# Unknown Frequency in Planetary Gearbox

Hi, I have been watching a frequency that I can not identify for a year now. It has been increasing every time I take a reading. The frequency is at 55,195. It looks as if it could be a side band of 2xGMF but I think it is an independent frequency. It has increased from .82 mm/sec in November last year to 6.697 currently. I can see this frequency in the motor and 2nd stage of the gearbox but has only increased in amplitude in the 1st stage.  Any help is appreciated. I will try to give all info but if you need something else please let me know.

#### Attachments

Files (1)
Original Post

Is the peak at 1.7 of GMF? If so that has been known as a "ghost frequency" that Taylor and others attribute to the gear cutting machine. I've only encountered it once. The increase of amplitude is concerning. I would contact the GB  manufacturer. Just a guess on my part.

()

What software generated the frequencies? And are you certain the 1st stage is a planetary arrangement (fixed ring or annulus)?  The reason I ask is in a planetary arrangement, and for equally spaced planets, the number of teeth in the ring gear (R) is equal to the number of teeth in the sun gear (S) plus twice the number of teeth in the planet gears (P).  So R = 2 * P + S.  For the 1st stage

S = 25

R = 308

P = 141

but 308 ≠ 2 * 141 + 25.  It is 307.  That would mean you do not have equally spaced planets or some other design criteria may have been applied, hence the reason I ask "are you sure it is a planetary arrangement.'

Also, the 1st stage ratio you have is shown as 13.32.  Normally one thinks of the upper limit of ratio in a simple planetary as about 12:1.  Just another odd fact.

The frequency you mention is very close to 2X of a frequency you show in the calculated results of 462.46 Hz (462.46 * 60 * 2 = 55495 cpm). If I use the penciled in speed of 1195, instead of 1200 RPM, then the frequency becomes 55264 CPM.

In light of your increasing amplitudes, I would suggest that you inspect the condition of the sun gear and planets when it can be done.  This usually does not require disassembly but will require the use of a bore scope.

()
rgf12 posted:

Is the peak at 1.7 of GMF? If so that has been known as a "ghost frequency" that Taylor and others attribute to the gear cutting machine. I've only encountered it once. The increase of amplitude is concerning. I would contact the GB  manufacturer. Just a guess on my part.

This comment makes it seem like "ghost frequency" will be at 1.7X.  This is not the case, although it might be the case for a very specific example in Taylor's book.  I suggest you read https://www.bksv.com/media/doc/233-80.pdf, specifically pages 10 & 11.

()

Thanks for your response. We use CSI 2140 and the Citrix machinery health manager software. I am sure it is a planetary gear with three planets in the first stage and three in the second. The frequencies I provided came from the manufacturer. Since I have been doing this for the last year I was told that the 23 in the order was the pinion but the information given to us from the manufacturer says 25 teeth. Have you ever experienced a difference like this in this style of gearboxes.  As far as the 2X frequency our database lines up our harmonics. In the fifth picture the triangle is over the 2XGMF with what looks like a sideband that is very high. Is it possible for this to be misalignment of the pinion?

()

if the tooth counts came from the OEM, then likely they are correct.

You will note that when I said R = 2 * P + S I clarified that with the statement "for equally" planets.  Let me amplify on that just a bit and state "For evenly spaced planets that are engaging the next tooth at the same time, then both the number of sun teeth and ring gear teeth have to be evenly divisible by the number of planets.  Since the sun tooth count is 25 (or even 23) and the ring tooth count is 308, then it can't have three planets unless something else has occurred in the design.  Again, back to that latest rule, note I said "engaging the next tooth at the same time".   Enter another epicyclic gear term, "non--factorizing".  A non--factorizing planetary gear train is one in which the number of teeth on either the sun gear or ring gear is not evenly divisible by the number of planets.  There can also be "factorizing" and "partial factorizing" tooth counts.This is all a complex subject, to the point that ANSI/AGMA has a 104 page document titled ANSI/AGMA 6123-B06 for Enclosed Epicyclic Gear Drives.

But for now, since this is 2X of sun mesh or a sideband away from 2X mesh, again in my experience a borescope inspection of the involved gearing is justified.  2X of mesh is often associated with increasing levels of friction; FYI, some feel an increase in the 3X mesh is also an indicator of increasing levels of friction.

Just for reference, there is a good article on wind turbine diagnostics using what is termed sideband enery ratio at https://www.gemeasurement.com/...-technical-paper.pdf that you may find useful.

()

The gearbox arrangement is parallel shaft – planetary – planetary. For planetary stages, fixed ring gears and rotating carriers are used. Number of teeth as below:

İNPUT : 1487 rpm

Input state – 61/26

Medium stage – 23/62/148

Output stage – 38/49/137

recaicev@gmail.com

()

"ADIN", are you referring to the same gearbox that the OP "jvaugha" was referring to?

()

NO

Then start your own topic rather than highjacking this thread and potentially confusing the issue!

()