Are sidebands always (roughly)symmetrical to fundamental gearmesh and its harmonics?
I have four identical gear reducers and only ratio 4.1 known. I tried to find gear teeth from spectrum. The most possible gearmesh are located at 10.5X from my opinion but it's impossible. If tried to assume gearmesh at 11X, all major sidebands located at lefthand sides, is that possible? What machine condition might cause it?

BOPP_II_Silo_Gear_Motor-B.ppt (454 Kb, 106 downloads) Gearmesh

Kevin,

With a 1799 rpm input and a 7.313 rpm output, then I suspect you have at least a 2-stage gearbox and not a single stage, assuming spur or helical gears. So I would not expect that the gearmesh that you have is an exact multiple of input shaft speed. My conclusion from the limited data, assuming that gear vibrations are dominant, is that the presence of output shaft sidebands indicates a gear fault (wear, cracked/broken teeth) on the final stage of the gearbox and not the first stage. You need more info on the gearbox. Two or more measurement points (radial and thrust directions) with PeakVue would help you trackdown the gear and bearing frequencies.

Walt

Walt,

7.313 hz and 1798 rpm 4:1

Even with the confused aspect, Walt is probably right. This looks like a broken or cracked gear tooth.

What type of gearing is it? Has the reducer been disassembled and reassembled without match-marking? These spectra are labeled as being on the motor. Why not on the reducer?

Another possibility is that 10.5 is actually 1/2 gearmesh frequency and you have a 21/84 combination.

I am glad Danny caught my foo-pah (Hz instead of cpm). Has the output shaft speed been confirmed to be 438.8 rpm? I have seen sidebands at 2xSS, so that is why I ask. I assume you know how to use the freqency interpolation feature of the cursor in the spectrum view to get accurate frequency/speed values. Time synchronous averaging is a good technique to separate gears from bearings and to identify which shaft has the fault.

Walt

Thanks for providing opinion.
Speed were measured minutes ago with CSI Speedvue and got 1784/436 rpm. DO you still think it's 2 stage gear?
And I don't understand "Time synchronous averaging is a good technique to separate gears from bearings and to identify which shaft has the fault". Can you explain more?

Thanks.
Kevin

Danny could be correct about the 1/2 gear mesh. The gear appears to be a right angle drive, but the casing is rather complex, possibly indicating more than a single reduction gear. What is the final drive inside the guard; belt or chain? Your vibration levels are very low at 1.6 mm/sec pk or 0.06 in/sec pk, so have the vibration levels increased over time or are high compared to an identical unit? Does the gearbox nameplate confirm a 4:1 ratio or could there be an integral slip clutch for overload protection? The sidebands could be from frequency modulation from the output shaft and not amplitude modulation that is typically from gear and bearing faults. I would NOT open up the gearbox based on this data presented, unless you can confirm a significant change has taken place.

You need a regular optical tach (not SpeedVue) for time synchronous averaging. The CSI web site should have one or more application notes on this subject.

Walt

I think it's a two stage reducer with 2*2.05=4.1 ratio. The 2nd stage have 21:43 teeth. This explain 10.5X gearmesh which is 2nd stage mesh frequency. I will keep monthly monitor to trend the vibration level. Thanks you all for valuable comments.
Kevin Ku.

Gear mesh sideband - the sideband value usually indicated the speed of the shaft supporting the gear with the problem. If you known the tooth combination of the gear set I am able to calculate the patterns developed which may help to understand some of the odd frequencies provided by a gear box
William

The tooth count from gear vendor is:
i = (64 / 32) x (43 / 21)
The 10.5X gearmesh confused me on spectrum is from 2nd shaft. X*(32/64)*21=10.5X.

Thanks
Kevin Ku.