pic: DeWalt Gearbox with two CIMs

[dlavery]

This is a gearbox for a direct drive system I have been working on. It was inspired by team 1097s direct drive system (http://www.chiefdelphi.com/media/photos/20620). The gear ratio between the CIMs and the dewalt replaces the "final drive" ratio in a standard system. The gearboxes will be mounted on one end of the robot and a .5" drive shaft will run the length of the robot. The wheels will get power from the shaft through a several sets of miter gears. To avoid using U joints like team 1097 did I plan on using different thicknesses of tread or different sized wheels to get the center wheel offset.

OK, not to be too pedantic about this, but "direct drive" and "gearbox" are oxymoronic when used to describe the same system. A direct drive system is, by definition, one in which the load is coupled directly to the output shaft of the motor with no intermediate gear box or other torque/speed modification in the linkage.

Other than the terminology issue, this looks like a nice design. If you are going to go with the idea of a 0.5" drive shaft running the length of the robot, I would suggest that you consider using a torque tube rather than a solid shaft. A 0.1" wall torque tube 24" long with a 0.5" bore will save approximately 40% of the mass of a solid 0.5" shaft made of the same material (and even more if the torque tube is aluminum vs. a solid steel shaft).

-dave

[Jeff Pahl]

This is still a conceptual design at this point. We have not run the calculations yet to determine if the Dewalt transmission is going to hold up to this kind of abuse. The idea is to see if we can come up with a two-motor, shift-on-the-fly drivetrain this year. Matt decided to try this approach since we have some extra Dewalt parts left over from a couple of years ago, so the cost to experiment is low.

Dave- it absolutely will be a torque tube design for the power transfer to the wheels, with an aluminum tube. Once we have the torque calculations we will decide just how light the tube can be. (He's working with a bunch of mentors that design spaceflight hardware, the first question is always "so just how much does that weigh?" )

[Bill_Hancoc]

Last year i made a concept using the opposite of what you have here...i never made it but this is what it looked like. Here.

[Kingofl337]

Team 40 beat the living snot out the Dewalts and they never even marked a gear.

They are hardened to very high rockwell, we lost one of the mentors to Dewalt and the engineers couldn't believe we even were able to bore out one of the gears from what she told us. From what I understand the rings or posts for the gears will break if anything does. In which case they should be easy to swap out.

[newton418]

Are you trying to save weight, space, or are you just throwing around ideas?
Looks cool, though with the torque you have going into the DeWalt transmission I would definitely test it first.
If you are trying to save weight, how much do you expect the whole thing to weigh (minus the motors)?

[MAteo9944]

http://www.chiefdelphi.com/media/photos/26104
There is the updated version.
This gearbox is mostly me just throwing ideas around. I like the ease of a prepackaged 3 speed gearbox in the Dewalts, but I wanted to use two motors per gearbox for torque.
Without the motors the gearbox is just under 2.5 lbs. More weight can be removed though if necessary.

[Lil' Lavery]

(the delete button seems to have vanished from me.... it's not where it was)

[newton418]

To specify my question, why not get the extra torque after running each CIM through their own DeWalt transmissions? Though I have to say, if it only weights 2.5 lbs without the motors, that would be a good reason. If you build it make sure to post pictures and results!

[Greg Needel]

Originally Posted by newton418

To specify my question, why not get the extra torque after running each CIM through their own DeWalt transmissions?

Running each cim into it's own transmission would not give you extra torque. In most if not all cases the total torque is the same. Since each motor has a maximum amount of power it can generate at 12 volts within the given amperage requirements all you would be doing by making separate gearboxes is spreading out the load. In this situation the only reason I would consider doing that is if I was concerned with the torque of 2 cims breaking the deWalt transmission (yes I am thinking that).

From my experience in drive train design I will always advise that people go with a central gearbox and power transmitted through chain or another means. This is due to your overall friction force (another discussion) but basically if you spread out the load to all of your wheels each wheel can take the total torque produced in the situation where another set is off the ground (assuming the mass is over the wheels touching the ground) that is where the normal force/ weight of the robot plays in (yet another discussion) either way keeping your motors together will save you weight in the long run and give you the nice ability to put the torque everywhere you want it to be in the correct situations.

[Cody Carey]

Originally Posted by Greg Needel

From my experience in drive train design I will always advise that people go with a central gearbox and power transmitted through chain or another means. This is due to your overall friction force (another discussion) but basically if you spread out the load to all of your wheels each wheel can take the total torque produced in the situation where another set is off the ground (assuming the mass is over the wheels touching the ground) that is where the normal force/ weight of the robot plays in (yet another discussion)

I'm confused... I don't quite see where this paragraph fits in with proving that there will never be more torque if you put the motors through separate gearboxes. Maybe it's just me only having a highschool physics background, but are some of the words taken a bit out of context? maybe a diagram would help me? I understand the terminology, this paragraph just confuses me.


