We have reciprocating compressor piping for which recommendation is to use the hold down type of support with beleville washers (disc springs).
The recommendation is to resist the pulsation load with friction generated at supports.
Friction will be varied by imposing higher vertical load by tightening the bolting against the beleville washers.
How does one simulate the beleville washers and the additional vertical load in Caesar. Pls provide guidance.

Manoj,

Bellevilles are extremely non-linear, may be you can make it linear and used Caesar-II spring model for linear spring.
Also try to utilize the stiffness in analysis, so that the 25% may be or may not be the area of concern.

Long time back, I remember one discussion in Caesar-II forum about the same. Unfortunately, I am unable to find that. Please try the search option and hopefully you will get some help.

One suggetion: - they are treated as shock arrestor with provided loads and spring stiffness.

Regards,
R.K.

Thanks RK,
Tried searching on the discussion forum.Found a couple of ideas.
One is to model bilinear restraint and the other is to impose the additional compression force as a Force input.
Which is advisable?

Another point is since beliville springs are non linear , modelling as springs will only be an approximation.

Regarding the suggestion of these supports being shock absorbers, does it mean that we can model them as snubbers in Caesar input.

Or model as sway brace ?

Manoj,
The shape of the Belleville spring characteristic curve is strongly affected by the relative height h0/t.

Where; h0 will be the inside height of disc while t can be the thickness of material used. If the values of the relative height is small say for e.g. (in the range of 0.25-1.0) the spring has nearly linear features. This you can confirm with your vendor. Also the arrangement of the spring is vital in this case, it all depends on how you want, say for e.g. if the spring arrangement you need for higher spring rate the go for parallel arrangement of these springs and so on.
One important aspect which I forgot to mention is the friction, friction will have a significant impact on these springs. Confirm this with your vendor; how he is going to design the spring, what are the surface treatment and lubricant type and all. Generally for internal friction we can use 0.03-0.05 as friction. (I rem asking this question long time back, but did not got the solution for this, I had observed that with using the inner friction value as less as 0.03, sometimes changes the springs types)

Yes, snubber can be modelled, but be cautious while defining the displacement (at CNODE & load range).
Regards,
R.K.