DRV8821 Motor PWM Frequency 45 to 55KHz

Hi Maxime,

The internal PWM frequency has nothing to do with the stepping rate of the motor. The internal PWM frequency is used to coordinate current chopping or curent regulation.

As you know, this is a regulated current stepper driver. When the H Bridge is disabled because an ITrip event has been reached, the H Bridge is disabled for a certain amount of time. A current regulation cycle duration is the inverse of 50 KHz (45 KHz is the minimum you will see on the DRV8821), which comes to 20 us. That means that the TBLANK, TIME ON and TIME OFF periods suming up to a cycle can not last more than 20 us. If an ITrip event took place 15 us after the H Bridge was enabled, the H Bridge will be disabled for 5 us, and so on.

We chose 50 KHz because this frequency is quite above audible range. If we had chosen a frequency lower than 20 KHz, the stepper would behave like a speaker (not a good one, by the way) and completely undesirable high pitch squealing would be present. We can not make it much higher than 50 KHz, because then the switching losses would cease to be negligible and these losses do not produce any gain in motion quality.

Now, back to your stepper. Steppers can not be stepped at any given frequency. If you look at your motor datasheet, you will find a speed/torque curve which details how fast this motor can be stepped. In other words, how fast you can pulse the STEP pin on the DRV8821 multiplied by the degrees of microstepping. If it is full step, then the STEP rate is the same speed the chart provides. If it is half, then you need to multiply the STEP rate by two, if quad, multiply it by four, and so on.

The chart will show you that torque capability decreases as speed increases. Hence, there is a speed at which the motor will not be able to run. For what you describe, that speed is 2.5 KHz. In other words, if full stepping, a frequency on the STEP pin larger than 2.5 KHz will cause the motor to enter resonance and it will not be able to move. This is cause by the physical construction of the motor and no driver on this planet will be able to reliably drive this motor past this frequency at the stated voltage and winding current. Unless we are dealing with a superbly fancy drive system with complex algorithms. But I wouldn't hold my breath on that one...

There are a few reasons why you may be loosing steps:

1. Too little current will cause the motor to not be able to respond to the requested torque. The stepper chart will tell you the speed and the torque it can supply at a given current and voltage. If you increase the current, this line moves up a little bit. But if you decrease the current, then the line moves down. In order for the motor to operate reliably, you must be operating with limits below this line. That is, the speed and torque must be contained within the graph.

2. Resonance will also cause missed steps and quite erratic behavior. Steppers ring quite a bit, and there are some speeds at which they ring more than others. Ideally, you do not want to operate on these speeds. This is again a physical reality of stepper motors and they all suffer from this malady. Techniques to improve performance are to use microstepping. Full step is pretty bad at those speeds causing resonance. When you introduce microstepping, the ringing is softened and the effects are practically removed.

Hope the info helps!

Best regards,

JIQ

Thanks a lot Jose !

Those info helped a lot !