[FTC]: Chain-driven drive trains

[madz]

Hi everyone,

We're a 2nd-year team, and up until now we have been using wheels driven by a single gear (last year) or a chain of gears on one side (two motors powering three wheels). We're having a hard time getting the gears to mesh without causing them to strip each other out with lots of metal shavings, or to be too loose if we add some spacers.

We want to switch to a chain-driven system. We have three wheels on each side of the robot and two motors to power them. Should these all be powered by one long chain? Or should one motor run one of the wheels and the other motor run two? What are the pros and cons of each?

Thank you!

[GeeTwo]

The biggest problem with a single chain is that unless you use a lot of idle gears, you will only engage a few teeth on most of the sprockets, which is not at all proper for transferring torque. It would also become completely useless if one link breaks. Coupling each axle to the next with a short chain will give better engagement, and some redundancy. It will require more sprockets and a bit more chain, of course.

[MattRain]

The best option for a chain driven base would be a "West Coast Drive." Do some research on this drive-train, as it is mainly used in FRC.

The Middle wheel is normally geared and driven by one or two motors, located in the middle, and bottom of the robot. From the Center wheel, two separate chains going to each outer wheel. If your chain was to snap, or fall off, then the center wheel would still be moving.

Also, go with bigger sprockets on your wheels. Do not go with a 16 tooth sprocket. If at best, only 8 teeth of the sprocket are engaged with the chain. I would recommend going with a sprocket that has 32 teeth or more.

While we did not create a true WCD, you can look at our drive-train on our twitter and YouTube pages. The main issue with our chain: If the chains snaps that are connected to the motors, we loose the whole side of the robot. Also, I would recommend having both sets of chains on the between the wheels and frame. (We had to move one set outside of the wheels due to motor length.)

https://twitter.com/trojans8640?ref_...Ctwgr%5Eauthor
https://www.youtube.com/watch?v=Cjy5xm3vSQk

[madz]

Thanks, guys.

We have it set up where we can't run two separate chains off the same axle because our robot is almost too wide with no good ways to save any more room, so they have to be in a line. We are using overlapping 6-inch wheels.

We designed it with idler gears so that all the sprockets should have half or more of their teeth on the chain if we use one long chain. But one question is how often do chains break? Ours in on the inside and protected from hits, and the axels are supported on both ends so we hope nothing will bend.

Or if we drive two wheels with one motor and one wheel with the other motor, will that cause problems driving and turning? One end of the robot is heavier than the other by design so we thought putting the one wheel drive on that end. It drove fine with the 2-motor-3-wheel tank gear train and no drop center.

We can test these thing later this week, ourselves, but want to make sure we aren't doing anything too silly.

[orangemoore]

Originally Posted by madz

We designed it with idler gears so that all the sprockets should have half or more of their teeth on the chain if we use one long chain. But one question is how often do chains break? Ours in on the inside and protected from hits, and the axels are supported on both ends so we hope nothing will bend.

If someone ever breaks a chain in FTC they are doing something very very wrong. 25 chain shouldn't break.

I've attached a picture showing a relatively simple way to run a single chain for 3 wheels and 2 motors.

[RRLedford]

Originally Posted by orangemoore

If someone ever breaks a chain in FTC they are doing something very very wrong. 25 chain shouldn't break.

I've attached a picture showing a relatively simple way to run a single chain for 3 wheels and 2 motors.

There are two ways that a chain can "break" in a match:
1) the rare, unlikely way where chain tension limit is exceeded
2) the more frequent and likely way is for a master link clip disengaging & falling off. Then, soon after, the clip side link plate slides off the pins, leading to the rest of the link popping out and effectively "breaking" the chain..
Either way, with a single chain per side, this essentially kills your bot.

[RRLedford]

Originally Posted by madz

Thanks, guys.

We have it set up where we can't run two separate chains off the same axle because our robot is almost too wide with no good ways to save any more room, so they have to be in a line. We are using overlapping 6-inch wheels.

