How are the allowables for the circumferential stress at the tip of the wear plate and the circumferential stress at the horn of the saddle determined in PVelite?

In the manual it suggests that these values should be 1.5 times the tensile allowable, and in fact when looking at the example provided for the horizontal vessel module that is the case. However, in my case, using SA 240 316 at 400 deg. F, my tensile allowable is 19300 psi but the allowable given for the circumferential stresses mentioned above is -19260 psi.

I ran the problem in both the standard analysis and in the component option. The allowables given in each analysis were the same; but, the stresses themselves were widely different. The different analyses calculated different Q loads, which accounted for some of the difference in the solutions, but not much. I compared the equations and substitutions and found that the K values substituted in the equations were very different also, even given the same inputs. I ran numerous cases and the component module stresses were significantly higher for each case. Are the programs using different K's for the same equation? If so why and which one is more correct?

Thank you,

Charles Babcock

The circumferential stress at the horn of the saddle is bending type compressive stress. Hence, the allowable stress for this is the minimum of 1.5 times the allowable tensile stress and 0.9 times the yield stress. In your case the yield stress is 21,400 psi and the allowable tensile stress is 19300 psi. So,


    Compressive Allowable Stress = Min (1.5* 19300, .9*21400) = 19260 psi.



As already discussed this morning the difference in the K factor was due to the inconsistency between your input into the component part and into the main part of PVElite.



Thank you for your inquiry, additionally we also found an error in the main part of PVElite, in the calculation of wear plate extension. This has been fixed now and will be available in the next version (4.1) of the program.


Regards,
Mandeep Singh
COADE, Inc.



[This message has been edited by msingh (edited November 03, 2000).]