PDA

View Full Version : Exploding 884s


Alan Ing
02-21-2003, 09:26 PM
Okay, now that the robot shipped, can someone tell me what happend?

This year we are using 4 motors to drive two seperate gearboxes. Everything was running fine and dandy until we decided we wanted to try and recalibrate the speed controllers. Trying to be expedient about it, we pressed the calibration button simultaneously on two speed controllers that control one gear box and after a second or two, one of the speed controllers blew up. Not having a clue about what happend, we tried pushed the calibration buttons on the remaining speed controllers and another one blew up. What gives? In three years, we've never had this happen before and to my knowledge, there are no metal chips in the electronics as we carefully assembled the electronics after we machined everything?

What gets me, is that it was running fine and only when we pushed the calibration buttons that two speed controllers decided to die. I can see one speed controller failing out of a manufacturing defect, but two?

D.J. Fluck
02-21-2003, 09:52 PM
Wow, Ive never heard of anything like that before...but I think you should contact IFI about that

Josh Hambright
02-21-2003, 10:20 PM
not to make light of this situation because i can only imagine how much this must of sucked and how much this musta freaked you out....

but did they actualy explode? Cuz i would love to see a video of a victor actualy EXPLODE like violently with pieces flying everywhere....

On another note...
We did calibrate our victors but we had issues with it and so on the suggestion of users on this board we reset them to default calibration because we found it really didn't help but instead hurt the response of them, especialy in autono mode.

Al Skierkiewicz
02-23-2003, 09:17 PM
We calibrate speed controllers each year in the current robot, to be sure they are working properly with the interface and all parts. I have never heard of a team trying to calibrate two speed controllers at the same time. At best this sounds like a dangerous practice. It is best to remove all circuit breakers except for the the controller that is being calibrated. Then you must remember that the controller does not store the calibration points until the button is released. The cal button must be pressed and held while the joystick is moved from extent to extent and while it is allowed to rest in the center position. We will not release the cal button until a "hands off" (no one touching the stick)verbal acknowledgment has been received.
Can I assume that since these controllers are feeding the same gearbox that you are running them from the same PWM output? Is your mechanical design capable of running with one motor disabled? Of course, if one motor is calibrating and the other motor is powered down, there should be no problems. I suspect that what may have happened is that one of the cal buttons was momentarily released during the cal process and it would have tried to supply full output to a locked gearbox.

f22flyboy
02-23-2003, 09:34 PM
Oh crap....

this post just reminded me that I totally forgot to calibrate the victors this year. Oh well, yet another thing that needs to be done at regionals

Oh and sorry for this useless post, I have no idea why your victors are exploding, although I have experienced a similar occurrance. Last year we had a short and the victor went up in a big puff of smoke.

Josh Hambright
02-24-2003, 07:21 AM
yah smoke sucks...we had one go up in smoke last year the first time we fired up the robot....first time our team had seen magic smoke in its existance and scared the crap outa most everyone....and that smell...that horrible Victor smoke smell...

SuperJake
02-24-2003, 08:26 AM
Smoke is bad? Nnooo - Electronics run on smoke! The trick is not letting the smoke out. If you let the smoke out of the electronics, they don't work any more.

Iain
02-26-2003, 09:22 PM
They don't call it "the magic smoke" for nothing, yep.

Clanat
02-26-2003, 10:51 PM
What do teams calibrate the victors for?

Jnadke
02-26-2003, 11:37 PM
Originally posted by Clanat
What do teams calibrate the victors for?


I'm going to try to explain this from a programming standpoint.

The speed controllers are already set to have a value of 127 being neutral, 255 full forward and 0 full reverse. However, if you look at the debug output by the joysticks, they usually hover around 245, because the plastic housing is in the way. Therefore, unless you calibrate the speed controllers, your robot will never see full power.

By calibrating, it sets the full power of the speed controller to be whatever you tell it to be. If you go into calibration mode, and move the joystick full forward and full reverse (example: joystick of 245 and 10), then those values will be the new range. Anything beyond those values will also be full forward (so it would still be safe to send it a value of 250).


Remember to return the joystick to center position before releasing the button, because the very last value it reads is recorded as neutral.

Simeon
02-26-2003, 11:50 PM
our victors haven't exploded, but one of our spikes did
something smelled funny
then after like 1/2 an hour we found that the little resistor thing on the side had totally melted...
the fuse was perfectly fine tho.
:confused:

Josh Hambright
02-27-2003, 07:17 AM
are u sure you had the polarity of the wires hooked up right?

Manoel
03-01-2003, 08:03 PM
Originally posted by Al Skierkiewicz
I have never heard of a team trying to calibrate two speed controllers at the same time. At best this sounds like a dangerous practice.

We've done this for the past two years and never had a problem. Why would it be a dangerous practice?

Jeff Waegelin
03-01-2003, 08:10 PM
Originally posted by Manoel
We've done this for the past two years and never had a problem. Why would it be a dangerous practice?

It's just not a good idea to be messing around with lots of things at the same time. It's safer if you just calibrate one at a time.

Manoel
03-01-2003, 10:20 PM
Originally posted by Jeff Waegelin
It's just not a good idea to be messing around with lots of things at the same time. It's safer if you just calibrate one at a time.

Yes, but picture this:
We use a drill and an atwood to power each track of our robot. Those joysticks have some travel issues (eg. sometimes full forward is 254. Now center it back and put it forward again. You may get something like 240) and it wouldn't be nice to have the motors carefully set to the same speed with gearboxes and see everything fail because a victor ain't sending the same signal to each motor. That's why we calibrate FOUR victors (two for each side) at a time - we need many hands, though:D

Al Skierkiewicz
03-02-2003, 12:32 PM
You are making some grave assumptions here. The design free speed and the loaded speed (in actual production models) are two different things, not to mention the distinct difference in speed between motor forward and reverse. Added to the fact that the spacing of the commutator and the switching frequency of the speed controllers affect the average power delivered to the motor, etc., etc. When you calibrate a speed controller, you are merely telling it that the maximum value and the minimum value you have sent it are to be interpreted as 255 and 0 respectively and that the position of the joystick at the moment you release the calibrate button is 127. I can see that calibrating both at the same time might be thought of as an ideal situation, I just think it sounds dangerous from a safety and operation standpoint. If you do not test the motors (and calibrate them) independently, how do you know that either of them is functioning properly?

Bduggan04
03-02-2003, 02:00 PM
Originally posted by Simeon
our victors haven't exploded, but one of our spikes did
something smelled funny
then after like 1/2 an hour we found that the little resistor thing on the side had totally melted...
the fuse was perfectly fine tho.
:confused:

You probably reversed the polarity to the spike. I believe that the "little resistor thing" is actually a diode meant to allow the current to travel one way only. It melting would make sense if you reversed the polarity.