I certain behavior has happened to many of my student groups intermittently. Using the Cortex and RobotC 3.04, there are times when code will compile and download but appear to not run. On inspection of the debugger, it appears all sensors are working and in fact the Cortex thinks the motors are running (I can see the motor power levels go on and change in response to the sensors via the debugger). It has been difficult to pin down because there are times when the robot doesn't work for other reasons (wi-fi drops, usb port drops) but a common symptom in this particular situation seems to be the game light on the Cortex alternating between yellow and red. Here are some things I try when I get this condition:

Check and/or replace battery
swap a motor
unplug and plug usb
cycle power on Cortex
restart RobotC
download a different program
swap Cortex
wait

The only thing that works on the list above is swapping the Cortex or waiting awhile; later the original Cortex always works again which makes me wonder if if could be a heat issue.

I can find no reference to game light alternating between red/yellow in documentation. We are not in competition mode in any way and these robots are completely autonomous, no controller (unless be programmed via VexNet). This has started happening recently and the only significant change we've had is the introduction of the Sharp GP2D12 IR sensor which we have wired directly to an analog port without any supporting components or circuits (just a simple molex connector on the end). The sensor seems to always work fine and is set as a potentiometer.

Thanks in advance for any thoughts.

So, the issue you said is that "there are times when code will compile and download but appear to not run"
What do you mean the code doesn't appear to run?
None of the motors move? The screen and buttons don't respond? Variables don't change in your code?

If you have an LCD Screen (and aren't doing this already), try making some basic code that will react when you press a button. (I.E. Make the screen backlight go on when you hit the center LCD button, and make it go off when you let go)

Now for the dumb questions: (take them or leave them, you don't have to do all )
Have you made sure the Mastercode is up to date?
Is the ROBOTC firmware up to date?
Is the robot power switch turned on when you try to run the code? (because the code will run but motors won't move if it's plugged into a computer via usb, but switched off)
Do you have a fresh battery? (~8V on a voltometer)
Do you have a fresh backup battery?
Can you post a video showing this happening?
Have you tried running the code with the Joystick and the Cortex linked over VEXnet?
Have you tried programming with a different computer?

Sorry, I wasn't clear. The motors do not move. Debugger says they are moving. Sensors appear fine and motor levels respond properly to sensor input. In short, everything seems to work but the motors do not move.




Yes to all the questions except for posting a video.

Update:
I've found on one microcontroller, that if the motors are plugged into motor ports 6,7,8 or 9, they work. If 2,3,4 or 5 they do not.

Weird.

There are two internal resettable thermal fuses on the VEX Cortex. One for motors 1 to 5 and one for motors 6 to 10. They way they work is that they heat up under high currents and open the circuit, when they cool down the circuit is closed.

The fuse is sized so that all five motors that it controls should be able to run at full power. It is designed to set off if there's a short circuit on the motor leads. It will also likely set off if multiple motors are stalled where the current drain is much higher. It may also set off if you're trying to drive more than 5 motors through use of Y cable.

If none of above conditions apply to your robot then it sounds like one of your fuses are bad. The fuses are not socketed -- they're soldered onto the board -- and are not designed to end user replacement. You might want to call manufacturer to discuss your next steps.