Back-driven Worm Gear

[joandrews]

Hi,

I'm hoping a more experienced ME can shed some light on my problem.

The projet I'm working on involves a simple gear train (all spur gears) that is being driven by a DC motor with a worm gear. The last gear in the train (which is the output gear) is connected to a shaft in the center, but also has a handle attached to that shaft so the user can manually turn that gear and rotate the shaft. All gears here are plastic. This is an electronically operated deadbolt -- same as ones you'd have in your house -- as pictured below.

My problem is this:
As far as I know worm gears can't be back driven, or at least not without seriously damaging the gear directly connected to them, so how do I design my system so that the deadbolt can be turned by the DC motor but also turned by the user manually?

I even bought an electrically operated deadbolt and took it apart. They have a DC motor, with a worm gear, and a gear train very very similar to the design I have in mind. How is it that their product lasts in the field without the gears being destroyed? Their spur gears are very standard plastic, straight cut. Their worm gear seems to be standard as well. I can include pictures if needed.

Was I misinformed? Are worm gears allowed to be back driven without any trouble if they are not helical? Is there something here that is dictated by the height of the threads on the worm gear and the depth of the spacing between the spur gear connecting to it?

Thank you in advance.

 

[MikeHalloran]

Peugeot used a worm gear set in car axles for decades; perhaps they still do. The trick, if you will, is that the gear ratio is somewhere near the typical 3:1 or 4:1, so the helix angle is quite large, and the axle can be backdriven. So it's not true that _no_ worm gears can be backdriven, just the sets with more typical ratios of 10:1 and up.

There is another possibility. Gears normally produce a separating force under load. I have seen a few mechanisms where one gear is mounted elastically, e.g. on springs or on a structure that is intentionally flexible in the direction of the separating force, so at a particular torque level the teeth just 'ride over' each other without damage. It usually makes a soft ratcheting noise. ... sometimes not so soft.

I _think_ I have seen a few mechanisms that do the same thing without the ratcheting noise. I'm not sure how that works, but I'd guess axial travel of a worm, or maybe some funny business with cams and clutches.



Mike Halloran
Pembroke Pines, FL, USA

 

[joandrews]

I have taken apart an off the shelf electronic deadbolt with the manual latch handle (as pictured in the first post). The first spur gear has 36 teeth so the worm gear ratio is 36:1! Somehow they're still allowing the latch to close.

 

[BrianE22]

Are you sure that's the worm gear ratio? For worm gears, the ratio = (number of teeth in gear)/(number of threads in worm).

 

[joandrews]

OK. Sorry. The ratio is 36:5 to then. I played with their model again and it seems they've put a lubricated o-ring around the handle/shaft where the output gear is attached so the gear isn't directly in contact with the handle. When the output gear rotates through power transmitted by the train it has enough force to 'grip' the shaft through the o-ring and rotate the shaft, but when the shaft/handle is rotated manually the gear doesn't rotate. The o-ring creates a low friction surface that causes the shaft to rotate without that power being transmitted to the gear and back up the train.

So I guess my question now is: Is this o-ring a special type?? Would any o-ring work? I'm guessing this is a common scheme?

Sorry I'm an EE by training and trade, I'm doing this for a personal project so I appreciate any guidance on where to find this o-ring or how to spec it, from my fellow MEs.

Thanks.

 

[MikeHalloran]

It's probably a standard o-ring. ... but at least dozens of materials are available in any given size. The only limitation here is that the o-ring compound has to be compatible with the lubricant. If you can identify either material, lube or o-ring, you can find something appropriate for the other.



Mike Halloran
Pembroke Pines, FL, USA

 

[Compositepro]

The proper term for the function of the o-ring is "clutch".

 

[Strong]

The "O-ring" could be a "one way clutch" that grip only in one direction of rotation. This door lock gadget looks a lot like one shown on the "Sharks" TV show where poor folks seek investment money from rich folks.

Walt

 

[joandrews]

Thank you all so much for the help. I'll try and experiment with different o-ring "clutches"