Can the 'source' for the pulse train generation be a position as calculated by a counter from a quadrature encoder ?
No, not exactly. You can use either individual channel of a quad encoder as the source, but you'll be limited to 1 active source edge for every cycle of 4 quad states. There's no "signal" available that corresponds to the 4x multiplier you can get for position decoding.
Can the board 'autonomous' execute these multiple functions ie keeping track of both positions as well generate a pulse train ?
Yes it can. If you only need to query the instantaneous positions one right after the other, there should be no problem. If you need them to be hardware-synchronized, you *may* have difficulty due to limitations of the 6601. Only 1 of the 2 encoder tasks will be able to use high-speed DMA for buffering, the other will need to use interrupts which will limit the possible acquisition rate. If you haven't yet purchased the 6601, consider the 6602 which can support DMA for 3 simultaneous buffered tasks.
In either case, the pulsetrain generation is handled entirely by the hardware on the board with no DMA or interrupts needed.
I read in this forum that I can expect some troubles if a zero pulse (for the R-axis only) does not occur at a fixed state of the quadrature A/B inputs
Yes, there can be trouble IF you NEED to use the Z-index feature. There may be other workarounds...
If somebody did something simular, did you succeed (do you advise) to use traditional NI-DAQ or NI-DAQmx ?
Can be done with either, but I would steer you toward DAQmx whenever feasible. That's where NI is putting there development effort.
-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.
