Please could anyone advise on the most correct method of modelling trunnions (and duckfoots) on elbows. There appear to be 2 schools of thought :

a) The "Kellog" method which involves applying "notional extra flanges" to stiffen the elbow but ingores the increased SIF of the trunnion in CAESAR calc. The Kellog formula is then manually applied to the forces and moments calculated by CAESAR.

b) The direct method which also adds the extra flange to stiffen the elbow but applies the SIF as calculated using the Hankinson / Budlong method as implemented in CAESAR.

The SIFs are relatively high in the Hankinson method and thus allowable loading is low. The Kellog method generally allows much higher loading and is widely used.

[ February 03, 2002: Message edited by: Andrew Weighell ]

[ February 04, 2002: Message edited by: Andrew Weighell ]