Interpretation of MW Kellogg Method Examples

[Robster1us]

After long study of the MW Kellogg methods, and setup of the various form (which lend themselves to spreadsheets quite nicely), I have a much better understanding of how to get at least some result for simple systems with the constraints shown in the book (two places only, although I'm working on the branched systems approach). I like the ability to check the answers, which even if it doesn't make finding errors any easier, at least points out that something's wrong.

My question is this. Although in another thread, I got many good answers on the applicability of the method, now that I have struggled through obtaining the answers, I am trying to understand the extend of its practicality in any system unless it's a very small one of short length and only two constraints at the ends, and resting on nothing in-between.

The reason I ask is that every example seems to presume that "Yes, you can have a 100-ft section of pipe unsupported in any way, and here's how you find the reactions under uperating temperature conditions." At least in my field, we typically support piping on 10-ft (or shorter) centers, and at times U-bolt the pipe to the support, depending on what the structural engineer tells us us required for uplift, etc.

Clearly, this method has been used for real piping systems for a long time prior to the advent of reasonably-priced computer FE modeling. Are there some underlying assumptions that I'm missing that would tell you that your system stresses and end reactions, modeled without supports as in the first 7 examples of the Kellogg General Method, are conservative? Otherwise, how can we know that adding a support, even just a hanger, isn't causing additional, unacceptable stress in the system?

Any help is appreciated.

 

[StevenHPerry]

I don't have the Kellogg book with me, but I'm fairly certain the point you're getting to is this:

The calculations you're looking at are used to calculate the displacement stress range (S_E) only. It is a 'free thermal' analysis.

In the simplified method, lay out your weight supports to satisfy the sustained stress criterion (support per a span table perhaps) then consider the displacement stress range assuming that sustained will be okay.

If you're lucky enough to find some vintage stress calculations, you'll see that approach applied. The stress groups, it seems, didn't even worry about sustained stress. They used code equation 1a (S_A without the S_L component) almost exclusively.

These are my observations from delving into some old facilities and listening to stories from the graybeards.

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.