View Full Version : Gear Boxes
I have been reading all of the threads about drive trains that have been posted here and as always it is amazing to see the different view points and ideas. I have also been looking at different white papers that posters have recommended.
Now my question is more about the various gear boxes that I see on different robots. Over the years we have stuck the the supplied gear boxes and they have served our needs well. This year, due to space constraints and just the desire to try some new things, we are thinking we should at least look at some other options. Soooo, can anyone direct me to to any threads, white papers, pictures on gear box designs. I would love to see anything like the system that the Bombsquad used last year (and posted this season with the new wheels). Or a system where each wheel is independtly powered with a cim motor mounted vertically.
Any help would be appreciated
A solution our team came up with is using the two FP's in the rear , and the two Mabuchi RS545's in the front, mated to BaneBots gearboxes, powering individual wheels.
I'll get a screenshot of the setup for you later this afternoon. The packaging is awesome: the gearbox/motors take up a space 1.5" square and ~ 4.5" deep from the inside edge of the wheel.
mathking
15-01-2009, 09:24
Last year for our mecanum drive we used two pieces of rectangular aluminum tubing, with two pairs of gears in each. One pair of gears connected a CIM to a sprocket and chain to the front wheel and the other pair connected a CIM to the back wheel. Thus two gear boxes actually housed four transmissions, allowing four wheels each to be driven by a separate motor. I think we still have a CAD of it, and if so I will have a student post it. This gear box was relatively easy to design and fabricate. We had to cut some openings in the top to allow easier access, but it does not require really advanced machining skills.
What ever you guys can send me would be appreciated. A CAD drawing would definitely help.
MrForbes
15-01-2009, 10:55
That's ironic...this is the first year that we decided quickly that the kit transmissions would be just fine.
mathking
15-01-2009, 11:33
Here are some drawings of the gear boxes we used last year.
Jon Stratis
15-01-2009, 11:41
This year is kind of funny. In the off-season, the mechanical team played around with the Dewalt shifting transmissions (there's a white paper on here somewhere for those), since being able to shift last year would have helped a lot. Now they announce the game and we say "hmmmm.... shifting transmissions don't seem to be so important this year." So this year our gearboxes will pretty much be determined by space constraints. Some gear boxes, like the Dewalt's, are longer but narrower, while the KoP gear boxes are much shorter and wider. Which one fits better within the size constraints? I'm just a lowly programmer, so don't ask me! Other than that, i don't think i can say much, or the team members will hit me tonight for giving away team secrets.
During the off season we spent time working on drive systems where our focus was on different wheels and different programs but using the same basic gear boxes. We even purchased a wide variety of wheels even though FIRST emailed all the teams about buying equipment before knowing the game.
Sadly all of our work goes to the back burner until (maybe) next year.
I do agree that the KoP gear boxes would do the job. For the team it is more the space 4 gear boxes take up along with the orientation. For me personaly I want to just try something different and see where it goes. Even if we have to throw out the idea and bolt on the KoP boxes in the end.
I do agree that the KoP gear boxes would do the job. For the team it is more the space 4 gear boxes take up along with the orientation. For me personaly I want to just try something different and see where it goes. Even if we have to throw out the idea and bolt on the KoP boxes in the end.Any particular reason why you can't use only two and run chain or belt to an extra wheel?
Jared Russell
15-01-2009, 13:21
www.banebots.com still sells a wide array of CIM gearboxes
www.andymark.biz now sells planetary CIM gearboxes, as well.
Jon Stratis
15-01-2009, 13:29
Any particular reason why you can't use only two and run chain or belt to an extra wheel?
You certainly could, and with the surface and wheels this year i doubt you get any additional power, speed, or acceleration from driving each wheel independently. However, for a traction control system, driving the wheels independently does increase your ability to detect and correct slippage...
Yes, because some of the programming work is set around comparing wheel speeds and adjusting individual motor speeds. So (hopefully) each wheel will have a different motor.
Also, the drive chains would pass through areas already designated for our
gathering system. Not an impossible fix but we need to get away from our standard drive system or change the gathering system.
As a side note a great example of a space saving system is at the top of the page:
http://baxterbombsquad.com/whatisfirst.htm
here you go:
http://xs135.xs.to/xs135/09034/banebots545331.png
that's a mabuchi RS-545 with banebots gearbox mated to a kit wheel
The Banebots planetaries do not like side loading-I would support the other end of the shaft if you want to direct drive them.
