I asked my students to create a Labview vi that, among other things, controlled a servo. After assigning the task, I found a post on here somewhere saying it was not possible because there is no hardware timing available on the device we were using: a USB6009. I thought perhaps we could work around this limitation using an external clock source and the counter, but that doesn't seem to work either.
The objective is to put out a pulse between 1 and 2 ms in duration every 20 ms. Our idea was to use the counter with a 100kHz square wave input to time the 1-2 ms interval, and enclose that timing loop in another loop executing every 20 ms. The inner timing loop first read the counter value and returned it to the next iteration through a shift register. The vi then repeatedly read the counter, took the difference with the initial value and, while the difference was less than the count we were after, it turned on a digital output pin (input to the servo). When the difference became greater than some value, it turned off the output pin and exited the inner loop. Attached are two images showing the vi we're using. They show what happens on the first trip through the timing loop (shift the original count) and on each iteration after (wait until the count goes past some value).
The concept works, but it's way too slow. It seems the inner timing loop only executes about every 2.5 ms. The output pulse width changes nicely, but only in increments of around 2.5-3 ms.
Is there some inherent timing limitation with loops in Labview? Or did we in our ignorance choose some terribly slow element to put in the timing loop? We assumed not putting a wait timer in the loop meant it would execute at something like 'normal' computer speed, which should be more than fast enough to read a 100kHz clock. Can anyone shed some light on why this thing runs so slowly? Thanks, and appreciate your help.
