The datasheet for the DRV8824 indicates that the minimum pusle width for the step input is 1.9 usec high and 1.9 usec low. This matches fairly closely the stated max frequency of 250 kHz.
To simply the programing of my motor control algorithm, I was simply raising the step pin then lowering it about 5 usec later (above the needed 1.9us high time). Since I am driving the motor somewhat slowly, the low time would be several hundred usec later. This is all well within the limits of the motor and controller. With this setup, the motor was turning appropriately.
But if I change the software slightly to provide a square wave rather than a series of short pulses at the SAME frequency, I get a different behaviour. Specificaly the motor ‘sings’ differently. By singing I mean the sound that stepper motors make as they move. Other parameters (torque, speed, …) seem to be the same.
Could someone explain this phenomena? I was unable to find anything in the datasheet that would indicate that a series of short pulses rather than a square wave was better or worst than the other. But obviously there is some difference or the motor should sound the same.