The Banebots planetaries do not like side loading-I would support the other end of the shaft if you want to direct drive them.That design uses an external bearing block with an R8 bearing (mcmaster PN 60355K15) to support the shaft. That would certainly help, wouldn't it?
That design uses an external bearing block with an R8 bearing (mcmaster PN 60355K15) to support the shaft. That would certainly help, wouldn't it?
Yep. That would work fine.
Yep. That would work fine.Just to clarify, the bearing block is the 1/4" piece on the front of the gearbox. So, the axle is still cantilevered, but the design doesn't rely on the gearbox's internal bearings to keep the shaft aligned.
Still probably fine?
MrForbes
15-01-2009, 22:43
It will help some, but putting that bearing block on the other side of the wheel will be much better (assuming you have chassis there to put it on)
Tristan Lall
15-01-2009, 23:28
It looks like you've got two hubs installed there. Typically one is sufficient.
=Martin=Taylor=
16-01-2009, 01:36
I see we're not the only ones considering using the 545s to drive the bot.
We were thinking of running them through the 3.67:1 AM planet (its more reliable than the BB ones) and then use a 60:12 chain reduction.
It seems kind of crazy using such a small motor, but considering how little torque is necessary it kind of works out...
If we were to direct drive the wheels from a banebots gearbox using the 545's what sort of reduction would we be best to go with?
MrForbes
19-01-2009, 22:46
Hachiban's planned gearing will be about 18:1 and it will get them somewhere around 10 feet per second robot speed, which is pretty typical.
So, if you can find a 16:1 gearbox, that should get you pretty close to where you want to be.
You might want to do some calculations about the torque you'll have available at that reduction, and maybe see if you can figure out your maximum acceleration, and whether or not you'll be able to spin the wheels at take off.
Okay thanks we will do all the math when it gets closer to our build season. (in June sometime)
Daniel_LaFleur
20-01-2009, 09:25
It seems kind of crazy using such a small motor, but considering how little torque is necessary it kind of works out...
Has anyone calculated the CoF of those wheels on the carpet, to see if these motors/gearbox setups will stall those smaller motors?
MrForbes
20-01-2009, 09:37
Stall torque claimed for the motors with a 16:1 planetary gearbox (Banebots web page) (http://banebots.com/pc/MP-36XXX-545/MP-36016-545) is 592 oz-in, or 3.0 lbs-ft. Looks like it'll be close on FRP and probably stall on carpet, eh?
=Martin=Taylor=
20-01-2009, 11:04
Stall torque claimed for the motors with a 16:1 planetary gearbox (Banebots web page) (http://banebots.com/pc/MP-36XXX-545/MP-36016-545) is 592 oz-in, or 3.0 lbs-ft. Looks like it'll be close on FRP and probably stall on carpet, eh?
Yeah, we went back and double checked the math, and with that reduction the motors would have been drawing too many amps. (19 A is okey for a CIM but NOT for a tiny 545!)
We're currently looking at a 2 CIM drive geared to around 19 fps. This may sound like a stupidly high number, but with this setup we won't be producing as much torque and therefore won't be spinning the wheels as much.
We've tried it out on the proto and acceleration was improved a great deal.
Of course every drive train we've ever built has sucked... so bear with me:rolleyes:
MrForbes
20-01-2009, 11:16
just make up for it with one of your typical incredible ball handling thingys!
Stall torque claimed for the motors with a 16:1 planetary gearbox (Banebots web page) (http://banebots.com/pc/MP-36XXX-545/MP-36016-545) is 592 oz-in, or 3.0 lbs-ft. Looks like it'll be close on FRP and probably stall on carpet, eh?
This is why we're going with a 64:1 reduction on the 545's for our drivetrain. At 100% efficiency this slips the wheels on a full weight robot at only 1/4 of the max torque, which is more than enough safety margin to keep the Banebots running cool and fine on FRP. Carpet will be closer to the limit but still under it I believe, and the amount of driving being done on carpet should be little enough to avoid problems.
Also, it seems like a 16:1 would be much to high of a wheel speed to be useful. Unless my math is horribly off, the Banebots site specs No load RPM at 988, and with a 6" wheel this gives a theoretical top speed of 25 fps!
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.