We designed it with idler gears so that all the sprockets should have half or more of their teeth on the chain if we use one long chain. But one question is how often do chains break? Ours in on the inside and protected from hits, and the axles are supported on both ends so we hope nothing will bend.

Or if we drive two wheels with one motor and one wheel with the other motor, will that cause problems driving and turning? One end of the robot is heavier than the other by design so we thought putting the one wheel drive on that end. It drove fine with the 2-motor-3-wheel tank gear train and no drop center.

We can test these things later this week, ourselves, but want to make sure we aren't doing anything too silly.

Our FTC team (3507 Robotheosis) avoids using Tetrix gears anymore if we can avoid them, or only for prototyping purposes. Tetrix axles bend too easily, the framing system is too flexible, and the motor mounts are too imprecise for keeping a pair of gears properly spaced and aligned as robots get used, abused and banged up while competing.

Chain drive designs tolerate greater assembly imprecision and abuse, but they do put more stress on the axles/bearings. Note the stiffer overlapping channel mini bearing modules in the PICS below (1st two from prior years). We also never torque the driven items via the axles - we just attach driven sprockets to driven item and let axles free spin in bearing blocks. Note also the use of long heat treated 10-32 socket head cap screws as axles. With just a bit of polishing of the unthreaded zone they make terrific and very rugged high performance axles.






Our chained drive train this year sounds similar to yours, but with eight 6-inch "wheels" having two overlapping pairs per side pattern, with a single motor for each pair.



A previous year's design:


However for your chain runs, I strongly advise you to to NOT use the Tetrix set screw motor hubs (not even with a pair of them per sprocket), because of the set screw loosing issue with the frequency of reciprocating high torque that a drive train demands.
Get the new 2-piece Tetrix clamping motor hub from from and make sure to lube the set screw with just a tiny bit of grease so you do not get any on the bore hole. They will stay tight much better than a set screw hub and they better handle the frequent reversing torque that a drive train generates.

With six wheels and four motors, any weight imbalance (front/back) from having the longer/heavier chain of the two wheels paired with the one motor on each side will serve to help climb the mountain if it is toward the front. Our front-to-back weight split point is located just behind our 2nd (from front) wheel axle, which helps enormously for climbing the churros.

-Dick Ledford

[wgardner]

Originally Posted by RRLedford

Get the new 2-piece Tetrix clamping motor hub from from and make sure to lube the set screw with just a tiny bit of grease so you do not get any on the bore hole. They will stay tight much better than a set screw hub and they better handle the frequent reversing torque that a drive train generates.

Hi Dick,

Can you provide a link to the clamping motor hub you're describing? I can't find it on the Pitsco site. Thanks in advance!

[RRLedford]

Originally Posted by wgardner

Hi Dick,

Can you provide a link to the clamping motor hub you're describing? I can't find it on the Pitsco site. Thanks in advance!

My Bad - this is an AndyMark product and it is NOT a hub but a double split sleeved 16-toothed sprocket and a matching split clamp collar. :

S25-16T 6mm Bore Aluminum Sprocket (am-3185)
LINK→http://www.andymark.com/product-p/am-3185.htm


So as long as you can get the gear ratio you need while being locked in for 16 teeth at the motor sproccket it is an excellent performing item.

BTW- my complaint on the Aluminum alloy is also BOGUS, as it is listed as being 7075, so the one shaky clamp screw thread may have been caused by either a bad hole drilling/tapping or a bad screw thread.

If you can make it over the hump of the high $20 price, you will not be disappointed. The ability to reposition the sprocket axially along the shaft with out having to move the motor or incurring progressive shaft damage from cycled set screw gouges, that eventually lead to hubs jammed onto motor shafts, is an added bonus.

-Dick Ledford

[madz]

Thanks, everyone. For now we've gone with two chains on each side and it works pretty well in testing!