Jaguars failing

[Al Skierkiewicz]

Tom,
Can you give us a good way to know if the Jaguars have the latest software updates? The KOP box says rev A on it, does that imply a certain level of software or hardware revision?

[jhersh]

Originally Posted by Al Skierkiewicz

Tom,
Can you give us a good way to know if the Jaguars have the latest software updates? The KOP box says rev A on it, does that imply a certain level of software or hardware revision?

The way to tell is by plugging it into CAN. If using a black jag as a bridge, you can use the bdc-comm utility to query the version. If you intend to use the Servo PWM interface, then you don't need to upgrade the firmware.

As for hardware revisions, they are either black or gray plastics. All gray jags are the same hardware inside.

[Tom Line]

Originally Posted by jhersh

The way to tell is by plugging it into CAN. If using a black jag as a bridge, you can use the bdc-comm utility to query the version. If you intend to use the Servo PWM interface, then you don't need to upgrade the firmware.

As for hardware revisions, they are either black or gray plastics. All gray jags are the same hardware inside.

Thanks Joe! Unfotunately I'm working afternoons and midnights right now. Getting home at 4am is killin' me, and putting a crimp on responding to chief delphi questions in a quick manner is a bit tough.

[Peter Randall]

It has started again. We have already toasted two Jags from last year. We purchased a number for last year's bot and removed them all from the robot after toasting 4. We just built a prototype drive base and in two days have smoked 2 of the "replacements" we got last year. Same symptom, it only loses one direction after issuing the magic smoke. I really wanted to move to CAN, but I CAN'T if I CAN'T rely on the Jags. Has anyone smoked a Black Jag yet???

PGR

[Tom Line]

Originally Posted by Peter Randall

It has started again. We have already toasted two Jags from last year. We purchased a number for last year's bot and removed them all from the robot after toasting 4. We just built a prototype drive base and in two days have smoked 2 of the "replacements" we got last year. Same symptom, it only loses one direction after issuing the magic smoke. I really wanted to move to CAN, but I CAN'T if I CAN'T rely on the Jags. Has anyone smoked a Black Jag yet???

PGR

Peter, are you covering the open connectors on the jags with tape, a plastic jumper, or something else? It's been shown that direct static to the open jag connectors can cause issues.

Have you tested your frame to see if you're running electricity through it?

Are you mounting your electronics on a conductive or a non-conductive surface?

[Al Skierkiewicz]

Tom,
The direction failures are attributable to one of the power FET gate drivers shorting. I don't remember if the cause of the failures was ever found or reported.

[Chris_Elston]

We smoked our first Jag today. I was a bit surprise how it decided to go up in smoke.

We did not have any failures last year. This was a brand new one just out of the box. It was not under any stress, and decided to smoke. It's less than two days old. That's too bad because I am sure Luminary probably doesn't replace them because they assume you abused them. I hope we don't have any more that die like this one. It was almost in "coast mode" and poof...

[Joe Ross]

Originally Posted by Chris_Elston

That's too bad because I am sure Luminary probably doesn't replace them because they assume you abused them. I hope we don't have any more that die like this one. It was almost in "coast mode" and poof...

That seems like quite an unfortunate assumption to make considering that Luminary Micro's Failure Analysis Report states that Luminary Micro promptly replaced every failed Jaguar last year.

[Chris_Elston]

Thanks for this report.

Looks like ours is a U6/U7 failure.

[omalleyj]

Team 1279 worked on the Black Jag Beta program and we did manage to blow up two with static discharge, one by accident, and one on purpose (under TIs direction). Last years robots were very static prone, and the encoder and potentiometer inputs are fairly exposed. I would recommend covering the inputs with empty connectors when not using them to prevent that type of failure.

I would note that we also blew up our Kwik Byte driver station's ethernet connectors when running tethered at a fundraiser, so the Jaguars are not the only static sensitive devices. Use due care with all your sensors, etc..

TI was very good to work with about the failures, and they seem committed to improving their product. I would say that any team who has a failure should contact TI, document the conditions, and return the blown part. The Jags have many cool features and I would very much like to see them become a reliable component.

[Bryscus]

Originally Posted by omalleyj

Team 1279 worked on the Black Jag Beta program and we did manage to blow up two with static discharge, one by accident, and one on purpose (under TIs direction). Last years robots were very static prone, and the encoder and potentiometer inputs are fairly exposed. I would recommend covering the inputs with empty connectors when not using them to prevent that type of failure.

I would note that we also blew up our Kwik Byte driver station's ethernet connectors when running tethered at a fundraiser, so the Jaguars are not the only static sensitive devices. Use due care with all your sensors, etc..

TI was very good to work with about the failures, and they seem committed to improving their product. I would say that any team who has a failure should contact TI, document the conditions, and return the blown part. The Jags have many cool features and I would very much like to see them become a reliable component.

Oh you silly Northerners! There is a simple fix to your whole static discharge problem...MOVE TO FLORIDA! Sure, the education system isn't as good and there's a constant threat of hurricanes, but it's not very often my car shocks me!

[Tom Line]

Originally Posted by Al Skierkiewicz

Tom,
The direction failures are attributable to one of the power FET gate drivers shorting. I don't remember if the cause of the failures was ever found or reported.

Al,

I stated that because there was some discussion previously regarding the static. I wasn't sure if it had been on the beta test boards or not, but the summarization was that static charge directly applied to the exposed terminals could create issues in the jags. I couldn't remember who said it (fortunately someone spoke up here).

We'll be using Jag's again this year on our practice robot. I'm interested in seeing how they behave. We used them last year on our sparring-partner in the drive train and never experienced any issues with, though I know that anecdotal evidence is really of no help in determining their reliability.

Tom