Hi all,

This is my first post on this forum, as i am fairly new to Caesar.

Within a closed loop piping system, i have modelled several valves as rigid elements, with a representative mass assigned. These valves are to be supported by brackets which fit to the bolting flanges of the valve, (the valve is a 3 piece isolation ball valve). The brackets are fitted to a base which is then located on a sliding bearing surface.

I have applied 2 +Y supports to each valve, one at the inlet and one at the outlet of the valve. On some of the valves, i am getting lift-off at only one of the supports, i.e. the pipework is pivoting about the support.

My concern is that the two pipe supports are not linked together in the caesar model, whereas in reality they are connected to the same base plate.

Should i simply model these supports with a single support on the valve midpoint?

Any help will be appreciated

They look more like Y supports rather than +Y.
Or, you could add a moment restraint at one of those +Y's. For example, if you pipe is running in X, add an RZ.

The base plate on which the valve support brackets are attached is merely sitting on the slider bearing which allows the support to move in the X-Z plane and there is nothing in theory to stop the base plate from lifting away from the slider bearing in the +Y direction. Therefore i wasnt sure that it was accurate to apply a moment restraint.

Identify your limits... Read Piping Design by Kellogg.

Your model is ok. Is it important if one edge will lift-off? If you have PTFE on your base plate, maybe. If not, it is not so important as long as the contact is maintained on the support. In this case you can place a single restraint at the middle.

Regards,

When using beam element programs such as Caesar II, close coupled supports as shown and MODELED AS SHOWN WILL ACT AS A COUPLE and one load will be UP the other DOWN in many situtaions. AS the distance betwwen the supports is small this effect is amplified and quite UNrealistic. Remember that CII is a MATHEMATICAL model of a piping system, not a physical model.

Use ONE support to get more practical results. It is your decision to try some what-if cases to see how sensitive your calc is to the support.

Practical rule of thumb is to combine supports at one point in CII when the supports are within Factor * Diameter. Factor by company, experience, or recommendation. From my Nuclear days, a factor was used and at least 1.0 but I cannot recall what was used/recommended.

If i look from your pictures the valve bracket is bolt to the base, then you should apply X, Y, Z. It is not move axial, lateral and up but it can rotate.
And apply only support in the middle as Bob said, it is more practical.
I have also have experienced with this kind of modelling several years ago, but it was a Firedos pump that was uused in Fire Water Foam System of LNG Plant as you can see in figure below.