WRC 497 Calculation (Nozzle Edge and Pad Edge)

[Trestala]

I am trying to evaluate some nozzles with applied external loads (d/D = 0.47 and 0.55) using WRC 497. These are nozzles with reinforcing pads.

When commercial softwares use WRC 107/537, they evaluate stresses on both nozzle edge and pad edge using corroded thickness. However, WRC 107/537 considers the nozzle as rigid.

WRC 497 considers the thickness of the nozzle in its evaluation. I am evaluating the stresses at the nozzle edge using shell thickness with pad less corrosion allowance, and nozzle thickness less corrosion allowance. However, for the pad edge, I need a nozzle thickness value to get the stress values at this location.

Part 1, Sec. 15 of WRC 497 mentioned the use of pads and consideration of pad thickness, but no mention of evaluation at pad edge.
Also, on Part 8 of WRC 368, they have cited examples with reinforcing pad and did not mention any evaluation of stresses at pad edge.

Question 1: Do I need to evaluate stresses at pad edge for WRC 497, like how commercial softwares do for WRC 107/537?
Question 2: If so, how would I consider nozzle thickness for t/T value.

Design Code: ASME VIII, Div.1
Acceptance Criteria: ASME VIII, Div. 2, Part 5 as per U-2(g)

 

[MrPDes]

The d/D nozzles that you are working with are well beyond the maximum d/D allowed by WRC 107/537 and WRC 297. The stresses related to large d/D nozzles are very different to small d/D nozzles, therefore these bulletins will be of little or no help to you.

I understand that WRC 497 is a limited parametric investigation of simple large d/D nozzles. I doubt it is considered as a comprehensive design method and does not cater for reinforcing pads.

"The configuration employed in this parametric study is idealized and consists of two thin shell cylinders intersecting normally with no transitions, reinforcements, or fillets in the junction region."

It sounds a bit crazy putting such a huge reinforcing pad on such a huge nozzle. Perhaps increasing the shell thickness is wiser, and then WRC 497 might provide you with some direction.

I would then still conduct FEM with or without pad.

 

[Trestala]

Yes, we are using WRC497 for these nozzles because of the large d/D ratio. And my concern is if there is a need to do an assessment for the pad edge if the intersection happens to have a reinforcing pad because there is a change in thickness.

 

[MrPDes]

For Nozzles with very wide pads, WRC 497 may provide you with some guidance on the initial thickness sizing of the Shell-Nozzle Junction, however WRC 497 will likely not be applicable at all to medium width pads. However this guidance would only be useful for an initial guestimate before conducting FEM. The stresses at a junction without a pad are quite different to with a pad, therefore as explained above the idealized WRC 497 investigation is not applicable to comprehensively designing any nozzles with a Pad.

There is no method for establishing stresses at the edge of a pad with a large d/D that I am aware of. FEM will need to be used.

Stubbornly trying to apply a design method which is not suitable for your application risks failing to achieve safe reliable results.

 

[Trestala]

Please read WRC 497, pads are mentioned and not prohibited. I am simply asking if the way WRC 107/537 is used also applies to WRC 497 where pad edge is evaluated.

Local stresses even without pad would still pass for this case.

"Conduct an FEM" is a too common suggestion now; and I believe it's unnecessary for this simple case, and doing FEM would involve additional cost and manhours.

 

[MrPDes]

In WRC 107 the solid bar graphs are used to establish stresses at the edge of the Re-Pad. These solid bar graphs are also used for stresses at the OD of nozzles with large t/d. I expect WRC 497 doesn't have these solid bar graphs.

At the OD of a pad, the stresses relating to the hole are less pronounced (or even negligible if the nozzle neck is thick), therefore if the Pad is chunky, you could set the WRC 497 nozzle diameter to the O/D of the Pad and set the nozzle thickness to be very thick. But this is a bodge job based on guestimates.

If your nozzle is such a "simple case", why are you having difficulty designing it? In the olden days if you didn't have previous experience with a similar nozzle or didn't conduct an expensive test, you simply wouldn't have been able to do it. It sounds like you are in that position now. Fortunately for you it isn't the olden days any more. Today you have the option of a safe and relatively inexpensive option of FEM when no other design method exists.

Blasting ahead with dodgy guesses because it is too "expensive" to use available and appropriate design tools sounds like shoddy short cut taking to me.

Using ASME B31.3 process piping code may help you. It has simple yet conservative loading factors for large d/D junctions in piping systems.

 

[Trestala]

In WRC 107 the solid bar graphs are used to establish stresses at the edge of the Re-Pad. These solid bar graphs are also used for stresses at the OD of nozzles with large t/d.

Solid bar graphs? I have already read WRC 107 and it is silent when it comes to shell reinforcement. Could you be more specific which "solid bar graph" and the figure number? WRC 107 study uses an un-penetrated shell and with no pressure, why would they conduct their tests with reinforcing pads?

I expect WRC 497 doesn't have these solid bar graphs.

It seems you haven't read WRC 497 which is probably the reason you automatically dismiss it and rather use pretty pictures from FEM. WRC 497 in itself is based on FEM data. Why do you claim you are qualified to answer my query?

If your nozzle is such a "simple case", why are you having difficulty designing it? In the olden days if you didn't have previous experience with a similar nozzle or didn't conduct an expensive test, you simply wouldn't have been able to do it. It sounds like you are in that position now. Fortunately for you it isn't the olden days any more. Today you have the option of a safe and relatively inexpensive option of FEM when no other design method exists.

There is no difficulty with this if you have read my question clearly. Compress, PVElite and AutoPIPE Vessel have WRC 107 calculations which establish stress on the edge of the pad even if WRC 107 is silent on the used of reinforcing pads. I am simply asking if this could also be done with WRC 497.

Blasting ahead with dodgy guesses because it is too "expensive" to use available and appropriate design tools sounds like shoddy short cut taking to me.

Application of FEM on everything can be uneconomical and impractical. It costs additional manhours for us. Are you strictly working on the technical side of a project and have no consideration of the economic side?

WRC 107/537 are still being used up to this day. WRC 297, WRC 368 are also being used and accepted by many engineers. If you think it is inappropriate for design, then why Compress, PVElite, AutoPIPE Vessel and other softwares still use it? Why does the AI still accept it? WRC 497 was established to supplement these bulletins, why should this be any different?

WRC 107/537 is silent on reinforcing pads, yet they have developed a method on using it for evaluating stresses on the pad edge. This what I am asking if there is a method on using WRC 497 on calculating stresses on pad edge. Please read WRC 497 first before shutting it down.