Section 8 Div 2 Alternating Equivalent Stress Amplitude

[McT178]

I am checking a report and it appears the author is multiplying equation 5.36 by Young's Modulus ambient divided by Young's Modulus at operation temperature. Is this from an older version of the code? I can not find the Young's Modulus ratio anywhere in the code.

.5 * (Kf * Ke * deltSpk) * E/E

 

[TGS4]

You are correct that such an addition is not in compliance with equation 5.36.

Is it possible that the original calculation of delSpk did not have a Young's modulus correction, and this was a convenient place to do such a correction? Seems odd, though...

 

[McT178]

I can not find E ambient divided by E operation anywhere in the code. Do you know where in the code I can find the Young's modulus correction? This is a bimetallic joint; maybe it is a correction for that, but the modulus used is for carbon steel only.

 

[TGS4]

OK - something just occurred to me - exactly what edition of the Code is the analysis performed to? Because prior to the 2010 Edition, the calculation of the fatigue life in Part 3 was incorrect because the modulus correction was in the wrong location. Is it possible that this author of the report that you are reading was aware of this issue, and rather than put the modulus correction in the S-N curve, put it here?

 

[McT178]

I see what you mean. I looked at the 2007 code and method for determining the number of cycles has changed. I am reworking the analysis and it does not seem to be an added correction for that. This may come from an even older version of the code because N is being calculated from the natural log of Si and Sj. I am new to Sect 8 Div 2, so I could be missing something also.

 

[TGS4]

You should check with the author of the report exactly which Edition of the Code they are working to.

Also, it should be noted that although there is nothing mathematically wrong with using natural logarithms in the interpolations to find S-N from Si,j-Ni,j, the Code specifically says to use logarithms with base 10.

I must compliment you on your thorough checking job. Finding errors like this sometimes requires a fine-toothed comb - something most engineers who do ASME Section VIII, Division 2, Design-By-Analysis may be capable of, but even experienced vessel engineers have troubles with.

I recommend asking the analyst directly your questions. You will be able to tell right away if you should trust them or not. If they are defensive, our try to apply the "Position of Authority" card with you, run. If they engage you, explain what they did, and how/why they did it in a collegial manner, you have someone that you can work with. Trust your instincts in this. They _may_ have more knowledge about FEA for vessels/piping than you do, but you're no slouch/idiot.

 

[McT178]

Thanks, I really try to make sure I understand something 100% before I use it or sign off on it. I pulled out the 2004 code, and it looks like this is what is being used to do the fatigue analysis. I still cannot find the E/E correction, so I think I will take your advice and call the author. Thanks for your help.

 

[McT178]

As an update, you are correct about the modulus correction. The report I am checking is a hybrid of old and new code - the author never took the time to completely update an old spreadsheet. I reworked the report with the 2010 code and get about 30% more cycles. The older codes seems to be more conservative.