Helical gear noise reduction

[northislander]

We have a 20deg pressure angle helical gear-pinion set (17deg helix angle), AGMA 12 quality, with pinion rotating at 9000rpm, causing an annoying noise frequency ~3700Hz.
Any pointers on noise reduction? Our housing is aluminum. We've been told that perhaps 14.5deg pressure angles would reduce the noise...can anyone concur? Thanks.

 

[TPL]

You need to provide a little more information, otherwise any change such a a different pressure angle is just a 'stab in the dark'.

Firstly, how many teeth are there on the gear/pinion - knowing this would allow the gear mesh frequency to be determined - this in turn will provide a little better information of the source of the noise at 3700Hz.

I would guess that you are looking at some sort of gear mesh related noise. In general, if you can hear this noise then the gears are unlikely to be rolling smoothly over each other (in an old gearbox, this noise is often a symptom of gear wear)

Poor gear mesh can result from misalignment between the gear shafts, so that is something that might need to be checked.

One possibility for consideration is that you have a resonant gearbox casing, where the a surface/panel resonance is excited by an acceptable/normal level of gear meshing vibration.

Have you measured gear mesh vibration and assessed against recognised standards?

Have you tried a contact (blueing) check of the gear teeth?

 

[northislander]

Thanks for the info!
The ~3700Hz frequency corresponds to our calcs for fundamental frequency on this pinion (24tooth,9000rpm). We also have multiple needle roller bearings in the system, but their fundamental frequencies dont appear to correspond to this noise frequency occuring and thus we've narrowed it down to the helical gears. After assembly, the noise level begins to occur, ie it is not wear related. We will try the blueing procedure, thanks. The 14.5deg was suggested by a manufacturer as a recourse to sound reduction, but we wanted to get some feedback before maybe attempting this change.

 

[Strong]

I strongly suggest that you test for a natural frequency in the gear case or other components. A gear case resonance may be a lot easier to control by stiffness, mass or damping change. Controlling resonant amplification can be far more cost effective and provide greater sound reduction as compared to a gear design change.

I would also do a high resolution frequency analysis and/or a demoduated spectrum analysis of the sound/vibration to identify side-band frequencies around the gear tooth frequency. This could indicate which of the two gears are misaligned or have loose bearing fit.

I simply can't imagine going from 17-degree to 14.5-degree helix (assuming same gear quality) would reduce sound, but I'm always willing to learn something new.

Walt

 

[Wicsteve]

You might contact the engineering department at Ohio State University. They have a group that specializes in gear noise and run a consortium with various manufactures in industry.