Mal,
could you please explain a bit more in detail what exactly you are planning to do?
The vi that you have attached runs an untimed loop (which by the way causes 100% CPU load...), moving the motor by one step in each iteration (depending on your calculations). This is completely undeterministic behavior and I'm sure that you don't see smooth moves. This way the velocity of your motor is just related to the speed of your PC and has literally nothing to do with the frequencies of the waveforms that you have calculated before.
In fact you don't really use the board's motion capabilities. Why don't you let the board generate the steps by it's own?
Just to answer your question: You can change the motors direction by providing a negative target position but I don't think that this is the most important question that you should have.
Please give me a better idea what you want to do because it really looks like that your current approach won't bring you the results that you are looking for.
Best regards,
Jochen Klier
National Instruments Germany