Vibration limits of machinery on soft bases

[geesamand]

There are a number of vibration limits based on machinery mounted to rigid foundations. In this case I have good experience and trusted vibration limits for a new machine: 0.3 in/s peak and 3 mils pk-pk.

I have an application where we are mounting a machine on a floating platform. The mass of the rotor is a significant portion of the total structure weight, and we're finding that the normal vibration levels are a bit higher. The literature and common sense seem to indicate higher vibration is normal in this situation.

The question is: how damaging is it really? I envision that at higher levels, the deflections will generate fatigue failures and significantly reduce bearing life. Any methods or rules of thumb to estimate a safe and practical vibration limit here?

Thanks,

David

 

[Tmoose]

If the larger vibration amplitude is causing some part of the metal structure to bend more, then I guess fatigue and cracking could result, if the resulting stresses are above the endurance limit, or there are details that cause even higher local stress due to poor geometry.

The bearing loads often are reduced by the lower resistance of the more flexible structure.
Consider the opposite. Imagine an infinitely rigid structure. A machine could have immense unbalance, exerting forces 20 times the rotor weight or normal process loads on the bearings. The rigid structure would restrain the bearing housings so well the world would have no idea the hell the bearings were having to endure.

Auxilliary Components like switch boxes, wiring and connections going along for the ride may not enjoy the increased vibration levels.

 

[drawoh]

geesamand,

Is the rotor causing the vibration?

Do you know the spring rate of your mounts? If you have the acceleration and the peak to peak movement, you should be able to work out the forces. If you know the mass rating and resonant frequency, you can approximately work out the spring rate.

--
JHG

 

[geesamand]

The rotor is the cause of the vibration based on spectra and common sense. We have the usual contributions from imbalance and misalignment. We have another contributor at 3500cpm but it's much smaller and not appearing on the overall vibration much.

The "mounts" are not to ground - the machine floats in liquid. I can see how acceleration data might lead to reasonable estimates of internal forces. As well, I can see how looking at the attenuation of vibration as it migrates away from the source might give some idea where the vibration is applying the most load.