TGS4. Careful analysis is warranted.

There is a split flange that I've used with success (in much smaller sizes) made by Marzolf Co.

Regards,

donf

 

[McDermott1711]

It seems ASME B16.47 does not strictly disallow using such split flange, isn't it?

Nothing is more fairly distributed than common sense: no one thinks he needs more of it than he already has. Rene Descartes

 

[saplanti]

In case you are requiring the full strength capacity of the loose flange this is not the way. Under the full section bending moment of the pipe this system will not be behaving adequately and will fail earlier. As TGS4 has indicated the torsion will also be problematic.

I additionally take your attention into the tightening process of this flange. How you will tighten it? Something are very good as images but in practice this will be a long delay of your process and you will do the right thing at the end.

 

[LittleInch]

This image is, to me, incorrect. The ring has no offset of the holes to allow studs to clear the collar or indeed bearing surface for the nuts.

If you draw it to the correct dimensions, then some of the issues might become more apparent. Clearly there is some strange reason why you can't put the ring in place before welding on the collar, but just bravado something is not specifically stayed as a negative, doesn't make it a positive.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[KevinNZ]

LittleInch

I think the drawing is OK. the studs fit thru the holes, but there will be more space on the back of the flange for the studs.

I agree with the other posters. B16.47 only covers flanges to the dimensions given and none of the dimensions include a cut thought the flange.

 

[McDermott1711]

Thanks a lot for your responses!

Nothing is more fairly distributed than common sense: no one thinks he needs more of it than he already has. Rene Descartes

 

[JohnGP]

I agree with KevinNZ regarding the drawing.
How the flange performs will depend to some extent on the configuration of the matching flange, whether also split, one piece backing flange, or solid flange. If split, offsetting the splits by say 90deg would go some way to mitigating any negative effect. I think as drawn, good clamping action can be obtained.

 

[LittleInch]

McDermott1711,

We could do with a better bit of feedback other than just "thanks".

Now I've had a time to look at it again in the new drawing, are you trying to clamp the bolt face to the pipe and hence resist the axial forces by friction alone???

I though you were looking at a backing ring for a stub end.

All B16.47 has are weld neck flanges and blinds, hence your design does not comply with ASME B 16.47

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[McDermott1711]

Actually, this is a loose flange to secure a GRP collar to a sandwich type butterfly valve. Service is sea water. The valve has four holes (two top, two bottom) that stud bolts go also through them. These four holes lay on a circle acc. to ASME B16.47 series B.
If loose flage is not under ASME B16.47, then which standard governs it?

Nothing is more fairly distributed than common sense: no one thinks he needs more of it than he already has. Rene Descartes

 

[LittleInch]

Aaah, Now I get it. Remember we can only see and know what you tell us, so that would have been useful in the OP, but never mind.

The loose flange is a backing ring to the bolting dimensions of ASME B 16.47. It is not a described flange in B 16.47 which would prescribe OD, thickness, material strength etc.

Essentially it is normally a vendor supplied item who has "designed" it based on experience.

The backing ring, as noted above, needs to resist torsion as the loads are not centered. This menas it should be quite thick and made of strong material otherwise when tightened it will flex and bend.

Once you cut it in half, its torsional strength is much reduced.

You are correct to worry about the flatness of the ring - distortion will create high loadings on the shoulder and for GRE in particular, this can very easily damage the stub. I've had backing rings bend on me before now and had to replace them with stronger material / thicker (they provided thin aluminum ones instead of cast steel).

If , for some strange reason you cannot do what you should do and chop off the stub and insert the backing ring before final assembly, I would go for a backing ring which is about three times as deep as it is high if you're going to bolt it in half.

see attached


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.