Inspection Template for Ellipsoidal head

[glothex]

We made ellipsoidal head refer to an acceptable approximation of a 2:1 ellipsoidal head (one with a knuckle radius of 0.17D and a spherical radius of 0.90D).

What is the of inspection template which shall be used? Using true shape of ellips 2:1 or shape with a knuckle radius of 0.17D and a spherical radius of 0.90D.

Thanks before.

 

[JStephen]

If you require a head to meet the dimensions and tolerances of an elliptical shape, I can't think of any reason why that knuckle/spherical statement would even be in there. I assume that you can form to that knuckle and spherical shape, and measure tolerances from that shape, and that is the reason the code allows the knuckle/spherical shape.

 

[codeeng]

I would say that either shape would be acceptable for measurement.

 

[glothex]

I design pv with ellipsoidal head size ID=3500 mm, we calculate as per UG-32(d) t=(PD)/(2SE-0.2P).

Then we order that head to sub-vendor, but sub vendor request R crown and R knuckle sizes, so we inform sub vendor that R Crown is 3150 mm and R Knuckle is 595 mm (Refer to UG-32(d) 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).

When we simulate both head sizes on CAD, and found that on knuckle size has big different over 30mm between them, knuckle size of Approximate ellipsoidal smaller than True Ellipsoidal head.

UG-81(a) said that the knuckle radius shall not be less than that specified.

So, if our sub vendor make exactly the same sizes with approximate ellipsoidal head and we inspect the head using true shape of ellipsodal head, it is be out of tolerance, isn't it?

The answer is:
1. We change our design calculation for head using Appendix 1.4(d) Torispherical head. Is it correct?

2. then we check that head using Approximate ellipsoidal template. Is it correct?

Thank you.

 

[JStephen]

I think the intent of the code is that an elliptical head can be constructed as a knuckle-spherical head with certain dimensions, but when this is done, you still design by the elliptical head formulas.

I'm not sure what you mean by the knuckle radius on a true elliptical head. A truly elliptical head does not really have a knuckle (seems like ASME defines a particular radius as the "knuckle" portion, though, but that is arbitrary). The radius varies continuously from crown to tangent line. No two points have the same radius. This could be a problem drawing or interpreting your CAD drawing.

 

[codeeng]

Since the Code accepts both the 2:1 ellipsoidal shape and the 0.9/0.17 alternative shape I see no reason why you can't use either method for measuring the shape.