Re: Support for Unconventional valve

The last time I posted a poll about supporting a valve midway for analysis. Now upon further review, we discovered that the valve's centre of gravity is not located at its midpoint. In preparing my model, would it be a true representation if I model two valves with their lengths corresponding to the ends distances from the COG and proportionate weights based on these distances, or would the effect in C2 be the same as modelling two equal valves with the support located midway?

Best regards
Amachree

What I do when I have a large valve (usally w/actuator) and the center of gravity is important and not near the center of the valve, is to model the valve and up to the location of the center of gravity with multiple weightless rigid elements. Then at the location of the C.G., I model a short (1", 1/2" or so) rigid element and give that short element the mass of the valve.

Amachree,

Kevin's method will work, or you can model weightless rigids with a node at the CG and then input a force (equal to the entire weight of the valve) at the CG node and include F1 in your OPE and SUS load cases. That will give you the weight acting at the CG. The method you outline above will put only half the weight at the CG and the other half split between the two ends of your rigids away from the CG.

Loren & Amachree,

I don't recommend using FORCES. If you use FORCES, the MASS of the Valve won't be considered for acceleration loading (G's).

I code a little short RIGID element at the location of the CG and on that short element I input the WEIGHT of the whole valve assembly.

Regards,

2cd Kevins Method....