Hey, our team completed our rookie season but are looking to fix up our robot this summer for off season competition this fall. We are interesting in adding mechanum wheels and we want to try them out when we have plenty of time. However, we’re not really sure what the pros and cons of chain driven and direct drive systems are. So, if you could enlighten us a little bit, that would be awesome. Also, a list of parts would we need for each would also be super helpful. Thanks!!
Pro’s of Direct Drive
- Less components
- Lighter Weight
- No chain tensioners to worry about
Con’s
- If you don’t know the exact drive ratio you want for the bot, it’s harder to adjust speeds with direct drive.
- You’re locked in to putting your transmissions at the 4 corners of the robot. (This year that would have been bad - you would have really limited your space for the ball and your kicker.
- Servicing can be difficult if you aren’t carefull with the design with direct drive.
In your case, if this is a learning situation, I’d definitely say go with chain driven so you can play with your drive ratios until you’re happy.
Tom Line is right on the money for a development robot.
If i knew the gearing I wanted for a *competition *'bot I’d go with direct drive (if at all practical) just to not have to deal with chains, designing tensioning systems, and the keys/set-screws generally associated with sprockets. We used a chain-driven mecanum driveline this year, it was not hassle-free.
Since nobody has said anything about chain-drive:
Pros:
easy to get
very resistant to shock loading
Very reliable when done right (notice the qualifier please)
Cons:
can take up large volumes within the robot
can be unreliable if not done right
can be heavy
Direct drive has the pros of less weight and you don’t have to deal with roller chain, which is really really annoying in my experience. The cons are mainly packaging, if you don’t like where direct driving makes you put your gearboxes you shouldn’t use it. You also have more trouble changing your gear ratio if need be.
Chain drive has the pro of letting you put your gearboxes in the center of your robot and quick reduction adjustments with different sprockets, with the con of a lot more weight, and dealing with the tensioning hassles of roller chain.
If I remember you guys from CT correctly you’ll need to add two more gearboxes to your robot either way.
Well, for issue one. I would recommend a AM SuperShifter. The shifters themselves offer lots of options when it comes to switching out gears and playing with different ratios.
As for the second issue, this is assuming you have a 4 WD. I’ve seen many teams place their direct drive gearboxes right smack dab in the middle of their 6WD, and they get stellar results.
The general preference in a 6WD is that one of the wheels is direct driven, and then the other wheels are either chained or belted together. Either way, you’ll need 8 sprockets/pulleys total (if you go DD) and still a decent amount of chain/belting.
But, Direct Drives do require precision. Gearboxes must be properly integrated into the chassis in order to get the appropriate height off the ground/ample support for the output shafts.
- Sunny
Mecanum drivebases with anything besides four wheels are rare, to say the least.
Oh, I wasn’t aware we were speaking strictly in mecanum drive train.
I’d recommend against Supershifters. For the competitions, if you are driving a mecanum base, you won’t be winning any pushing battles. Having a shifting system won’t help you there. In terms of being able to play with different ratios, Shifters would only allow you to try two. It’s almost better just to make your base in a way that it’s not impossible to change out the sprockets.
For a first time mecanum base, I’d recommend chain drive. Because this won’t be a competition bot, you don’t have to worry about the weight or the volume, and chain will give you numerous benefits as you experiment.
That said, depending on the challenge next year, you’ll most likely want to reevaluate your options should you choose to go with mecanum drive.
Can’t the drive ratio thing just be solved by the software? Just reduce the maximum PWM output
Nope, that just reduces the speed. The drive ratio changes the available speed and torque at the wheel, which affects other things.
Change PWM==change speed only, which is how you drive anyway.
Change gear ratio == change available torque and available speed.
If you fix it in software, you reduce the torque (pushing/accelerating power) necessary. If you have trouble steering straight, you CAN fix that in software. While a mecanum does not need pushing power that much, since the wheels slip, it has to accelerate.
Edit: Eric beat me to it.
