[PhilBot]

Yeah, I really did suspect the serial port, so I went out of my way to test that aspect..

I coded up a simple state machine so that the only thing that was printed was when the gyro stated moving and when it stopped. This way you could see exactly when the first and last change occured... typical output was:

Started 901
Stopped 1805
Started 1801
Stopped 908

This indicate two 90 degree rotation sequences (First CW then CCW, showing tenths of a degree). So no chance of overflowing any buffers etc..

What I find odd is that the angle appears correct although the "Stopped" value was printed a second after I actually stopped. (Despite trying dead-bands all the way from 4 to 20).

Also, for the record, if you output too fast using Printf's you just lose characters... they don't back up.

Has anyone actually looked at the output of one of the lastest KOP Gyros on a scope? Does the output cut off immediatley after a rapid rotation?

Phil.