Reducing backlash without antibacklash gears 2

[Proctor43]

Hi all,

I am trying to design a gear train for an encoder system (negligible force exerted on the system). The gears are fairly small though. Right now I am using a dp of 48 and was wondering if going to a higher number could possibly decrease backlash. I think going to higher than AGMA Q9 is out the question price wise. Any thoughts would be appreciated.

Thanks

 

[MikeHalloran]

Use a timing belt, or, better, direct couple with a zero backlash coupling.


Mike Halloran
Pembroke Pines, FL, USA

 

[plasgears]

You may want to experiment with zero clearance between steel pinion and plastic gear. A small interference will run in after a while. Use nylon 6/6 gear for resilience. Face width should not be minimal.

 

[williedawg]

You didn't mention speed..
But, if there is negligible force, can one of the gears be on a spring-loaded pivotable mount, so that the engagement is forced?

 

[Clyde38]

negligible force exerted on the system

Why not try a friction gear arangement instead of toothed gears?

[link

http://www.linkedin.com/in/clydehancock

]

[/url]

 

[Proctor43]

@plasgears What about acetal gears? One of the design goals is zero maintenance.

@Clyde38 I am not sure I fully understand a friction gear arangement. Is that basically no teeth, just rollers? I have been thinking some fairly rigid rollers would do fine.

@williedawg Not sure that would work only because of size issures. Otherwise the max rotation is only +/-170.

I think it would help if I gave a little more scope for the project. those gears all have a face width of less than .1" The black cylinder is a magnet and the drive gear is the larger one in back. The goal is a 1:1 ratio all within this space. Oh and it will all be submerged in mineral oil.

 

[Proctor43]

Forgot to attach the image.

 

[Clyde38]

Proctor43:
Yes, just rollers (zero backlash and no lost motion). Find a material that will deliver the performance you need when submerged in mineral oil.

[link

http://www.linkedin.com/in/clydehancock

]

[/url]

 

[Proctor43]

@Clyde38:

I just realized the problem with rollers is that it might be difficult to keep the required ratio exactly 1:1 especially since they would be deforming at the contact point.

 

[Clyde38]

Again, this would depend on the material you have selected for the rollers. Just for fun, take a look at this:

http://www.fallbrooktech.com/02_demo.asp

[link

http://www.linkedin.com/in/clydehancock

]

[/url]

 

[Proctor43]

Pretty cool product. Ive followed the company since I first read about it in Popsci a few years ago, too bad it seems it only has found a foothold in Bicycles.

 

[plasgears]

Proctor,
Acetal is best for engineered plastic gears, but this application is begging for resilient gears like Nylon 6/6. I had good experience with gear trains running into stall when using 6/6.

 

[Proctor43]

plasgears:

Is stall a good thing? Honestly confused about your sentences. Can we get Nylon 6/6 gears at a good cost(competitive to machined bronze gears, the current design choice) in low quantities?

 

[plasgears]

Proctor,
In some designs the powers that be will not allow mechanical stall torque protection, so you have to resort to thermistors and flexible gears. I have had some success with bellville spring slip clutches, but the cheapest way is thermistors and flexible gears. Your car seat gearmotors probably are designed that way.

By the way:
thermistors provide quick resistance rise when hit with high stall current in a DC circuit.

 

[Proctor43]

plasgears: I think that you are moving away from the problem at hand. There are no loads on the gear set (besides sliding friction of the gears on their shafts and the mass of the magnet).

 

[Clyde38]

Proctor43: If you found NuVinci interesting, you may enjoy this..

http://www.torsen.com/products/T-3.htm

[link

http://www.linkedin.com/in/clydehancock

]

[/url]