Hi,

I am looking for some advice re HTF/APF on a particular gearbox in my workplace.

The input gear has 20 teeth and the output 100, giving a common factor of 20 (20 assembly phases). The attached image details all the relevant frequencies for fault finding, given in Hz unless otherwise stated.

My understanding is that it is desirable to design gear combinations which have a GCF of 1, such as 92/35 teeth. As mentioned above, the GCF in the gears I'm looking at is 20. Does this imply that every 20 teeth on the 100 tooth gear could be exposed to greater wear? And would the wear on the 20 tooth gear be even? Or would a single tooth potentially be exposed to greater wear?

The other interesting thing this combination raises is that APF will always be the same frequency as input speed, and HTF will always be the same frequency as the output speed. This could make fault finding for faults specific to these frequencies more difficult, I assume.

Anyway, on topic again. Last year I had reason to believe that the gears were worn and potentially a tooth was chipped. A visual inspection was carried out and damage was found to teeth on the input shaft. I received no feedback regarding the output shaft although I suspected it would show damage also. The input shaft only was replaced at a shutdown due to time constraints - the output shaft is much more difficult to replace due to the arrangement of this gearbox. Readings obtained after start-up still indicated the same issues, so I am awaiting the output shaft being changed also.

My question is, assuming that what I said above is correct - that every 20th tooth is exposed to greater wear, is it possible that a different 'set' of teeth are now being worn given that the input shaft has been changed? For example, with the old shaft the same teeth were meshing every 5 rotations potentially causing excess wear. One new gear has been introduced, would the wear continue on the original set of 5 teeth on the output gear, or could it theoretically wear a different set of 5 teeth?

Finally, would it be worthwhile looking into changing the configuration of these gearboxes to prevent this setup? For what it's worth, there are several gearboxes on-site that have this configuration and this is the only one which has shown these issues. 

Hopefully this all makes sense, thanks for any help you can offer.

Regards,

James

 

Attachments

Images (1)
Original Post

You might want to review the paper I authored at https://www.maintenance.org/fi...enerated_Spectra.pdf.

As to your pinion showing wear and minimal wear on the gear; each gear of a set has what is termed a surface durability rating and a bending strength rating.  The durability rating refers its resistance to pitting; the strength rating refers to the power (or torque) at which point the tooth fractures at its root.  In your case the pinion may have a durability rating significantly less than the gear and thus wear would be anticipated on the pinion, but not necessarily the gear.  In an ideal design you would want the durability rating to be well balanced between the pinion and gear.  Another factor is the number of cycles; in your case a pinion tooth is subjected to 5X the number of engagement cycles as a gear tooth.

When you introduce only a new pinion you are going to essentially set up new wear patterns based on the patterns on the gear, which exist to some degree, even though they may not be visible.  Initially you may get higher than desired vibration and audible noise, but over time, as the two rotors wear in, the levels of vibration and sound should decrease to some baseline.  Unfortunately it isn't easy to put a timeline on when this wear in process is complete as it is dependent on power transmitted relative to the rating, operating hours, etc.  Lube oil analysis is recommended and during this wear in phase I would be doing such testing more often than normal.   

You might want to review the paper I authored at https://www.maintenance.org/fi...enerated_Spectra.pdf.

As to your pinion showing wear and minimal wear on the gear; each gear of a set has what is termed a surface durability rating and a bending strength rating.  The durability rating refers its resistance to pitting; the strength rating refers to the power (or torque) at which point the tooth fractures at its root.  In your case the pinion may have a durability rating significantly less than the gear and thus wear would be anticipated on the pinion, but not necessarily the gear.  In an ideal design you would want the durability rating to be well balanced between the pinion and gear.  Another factor is the number of cycles; in your case a pinion tooth is subjected to 5X the number of engagement cycles as a gear tooth.

When you introduce only a new pinion you are going to essentially set up new wear patterns based on the patterns on the gear, which exist to some degree, even though they may not be visible.  Initially you may get higher than desired vibration and audible noise, but over time, as the two rotors wear in, the levels of vibration and sound should decrease to some baseline.  Unfortunately it isn't easy to put a timeline on when this wear in process is complete as it is dependent on power transmitted relative to the rating, operating hours, etc.  Lube oil analysis is recommended and during this wear in phase I would be doing such testing more often than normal.   

Many thanks for your quick and detailed response John. I do wonder now if I have perhaps been hasty requesting the output shaft also be replaced, given the 'bedding in' period you mentioned. I did think about this but given what I am seeing in my spectral data I decided to call it. The paper you published is a good read - I actually think I've happened across it before a few years ago! Thanks for sharing.

Spectral data on the output shaft shows what I assume to be the GNF (peak not related to any other frequency in or around equipment) with 1x sidebands, as well as fairly high sidebands around GMF. PeakVue data on the output shaft suggests an issue with the gear teeth. The circle plot clearly shows 5 teeth with higher impacts. A second set of 5 teeth can be seen at lower amplitudes, which I assume could be as a result of the input gear being replaced, obviously changing the teeth which mesh. I included the circle plot as an attachment - I tend to take them with a pinch of salt but this one seemed to tie in with everything I was thinking nicely! With regards to the lube oil analysis, this is handled entirely by external contractors and despite my job title being a 'Condition Monitoring Technician' we have little input in this - we are primarily Vibration Analysts who occasionally get to dabble in other technologies!

Attachments

Images (1)

Add Reply

Post
×
×
×
×
Link copied to your clipboard.
×