But then again, maybe it is because it is 2:30 in the morning.

[Greg Needel]

Originally Posted by Cody C

I'm confused... I don't quite see where this paragraph fits in with proving that there will never be more torque if you put the motors through separate gearboxes. Maybe it's just me only having a highschool physics background, but are some of the words taken a bit out of context? maybe a diagram would help me? I understand the terminology, this paragraph just confuses me.


But then again, maybe it is because it is 2:30 in the morning.

not a problem, I am only half awake myself. That paragraph is talking more about the design system then proving that there won't be more torque. But since you asked I will explain why you don't get more torque.

Torque from dc motors is theoretically limitless, the more voltage you put to the motor the higher the torque you can create. Now in a real world (FIRST included) motors can only take so much voltage before they burn up. So especially in FIRST where your voltage is limited so is your maximum torque. The best way that I can think of describing the scenario in question is by comparing it to water. If the gearbox is designed correctly to maximize torque without drawing more then 40 amps then you have 1 full water container. This container got it's water (torque) from both motors combined (the max before tripping breakers). Separating the motors is like pouring the water into 2 glasses, you still have the same amount of water it is just in a different place. Because torque is limited by voltage and amperage it doesn't matter how you split it, it will remain the same.

Assuming the same gear ratio in each gearbox consider the following

max torque out of 1 transmission with 4 motors
=
combined max torque of 2 transmissions with 2 motors each
=
combined max torque of 4 transmissions with 1 motor each.

Now there are other things to consider like gearbox inefficiencies and other losses but for all intensive purposes you can pass the water around all day but in the end you are still left with the same amount of water.

[MAteo9944]

Originally Posted by Greg Needel

Now there are other things to consider like gearbox inefficiencies and other losses but for all intensive purposes you can pass the water around all day but in the end you are still left with the same amount of water.

Another thing that you hinted at was that if you have 4 transmissions, if some of your wheels lose grip or leave the ground then you suddenly have less torque available. (If your wheels are leaving the ground during a pushing match you have other problems.)

[newton418]

Originally Posted by newton418

To specify my question, why not get the extra torque after running each CIM through their own DeWalt transmissions? Though I have to say, if it only weights 2.5 lbs without the motors, that would be a good reason. If you build it make sure to post pictures and results!

Originally Posted by Greg Needel

Running each cim into it's own transmission would not give you extra torque. In most if not all cases the total torque is the same. Since each motor has a maximum amount of power it can generate at 12 volts within the given amperage requirements all you would be doing by making separate gearboxes is spreading out the load. In this situation the only reason I would consider doing that is if I was concerned with the torque of 2 cims breaking the deWalt transmission (yes I am thinking that).

From my experience in drive train design I will always advise that people go with a central gearbox and power transmitted through chain or another means. This is due to your overall friction force (another discussion) but basically if you spread out the load to all of your wheels each wheel can take the total torque produced in the situation where another set is off the ground (assuming the mass is over the wheels touching the ground) that is where the normal force/ weight of the robot plays in (yet another discussion) either way keeping your motors together will save you weight in the long run and give you the nice ability to put the torque everywhere you want it to be in the correct situations.

What I meant by "the extra torque" is the extra torque you are getting from the second motor. Basically, you have two different scenarios:

(1) You have a transmission with a given ratio, and you have two identical motors inputting power (as in your design), then you get a specific max output torque from the transmission. This can then be connected to a drive train (preferably where all wheels are connected, so you power all wheels), determining the max output torque on the wheel axels.

(2) Now consider taking the same motors and the same transmission, and having one motor input power into one transmission. If you take two of these single motor-transmission assemblies and connect the output shaft of each transmission to a drive train (where all wheels are connected), you end up with the same max output torque on the wheel axels.

So my question is, what is the advantage to the former setup (besides potentially saving weight)?

[Greg Needel]

Originally Posted by newton418

So my question is, what is the advantage to the former setup (besides potentially saving weight)?

well there are a few, but by no means would they outrank a centralized gearbox in my book.

weight distribution - with each motor (cim) weighing about 5 lbs you might want an even weight throughout the chassis as apposed to a centralized one.

swap ability - for some reason if a motor burns instead of taking a transmission apart disabling your drive train (with a centralized tranny) you can just quick replace the single module that is broken.

Robot layout - sometimes it may be easier to mount 1 single motor and tranny then a full gearbox. The example that comes to mind for me is a planetary set where they are really compact.

I am sure there are more....the main thing to remember is that there is no "one and only solution" and it is all about evaluating your specific situation, and making the best decisions you can. Every team has their own reason for doing things, all of which are valid.