2:1 Elipsodal Head Profile

You are describing a torispherical head, which is not the same as semi-elliptical.

As Tornado indicated, a torispherical head has a head and knuckle radius, but a semi-ellipsoidal head does not, the radius varies continously. Naturally if you plot both there's a difference.

For SE, head height h, the radius varies from maximum (~ corresponding to head radius) r_max = D²/4/h to minimum (~ corresponding to knuckle radius) r_min = 2*h²/D.

In the special case of 2:1 SE, h = D/4 and r_max = D, r_min = D/8.

As far as I'm aware ASME VIII covers both SE and torispherical heads.

thanks for your reply tornado n guru .

As i have already refered , im talking about ellipsoidal dish end only.

Kindly refer UG32 (d) last para stating that "An acceptable approximation of a 2:1 ellipsoidal head is one with a knuckle radius of 0.17D and a spherical radius of 0.90D"

Taking this has reference the dishes are being getting formed.

Hence ,my question..which profile to taken as reference for dish inspection.?

You may be talking about ellipsoidal dish end only, but ICR =0.9 D and Knuckle radius = 0.17 D describes a torispherical end.

I don't have ASME VIII to hand but if it says that is an acceptable approximation I don't see a problem. Both types give a precise profile, which I assume you have laid out, so whatever the deviation is must be OK. The actual deviation is proportional to the head diameter, which you haven't told us. Also you only now say it's a 2:1 end. It's possible the calculated metal thickness might be slightly less with SE.

If that doesn't cover it, I'd say if you have a torispherical end, use torispherical as reference, if you have SE, use SE as reference. Might also be worth having a talk with the inspector.

that was only i was thinking . but the layout for a 2700mm diameter ellipse dish end differs by almost 25 mm .

Because of this im finding it very difficult to make my management to glitch in the code .

Hence need to clear whether thers a glitch in code or some other things are there to understand for me

It's been a while, and I don't have access to current code - but in my past flanged and dished (or torispherical) heads would have to thicker than 2:1 elliptical for the same pressure.

didnt understand ur point sir

I think he's just confirming my point about design thickness being greater for torispherical than SE. Not only 2:1, same for any height. To state what might be obvious, a head can be 2:1 without being SE, and can be SE without being 2:1.

I still don't see the problem. If either SE or TS is acceptable to the code, it doesn't matter how much they differ. Why lay them out, showing the 27mm difference, and giving your management something to query? It's not significant IMHO, and if you hadn't bothered they'd be none the wiser! Clearly the curves will be different, for the reasons given.

Upon further investigation, I think (still don't have Code to study) the code is saying a torispherical shape with crown radius=0.9D and knuckle radius=0.17D is close enough to 2:1 elliptical to use the same thickness. A 'standard' flanged and dished head with CR=1.0R and KR=0.06 would have to be thicker. Like someone said, 27mm on a 2.7m head is pretty small. If I were laying out nozzles on the head, that might be a different matter.

You could be right, but I'd have guessed the code was saying the shape is close enough to 2:1 SE.

But for 2:1 TS, and KR = 0.06D, I make CR = 0.664D, not 1.0D (or 1.0R). If KR = 0.06D, I make CR = 1.0D if h = 0.17D, vs 0.25D for 2:1.

BTW have you been having problems with CR4? Down several hours yesterday and same today, only just came on.

I'm used to folks saying 'elliptical' to mean 2:1 semi-elliptical, so yeah, that's what I meant. And yes, it has become quite common for my CR4 page in the morning (7 am CT) to be blank. It seems other people are having the same problem. (I think they're trying to shake Lyn off the hook.)

OK thanks

As usual, the OP has gone quiet!

Point is this - if your customer specified a 2:1 elliptical with x" thickness, you can't supply a torisperical head with the same thickness.

This might help....

http://www.inspection-for-industry.com/pressure-vessel-heads.html

Are you saying I put a crow's foot on the torque wrench?

Are you happy now? Thinking about it again, do you mean a SE head was specified, and you want the inspector to accept torispherical, on the grounds of UG32 (d)? If so the inspector will accept it or he won't. There is bound to be difference when you lay both out. 27mm in a 2.7m dia head doesn't sound too bad to me, but I wouldn't think the amount of deviation is too important, it's the principle.

Also, I assume knuckle radius of 0.17D and spherical radius of 0.90D give minimum deviation from SE, because ASME have laid it out, but various other combinations are possible, and you might find one that looks better. In general, for 2:1, spherical radius factor = 0.5 - 1/32/(k - 0.25). Eg for k = 0.2D, R = 1.125D.