Re: further: pump connections not infinitely rigid anchors

On Jan 19, 2006, at 10:58 PM, aluser2 wrote:

> many caesar users insist that the endpoints of piping as connected to
> flanges on vessels or pumps MUST be defined as "rigid anchors". They
> refuse to accept any other alternative.
>
> they do not alter the program default of 1e12 or 1e14 lb/in default
> anchor stiffness in all 6 dof.

Gotcha. I didn't quite understand what you were getting at. I do now, if only because I spent a lot of time trying to straighten out one of my client's customers on this very issue a few years back.

My client supplied his customer with a light air-to-air heat exchanger. Customer comes back and says the heat exchanger which had already been delivered _must_ be designed for a load set which he'd sent along and which looked high. After a while it occurred to me that the analysis monkey probably did just what you said--took the heat exchanger as a point of rigid fixity, then whocked up the calculated reaction loads as a design requirement. After buying the heat exchanger.

Much discussion was exchanged, which was not helped by the need to go through some sort of expediter who was a nice person but a total no-hoper with technical discussions. We finally satisfied the customer with a calculation showing the foundation bolting would withstand the given loads. It was never obvious to me that the load set was actually carried by the foundation bolting. He wanted a calculation as if it was, so I gave it to him with the proper caveats, which I doubt he understood. Although I provided a couple of letters and a phone lecture on support stiffness, I also never figured out if the customer realized that he'd badly mis-judged the support stiffness and that his piping stress was probably way over what he thought it was. I couldn't even get orientations for his load set, which might have made sense as reaction forces at a single point, but not with regard to equilibrium of the heat exchanger. I think the notion of static equilibrium went right over the guy's head, like the notion of finite support stiffness. The guy had his Caesar or Coade or ADLpipe analysis, and that was that.

Is it any wonder I preach about first principles at such length?

Christopher Wright P.E. |"They couldn't hit an elephant at chrisw@skypoint.com | this distance" (last words of Gen.

.......................................| John Sedgwick, Spotsylvania