"Backdriving" Motors -- Safe for Victors?

[Ken Streeter]

Last year due to wiring errors our team destroyed far too many Victors for our team's budget to handle. Our troubles at the time were due to connecting Victors "backwards" with the +12 Power/Ground "input" connections instead being placed on the M+/M- "output" connections. We found that doing this and then turning on the power to the robot immediately results in a big popping noise and bright flash -- later followed with the need to call IFI with a credit card number... (Call this a lesson from the school of hard knocks!)

However, enough of history... back to the present time, when we are being extremely cautious in order to keep our Victors working properly...

Our question is whether or not it is likely that the Victors can be "burnt out" by being backdriven. A search through ChiefDelphi turned up two previous threads that discuss this issue, but I don't feel that the question was resolved of whether or not backdriving the motors is generally safe for the Victors:

2004 Thread on "Victor burnouts" (http://www.chiefdelphi.com/forums/showthread.php?t=28137)
2006 Thread on "Back EMF Safe for Victors?" (http://www.chiefdelphi.com/forums/showthread.php?t=42696)

The two scenarios of which we are concerned are as follows: (The first is the primary concern, as we are not aware of a good way to avoid it.)

Scenario 1: We plan to have Fisher Price motors connected through a gearbox (either the KOP FP Gearbox or a BaneBots planetary gearbox) to drive a robot lifting mechanism. In order to "reset" the mechanism to get it to fit back in the robot volume after completing a match, the mechanism would be backdriven (presumably with the robot power off). This would backdrive the motor for just a few seconds, surely for no more than a couple thousand rotations.

Scenario 2: We plan to have small CIM motors connected through the BaneBots planetary gearboxes supplied in the kit as the primary drive for the robot. Pushing the robot with the wheels on the ground will backdrive the motors through the gearbox. We have always sought to avoid this situation in the past by simply carrying the robot, but we figured we'd ask if pushing the robot in this way is likely to be a problem, since many other teams could benefit from a definitive answer on this scenario.


Any conclusions or additional experience from the electrical gurus out there?

Thanks!

--ken

[Kris Verdeyen]

We've done what you're describing (both cases) for a number of years, and I can't recall ever having destroyed a Victor. In any case, I wouldn't recommend pushing your robot to the match backdriving its wheels, although short distances would be fine. I would reccomend putting the drive Victors in coast mode if you're planning on backdriving them.

[dtengineering]

Both the scenarios you describe should be fine based on our experience. Don't be suprised as you push the robot around that the control board lights up... however the Victors don't seem to have any problem handling this little bit of current.

I do suggest that you try moving the victor jumper from coast to brake, at least on the drive motors... not for any electrical reason, but simply if you are running an open-loop (no encoder speed feedback to the RC) control system that having that active braking seems to make driving the bot easier.

I would also check to ensure that the CIM involved in lifting the ramp will not backfeed on its own once the match ends and power is lost.

Good luck on keeping the smoke inside the Victors this year... we hooked one up backwards our first year and have been extremely cautious since... I still keep the burned out shell of it on my desk as a reminder to everyone.

Jason

[DonRotolo]

This would backdrive the motor for just a few seconds, surely for no more than a couple thousand rotations.
:eek: A couple thousand in a few seconds?

The back EMF produced by a motor is proportional to the motor shaft speed. If you turn the motor at 2 or 3 revs per second, you will see only a small voltage, but at 6000 RPM (100 revs per second) you'll see quite a good amount of power.

The Victors can handle reasonable energy from backdriving, as dtengineering stated, but don't go wild and expect it to last.

It is more likely your Victor failures are due to other factors, such as metal shavings in the Victors.

Teams should make it a practice to protect their Victors (all the electronics, really) from the mechanical folks, with their saws and files and stone axes...

Team 1676 uses a blanket to cover everything when there's work being done.

Don

[Mike Betts]

Ken,

We have repeatedly performed both of your scenarios over the years and have never damaged a Victor in this way. We have only blown one Victor about 6 years ago and that was attributed to an aluminum chip which fell into the heat sink assembly.

I would not worry about it...

JMHO.

Mike

[JamesByrne]

Same here, never damaged a victor. We backdrive our robot without thinking to get it moved in the build area.

I do not think it will hurt them.

[Al Skierkiewicz]

No need to worry here, the diode protection in the FETs provide a current path when the power is removed that routes the current to the input terminals of the Victor. This current will light up the LEDs on the Victors and will move the fans.
Although pushing the robot with the power off will not damage the Victors, you might be putting undue forces on the mechanical systems and wheels.