Modelling or not the covers in a cabinet

[DavidDEC]

I'm doing the analysis of an electrical cabinet where a standard Rittal TS8 solution has been used.

The structure of the cabinet alone doesn't seem to reach the natural frequency that is required in this project.

We are considering taking into account the role of the covers and the door in the whole system more rigid. The door is made of 2 mm steel, and the covers use 2 mm aluminium.
Each cover is attached to the structure with 6 M6 bolts, and presses against the structure with a gasket in between.

My question would be first you would consider the role of the covers relevant, and we would not be adding any artificial stiffness.
Also, how would you model these covers? Would you just model the bolt connection, or also the contact between the cover, the gasket and the structure?

Any help or hint would be very much appreciated. Thank you!

David de Esteban

http://www.decnika.com

 

[rickfischer51]

Well, the covers are significant. You usually cannot include contact in a modal analysis because contact is a nonlinearity. I would try and bound the problem. Model it with connections only at the bolt holes, then model it with a gasket that is bonded to both surfaces. your real frequencies should be between those two solutions, probably closer to the second case than the first. If you have a cabinet of known performance, analyze that one first to figure out what works. I'm guessing this is a prelude to a response spectrum seismic analysis. My experience with cabinets like this is that most of the energy will be in the first mode, as indicated by the mass fraction and participation factors. You can take that frequency, plug it into your acceleration spectrum to get a g loading, add some missing mass if need be, and apply that as an equivalent static load to a nonlinear model with contact, plasticity, large deflections, etc.

Rick Fischer
Principal Engineer
Argonne National Laboratory