• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 29 subscribers
  • Views 602 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV8811 Audiable noise

Peter Sleeman
Prodigy 30 points
Other Parts Discussed in Thread: DRV8811

Hi

I'm new to the DRV8811 and have a first proto board up and running driving two small-ish hybrid steppers connected in series bipolar fashion.  They are 8-wire types with 1amp 5V phase windings (each of the 4).  So in series they are "10V 1amp".

They step OK after a bit of fooling around.  But they squeel and hiss all the time (like boiling kettles!).  The two driver chips are medium warm, not "hot" in my estimaiotn.  I can affect the hissing (and chip temp) by adjusting the current setting via a multi-turn pot controlling Vref level.

I have 30K and 1nF for the time constant components (was 47K and 1nF but reading some other threads prompted the move to a little higher chopping frequency).

In full step mode I *can* make them quiet, setting Vref to around 1.35V (Vdd is 3.3V).  I use 0.2ohm sense resistors and Vmot is 12V.  But they seem to be on the edge of non-operating when at this quiet point.  In micro step mode they are always noisey regardless of the pot (the squeeling just changes its fundamental frequency).  They do step OK in microstep btw.

So I'm a bit puzzled what to try next.  I'm using sync rectification btw.

TIA

  • 0
    TI__Expert 7455 points

    Hi Peter,

    Steppers that are current chopped (current regulated) will always be somewaht noiser. This is because of the current ripple at the motor windings. Although not meant to work as speakers, the effect is nonetheless present. Some people even take advantage of this "feature", and playfully transform their motors into music synthesizers.

    Obviously, we want the motor to move a load, not to play keyboard. Hence, this high pitch squealing must be eliminated at all costs. It can be done if you have access to looking at the current. If you can zoom in at the winding current by measuring it with a current probe, it should be fairly easy to see the frequency at which the current chopping is taking place. If you see frequency being below 20 KHz, you will most likely hear it. The lower it is, the more you will hear it.

    The problem is that the resistance at the RCx pins alone will not determine the switching frequency. At the end, the switching frequency is affected by the power supply voltage (the higher the voltage, the higher the current increases and reacheds ITRIP), the motor inductance (the higher the inductance the slower the current increases) and the decay mode (fast, slow or mixed). Each motor and application will be different so using somebody else's resistance and cap may not be enough. You may need to choose your own according to the motor and the power supply voltage.

    Here are some guidelines that can help:

    1. Mixed decay mode can help to reduce the high pitch squealing.

    2. The higher the switching frequency, the least auditive are its effects on the winding.

    3. Fast decay mode always result in the most audible as the current ripple is the largest.

    4. The higher the stepper current, the more audible it tends to be.

    The fact that your noise levels are not as bad on full step points out to the fact that your noise is coming from your fast decay. Under full step, current is always regulated with sloe decay. I would try to maximize the effects of the slow decay as much as possible, although under microstepping you will want a percentage of fast decay. In other words, you want mixed decay during sine wave quadrants 2 and 4 (angles 90 to 180 and 270 to 360). You mention adjusting a pot. Is this with the DRV8811 EVM? If so, make sure the jumper has been moved to the respective position.

    The DRV8811 offers a great deal of flexibility when it comes to selecting the switching frequency so I am positive there should be a combination in which audible noise can be minimized greatly. I am not going to say it can be 100% removed as this may not be possible depending on the applications particulars. Steppers are by definition prone to vibration, which in turns means noise. In order to fully remove most of this vibration, different algorithms, like 256 degrees of microstepping or AC Servo drive, may be needed.

    Hope the info helps. Best regards,

    JIQ