Multiple Motors

[phrontist]

So what are the various schemes for using multiple motors to drive your robot? Specifically using the drill motor's and chiapuas together to get better torque/speed. I am asking because my team intends to use this design as part of our tranny next year. Is it possible to add another gearbox that shifts between the two motors, or do you just gear the chia down? I'm really confused and I'd like to hear a comparison between the various approaches, and what everyone think's is the best. I'm also asking because I want to create the FIRSTwiki entry for multiple motor drivetrains (though I'd be delighted if someone more knowledgeable stepped up to do it, or any FIRSTwiki entry for that matter.)

[dez250]

here is a White Paper from team 716 the who'sCTEKs that is a dual speed transmission that runs with the drills and cim motors... I would suggest taking a look at it and try to build it if you have the materials and shop available.
Good Luck!!!

[Hinkel Y.]

Originally Posted by phrontist

So what are the various schemes for using multiple motors to drive your robot? Specifically using the drill motor's and chiapuas together to get better torque/speed. I am asking because my team intends to use this design as part of our tranny next year. Is it possible to add another gearbox that shifts between the two motors, or do you just gear the chia down? I'm really confused and I'd like to hear a comparison between the various approaches, and what everyone think's is the best. I'm also asking because I want to create the FIRSTwiki entry for multiple motor drivetrains (though I'd be delighted if someone more knowledgeable stepped up to do it, or any FIRSTwiki entry for that matter.)

Don't quote me on this, but the it you have to have two secton for the gearbox. One having a single output after you sync the two motors. Then the ouput shaft connects to the shifting part of the gearbox. Our Gearbox here has two sections or you could say, "two gearboxes". One section (on the right) just syncs the three motors together and then having a single output... Then the other section (on the left) has the output shaft from the right section connected to other gears that have in a total of four gears in that section to shift. We use a "dog" to engage the two gears on the bottom.

[phrontist]

Okay, new question.

If I have a 1000 rpm motor gear down 1:2, and a 500 rpm motor running in tandem, and the system is put under load, won't situations occur in which one motor is driving the other? Wouldn't this mean that a two motor system set up this way can acutally decrease speed and/or torque?

Even worse, if one motor is, in effect, "spinning" the other, won't this cause backcurrent to damage the victor?

[Paul H]

We actully used #35 chain and sprockets to match the Drills in low (with threir gearbox) and the Chips before putting it into our transmission. I wasn't on the chassis team, so I don't know what the number of teeth on each sprocket was (something makes me say 3:1, but that's kind of a guess), but I do know that this is how 494 did it on their 2003 robot, maybe someone from their team knows.

Also, if you plan on building our gearbox, make sure you heat treat the gears. We had some problems with that before we heat treated them.

I haven't talked to Jim about it, but hopefully the whitepaper will be up soon with detailed instructions on how to build it. I'm not sure if he'll put instructions on how to make it multi-motor.

Let me know if you have any questions and I'll try to answer them or ask Jim...or I could ask Eric, he built it.

[phrontist]

Originally Posted by Paul H

I haven't talked to Jim about it, but hopefully the whitepaper will be up soon with detailed instructions on how to build it. I'm not sure if he'll put instructions on how to make it multi-motor.

Let me know if you have any questions and I'll try to answer them or ask Jim...or I could ask Eric, he built it.

A whitepaper would be great, but I'd be elated if someone from your team made a really nice page for it on FIRSTwiki. I know a lot of people are interested in your design, and many more would be if they heard about it. See above link.

[rswsmay]

There have been quite a few papers put out about using multiple motors in drive trains, matching their speed/torque, gearing, observing max amp draw, etc. I won't bother to regurgitate what has already been stated. I'll just point you to the references you need.

1. JVN's FIRST Mechanical Design Calculator http://www.chiefdelphi.com/forums/pa...er&paperid=252

2. Discussion on joining 2 different motors http://www.chiefdelphi.com/forums/sh...ghlight=motors

The spreadsheet is quite extensive and should be a great tool for you to use in your design.

Originally Posted by phrontist

Okay, new question.

If I have a 1000 rpm motor gear down 1:2, and a 500 rpm motor running in tandem, and the system is put under load, won't situations occur in which one motor is driving the other? Wouldn't this mean that a two motor system set up this way can acutally decrease speed and/or torque?

Even worse, if one motor is, in effect, "spinning" the other, won't this cause backcurrent to damage the victor?

[sanddrag]

You may want to look at our 2003 design here http://team696.org/forum/image.php?iid=164 There are also other various pictures of it in the gallery. Very simple yet fast and powerful. It did have a lot of weight and binding issues because of multiple pillow blocks and chain tension and whatnot.

