coupling Bosch & Atwood motors

[caffel]

Any theories as to the most efficient method for coupling the two together to get more power for driving the machine?
Most people that do this seem to create a gear ratio that matches the no-load output speeds of the two.
(then feed them the same voltage)
Is there another way? If so what would be gained?

[Matt Reiland]

You can also drive independent wheels with independent motors and let the wheels pick up the speed/power differences. We went with gear drive between the two with the bosch (18t) driving a 48 tooth gear mounted to the Atwood, worked very well as power goes.

[Jnadke]

People continually think that combining motors is black magic. It's really not that bad once you understand the Physics involved. Just as long as you get the free speeds approximately the same (about +/- 10%), the motors will perform as individual components. If the free speeds are not exactly same, the only time you'll run into problems is when the motors are operating at their free speeds.

This year matching motors is so much easier because the power output of the CIM and drill motors are nearly identical. Watch out with the drill motor though, you lose 10-15% when running them in reverse.

[Matt Reiland]

Jnadke is right, there is a very noticable change in the power on the drills, enought that without this:

IF ( (HighGear=1) AND ( ChipDriveLeft > 127 )) THEN 'When in High and driving forward reduce Left Drill Power
DrillDriveLeft=127+(((ChipDriveLeft-127) *9)/10) MIN 127 MAX 254
ELSE
DrillDriveLeft=ChipDriveLeft
ENDIF

IF ( (HighGear=1) AND ( 127 > ChipDriveRight )) THEN 'When in High and driving reverse reduce Right Drill Power
DrillDriveRight=127-(((127-ChipDriveRight) *9)/10) MIN 0 MAX 127
ELSE
DrillDriveRight=ChipDriveRight
ENDIF

Our robot wouldn't track straight, next year we will try to make left @ right trun the same direction to get max power out without software limiting.

[Greg Ross]

Originally posted by Jnadke
If the free speeds are not exactly same, the only time you'll run into problems is when the motors are operating at their free speeds.

Can you elaborate on the kind(s) of problems you would run into when the motors ARE running at free speed?

[Matt Reiland]

I think that even though both our motors were not geared exactly the same (as far as free speed goes), we never could reach free speed because of the losses and drag of the drive system, therefore the problems of two motors trying to spin at different speeds never surfaced in our setup.

[JVN]

Originally posted by Matt Reiland
Our robot wouldn't track straight, next year we will try to make left @ right trun the same direction to get max power out without software limiting.

You can do this relatively easily, even if you have "mirrored" gearboxes. We do by tweaking our chain wrap. On one side, the chain runs along the bottom of the output sprocket, while on the opposing side, the chain runs along the top of the output sprocket. Thus, with all 4 motors running forward, the wheels spin the same way.

Although, this probably won't help you, because if I'm not mistaken 226 uses a direct drive system to the tracks.

Perhaps you could simply put in an additional 1:1 gear reduction on one side to reverse the output spin? You would lose a little bit to energy loss, but it would allow you to spin all motors the same direction.

Some teams design different gearboxes for left/right where the motor orientation is different on each side, this fixes the forward/backward problem.

I've been playing around with some different designs this off season, although we'll probably stick with chain next year, I'm curious to know how other people solve this problem.

[Jnadke]

Originally posted by gwross
Can you elaborate on the kind(s) of problems you would run into when the motors ARE running at free speed?

Basically, the faster motor will try to spin the slower motor, and the slower motor will act as a generator (slowing the faster motor down and wasting electricity). Although drivetrain friction is only a controlling factor, Matt is correct. Now for an explanation.

Remember that Torque is a force that is rotating. Like all forces, it tries to accelerate things (make them go faster). Lastly, remember that a force accelerates according to the equation Force = Mass * Acceleration. Also remember that, when dealing with motors, torque decreases as speed increases. Therefore, as robot velocity increases, the ability of the motors to increase this velocity goes down.

As Matt said, drivetrain friction will, for the most part, keep these differences under control. Drivetrain friction is a negative force, meaning it tries to decelerate things. For the most part, the faster motor will pick up this slack. While the slower motor has already reached its free speed (and cannot produce any positive force), the faster motor will still try to accelerate the system. The generator effect may or may not happen depending on the severity of your drivetrain friction. Even then, you won't notice any severe negative effects (staying under the 40A maximum) unless the free speeds are more than +/- 30%.


In the end, no matter what, you're not going to come up with a gear ratio that matches the free speeds perfectly. I'd error on the side of making the drill motor being the faster of the two for three reasons:
1. It is more powerful, and therefore it will pick up more slack.
2. Since it has timing, you'll incur less negative effect when operating both motors together in reverse. (Internal friction due to the timing will cause the drill motor to be slower, we want to minimize the difference)
3. The CIM motor has a higher internal wire resistance, so the generator effect will be less severe if you spin it faster than its free speed.

Overall, I'd aim for a gear ratio in the area of 3.4 or 3.5 to 1 when gearing the drill (raw) to the CIM. I wouldn't use the drill gearbox when gearing motors together because internal gearbox friction can be unpredictable. The less gears between them, the better.

