The plan is to have a integrated comparator and h bridge driver in 1 IC. I had a setup with lm339 for current limiting in fast decay mode. Switching frequency of the bridge at 20khz. Oscilloscope is not available so I can't really tell what happened, but the circuit was working. However to get any useful current, comparator + or dac setpoint input had to be set higher than expected value (don't remember exact values because I was working on that circuit a month ago). Shunt resistor is 0.25 Ohms and expected current through the bridge was 5 Amps (4x1 Ohm resistor in parallel). Without oscilloscope I couldn't see the waveform so maybe it was fast decay that was messing with it, because comparator sets the threshold not RMS current and it was decaying so quickly that the average was really low (motor inductance is unknown because the DC motor is from a car fan, but it is a big one with weight between 1 and 2 kg so it's inductance is pretty huge).
I'm thinking that with something integrated there will be better regulation. Inductive spike on the resistor probably messed up comparator at low values, but I'm not sure if lm339 is even fast enough to catch the spike (1k pullup on the output, 5V Vcc). When going through the l6505 datasheet, it said that there was a delay in the comparator and that there was a minimum current because of it. I was like, YES! Inductive spike is definitely avoided because of that.