The Z pulse must be in the high state simultaneous with the A,B signals being in their configured state. There are 4 possible High-Low permutations to pick from for A,B and I don't recall the default offhand (H-H maybe?). So it's not exactly a minimum pulse time that's needed, it's more of a pulse sync which may not be possible for you to do without some interface circuitry.
This restriction works fine with industrial encoders that often generate a Z pulse that is synced to A,B and lasts either 1 A,B state or 1 quadrature cycle of 4 states. It doesn't work well when you'd like to reset a count value based on some kind of asynchronous external signal (home sensor, etc.). Another note - the behavior is *level*-sensitive, not edge-sensitive. If you hard-wire Z to a 5V High, the count will simply cycle 0,1,2,3,0,1,2,3,0,1...
There really isn't any simple workaround that I know of. You might want to visit the Product Suggestion Center to add your voice. Personally, I think counters should be allowed to hw-reset their count based on an asynchronous edge, even in modes other than position measurement.
-Kevin P.
ALERT! LabVIEW's subscription-only policy came to an end (finally!). Unfortunately, pricing favors the captured and committed over new adopters -- so tread carefully.
