Thanks for your reply Mandeep. I was out on vacation and I just returned so I could not get back with you last week. the example configuration I was looking at is as follows:

Vessel: 96" dia. 1/2" wall
Nozzle: 24" dia. 1/2" wall
R-Pad: 42" dia. 1/2" thick
Material SA516-70
Temp: 500F

As you can see the R-pad is 9" wide.

The above configuration came from a FEA analysis report I found on the internet at the following site which you might want to look at: www.dynamicanalysis.com/pvp96a.htm

The WRC 368 equation 1.65 SQRT (RT)= 11.43" with R (vessel)= 48" and T=1" using the thickness of the vessel + R-pad. Note that in the sample FEA calculation even though the pad was only 9" wide the local membrane plus bending stresses were still attenuated at the pads edge indicating that the WRC 368 equation above gives conservate pad widths.

I ran the same configuration on the Caesar WRC 107 and 297 and got very high results compared to the FEA report as I indicated in my first post. It appears that the way the Caesar WRC programs models the configuration is what leads to the erroneous results. The Caesar WRC models the R-pad, when calculating the stresses at the edge, as if it were just a rigid pipe attachment with an OD equal to the R-pad OD (in this case 42"). Since the highest local membrane and bending stresses are always located at the nozzle connection to the vessel, then the R-pad considered as just a large nozzle results in these very high bending stresses at the pad OD.

However in the FEA it is modeled just as it is in reality, i.e, a 24" nozzle connecting to a flexible and relatively thin vessel + pad which in this case the local bending stresses are attenuated very quickly so at the edge of the pad they are no longer present (in other words the pad does not act like a rigid attachment as assumed by the Caesar WRC calc).

Therefore to me it appears to be not correct to model the pad as just a larger and stiff nozzle and will lead to beefing up the vessel or trying to lower piping loads when not really necessary.