FEA on PV to know fatigye life cycle using PD5500 std

[brightinstinct]

Hi Everyone
I need to carryout a FEA using ANSYS on a designed PV to know the number of cycles that the vessel body and Locking ring can be subjected to under Fatigue.I am looking for some references for similar kind of analysis done.

VESSEL SPECIFIACTION
Design Pressure: 11,000 psi
Max Working Pressure: 10,000 psi
Volume: 7.5 Litres
Temperature: -20 to 150 Degrees Celsius
Internal dims (approx): 126mm bore x 600mm long

DESIGN CODE
PD5500 2003 Cat 1, Asme VIII Div 2 (For Fatigue)

I have never done any FEA for fatigue, so i needed some clue very urgently to start the work.Also please let me know where i can find code PD500.
could you please shed some light on this.
Regards
sanjay

 

[Mandeep1]

Dear Sanjay,

When you are performing fatigue analysis you are looking for peak stress. The peak stress is very high stress but is usually in a localized region. Such as, at the tip of the hole in a plate. Since it is surrounded by regions of lower stress, it does not lead to failure. But, under cyclic loading, a crack can originate from this region of high stress and can lead to failure.

So, you want to find the peak stress. This can be computed using FEA. But, you will have to model the region of high stress such as welds, openings, change in geometry etc, to find the highest stress. This can get very complicated and difficult. For example, in FEA models a sharp corner can cause a singularity, which is to be addressed.

Another approach that is given in the codes is the use of the stress concentration factors. ASME Sec. VIII Div. 2, PD5500 and other codes provide these SCFs by carrying testing using strain gauges. So, once you compute the basic membrane stress in the shell away from the region of high stress then you can multiply this stress by the code SCFs to get “peak stress”. The code provides assumptions and limitations for the use of their SCFs.

You can use either FEA or basic code equations to get the membrane stress.

Once, you get the peak stress you can go to the respective code to find the cycle life. There is an example in Appendix G of ASME Sec. VIII Div.2 code.

I would suggest doing a hand calculation to get the idea of stresses. It will make sure that you are in the ballpark when you get the results from FEA.

Hopefully this will get you further along.

Mandeep Singh

 

[JohnGP]

brightinstinct,
Just to answer the other part of your question, if you really want it, PD5500 is available from BSI, see

http://www.bsi-global.com/Manufacturing/Pressure/pd5500.xalter

Cheers,
John

 

[davefitz]

Out of curiousity, I recall that there is a phenomena in finite elements wherby the apparent stress at a disconitnuity increases as the mesh size decreases, and that this can be addressed using fracture mechanics theory. Can you refresh my memory as to what circumstances lead to this result, of an unbounded increase in apparent stress as mesh size goes to zero?