Flickering Victors

[jgannon]

I don't have my hands as deeply in the code as in the past, so this may not be all that descriptive, but hopefully someone will recognize the behavior I saw yesterday, and point me in the right direction for debugging today.

My students are programming the robot in Java, and it sounds like what they've put on the cRIO is very close to the default code, with the notable exception of configuring the drive code to use Victors rather than Jaguars. When the code is running, and no one is touching the joystick, the lights on the Victors change colors rapidly and chaotically. The joystick appears to be configured correctly, because I see the messages about switching from arcade to tank when I move the throttle. The pattern seems to change when I move the joystick; we see brief periods of solid green (but not red) mixed in with the flickering. The behavior is the same whether we're in arcade or tank. The odds that all four of our Victors are on the fritz seems unlikely, so I'm leaning toward some software configuration issue. In any case, I'm totally perplexed... any suggestions?

[Greg McKaskle]

If you look at the Diagnostics tab on the DS, it will show some/most of the errors that are happening in the SW.

In particular, are user watchdogs happening? You may need to clear the display and see what new stuff shows up.

Greg McKaskle

[Joe Ross]

It sounds like it's still using the Jaguar class. The faster update rate and wider pulse widths that the jaguar class uses cause problems for Victors.