Reduce Noise and Vibration of Planetary Gearset 1

[99ishvr4]

I am wanting to improve a planetary gearset in a cordless power tool. The current gearset creates a lot of noise and vibration. The motor spins at ~17,500 rpm, and the planetary gearset reduces it to ~4000 rpm at the output. The gear ratio is .228. The sun gear has 13 teeth, fixed ring gear has 44 teeth, and the three planet gears have 15 teeth each.

Currently the sun gear is on the motor shaft, the ring gear is fixed, and the planet carrier acts as the output. I am not tied to this configuration. I would like to keep a similar gear ratio but drastically reduce noise and vibration, while maintaining strong gears. The maximum motor torque (stall torque) is 0.23 N-m, so output torque is ~1 N-m. The current gears appear to be cast steel.

I have included pictures of the current gearset, including some pictures under a microscope. My initial research is pointing towards improving the manufacturing or the gear teeth and also applying a meshing phase difference to the planet gears. However, the world of gears seems to be very robust and intimidating. I am trying to read what I can but I quickly become lost. Can anyone help point me in the right direction to start my investigation? Also can anyone comment on the quality of the current gear teeth?

General assembly:

Planet Gear (magnification):

Ring Gear (magnification):

 

[mfgenggear]

Please advise the material and hardness requirement.
The issue is poorly designed gears teeth att r ibutes. There will be a significant cost increase.
It will depend on the quantity produced. Look at improving gear tooth agma quality. The sun and planets can be cnc gear cut. High precision agma quality. And the ring gear can be broached. It's all about the manufacturing method and cost permissible. The material will produce better finish if heat treated to 32 HRc hardness.
If this is to expensive discuss with the die cast manufacture for better agma quality. The more precision and the better finish, pitch diameter run out less vibration.

 

[3DDave]

First do a sound check on the noise and vibration to see how it matches to the gear teeth ratios.

We had a gear reduction drive that was too loud and it was replaced with a rubber belt drive and it made the same racket.

The cause was the chopping frequency for the motor drive. Replaced it with a better motor drive and all the sound went away.

The vibration may also have nothing to do with the gears. They are close to the center and are small in comparison to the rest of the work. See if the system vibrates without the reduction installed. An imbalanced motor rotor would be the biggest contributor. Again, that will be seen in a frequency plot as being in-phase with the vibration. Run the motor at 1000 RPM and see if there is a 1000 RPM spike in the noise/vibration measurement.

In general using even a slight helix will more smoothly transfer motion from tooth to tooth, but those are more expensive that straight tooth gears. You might look at making a separate pilot for the ring gear rather than depending on the gear teeth in the one end of the gear box. The present arrangement lets in a small amount of room for the ring gear to rattle. You could also be mean spirited and use a UV cure adhesive on the teeth on the gear box and curing that with the ring gear in place to ensure it doesn't rattle.

 

[99ishvr4]

mfgengear,

I have not determined the hardness or material requirement yet. I want to confirm, from the pictures I posted, you believe the issue lies in the quality of the gears themselves? This is a concern of mine. As I looked at the teeth under the microscope, I did see that they shapes are not 100% identical from tooth to tooth, but I have not began measuring yet. Do you think higher quality gears in the same orientation/geometry as the current gearset would resolve the issue?

 

[99ishvr4]

3DDave,

Thank you for your reply.

I should have mentioned that I did run the tool with the gearset installed and not installed. The vibration and sound is greatly reduced without the gearset installed, as I perceived it. I do not currently have any means of measuring the vibration. What equipment would you recommend for this?

I can temporarily glue the ring gear in place for testing to see if it makes an improvement. If so, we can look into modifying the design so the ring gear is fixed in place.

 

[99ishvr4]

Here are some videos to compare the sound/vibration with and without the gearset.

Tool with the gearset at max rpm:

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1706111268/tips/WITH_GEARSET_-_MAX_RPM_ispuc1.mp4[/url]

Tool without the gearset at max rpm:

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1706111420/tips/WITHOUT_GEARSET_MAX_RPMS_pikkha.mp4[/url]

 

[3DDave]

You might hold the phone directly against the housing and making a recording and then use audio software, such as Audacity, to look for

https://manual.audacityteam.org/man/plot_spectrum.html

to see which frequencies dominate.

