Designing my own anti-backlash setup

[jvo1]

Hey,

I'm a young engineer and am dealing with a gear problem for the first time. I pretty much have the gears and their specifications down, but the problem is designing a simple anti-backlash system.

My setup is the main (input) shaft with a 24T to 90T gear reduction. Parallel is the second gear reduction of 24T to 144T. All the gears are 96 DP. The centers of all the gears are in the same vertical plane.

I cannot use anti-backlash gears, nor can I use worm gears, due to cost and efficiency. I'm looking to incorporate a method I know as "gear interference" in which I was attaching each of the non-input shafts to a carrier arm whose rotation point was stationary, allow the floating (non-input) shafts some movement leeway, and attaching them together with some sort of spring, so as to keep them engaged. (I've seen this dont with a solid plate, but it seems a spring system would work better).

In short, I'm making my own gearbox. I'm also under the impression that this will adequately solve the backlash problem. I'm not sure, but I think the RPMs are around 12,000.

So my question is: am I making any sense or am I totally off my rocker? Is there a better way to do this (eliminate backlash) with my limitations? If I need to clarify something, please ask, and ANY input is appreciated, thanks.

 

[JakubMech]

Do the input torques/frequencies establish seperating forces, which at the right frequency could setup resonance in the system (speed related).

With the gearboxes I see, we live with a small amount of backlash and ensure it remains within a prescribed limit.

Can the design be adjusted to minimise the backlash to a very low level, without backlash adjusters?

 

[israelkk]

If your gear RPM is 12000 than you need lubrication and for lubrication to work you need some backlash. I do not see how backlash will harm your gear if it works at 12000 RPM. Usually you want to minimize backlash when you you switch directions of the gear rotation and you need precise positioning but not usually at 12000 RPM.

 

[jvo1]

Ok, so assume the RPMs arent 12000 and I cant have backlash (because I'm told I cant).

 

[Philrock]

By spring loading the gears you are generating increased contact forces/stresses. You are essentially driving a wedge into a notch. Check your contact stresses very carefully.

 

[diamondjim]

I have heard of customer using no
backlash in slow moving equipment
to increase the rigidity but these
were steel gears and lubed. What
material are you using. The gear
accuracy was about 10 agma.