Some test results to measure the performance of the 2007 KOP gyro are attached. Basically it works well up to about 148 deg/sec but the output is delayed from 100 to 250ms depending on the rate and the duration of the rotation. These measurements were taken because of robot rotation control problems - see
You know… I think the impedance on the analog devices’s sensors are too high for the PIC’s ADC. I’ve never seen the kit gyro-- does it include an op-amp, or some other means of lowering its output impedance, to allow the ADC to quickly get the gyro reading?
There is an op-amp on the module. We did some analysis on the A/D input impedance to try to figure out why a 100K pot we were using looked more log than linear. Turned out that it wasn’t a problem charging the A/D 25pf cap it was the pot. IFI did slap a .047uf cap with a 1K series resistor which has a time constant of 47us. Even with a 100K source impedance the time constant is still under 5ms. The A/D data sheet shows about a 40ms delay for the gyro output to come up to full rate. Our 100Hz sample rate adds up to an additional 10ms of delay. If I have some time I’ll monitor the response of the analog output on a scope and see if it does have anything to do with driving the RC input pin.