Motor Imballance

[jpsaul7usa]

Hi. We're a newbie team and just found out about this website. I didn't see any other posts on this board so i figured I'd start a thread and either someone would kindly answer my question or angrily post a link to a previous thread. Either way it's a win situation for us.

Anyway, we're having a lot of trouble trying to achieve a ballance between our two Bosch motors. We removed the pins a long time ago by accident, and since the drills ran smoother, we kept them out. However, whenever the bot's on our proving grounds (aka col-de-sac near our garage) it drives anywhere but straight. sometimes it's only a slight difference easily corrected by the driver, but other times it gets really out of hand.

Yes, we are still using the plasic mounts, as it never occured to us that they weren't sufficient, and they're doing the job well enough. We have had to repair one of the brushes so far, and have taken the drills apart several times. Aside from programming (we're already working on it) are there any simple solutions to making the system more mechanically sound? Thanks for you help.

[Cory]

this may be a really dumb question, but are both motors in the same gear? One may be running in High while the other is in low. Have you removed the clutch? If so, two metal tabs that fit into the housing below the output shaft should be in the motor. It runs very poorly without them.

Cory

[Carl Owenby]

If you're like us, during forward travel, one motor runs forward while the other runs in reverse, since they're on opposite sides of the bot.

We've noted up to a 15% difference between forward and reverse speeds. After some research, I found this has been documented by others in prior years.

Could this be contributing to your problem?

[sevisehda]

If your just using the drills then more than likely your running into the forward/reverse difference. The motors are tuned to run in forward slightly faster. Its possible to mount the motors or run the chain so they run together. Or you could just adjust the program to cut forawrd by 15%. I'm not sure what really chaotic means but bots act weird on concrete, tile floor always try to test them on a carpet similar to a playing field.

[Adam Y.]

We've noted up to a 15% difference between forward and reverse speeds. After some research, I found this has been documented by others in prior years.

I forget it has something to do with the way they place the brushes. It is called timing. There can either be ccw or cw timing which tells you which direction the motor is turning. They do that with many types of motors especiallly ones where the motors are going to be turning in mainly one direction. Unfournatly problems arise with using motors with the same timing on each side of the robot. It will mean that it tend to go off crocked.

[Joe Ross]

If you are using casters, the 15% speed difference will be much more noticeable then if you use 4 driven wheels.

[jpsaul7usa]

Yeah, the motors are in the same gear. I never considered before the fowards/backwards thing before, and at this stage it will be very difficult to solve., as this weekend we're integrating all of our systems to the frame. I think programming will be our only fix. We are using casters sadly as we don't have the time and the space in the bot to put in 2 more gearboxes.

Cory, or anyone who knows what he's talking about, which metal tabs are you referring to? And which section is the clutch? (I think I know but I'm not possitive?) Our motors are fully assembled minus the pins in the tranny.

Also, on the work bench, the motors noticably differ in how they turn when we rotate the shafts they are connected to. One of the motors is stiffer than the other. This could be that foward backward thing again, but we're not sure. We know it's not the gearboxes because we've balanced those so there's little binding.

Thanks everyone for you help!!!

[Adam Y.]

I think programming will be our only fix.

That is pretty much your only fix because the only other thing you can do is get opposite pairs of timmed motors cw and ccw.

Also, on the work bench, the motors noticably differ in how they turn when we rotate the shafts they are connected to.

Yeah I think it is because the motors are not neutrally timed.

[sevisehda]

Without more details about your drivetrain, it hard to say what to do. I'd recomend running the motors forward/reverse at the same time. If you do it by software you cut 15% of your max speed/toque. Check all the alignments because this may cause the stiffness. You may also want to see if one of the motors and transmission runs louder than the other this could also indicate misalignment.

[Iain]

So, can anyone say what the best way of fixing this through programming might be? I ask because our team is in exactly the same boat as jpsaul's, right down to running on two wheels + two castors and having to fix a warped brush mount.

(We noticed this imbalance problem in our robot last week but assumed it was an early sympton of the brush failure (which just happened to be on the motor running in reverse). I guess not.)

[Cory]

Originally posted by jpsaul7usa
Yeah, the motors are in the same gear. I never considered before the fowards/backwards thing before, and at this stage it will be very difficult to solve., as this weekend we're integrating all of our systems to the frame. I think programming will be our only fix. We are using casters sadly as we don't have the time and the space in the bot to put in 2 more gearboxes.

Cory, or anyone who knows what he's talking about, which metal tabs are you referring to? And which section is the clutch? (I think I know but I'm not possitive?) Our motors are fully assembled minus the pins in the tranny.

Also, on the work bench, the motors noticably differ in how they turn when we rotate the shafts they are connected to. One of the motors is stiffer than the other. This could be that foward backward thing again, but we're not sure. We know it's not the gearboxes because we've balanced those so there's little binding.

Thanks everyone for you help!!!

The clutch is the black rotating piece on the front with lots of numbers on it. If its on there, those tabs are too, however if its not, if you look down at the motor, shaft up you will see two metal pieces sticking out of it directly across from each other, under the shaft housing and on the outer edges of the motor. if you have the clutch off and you turn the motor upside down, they will probably fall out. they have little springs on the ends of them. We took the clutch off our motors and they ran very poorly without these little tabs in, so we manufactured a device to hold them in. It was a very temporary solution, but I could describe it to you If you think it would help. If you need any more help, you can PM or email me.

Cory

[Adam Y.]

Without more details about your drivetrain, it hard to say what to do. I'd recomend running the motors forward/reverse at the same time. If you do it by software you cut 15% of your max speed/toque.

Yeah but what you are saying will not even fix the problem. It will only cause the problem to occur when you do a hard turn(one motor foward one reverse).

[jpsaul7usa]

ok. So I got to the meeting a little while ago and I've been looking at our drive system for awhile. After a little head scratching I have come to the conclusiong that both motors are spinning in the same direction when the robot is moving fowards and backwards. Did I mention that we have our motors connected to shafts tha go into our gearboxes and spin the wheel shafts? yeah...

It's not allignment and it's not the gearboxes binding. I took the motors off and put them side by side (we have not removed the clutches, which i previously thought were called transmissions and vice-versa.) One motor (by motor i mean clutch, tranny... the whole assembly) spins more freely than the other, and each has had the pins removed from the clutches. So... either we now take the motors apart again to see which component is stiff, or we go out and try to buy another motor.

So does anyone know something about the motor assemblies that makes them stiffen and/or loosen, and a way to correct these deviations? eh?

[Adam Y.]

Are you sure something isn't stuck in there?? A metal chip could be the problem and be very easy to miss.

[Al Skierkiewicz]

jpsaul7usa,
What you have run into is the variation of production and your own design. If you are absolutely sure both motors are running exactly the same direction then the problem has to be something else and here is the list.
1. There is more grease in the transmission and therefore more drag. Some teams clean the trans and repack with better grease.
2. The speed controllers are not calibrated, so calibrate following the procedure in the data sheet.
3. You have more wire in the path to the slower motor (or a poor crimp) and therefore more resistance in series with the motor/speed controller/breaker panel. Make wire lengths the same or at least as short as possible for high current systems.
4. The thumbwheels on your joysticks are not centered or the joysticks are not tracking.
If you cannot isolate any of the above then you may have to compensate in software.
5. Practice, practice, practice!
Finding a bad crimp could be hard but may show up as a hot connector. We solder all crimps that feed high current devices and all connectors on #6 wire. Even if you loop two breaker panels together with #6 you are still adding resistance that limits the current.
Let us know what you find.