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.

DRV2667: How to reduce the rippling noise in the output voltage

Part Number: DRV2667

Hi, 

    I am using DRV2667 to drive a disk piezo actuator with maximum of  +- 100 V output range. We have another separate PCB with gyro, MCU and some other driver sitting on top of this piezo actuator. Of course the PCB and piezo actuator are insulated.

    problem observed : when I drive the piezo actuator with a constant voltage(say 50V), I noticed my actuator is buzzing with high frequency(not the sound caused by the switching of the driver) and my gyro on the PCB is drifting a lot even through the platform is not rotating. This problem always exists no matter whether I am using the evaluation kit or my PCB board(with DRV2667 on it) to drive the piezo actuator. However, if I use the DC power supply equipment in our lab to direct drive it with same voltage, the buzzing and drift of gyro disappears. 

    Possible cause: I used oscilloscope to compare the voltage generated by the driver and our DC supply equipment and noticed the voltage generated by the driver has a large high frequency noise. And this noise level increases with the output voltage. Seems it has a frequency around 2.5 KHz which may corresponding to switching frequency. The noise of our DC supply equipment is significantly smaller. Thus we think the buzzing is caused by ripping in the output voltage and the gyro drift is due to the electromagnetic interference caused by the noise. 

    Questions:  I am wondering if you have any suggestion about reducing the ripping in output voltage, such as changing inductor/capacitor values, adding filtering in the output voltage. What's the PWM switch frequency of the boost converter?  Any suggestion about reducing the electromagnetic interference of this chip? 

I also attached the schematic of our own driver board as a reference. Let me know if you need more information. Really appreciated. Thank you very much. 

 

  • Hi, Yanhai,

    Welcome to e2e and thank you for your interest in our products!

    This is a common problem on the piezo drivers with integrated boost. Due to the high voltage that these devices handle, the high ripple noise could be faced if the supplies are not regulated correctly.

    The first thing that I would suggest is to increase the capacitance level at all the power pins of the DRV2667. This involves the VDD pin and PVDD pin. The most important thing here is to place the capacitors as close as possible to the pins (I would suggest to increase the capacitance 10x times). Since the power supplies handle a high voltage level from the boost, there could be a significant drop if the capacitor is not close to the device. This practice may help to reduce the ripple.

    In addition, is the high frequency noise part of the frequency band that you will use at the output? I mean, is it acceptable for you to include a low pass filter to mitigate this high frequency noise?

    If that's the case, you may use our LC filter calculator tool in order to include a filter circuit on the output stage:

    https://www.ti.com/tool/LCFILTER-CALC-TOOL

    Best regards,
    Luis Fernando Rodríguez S.

  • Hi Luis, 

             Thank you for your prompt reply.  Seems LC filter design is for audio load by specifying the its resistance. Can it be used for a piezo actuator load with a capacitance load value? If so, how should I accommodate  this change ? Thank you 

    Yanhai

  • Hi, Yanhai,

    This is a good question. Let me perform some tests with the LC filter and I will come back with my results and recommendations.

    Best regards,
    Luis Fernando Rodríguez S.

  • Hi, Yanhai,

    I made a mistake regarding my recommendation with the LC filter. As you mentioned, the capacitance of the filter will be considered as part of the piezo load. So, I think this won't apply to this case.

    However, I strongly suggest to follow my suggestions about the decoupling capacitors.

    Best regards,
    Luis Fernando Rodríguez S.