Limit Switches on CAN freaks out

[nighterfighter]

We are using CAN to control our Jaguars this year, but we get an error every once in a while when we boot up the robot.

If the limit switch is pressed when we boot up, every now and then (it appears to be random, we can't find a pattern) NetConsole throws a bunch of errors, called "Code: -44087 ERRROR Status == -44087(0xFFFF53C9) in SetTransaction() of CANJaguar at line 392" . When this occurs, we can't drive any motors, as the entire CANBus dies.

Like I said, it doesn't occur everytime the limit switch is depressed, but if that were to happen in a match, we'd be screwed.

If don't know if it makes a difference, but we have a total of 4 limit switches. Two are on a Grey Jaguar, and when those switches are depressed, we have never had a problem. The other 2 switches are hooked up to the same Black Jaguar, and that is when the error occurs.

[virtuald]

Interesting. We have the same problem -- and we're using limit switches on one of our Jaguars. However, we would only lose control of our motors for maybe a half second or second or so before it would come back. Out of curiosity, why do you blame the limit switches -- does it not have this problem without them?

Someone had mentioned on a different thread that this error is due to improper termination. But ours seems to be properly terminated.

[nighterfighter]

We have never encountered the problem when the limit switches weren't pressed.

And yes, we have a proper terminator on the final jag.

[WizenedEE]

The labview explanation for that error is that the C RIO doesn't have the files it needs (either the FRCBlackJagBridge.out or the FRC2CAN.out.. or whatever their exact names are).

I know from experience that it does give that error when it doesn't have the files, but it seems to sometimes give the error if you do have the files.

Our team is using limit switches too. One of them is almost always pressed, and even when some are pressed it doesn't give the error.

I don't recall the problem ever happening on our competition robot, with the 2CAN. Maybe I'll try moving the starter of the bus to a different black jaguar, without limit switches.

[nighterfighter]

We formatted our cRIO using the BlackJagBridge option on the cRIO image updater.

Do we need to FTP the files in, like last year?

[WizenedEE]

Originally Posted by nighterfighter

We formatted our cRIO using the BlackJagBridge option on the cRIO image updater.

Do we need to FTP the files in, like last year?

I think you do; there's no harm in updating your software.

Remember to change the .ini file also.

[jhersh]

Originally Posted by nighterfighter

We formatted our cRIO using the BlackJagBridge option on the cRIO image updater.

Do we need to FTP the files in, like last year?

Originally Posted by WizenedEE

I think you do; there's no harm in updating your software.

Remember to change the .ini file also.

NO!

Please do not replace the files in your image. The files that now come in the image are far newer than any that are posted separately on the CAN Jaguar Project on FIRST Forge. Please note that those files are all marked as obsolete. The imaging tool takes care of everything you need.

The only exception is if there is a new release from CTRE for the 2CAN plugin, then you may need to update that. In that case, only replacing the file is needed. The imaging tool will still modify the ini file for you.

-Joe

[nighterfighter]

Okay, thanks. We won't change anything.

Just hope that the CAN doesn't freak out during a match, or else we will have to do a rapid switch to PWM cables...

[jhersh]

Note that there is another thread discussing an issue that may be the same thing.

http://www.chiefdelphi.com/forums/sh....php?p=1036479

Can you give any additional information about your setup, such as what language you are using, if you ever see this part way through a match or only on start-up, what CAN interface driver are you using, and any other information you think might help?

Thanks,
-Joe

[PranavSathy]

Hey, we were having getTransation() and SetTransaction() issues during our regional, our simple fix was to add a Wait(.01) before the drive was called, that fixed our errors, we added this inside our drive forward loop for autonomous, so it doesnt affect our teleop, but it seems to work like a charm, try it out and see if it works. Good luck!

[nighterfighter]

We experienced the scrolling CAN issue during one match at our regional.

Despite rebooting the cRIO 4 times, it didn't help it. After the 5th reboot we were able to drive, but there were only 4 seconds left in the match.

We decided to trash the limit switch.