Yes MoverZ,as usual you’ve pointed out quite well.

In fact, I’m in line with your opinion about Kellogg formula, ASME III Nuclear Code, EN1591 and EN13445.
What about VIII Div2? We avoid it just thinking are too many unknowns?

It would be difficult to convince a client your calculation is following B31.3 code with flanges calculation based on EN code, just to follow a modern approach. A clever client will ask you "what about an ASME code"?
And in my understanding, the EN code is specifically counting all the stuff the Kellogg formula ignores, while the quantity of unknowns for what is in field is fair the same.

We are using Kellogg/Nuclear Code formula as a prudent approach, preferring to stay on the safety side. Yes, the Kellogg formula doesn’t count the flange is not a simply circular plate, doesn’t count that often the gasket is not an O-ring, doesn’t make a difference between a tensile and a compressive axial force.
And finally, how much safety is in?
The answer is not important for nuclear field. The safety is important.
Maybe for other branch of activity it would be reasonable to know a safety factor and to make a sound engineering analysis.

Let me make a remark. Suppose you are over the traditional PEQ limit and you are qualifying the flange based on Coade calculator. This is quite the same as qualifying the flange based on J factor of the Section VIII Div2.- this is my opinion.

My best regards