Re: Prototype transmission for potential use this year
 

They also used 4 lead worms for increased efficiency, which made them backdrivable. This let them coast to stops, and even theoretically get pushed

The other thing to keep in mind about worm gears is that most of their non-back drivability (even for 1 lead worms) comes from when they're stopped. When they're stopped, you've got a ton of force pushing the worm gear's teeth into the worm (if the output's under load), and because of the angle of the contact between the teeth, it usually can't slip. However, when the tooth surfaces are turning relative to each other, you're dealing with kinetic sliding friction, which is a lot less. You'll actually see the output pushing the input a bit when some worm gearboxes aren't under power, but moving. Because they no longer have to deal with the huge static coeficient of friction, they become essentially temporarily backdrivable. This means that worm gearboxes can coast for a little while, especially when they've got a fair amount of force on the output.

Re: Prototype transmission for potential use this year
 

That seems brilliant! Do you know how that would work? It seems quite tricky!

Re: Prototype transmission for potential use this year
 

From what I've heard they weren't backdriveable until the gears wore in, and they had no problems.

Re: Prototype transmission for potential use this year
 

What do you mean by omni-drive? I think the answer is yes. The sudden stop I don't really think will be an issue, yes the wheels will lock up but worse case the energy will go into the robot frame and might tip the robot slightly or make the wheels just skip over carpeting before anything bad happens. It takes considerable force to strip these teeth but we will see how much of an issue it becomes. We already have several contingency plans in case it becomes an issue(such as a rubber interface as a clutch system that would slip on high force changes).

Re: Prototype transmission for potential use this year
 

Hi sanelss,

Thanks so much for sharing the details of your design; the youtube video was terrific!

We're working on a similar design for sim motor/Jaguar/encoder control/Mecanum wheel. Here're some links that describe gear design that have been discussed in this thread:

Surface hardening and polishing of gear wear-in
Gear tooth strength
Worm gear "anti back-drive" or "self locking"

We looked at worm gear to allow getting the CIM motor 90 deg to wheel but are pursuing a 2-stage design with planetary on the wheel shaft and bevel for 90 deg transition:



Do you have advice on where to source gears? I've only found expensive sets for industrial fabrication and very cheap hobby gears.

Re: Prototype transmission for potential use this year
 

McMaster sells some, or you could go with one from boston gear. I've done a swerve with the L110y. They're pretty expensive.

As for the drive design, bevel gears are going to be more durable than the worm gear setup. However, you do have to take into account that the two gears will be pushing against each other in every way possible, so you'll need thrust bearings, as much support as possible on your shafts, and you should use shorter shafts to minimize bending. It's also advantageous to run the gears at a higher speed if possible. When the gears run slowly, they can bend away from each other and slip.

Also, whenever you use gears, REMEMBER TO LUBRICATE THEM!

Re: Prototype transmission for potential use this year
 

Thanks for the feedback!

I have a couple of L93Y miter gears I bought on Amazon ($17) to play with. Our design has changed and requires a gear reduction and I'm just learning about bevel gear sets. Any guidance on bevel gear selection would be greatly appreciated.

Re: Prototype transmission for potential use this year
 

Ok, there is a lot for me to cover here

First off, please lubricate your gears, and put some sort of cover on the gearbox so the gears don't get covered in debris. Your efficiency is already going to be really low using a two start worm and no lubrication isn't going to help.

I have to agree with many who've already stated this, you are at risk of snapping the worm gear teeth. You have the benefit of using a bigger pitch gear than we did (16 Pitch right? or metric?) but you'll also using a smaller diameter gear and bigger wheels. The gear teeth snapping was one of our main concerns and we would never have used worm gears if they didn't back drive.

As for getting the worm gear in a harder material the only other material that they're made in is cast iron and while thats harder it's also brittle and won't help at all with the teeth snapping issue.

One other comment on the design, you're going to want thrust bearings on both sides of the worm.

Our Gearboxes were always back drivable, we could always push the robot around but it just wouldn't roll by itself early in the season. By champs we could push it and it would roll just like it would if it had a standard gearbox.

We never had any problems with the worm gearboxes, we didn't have to touch them the whole season...or after.

Our worm gears were Boston gear gears's purchased through Motion Industries. Gears with the same specks are available from SDP-SI.
Boston Gear bevel gears are also available through Motion Industries but i would suggest getting Martin Sprocket and Gear gears instead. Martin bevel gears are case hardened and will last longer than the Boston gear ones.
The bevel gears on Mcmaster-carr are Martin but they have a limited selection so we bought ours through Motion Industries.
But yeah they're all pretty expensive...

Re: Prototype transmission for potential use this year
 

I brought up a few of the several concerns here, but I don't have experience with worm gearing for drive parts. My thought is that the teeth will just snap off before the wheel looses traction with the ground. In your gearbox, did you have your worm gear before or after any other reduction?

Re: Prototype transmission for potential use this year
 

I know you were the first one to bring up these concerns and I agree with you completely, I was just adding another voice to the argument.

In our gearbox we had a 1:1.4 spur gear (20 to 28 tooth) reduction from each of the cim motors before the 1:10 reduction of the worm gear (4 start worm and 40 tooth gear). The worm gear was on the output shaft. It can be seen here http://www.chiefdelphi.com/forums/sh...ht=192+gearbox.
The layout is the same as the AndyMark rawbox

Re: Prototype transmission for potential use this year
 

That's really cool. I've never seen a robot with this sort of layout. The gearbox looks pretty small/space saving. Thanks for sharing!

Re: Prototype transmission for potential use this year
 

The gear box that Joey linked to is the first generation of space saving designs that 192 has produced. This year we will be using our third generation of space saving gear box. Each year is different and this gives the students a great engineering project to work thru. The worm gear box was bulletproof.:)

Re: Prototype transmission for potential use this year
 

Joey,

Thanks so much for the info! I hadn't found the Martin company; I've ordered the a Martin set: 3:1, 16P, steel/CA to test. I found them on Amazon where I've have had excellent shipping service. I use Amazon Prime and all Prime orders are free (with subscription) and guaranteed delivery dates usually within 2 days.

Re: Prototype transmission for potential use this year
 

Good to know they're on amazon with prime, ours took a lot longer to get to us from motion. I don't know how much info amazon gave you on the gears but all dimensions and mounting distances are available on their website in the catalogs.
This is the one for their bevel gears http://www.martinsprocket.com/docs/d...s.pdf?sfvrsn=4

Re: Prototype transmission for potential use this year
 

Good point! I almost wrote: Amazon's technical information is very weak and sometime flat-out incorrect. I never select from their info; only part #, and they sometime screw that up.