LOL This is why I am a programmer, not a hardware guy:D
This is my first year on Team 2410 and we used a chain-based mecanum drive. From what I saw, here’s some specs on it:
Pros:
Easy to set up
Easy to fix
Gave us a lot of experimentation room as far as where the gear boxes went
More space for the edges of our robot
Cons:
Chain is like a woman. High maintenance XD
Chain would get caught on various things
Needed grease every so often
Sometimes would fall off if not tuned exactly
So that’s what I picked up on and could pull off the top of my head. I don’t believe we ever tried a direct drive mecanum so I can’t help on that. Hope my 2cents helps.
A correctly set-up chain drive should require zero maintenance. We use 25 chain and didn’t have a single break or issue, nor did we ever have to adjust it.
The trick is to hand grind (or cnc your chassis) so that you have good alignment.
The other trick is to develop a tensioning system that doesn’t add friction, and keeps the chain snug. Snug, not TIGHT. Tight chain is bad. Snug chain with just a hint of play is perfect.
Our 8 wheel traction drive train had more problems with the plaction wheel fasteners working free then breaking off. We were using grade 8 fasteners and locknuts. A bit too much forward and back working I guess.
I wonder if it would be a good idea for andy mark to put 8 more holes on their chain sprockets and 8 nubs on their wheels so the two ‘key’ together.
Seriously?
the fact that they were grade 8 was probably the problem, hard bolts don’t bend, they snap, and aren’t very forgiving of shock forces, grade 6 is probably a better option and 8 more holes would weaken the sprocket pretty severely.
i’ve never heard of a team who have had issues with bolts snapping before, we never have that problem, but then again, we’ve never used traction drive, only mechanum and a couple tank drives.
Grade 6? As far as I know there are grades 2, 5, and 8. Grade 8 bolts are marked with 6 lines though, it’s all very logical :rolleyes:
Fasteners do very poorly when they’re loose, regardless of grade. I would recommend using ny-lock nuts at the very least, and maybe add a split or star lock-washer. It might be an appropriate place to use a smaller fastener with a higher torque so that the fastener does not lose clamping force during operation.
We started with grade 2’s. The only type of nut we use is nyloc. Those worked for 1 competition, and started breaking and falling onto the field.
We went to grade 8 with nylocs. Those worked for 1 district, State Championships, then started snapping on the Newton field during the finals match.
We’ve replaced the grade 8’s that had broken at Championships and are running that way. I don’t expect a problem: we’ll check it after the Tardec competition this Monday, after Marc later this month, and then after IRI next month.
However, I think that this may be a larger problem for high-torque drives that are used a lot. The side force from the fasteners simply isn’t enough to keep the sprockets from working, and that eventially snaps bolts off or causes them to loosen (nyloc or not).
That’s why I’m suggesting some sort of mating system between the sprocket and the hub of the plaction wheel.
Admittedly, our robot got a lot of use this year at 65+ matches plus practice field time. In addition, we had a high amount of skid-steer drag during turning. We went with the longest wheel base that the motors could turn so that it would be difficult for other robots to spin us. It was ALWAYS the bolts in the double-sprocketed wheel that failed first (the rear wheel was chain driven on the inside and had an outside sprocket that drove the next wheel forward). We had one sprocket on the inside and one on the outside and were using laser-cut spacers that were essentially copies of the Andy-Mark ones.
It sounds as if the bolts are directly transmitting drive-line loads. To the best of my knowledge drive-line fasteners (or any fastener for that matter) does its job best when the clamping force it provides transmits loads through the friction produced from said clamping force. Are the bolts being re-tightened after every match? Are you using a torque wrench to ensure a minimum clamping force or just “TFT” (till it feels tight)?
An FSAE car I worked on used a single lug nut in the middle of the wheel and 6 aluminum studs on the wheel hub. There was no way the aluminum studs could handle braking or engine forces on their own. We did the math and determined that our center-lock lug nuts would need to be torqued to around 110ft-lbs to give enough clamping force to transmit engine and braking torque without relying on the studs, which were there mainly for aligning the wheel. Every time before the car went out, the lugs were all re-torqued, and we never had a problem.