Melexis gyro w/ voltage divider connection

[maxvtol]

After many successful flights using my custom mixer, it seems like my yaw gyro quit working.

Before putting instrumentation to it, after turning it on it would work fine for a few minutes, then the yaw gyro would cause my tilt servos to max out in one direction (than goodness all this happened while on the ground). If I turned the gyro gains off at the radio, the servos would return to neutral, so I know it's the yaw gyro.

Of course now that I have it disassembled with instrumentation on it, I can't repeat the problem. I'm only measuring 0.1mA being put on the gyro by the voltage divider, so I can't imagine that would cause it. This was the recommended method to get the voltage from 5v at the gyro to <3.3v required at the ADC on the ARMmite by Bruce from Coridium Corp.

A couple of questions:
Has anyone had any problems with the Melexis MLX90609 gyro?
Does anyone see any problems with my hookup?

I guess the most likely cause is a loose wire, but it's funny that it only happened while motionless after a few minutes on at least a couple of occasions. All the wires seemed to have a good connection upon disassembly.

[maxvtol]

Just found this in post #3 from this thread (http://www.rcgroups.com/forums/showthread.php?t=639405&highlight=voltage+divider) :

...If you need to extend the range of the voltage being measured so that it is higher than what the PIC's A/D converter can handle, you need to use a simple voltage divider circuit (2 resistors), taking care that the total resistance does not exceed about 20K, otherwise insufficient current will cause random fluctuations in the A/D conversion.

Would this apply to all ADC? My total resistance is 25k.

Also, I forgot to mention I'm using the breakout board (http://www.sparkfun.com/commerce/product_info.php?products_id=8372) from Sparkfun for the Melexis gyro, and the ARMmite (http://www.coridiumcorp.com/catalog/product_info.php?products_id=57) from Coridium Corp.

[maxvtol]

After over an hour of bench testing the gyro finally failed, the Rate voltage dropped to .6v with the input remaining a constant 5v. The other 2 gyros seem fine.

I guess the only question that remains is has anyone had troubles with the Melexis gyros?

Thanks