Re: Robot Crashes with voltage spike
 

yes, we are tethered. I think we tested it with the PWM cables removed yesterday, but the person who would know that has left - he might reply later. Folks went to dinner, but we can try some things a little later tonight.
We loaded our code on last year's robot, using the same CRio but different Jaguars, and it works fine. not su
thanks so much for your help!

Re: Robot Crashes with voltage spike
 

Ok,
I am heading out to robot class. I won't be back on line for a several hours if at all. Hope you find what the problem is.

Re: Robot Crashes with voltage spike
 

The Jag has an H-Bridge power output configuration.

Look at the diagram at the top right of this page. S1 and S4 are closed to go forward and S2 and S3 are closed to go backward.

Now imagine that the FET S4 has shorted. When the Jag is at neutral, no problem. When S1 is closed, no problem. But when S3 is closed, you get a condition known as a "shoot through" which is a short circuit...

Mike

Re: Robot Crashes with voltage spike
 

We've been running tethered. Radio is just one more thing that can go wrong ;)

I have just discovered that our electrical team made a mistake in how they power the Side Car. I believe this is the problem. I'm going to test that theory tomorrow afternoon and get back then.

Thanks for all your help!

Re: Robot Crashes with voltage spike
 

we did in fact find the problem

our electrical team hadnt hooked the Side Car to the power distribution board: it was reciving power only from the data cable from the cRIO.

Thanks again for all your help!

Re: Robot Crashes with voltage spike
 

Ninja,
Do me a favor and do as full a check of the Crio as you can before ship. I would want to be sure that something wasn't stressed while supplying power to the sidecar.
Glad that you have found something. We call that "a smoking gun" so one more symptom to add to the list.
Thanks,
Al

Re: Robot Crashes with voltage spike
 

When you have a strange problem, it's annoying to get asked "Are you sure it's plugged in? Are you sure it has power?" But that should always be the first step in debugging any electrical system.



Now that you know what the problem was, would you like to know why it was behaving as it was? Here's my guess. The PWM signals from the cRIO were powering up the Digital Sidecar circuitry. At neutral, they provide a 1.5 millisecond positive pulse, giving just barely enough power to make things work a little. It's obvious that one of your motors is configured as "inverted", so that pushing the joystick on one side forward would decrease the pulse width for one motor, and pulling back on the other side's joystick would decrease the pulse width for the other. Doing them both at once would remove enough power that the Digital Sidecar shuts down completely.

This is a symptom that should be remembered for future diagnosis.