drv8829 pwm noise

Hi Dan,

Can you provide more information, like the voltage you are using and the motor connections? A schematic of your board would also be helpful.

While awaiting your information, I have seen (and heard) audible noise under two conditions. One is that the undervoltage circuit is shutting down the drivers because VM droops below UVLO. Another condition is the overcurrent protection. After the blanking time (typically 3.75 us), if the current is too high the driver will shut down temporarily. This can create an audible condition.

Right now I am using 30VDC (lab power supply).  Here is a snip of the schematic (one coil, the other coil has the same circuit).  Right now I have the peak current set to 2.5A.  Motor is a 6A 0.5ohm 2mH.

BTW, the DRV8829EVM has a same issue.

Thanks.

BTW, I see this when the motor is idle and when I am I stepping.

Dan,

Thanks for the info. I have a few more questions pointed toward the power supply and motor since both your board and the DRV8829EVM show the same symptoms.

1) What is the current rating of the power supply? Is it possible that VM is drooping for short periods of time? You can look at nFAULT to determine this.

2) Does the motor have a capacitor across it? If so, what is the value? I have seen this cause problems.

3) Which decay mode are you running?

4) Can you provide scope captures of the outputs voltages, the current through the motor, and VM when this occurs. What is the state of the I[4:0] and VREF pins during this time.

Thanks.

Rick--

1) The lab supply is current limited to 3A.  PSU shows no indication of going into CC mode.  Fault line is always high -- during motion (> 10K pulses/sec) and idle.

2) No caps across motor leads.  There is ground fill on same plane and a ground plane under the motor leads.

3) Decay mode is mixed (Vdecay is 1.6V).  

4) Attached is Imotor(CH4)/Vmotor and Imotor(CH4)/Vsupply(CH1).   Imotor probe is 100mv/1A.  Vmotor is 20V/div.   Note there is no motion during this capture.  Same with motion.  Vref=0.96V.  I[4:0]=0x18.

Dan

Dan,

I am still investigating your issue. Based on your scope shots, I don't see the cause of audible noise. The PWM signals are operating at ~50kHz; well above the audible range.

Is it possible that the motor is vibrating internally, and this vibration is causing the noise? Can you add some type of dampener (bushings, pad) to confirm this?

Can you provide the motor p/n? If we have it here, we could try to replicate the issue.

I will continue to research it on this side.

Thanks.

Rick--

Subharmonics from the 50KHz is my guess.  If I hold the motor, they are still there.  At 0% and 100% current in one coil and it goes way.  If I unhook one of coils, the noise is gone.

The motor is custom.  Is there a way to send one to you or try one that is similar?

Dan

Rick--

Is there a way to apply an external PWM signal on the DRV8829?

Dan

Dan,

I am checking into the next step for your problem, as well as applying an external PWM signal to the DRV8829.

Will post an update soon.

Dan,

PWM'ing is possible using the ENBL pin. Refer to Table 2 in the datasheet. One must be careful when PWM'ing to set I[4:0] higher than the desired current, or the internal chopping circuit will interfere with your external PWM'ing.

You are now bypassing the internal current control, so an external op amp across Rsense may be needed to adjust your PWM input.