motor controller profiles

[lemonzap]

My programmer team and I just collected some data on the operation of all 3 motor controllers. Jaguars, Victors, and Talons. We put a motor on a wheel with a gearbox and an encoder. We collected the minimum value that had to be sent to the controller to start the motor moving forward and reverse, the maximum forward and reverse speeds, and a graph of speed over time. We had the program increase the value sent to the controller by 0.01 every 0.5 seconds until it reached full power then decrease the value by 0.01 until it reached full reverse power, then increases again until it reached a stop. I've made sure the controllers were properly calibrated and used the correct class for each.

Jaguar:
Start value: 0.110
Reverse start value: -0.060
Top speed: 543.478 units/sec
Reverse top speed: -543.478 units/sec
https://dl.dropbox.com/u/36843099/jaguar%20speed%20over%20time.png

Victor 884:
Start value: 0.090
Reverse start value: -.060
Top speed: 657.895 units/sec
Reverse top speed: -568.182 units/sec
https://dl.dropbox.com/u/36843099/victor%20speed%20over%20time.png

Talon:
Start value: 0.070
Reverse start value: -.060
Top speed: 543.478 units/sec
Reverse top speed: -568.182 units/sec
https://dl.dropbox.com/u/36843099/Talon%20speed%20over%20time.png

[Mike Copioli]

Were you using the correct class for each controller? Each motor controller has different input timing requirements.

[Ether]

1) Was each motor controller properly calibrated?

2) Did you use the proper driver for each motor controller?

3) You didn't mention which model Victor you tested. It looks like the older 884, not the new 888.

[lemonzap]

Ah. ya I noticed that there were different classes for the different controllers but forgot to swap out the code at the time. The talon was properly calibrated and I'm fairly sure the Victor was. Ill correct the data when I re-run the tests tomorrow.

[Ether]

The talon was properly calibrated and I'm fairly sure the Victor was.

Since you're not using a joystick, can you please describe your calibration procedure ?

(I know how I'd do it with a test like this, but for completeness it would be helpful if you'd document how you did it)

[lemonzap]

The reason I believe the Victor may have been improperly calibrated is because the lights didn't light up exactly how the instructions said they should. I think that on the Victor we may have just done a factory reset. I will confirm that everything has been properly calibrated before the test tomorrow so we can get more accurate data.

[cgmv123]

Here's the data that's present overlaid on top of each other (as best I can) for comparison:

13585

[lemonzap]

I re-ran the tests with the motor controllers properly calibrated and using the correct classes. the graphs have been updated but there was no significant change. I hope this data will be useful. also keep in mind that the values will change depending on the motor, gearbox, encoder, and battery life. These values are fairly arbitrary on their own and are mainly useful for seeing the comparison between the different types of motor controllers. For instance the dead zone around zero for talons seems to be significantly smaller than the the dead zone for the other 2.

[Ether]

I re-ran the tests with the motor controllers properly calibrated and using the correct classes. the graphs have been updated but there was no significant change.

Is this (http://www.chiefdelphi.com/forums/attachment.php?attachmentid=13591&stc=1&d=1358738345) a typo? It doesn't agree with the graph (https://dl.dropbox.com/u/36843099/victor%20speed%20over%20time.png).

[lemonzap]

I believe it was a fluke in the data. A random sudden spike.

[Ether]

If you have a Victor 888 and would be willing to run your test with that, it would make your data set complete and help put things in perspective.

[lemonzap]

I definitely would but I sadly don't have any, and I don't think that we'll be getting any because the talons seem pretty reliable. I do wish we did though so I'll probably have someone order them after this season.

[slijin]

I definitely would but I sadly don't have any, and I don't think that we'll be getting any because the talons seem pretty reliable. I do wish we did though so I'll probably have someone order them after this season.

If you haven't used your IFI PDV yet, you can get two 888s or two Jags with it.

[MisterG]

I definitely would but I sadly don't have any, and I don't think that we'll be getting any because the talons seem pretty reliable. I do wish we did though so I'll probably have someone order them after this season.

Lemonzap,

Thanks for posting, this data looks very interesting.

Would you consider sharing the code that you used for the comparison? If you did, perhaps someone else could repeat the test with the 888.

Al G.