jeffrz

I am trying to make repeated measurements with two counters on a USB-6251. I have a drive signal and two sensors. The sensors activate approximately 350ms after the drive goes high. The frequency of the drive is 2 Hz. A portion of the block diagram is shown below. The measured values are put on a queue and a GUI loop updates a waveform chart (scatter plot) and logs the values to a file. Taking out the GUI and logging doesn't improve the performance. The issue is that the counter is reporting 2 values per drive signal. One is a reasonable value (~325 msec) and the other is zero (which results from a timeout). I have set the timeout to 375 msec. That should be long enough for a normal event and short enough to time out if the sensor doesn't activate before the next drive signal. If I set the timeout longer (~ 550 msec), then I get one "correct" value per drive, but I don't really believe the value. It trends down ever so slightly so that over the course of 100k cycles, the value is clearly incorrect. Any suggestions would be appreciated. - Jeff http://forums.lavag.org/index.php?act=attach&type=post&id=6024