 

[hendersdc]

Sound meters are not very expensive and it sounds like this application would benefit from one instead of trying to do it all on your phone. Here are a couple in the $100 price range

Digikey

[URL unfurl="true"]https://www.mcmaster.com/10235T72/[/url]

 

[GregLocock]

Neither of those is a spectrum analyser.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[DAVIDSTECKER]

Have you considered Helical Gearing?

https://www.ondrivesus.com/resources/technical-notes-and-white-papers/helical-vs-spur-gears

 

[hendersdc]

True, those suggestions would be to just measure the sound level, not vibration.

 

[Tmoose]

Are you tasked with improving the product, or just wanting to your own personal example of the power tool ?

PXL_20240122_223334121_aexgd9.jpg
With the outer cover with the outboard sun gear shaft supports installed, the mesh due to incorrect center distances of the planets with the sun and ring gears looks to be WAY off. Even good quality involute gears can only stand so much excessive center distance.

The ring gear tooth form does not look much like an involute.
Though maybe it doesn't have to be ?

I'd think about:
- Grooving the ring gear back side for an o-ring snubber
- ditto the teeth on the sun gear
- Filling the cavity with crazy thick grease, like one of the Nye motion control/damping products for high end stereo equipment back when they had knobs and feel approaching a flywheel was desired.

https://www.nyelubricants.com/nyogel-975f

 

[mfgenggear]

I don't need to test these gears. The visual appearance of the gear tooth profiles are terrible., and at 17k rpm it will makes a racket and self destruct. For the gears to run smoothly there needs to be conjugate rolling of the gears.
These gears at that rpm must be at least agma class 8 or better. Theses gears are junk.

 

[TugboatEng]

Tmoose, Nye Lubricants Rheolube 377AL would be recommended for this application.

OP, start with incorporating needle bearings into your planet gears. You'll need this to maintain the tolerances recommended by the other users.

I do not suggest helical gears. These generate significant thrust loads on the planet gears that won't be tolerated by grease lubricated gearboxes.

I operate a brand of air starers on my engines. They have very finely toothed straight cut gears with similar ratios to what OP has described. The gears do not make noticeable noise even at very high horsepower transmission.

 

[3DDave]

As an aside - involute teeth do not have conjugate rolling, they do have conjugate action that is insensitive to center distance. Cycloidal teeth have rolling action and are very sensitive to center distance.

There is a large part missing - the output of the planet stage, which should provide better centering and take the other loads required. If the planets aren't centered when there is no load and no output bearing, that is merely a reflection of the amount of clearance, not necessarily some other defect.

A few degrees of twist makes the mesh much smoother; there's not need to run it to 45 degrees. The endplates will handle the thrust load.

Helical contact lessens the shock load of the full tooth coming into and out of contact all at the same time; it's that engagement that makes the most sound.

A big part that is missing in this is the rest of the assembly. If there are matching natural frequencies small inputs can be audibly amplified. The sound of an open gearbox doesn't tell much about how loud it is, particularly without something to compare it with. To me, with and without the planet gears sounded about as loud.

 

[GregLocock]

Just thinking about it, a spectrum analyser isn't going to be much good as the tooth meshing frequency is identical for all the gears (I think). I wonder if using nylon for the planet gears might work.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[3DDave]

Irregularities in the gears won't be. These will be different than from the poles in the motor.

 

[geesaman.d]

Helical contact lessens the shock load of the full tooth coming into and out of contact all at the same time; it's that engagement that makes the most sound.

While I agree with the fact that helical tooth contact is vastly less noisy than spur teeth, I think that only applies to good quality teeth with well controlled alignment and center distance. You wouldn't need a heavy helix angle (with high thrust) to make a big improvement, but wear would be a concern.

I don't have a strong option to offer the OP. Low quality teeth, noisy design, poor alignment/center distance, looseness, etc could all be significant noise contributors.

I do wonder what happens if you apply some lapping compound on the tooth surfaces, run it for a while, clean it out, regrease, and run it again? (You'd want to be very careful to not get it into the journal bearing surfaces)

 

[mfgenggear]

These are junk gears. The quality is the issue. 3ddave symantics