• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 692 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.

Original question:

DRV8701: DRV8701 interferences radiation

DRV8701: DRV8701 interferences radiation

Miguel Marin
Prodigy 30 points
Part Number: DRV8701

Hello,

we are using the DRV8701 and we have the same problem related here, and no resolved. In the EMI test certification the driver has a peak at 50 MHz, 100MHz and 150 MHz with level over the Certification limits. We tried with 1nF-100nF capacitors in SH1-SH2, SH1-GND,SH2-GND, increasing the decoupling capacitors on VCC, with filter in Vcc but the problem not change.

What can we do to improve this problem?

Someone can help us?

Thanks. Regards,

  • 0
    TI__Genius 14487 points
    Hi Miguel,

    I'm assuming the EMI test you're doing is radiated emissions. The first thing to do is to find what exactly is radiating. The EMI debug guidebook I have at my desk says that cables are likely the source or emissions for radiation below 200 MHz. If you can find a spectrum analyzer and make (or buy) an EMI probe, this might help identify the location of the emissions. The link below shows how to make a near-field probe:

    www.eng.mu.edu/.../aidi.pdf

    You can check your design to see if those frequencies are present in your equipment already. I also recommend turning off or disabling various devices on your board to see which one might be causing it. You can also determine if the DRV8701 is the device that is causing the issue by doing the following things.

    1) Disconnect the motor and test the board. If the emissions go away, then this issue may be related to the sizing of the bulk capacitor at the power-entry point of the board.

    2) If the emissions continue when the motor is disconnected, try changing the IDRIVE setting. A large IDRIVE current can cause the parasitic impedances around the FETs to ring and produce emissions.

    3) If the emissions are still there after step 2, then stop sending PWM signals to the driver. This will also help determine if the source of EMI is related to FET switching.

    4) If emissions continue when the device is enabled and no PWM signal driving the device, then the issue may be related to the device charge pump or the internal logic. You can test this by putting the DRV8701 into sleep mode by bringing nSLEEP low. If the emissions go away when you do this, then the DRV8701 charge pump or logic may be the cause. If this is the issue, then the board layout will need to be modified to reduce the size of the effective loop between the device pins and the capacitors on the charge pumps and and supplies.

    5) If emissions continue after nSLEEP is low, then another device on your board is to blame for emissions.
  • 0 in reply to James Lockridge19
    Prodigy 30 points

    Dear James,

    thanks for your reply.

    When I disconnect the motor the noise doesn't dissapear. I tried with all the possible configurations for Idrive and the emissions were still there. If I disable PWM the emissions don't dissappear. I'm sure the noise is produced by the DRV8701. If I put the driver in Sleep mode (with nSleep pin) the noise dissappears. If I remove charge pump capacitors the noise still there. Those capacitors are very very close to the device. I think the problem may be generated by AVDD and DVDD. If I remove the capacitors populated on those pins (1uF) the emissions decrease.

    Are  those capacitors needed for the normal runing of the DRV?

    Thanks

  • 0 in reply to Miguel Marin
    TI__Genius 14487 points
    Miguel,

    The capacitors on AVDD and DVDD are needed for normal operation of the DRV8701. AVDD and DVDD are LDOs, so the capacitor ensures they are stable.

    Are the caps populated close to the device so the loop between AVDD and ground is as small as possible?
    Are you using ceramic capacitors? They have the lowest ESR.
    Are there any vias between the capacitor and the IC? These act as inductors and result in poor decoupling on those supplies.

    Can you try soldering a 0.1 uf cap in parallel or on top of the 1 uF cap for testing? Also, try with a 0.01 uF capacitor.
  • 0 in reply to James Lockridge19
    Prodigy 30 points

    Dear James,

    thanks for your feedback. Then, I understand the DRV uses AVDD and DVDD internally (this not appears in the block diagram), is correct?

    The caps are close to the device, like in your develpment board, but we're going to put them closer and add another with 10-100nF.

    In our test, with capacitors very very close to the device, the EMIs level decreases, but is still present.

    It would be important to do a special mention in the datasheet to avoid problems with certifications procedures.

    Best regards,

  • 0 in reply to Miguel Marin
    TI__Genius 14487 points
    Miguel,

    DVDD powers the logic in the device, and AVDD powers the analog circuitry (like the sense amp). They can also be used to power external circuitry for light loads up to 30 mA.

    If you add smaller capacitors to the device in your final layout, I recommend putting the smaller ones closer to the device.