John –

Your responses to the questions of my alter ego/owner/manager were very helpful to me and I hope to others too. I’d like to respond to just a few issues that linger dispite such careful and thoughtful input.

1. Regarding the fact that the Code does not provide specific guidance for every possible configuration… the Code helps one evaluate tees, but not necessary every fitting-to-fitting group that doesn’t comply with D300, Note 13. You offer:
Quote:
“One of the possible "additional considerations" might comprise the application of knowledge of many years of successful service of similar constructs in other piping systems (i.e., engineering).”
This refers to 319.4.1(a) “duplicates, or replaces without significant change, a system operating with a sucessful service record.”

I have no doubt that such “duplicate” applications exist, but I started analyzing piping 25 years ago and I can’t say I’ve seen any duplicate systems, especially when it comes to branch connections. I’m also not sure that if I did know of a duplicate system configuration in operation for 20 years, how I would know it wasn’t just a few cycles away from fatigue failure, without performing NDE and as much analysis as it would take to just analyze the duplicate system in the first place.

So, I don’t find the use of “duplicates…with a sucessful service record” of much practical value in the design of branch connections in plant piping.

2. On the same topic that “the code can’t address every configuration” you comment:
Quote:
“Ken, would you rather that the phrase "...may require special consideration" NOT appear in the Code?”
No, I certainly see the importance of the statement. And I do understand the Code is not saying that it is a violation of the Code to weld other than straight pipe within two diameters.

But I don’t understand why in the same Note 13, the code doesn’t include a straight run distance for the branch of the tee also, and only addresses the distances from the branch centerline. Since the code is specific regarding the header, it seems to lead one to think that such an inaccuracy does not exist with the branch. Is that true? I don’t think so.

3.
Quote:
“Ken, would you suggest that it would be a better Code if Appendix "D" were deleted? With all due respect, is there an attitude of "well if the collective piping Codes cannot be totally comprehensive and perfect in all areas of piping design then they are of no value at all"?”.
No, that is not my attitude, but I do find that attitude more often than I’d like. It is in comments like “We have analyzed systems like this for years sucessfully without stress analysis programs, now with analysis programs like CAESAR II its not enough, we have to perform FEA to know whether a design meets code.” Of course, that statement holds certain assumptions: that the designers would have been informed of problems had their design not been sucessful; that the designs performed prior to when analysis programs were used, actually met the code.

4.
Quote:
“The body of research work (theoretical and experimental) is out there. TID25553 gives all the right references (later documents have supplemented those references). It cannot be expected that the Codes reprint all the underlying theory or reprint even the references to these works. Taken as a whole, the entire body of research work is not sufficient to cover all the design issues that we will encounter. Code Committee members would be the first to tell the world that the Codes are a distillation of the best knowledge that we have (albeit incomplete and imperfect) for piping design. In my view, that does not invalidate the worth of the B31 Codes for Pressure Piping.”
Though I “know” this, it was helpful to read it. I was beginning to think that the Code could reference some of the fundemental literature that is the basis of the code. Not as “required” reading, but as recommended reading in an Appendix. What better way to quickly impress upon the first time reader, and returning readers, that there is a significant knowledge base and history that forms the basis of the code.

In a broader sense, I guess what this whole thread demonstrates is that it is hard for me to accept that after doing this for enough years that I shuld have known better, that I didn’t know better. I didn’t know that stiffer components welded too near to tees can result in higher stresses than indicated by the standard SIF values.
_________________________
Ken