[DougHogg]

I have been trying to learn about multiple-motor drive trains (and single motor drive trains actually), so the above has been very helpful. Thank you to everyone who has been posting in this thread.

Originally posted by Jnadke
I'd error on the side of making the drill motor being the faster of the two for three reasons:
1. It is more powerful, and therefore it will pick up more slack.

If the drill motor is faster and more powerful, then the CIM motor will be dragging it back a bit, but I don't understand the reference to "slack".

Also, does the drill motor lose some power when it is geared down to the CIM's speed?

Originally posted by Jnadke

2. Since it has timing, you'll incur less negative effect when operating both motors together in reverse. (Internal friction due to the timing will cause the drill motor to be slower, we want to minimize the difference)

Could you divine "timing" in this context. (Does it relate to why the drill motor goes faster in forward than in reverse?)

Originally posted by Jnadke

3. The CIM motor has a higher internal wire resistance, so the generator effect will be less severe if you spin it faster than its free speed.

Makes sense.

Originally posted by Jnadke
Overall, I'd aim for a gear ratio in the area of 3.4 or 3.5 to 1 when gearing the drill (raw) to the CIM. I wouldn't use the drill gearbox when gearing motors together because internal gearbox friction can be unpredictable. The less gears between them, the better.

Okay, so here is the homework assignment I gave myself. Given the above information, to synchronize a Fisher-Price motor instead of the drill motor (in this year's game, we rotate our telescoping arm with a drill motor), you would want the CIM's to be a little faster, since they are more powerful. On the other hand, the Fisher-Price would have a higher resistance. Therefore I am surmising that it would be good to just get them as close as possible. How about a ratio of 2.75 to 1 for Fisher-Price to CIM? Does that sound like a good target?

[E Jones 234]

our team used those 2 motors together for our drive train this year... i'm not too sure about the gear ratios and whatnot, but maybe someone else on my team could help you there. we geared the drill and cim motors together, then ran them through the drill transmission. it worked really well for us, we were pretty slow, but almost no one could push us around.

[Jnadke]

Originally posted by DougHogg
I have been trying to learn about multiple-motor drive trains (and single motor drive trains actually), so the above has been very helpful. Thank you to everyone who has been posting in this thread.

Thank you. I will try to answer your questions, as I wasn't exactly clear in my reasoning.

If the drill motor is faster and more powerful, then the CIM motor will be dragging it back a bit, but I don't understand the reference to "slack".

Also, does the drill motor lose some power when it is geared down to the CIM's speed?

By slack I meant drivetrain friction. Since it's a negative force, the faster motor will focus more work on overcoming this. Generally, you want to "error" on the side of the more powerful motor being the faster of the two, so the torque at equivalent speeds will be more closely matched. The exception is if one motor has timing, because of reason #2. For example, if the drill motor has a 10% difference in speed from forward to reverse, if you make it 5% faster in forward operation, it will be 5% slower in reverse operation.

Yes, the drill motor does lose some power. Spur gears are anywhere from 95-98% efficient.

Could you divine "timing" in this context. (Does it relate to why the drill motor goes faster in forward than in reverse?)

Yup, you nailed it. Timing is when a motor is "optimized" in a manner of speaking. Remember that as a motor spins, it's coils are constantly pulling/repelling the internal magnets, which causes a force (torque). In a perfect world, we would switch direction of current in the motor windings at the point where the magnetic field is the strongest. However, because of the inductive (resistance to a change in current) and resistive properties of our coil of wire, the magnetic field takes time to build up. Therefore, we switch the direction of the current before we reach the point where the magnetic field is the strongest, so that, by the time the coil is at this point, it is starting to up its own magnetic field, and we can maximize its benefits (Magnetic field strength decreases as distance increases). Obviously, when you operate the motor in reverse, the magnetic field will be severely unoptimized, so you lose much of the power.

Motor timing is measured in degrees (how many degrees before the point of maximum magnetic field strength the current direction is changed). Timing is one way to maximize the efficiency of a motor. The drill motors we use have timing because it takes more torque to screw something in than it does to take it out.

Makes sense.

Yup. According to Lenz's Law and Back EMF theory, as you pass a magnet through a coil of wire, it creates a voltage potential which sets up a current that repels the movement of the magnet. Higher resistance means lower current, hence less back EMF.

Okay, so here is the homework assignment I gave myself. Given the above information, to synchronize a Fisher-Price motor instead of the drill motor (in this year's game, we rotate our telescoping arm with a drill motor), you would want the CIM's to be a little faster, since they are more powerful. On the other hand, the Fisher-Price would have a higher resistance. Therefore I am surmising that it would be good to just get them as close as possible. How about a ratio of 2.75 to 1 for Fisher-Price to CIM? Does that sound like a good target?

In general you'll always want to have the motors properly matched. It varies case-to-case depending on what gears you can find, etc. In this example, I'd error on the side of the more powerful (CIM) motor.