CMUcam 131 error, probable cause?

[efoote868]

This year our mentor assigned a couple of freshmen to learn everything about the camera, to configure it correctly, know the code, etc.

They pretty much fried the one from this year's KOP, and they broke the one from last year (only getting Initialization Error 131). I don't think its a software problem, because when we got the old camera out, it was very much functional, and after being in their hands for a day, it isn't. (And I have absolutely no idea what they've been doing :confused: )

Since my freshmen friends made mistakes twice, and this is the last camera we've got (until we get another from IFI), I would like to know any tips to prevent these errors / stop the camera from breaking. (I know the importance of hooking everything up correctly, along with a fully charged backup battery)

Well, lets hear them, and thanks in advance!

[Kevin Sevcik]

FYI, error 131 means the camera code failed at initialization state 3. 131=128+3. What this means to you is that your camera just isn't talking to the RC. This is the first state in the camera initialization routine that actually waits for a reply from the camera, and it's failed here, so you're just not getting a reply from the camera.

The possibilities are:
A. Your camera isn't getting power. You say the battery is charged, so not that.
B. The serial cable is unplugged, plugged in backwards, or th TTL adapter has come loose.
C. The cable from your TTL adapter is too long.

Those are about the only options I can think of. I'd really triple check all the connections, especially the TTL adapter, and try hooking up with a short PWM cable. Second bit of trouble shooting you can try is hooking directly to the the serial port with your computer, opening up Hyperterminal, and trying to talk to the camera directly.

[efoote868]

How easy is it to fry the camera, which would then cause the error (camera wouldn't send anything back, anyway)?

[Kevin Sevcik]

If the power LED lights up properly and everything, I'm willing to bet you haven't fried it, but I could be wrong. The easiest way would probably be to reverse the power polarity, since I don't think it has protection for that. I have managed to inadvertently short 5V to ground through the camera before and it was no worse for it. I have also had students pretty seriously mistreat it and it's come out ok. I'm not saying it's impossible to kill the poor little thing, but I think you should exhaust all of your troubleshooting options before you give up on it.

[DanDon]

You might want to try this piece of code: http://kevin.org/frc/frc_serial_port_diagnostics.zip

It's available at Kevin's site, and the description is:

Serial port diagnostic software that can be used to verify the proper operation of your robot controller's serial ports, CMUcam2, TTL to RS232 conversion board, and cables.