Hello All,
I am a relatively new to performing Elastic-Plastic Analysis and was hoping for some advice, the flange is not standard API 6A geometry, with the hub being significantly thicker. It looks like when it was first designed the dimensions of the studs correspond to a 5-1/8" 10ksi flange (Grade B7) whilst the ring gasket is for a BX-155 flange.
I am following the methodology given in API 17G (Appendix C), and after assessing the leakage capacity based off the contact pressure at the gasket, the structural capacity is determined using 8% peak plastic strain (Yield = 105ksi, UTS = 125 ksi) for the bolts or a through thickness average plastic strain of 2%. As mentioned the hub is significantly thicker than a standard flange so it is really the bolts that are limiting.
My issue with the current approach is that all the literature I have found for permissible bolt loads out with API 17G appear to limit bolt stresses to 83% of yield. This is logical to me as I would of thought plastic strain in the bolts is undesirable, especially in equipment that is safety critical.
I see that when using the 2% strain method ratcheting should be assessed for a design factor of 0.67. It states that ratcheting is okay if either:
- the loading results in only primary stresses without cyclic secondary stresses.
-protection against ratcheting is demonstrated by satisfying the the rules of ASME Sect VIII, Div 3, KD-234
I don't have sufficient data to carry out a full ratcheting analysis using KD-234, but due to the way the flange transfers bending loads (bolts placed into tension for positive bending and flange face transfers load for negative bending cycles) I would expect that in any perfectly plastic model ratcheting would occur in the bolts.
Is my understanding of this correct or am I missing something obvious? We are looking to use this capacity in the GRA but I am hesitant to base it on results where yielding in the bolts is acceptable, particularly as the flange is safety critical.
Thanks,
I am a relatively new to performing Elastic-Plastic Analysis and was hoping for some advice, the flange is not standard API 6A geometry, with the hub being significantly thicker. It looks like when it was first designed the dimensions of the studs correspond to a 5-1/8" 10ksi flange (Grade B7) whilst the ring gasket is for a BX-155 flange.
I am following the methodology given in API 17G (Appendix C), and after assessing the leakage capacity based off the contact pressure at the gasket, the structural capacity is determined using 8% peak plastic strain (Yield = 105ksi, UTS = 125 ksi) for the bolts or a through thickness average plastic strain of 2%. As mentioned the hub is significantly thicker than a standard flange so it is really the bolts that are limiting.
My issue with the current approach is that all the literature I have found for permissible bolt loads out with API 17G appear to limit bolt stresses to 83% of yield. This is logical to me as I would of thought plastic strain in the bolts is undesirable, especially in equipment that is safety critical.
I see that when using the 2% strain method ratcheting should be assessed for a design factor of 0.67. It states that ratcheting is okay if either:
- the loading results in only primary stresses without cyclic secondary stresses.
-protection against ratcheting is demonstrated by satisfying the the rules of ASME Sect VIII, Div 3, KD-234
I don't have sufficient data to carry out a full ratcheting analysis using KD-234, but due to the way the flange transfers bending loads (bolts placed into tension for positive bending and flange face transfers load for negative bending cycles) I would expect that in any perfectly plastic model ratcheting would occur in the bolts.
Is my understanding of this correct or am I missing something obvious? We are looking to use this capacity in the GRA but I am hesitant to base it on results where yielding in the bolts is acceptable, particularly as the flange is safety critical.
Thanks,
