Isolator or Hard Mount?

[JTMitchell]

Hello all,
I am designing an electronics enclosure which will hold a few circuit card assemblies as well as some other components. The enclosure itself has a first mode of ~300 Hz in the vertical, and the circuit card assemblies have a first vertical mode of ~700 Hz. The enclosure weight it approx. 60 lbs.

The enlocusre is being mounted on a frame which will have the attached vertical PSD input, 5-500 Hz, 3.84 Grms. Colleagues feel I need to mount these enclosures on isolators, although looking at the spectrum I am having difficulty deciding where I would place the isolator mode. Isolator modes from 10-60 Hz seem to do more harm than good. I can see that a stiff isolator (~70 Hz) wouldn't amplify the frame modes at 10 and 40 Hz, but the vibration is already starting to become attenuated by the frame at this point. Placing the isolator mode below 10 Hz will be a challenge due to sway space limitations.

The question is: Is there a need for isolators here? My preliminary model shows are higher output Grms on the isolated enclosure as opposed the the hard mounted enclosure.

Thank you.

 

[GregLocock]

You'll need travel limiters, mount at 4 Hz, but as you say, hardmounting looks viable.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IRstuff]

Will you have a reliability requirement? 3.84 Grms is pretty harsh, so any MTBF requirement will suffer.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[JTMitchell]

Thanks for the input so far. Yes, the 3.84 Grms is harsh, but unfortunately this is the input at the enclosure attach points that the frame designers have given us. A sub 10 Hz isolator may be a possiblility, I will just have to find one that can withstand the environmentals (temperature range, etc.) and not exceed sway space.

 

[Tmoose]

For pumps and stuff in potentially seismic areas it seems like snubbers are simply added to the mounts that will work ok 99% of the time.

http://ecx.images-amazon.com/images/I/51Tjv6cKWIL._SL500_AA300_.jpg

 

[boyze]

This posting has some age to it but I'll offer my 2 cents. A great reference for this type of situation is Steinberg's "Vibration Analysis for Electronic Equipment" book. The key take away is: knowing and accounting for the separation of the internal (card assemblies and other) natural frequencies relative to the enclosure natural frequencies and of course the excitation frequency range. The worst situation is if both are near other within the excitation range. Ideally they should be well separated. Steinberg recommends at least one octave.

I've seen PCB's dance away at ~ 25 Hz in a random vib environment and crack early on.

For isolation the comments on sway space are important for the reasons cited. Some isolators have built in snubbing which interestingly adds dynamic response complexity by a stiffening spring effect at the higher amplitudes. Consideration should also be given to durability of the isolator as it must flex to provide the "low" stiffness. I've seen rubber isolators get beat pretty badly for low fn situations in random vib. If this is a road situation there's also the "pot hole" and "speed bump" events that can impact isolator durability.

I've attached a sprdsht of a quick look see for an isolated enclosure to see it's response to the PSD for different isolation frequencies.

Good Luck.

james pike

 

[boyze]

I took this question as incentive to finish a macro and short study I've had on the burner for awhile. Attached is a link to a blog posting that summarizes an Excel macro and generic study I did for random vib response of a 2 dof system. Most of the results are intuitive but this analysis provides some specific results and the macro can be used for more general cases.

Have Fun!

James A. Pike

http://www.xl4sim.com

http://www.erieztechnologies.com