• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 1096 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DRV8701: 3oz copper and gate driver selection

Michael McCormack
Prodigy 160 points
Part Number: DRV8701
Other Parts Discussed in Thread: UCC27201, LM5109B-Q1

I need to drive a brushed DC motor in a 12V battery powered robot. I am thinking that I want to use the DRV8701 over the UCC27201 for my robotics application. My thinking was:

  • The DRV8701 has a wider operating range which goes down to 6V. With a 12V battery which will be discharging and adding in some cable drop when the attached motor locks I am somewhat concerned that the UCC27201’s 8V input might not be enough.
  • The DRV8701 uses a charge pump to allow 100% duty cycle operation of the high side FET. With the boost cap circuit on the UCC27201 I can’t just turn on and forget the driver as at some point it will discharge too much.

My problem is that I am looking at continuous current of about 30A and locked rotor current as high as 104A according to the motor spec I have today, but as a rule I like margin to change motors somewhat if suppliers change. The issue is with only 0.0086 inches between pads on the DRV8701 I can’t really use anything but 1 oz copper. Is there a way to use the DRV8701 with high current applications I’m missing? If I have to use the UCC27201 (or similar boost cap topology) how often does my micro need to cycle the FETs to ensure charging (there is nothing in the TI literature about what to do)?

  • 0
    TI__Expert 4650 points
    Hey Michael,

    The DRV8701 is a gate driver, so we actually don't require thick copper traces into the part for the high currents you need in your application. The maximum average supply current drawn by the DRV8701 is on average 6 mA. We can supply gate driver currents of 150 mA source / 300 mA sink into an external MOSFET rated for whatever you need your gate-driver current to be.

    For context, our DRV8701 EVM uses 2 oz copper across the entire top and bottom layers.

    www.ti.com/.../drv8701evm
  • 0 in reply to Phil Beard
    Prodigy 160 points

    Thanks for the condescending opening about the gate drivers' current requirements, I had no idea that the load currents were carried by the FETs.  So you can understand the problem please let me explain, PCB's are manufactured with the same thickness of copper on the entire PCB layer pair.  Hence, when one needs thick copper in one area it shows up on the entire layer.  A design which calls for a high current FET and something to drive it will use the same weight copper to connect them both if you mount the components on the exterior of the PCB.  For a cost competitive product, shipping in volume, thicker copper requires more than 8 mils of clearance but when I design my next low volume / high priced EVMs where my employer can fork over bushels of cash your guidance will be invaluable.  Now that I understand the target market for the DRV8701 I can move on.

    Do you have an answer for the second question, what duty cycle do I need to keep the high side driver of the UCC27201 charged?

  • 0 in reply to Michael McCormack
    TI__Guru** 114905 points
    Hi Michael,

    We have contacted the UCC27201 experts, and should have a reply soon.
  • 0 in reply to Michael McCormack
    TI__Intellectual 2805 points
    Hi Michael,

    I am the applications engineer in charge of the UCC27201 and I would be happy to assist you.

    In order to answer, "What duty cycle do I need to keep the high side driver of the UCC27201 charged?" I need to know the following:

    1) Selected bootstrap capacitor value. (The LM5109B-Q1's application section (www.ti.com/.../lm5109b-q1.pdf) can be used as reference)
    2) Gate charge of the FET
    3) Are you planning on using a gate-to-source resistor or any other leakage path for the bootstrap cap to discharge?

    regards,
    Mateo