I would also recommend the Team 716 2003 single speed as we (and 980) used it this year with much success. It is a little bit more complex than a pillow block, sprocket, and chain setup but it is more efficient, smaller, and lighter.

[rswsmay]

Although we didn't use mulitple motors or multi-speed, we used the same basic design shown in the pic of pillow blocks, chains/sprockets and a CIM motor for each side. Fairly simple design and no real machining or gear boxes to design. The design was to deliver 13.7 fps and actual was 12.9fps, so we hit about 96% of rated speed.

This year we may build our first gear box and use dual drive motors. I am sure we will not attempt (or need) more than 2 speeds unless we end up with a game in 2005 that really demands it.

Originally Posted by sanddrag

You may want to look at our 2003 design here http://team696.org/forum/image.php?iid=164 There are also other various pictures of it in the gallery. Very simple yet fast and powerful. It did have a lot of weight and binding issues because of multiple pillow blocks and chain tension and whatnot.

I would also recommend the Team 716 2003 single speed as we (and 980) used it this year with much success. It is a little bit more complex than a pillow block, sprocket, and chain setup but it is more efficient, smaller, and lighter.

[pryoplasm]

um, i only did electrical, so i cant comment all too much, but we had a nice system of gears last year with the chipahuas and bosch drills evening out, and didnt have any problem.....

[Mike Ciance]

gearing down the chips has worked great for us the last two years. (anybody who wants a good lookat our gearbox, come to Brunswick Eruption 3.0 on October 30th at North Brunwick Township High School)

for 2003 we made an all gears, 4 motor drivetrain which worked wonders (though we had problems with the drills at first) for 2004 we improved the design and it worked even better. talk to our engineer Mike L. PM me if you want to, i'll try and get you talking to him

[Andrew]

Originally Posted by phrontist

Okay, new question.

If I have a 1000 rpm motor gear down 1:2, and a 500 rpm motor running in tandem, and the system is put under load, won't situations occur in which one motor is driving the other? Wouldn't this mean that a two motor system set up this way can acutally decrease speed and/or torque?

Even worse, if one motor is, in effect, "spinning" the other, won't this cause backcurrent to damage the victor?

Unless you have a ridiculous mismatch between the motors' torque-speed curves (at the output of the matching gear-box), the back drive situation won't occur. You would have to have the motor going beyond its free speed in order to get no useful work out of it.

All motors have some "back emf." When back emf = driving voltage, you get your maximum speed. If you drive the motor beyond this speed, the back emf will exceed the driving voltage and you will be "back driving" the speed controller. Up to a point, this shouldn't damage the Victor. However, I'm sure if you created a net 15v at the Victor, you will find yourself into a condition that the designers didn't consider.

On the motor balancing issue, consider it this way.

The Victor supplies a current to your motor. The load provides a torque/speed curve into which the motor is driving.

When your motor acts on the load, it provides a torque which accelerates the load. As the load accelerates, your motor's speed increases, which increases the back emf. The back emf reduces the current supplied, which reduces the torque. At some point, your motor is supplying enough torque to run at the balance speed for a given current supply.

The load that your motor sees is a combination of the torque provided by the other motor and the other load torques provided by the robot.

As the load changes (for instance, your robot accelerates and then reaches steady speed for a particular PWM current), your motor will move up and down the torque-current curve until you reach a particular steady state.

You can actually have a pretty broad mismatch between motor speeds and have the system work. The down side is that your motors don't necessarily run at their optimum efficiency points.

But, where do you want the max efficiency point to be? When you're cruising at top speed? When you're pushing? Since load conditions in a FIRST robot are constantly changing, you will not be running at the most efficient point in any motor for any significant length of time. So, you might as well match the free speeds of your motors.

If you really tweak out your system, you might be able to balance the motors and gear boxes so that you get a little extra top speed or so that you get a little extra pushing power. But, that would take a lot of engineering effort (including verifying measurements to validate your theory).

Based on our experiences with multiple motor drive systems, I suspect that a little software load balancing would be worth the effort.

[Paul H]

Originally Posted by phrontist

A whitepaper would be great, but I'd be elated if someone from your team made a really nice page for it on FIRSTwiki. I know a lot of people are interested in your design, and many more would be if they heard about it. See above link.

Like I said, I wasn't on the chassis team, so I can't really elaborate much. Maybe I can add to what's there a little. I made a team page, but it says the format isn't right. I got sick of trying to redo it, so I just left it as it is. I have a low attention span!