Flanges Made from Plate Material

[Karloss12]

I am designing a manway in an ellipsoidal head. Material is ASTM A516 Gr 70.

The manway consists of a door that self seals under the pressure forcing it against the ring that is welded into the head.

My question is: Can this welded ring which is a reversed flange be machined from flat plate? Or must it be forged?

ASME VIII Appendix 2 says Hubbed flanges can't be machined from plate but I can't find any reference to welded slip on or slip in flanges.

The material is strong enough with and across the grain to pass the design. However my 3rd party design appraisers give a reason of fatigue in that if there are imperfections between the elongated grain structure then these may grow and the flange will split (like a piece of wood).

I am wondering if any of you have had any experience in this area?

 

[MrBTU]

Yes, you can use plate for a slip on flanges designed to Sect VIII Appendix 2. We would have to see your configuration to appreciate the concern of your 3rd party appraiser. Is he concerned about the flange faiiling or the head failing? The reverse flange will be in compression, correct? I wouldn't think that delamination would be a concern, but again - we need to see your design.

 

[vesselfab]

so this is going to look like the standard yoke type manway from LA boiler?

 

[Duwe6]

If you want to gain some confidence in the material, have the blank oval scanned with immersion ultrasound, preferably prior to machining. Ask your experts: "if 516 gr.70 is not acceptable for a M/W closure, how can it be acceptable for making pressure vessels?"

 

[Karloss12]

I have proposed an example but this is a general question about machining flanges from plate whether it be bolted or self sealing or compression or tension.

ASME and PD 5500 say plate can't be machined 'Weld Neck/Hubbed' flanges due to potential risk of lamination. It says nothing of plain flanges. I'm thinking my design appraisers are being extra cautious and excluding plate for all flanges.

This is more of a metrology/fatigue question.

 

[Karloss12]

Attached is an example. It is a flange in a cylindrical shell. The door is pushed agains the flange and seals with an O-Ring.

Does the arrangment really matter?

 

[trottiey]

As far as I understand, the reason you don't machine hubbed flanges from plate is because lamination flaws could be through wall in the hub. A similar logic explains why UG-14 doesn't let you make blind flanges out of rolled bar.

 

[gr2vessels]

I am a bit puzzled of what is the actual problem;- the ASME VIII does allow you to roll a plate 125 mm wide and 50 mm thickness into a ring, having the two ends welded with full penetration and then assembly/weld it into a shell as per your drawing (obviously the blank plate you roll has to be larger for a final machining to size after welding it into a ring and PWHT). Ditto for the cover as per your drawing. My interpretation is that the hubbed flanges are more sensitive than other flanges to the orientation of the grains/lamination, that's the reason for nominating them in Appx 2, section 2-2, clause (d)(2)(a) & (b). Your shell flange and cover conforms to this article.
Cheers,
gr2vessels

 

[Karloss12]

gr2vessels,

I agree with everything you say. Rolling plate into a ring is fine.

I should have written a better description in my first post. This flange in question is to be machined from 130mm plate without any forming. It is apparently less expensive.

 

[trottiey]

Whoa there, that would be your problem. If you machine 130mm thick plate down to that thin curved shape, then the laminar defects would not be parallel to your surfaces. As your 3rd party appraiser said, these could grow and split. trying to machine a curved sheet out of a wooden board is a pretty good analogy; the grain would cause it to split.

 

[Karloss12]

Thanks trottiey,

If you are to read ASME VIII and PD 5500 they both specifically specify that you can't machine hubbed flanges from plate (due to the thin hub possibly laminating). Surely in the same section there would be atleast some mention of plain flanges or studded pads and any limitations with using plate as opposed to forging.

The example flange I have put forward is quite thin like a hub so should be a forging where as wide flat flanges have more in common with a blind flange which is typically plate.

Engineering judgement.

 

[Jmoore1]

Just my two sense, it is strongly recommennded not to use slip on flanges at high pressures...

 

[trottiey]

If the code doesn't talk about it, that's neither an endorsement nor a prohibition. For example, the code doesn't say anything about nozzles placed on the knuckle of a head, but that's generally considered to be a bad idea. From the ASME Foreword: "...those aspects which are not specifically addressed should not be considered prohibited. The Code is not a handbook and cannot replace education, experience, and the use of engineering judgment."

But now that I re-read your original question, I think I might have been confused as to which part we're talking about. You're talking about machining the reverse flange in the vessel wall, not the door, right? So it only sees compressive stress? If that's the case, then I take back my last post. You don't get much fatigue damage under compression.