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.

DRV8305-Q1: High current after setting Gate_EN

Part Number: DRV8305-Q1
Other Parts Discussed in Thread: DRV8305

Hello,

we are using the DRV8305NQPHPQ1 as the gate driver for our
BLDC motor control board and are experiencing a jump in the
the total current to the board from 70 mA to 390 mA when
the GATE_EN bit is set. The DRV8305 also becomes hot
(58°C). This effect is seen all boards, so it seems to be
a systems phenomenon. The schematic and layout are taken
from the datasheet for the DRV8305 (SLVSD12C).

Here the schematic of the DRV8305:

and the schematic of the half-bridges looks like this:

The driver works and it it possible to spin motors (the MosFET
switching looks good), but the chip is hot. Any ideas where all
that current is going to?

Regards,

John

  • Hey John,

    I will consult with team and aim to provide feedback before the end of the week.

    Best Regards,,

    Akshay

  • Hello Akshay,

    thank you for your message. I will continue looking from my side and share any results here.

    At first, I thought it was maybe an issue with the boards, but now the same design has been
    used for a second board and it is showing the same issue. Plus, it effects the whole lot of boards.

    This issue is critical for us, since the board should also work at 80°C ambient. Although I have
    not yet tested in a tempered environment, I have some concerns, that this will be an issue.

    Look forward to the response from your team.

    Kind Regards,

    John

  • Hey John,

    I would like to know what Idrive setting you are using and would it be possible to lower it to see if this helps with the thermals? Could you also provide the datasheet of the MOSFETs being used?

    Have you measured what the phase current waveform looks like?

    Best Regards,

    Akshay

  • Hello Akshay,

    In the search of where the current is going, I have removed the gate resistors for all phases as well as the transistor connected to the Vcp. No improvement, the current remains at 360 mA. On this board, it is not possible to open the SH_X and SL_X circuits.

    As you asked, here are the settings for the Idrive and dead times:

        drvicVars_M1.Ctrl_Reg_05.TDRIVEN = TSour_2000_ns; // default 2000 ns (1780 ns)
        drvicVars_M1.Ctrl_Reg_05.IDRIVEN_HS = ISink_HS_0p500_A;  // default 60 mA
        drvicVars_M1.Ctrl_Reg_05.IDRIVEP_HS = ISour_HS_0p125_A;  // default 50 mA
        drvicVars_M1.Ctrl_Reg_06.TDRIVEP = TSink_2000_ns; // default 2000 ns (1780 ns)
        drvicVars_M1.Ctrl_Reg_06.IDRIVEN_LS = ISink_LS_0p500_A; // default 60 mA
        drvicVars_M1.Ctrl_Reg_06.IDRIVEP_LS = ISour_LS_0p125_A; // default 50 mA
        drvicVars_M1.Ctrl_Reg_07.DEAD_TIME = DeadTime_500_ns;  // default 60 ns (52 ns)
        drvicVars_M1.Ctrl_Reg_0A.GAIN_CS1 = Gain1_10VpV; // default 10 V / V
        drvicVars_M1.Ctrl_Reg_0A.GAIN_CS2 = Gain2_10VpV; // default 10 V / V
        drvicVars_M1.Ctrl_Reg_0A.GAIN_CS3 = Gain3_10VpV; // default 10 V / V
        drvicVars_M1.Ctrl_Reg_0C.VDS_LEVEL = VDS_Level_0p175_V; // default 1.175 V
    

    and the datasheet for the MoSFETs:

    NVMFS5C430NL-D.PDF

    I have measured the phase currents in order to check the current sensors and
    everything in my opinion looks good. The scaling factor for the current shunts
    is also correct.

    Here is a scope shot of the phase voltages (1: Ph_A, 2: Ph_B, 3:Ph_C) and the
    current on Ph_A. The PWM voltage of Ph_A is set to -0.4 V (pwmData.Vabc_pu.value[0])
    while Ph_b and Ph_c are set to 0.0 V. The motor is connected.

    Please let me know if you are seeing something that I am missing. :-)

    Greetings,

    John

  • Hey John,

    Thanks for providing additional information. I will aim to provide feedback before the end of the week.

    Best Regards,

    Akshay

  • Hello Akshay,

    thank you for the update.

    A new information. I have notice that C94 (4.7 uF on PVDD of th DRV8305) is running hot ( 75°C). I would have suspected high frequency ringing going over the capacitor to ground. However, the the scope image is clean on both sides of the cap.

    As I will be out of the office for the next two weeks, what we don't get solved tomorrow, I will have to pick up in calendar week 41. I look forward to any information you may be able to supply tomorrow.

    Greetings,

    John

  • Hey John,

     

    Thank you for your patience. 

    What is your PVDD? 

     

    I did a schematic review and here is my feedback:

    1) Please ensure CP1H cap is rated for PVDD and CP2H cap is voltage rated for PVDD x2. 

    2) The VCPH to PVDD cap is recommended to be 2.2uF not 2.2nF. This might be the major point of concern.

    3) Is there an external pullup resistor on nFault as recommended by the datasheet?

    4) I noticed that the Qgd of the MOSFET is 9nC and with your iDrive setting you will get a rise time of 72ns and 18ns for fall time. This is very fast as we typically recommend 100-200ns for fast switching. So, I recommend lowering iDrive.

    5) Is C94 rated for PVDD? I would suggest using a cap with even higher rating as sometimes caps get derated and it might affect their effectiveness.

     

    Best Regards,

    Akshay

  • Hello Akshay,

    thank you for your review and feedback. That helps me a lot!

    We are in automotive, so PVDD is 9 V to 16 V. Sorry, I should have mentioned that in the first mail.

    To you points:

    1) CP1H and CP2H are both 47 nF 50 V. Check: 16 V + 12 V = 28 V * 2 = 56 V -> borderline. What do you think, better to increase or okay?
         Part number TDK CGA3E2X7R1H473K080AA
         Datasheet here: TDK
    2) That was it! Wow! I think, I have looked at that on a (felt like) 100 times and did not see the mistake. I did not have a 2,2 uF part that fit, but was able to put in 10 uF 25 V and the current is down to 100 mA which is what I expected, the DRV8305 is much, much cooler and C94 is also not hot. So, I think this was the problem. Thank you, Thank you, Thank you. I am very happy.

    3) Yes, there is a 10 k pullup resistor R92 which is connected to nFault

    4) Good information and I will check. I originally increased these values due to the gate resistors at 10 ohms, but will lower them down again.

    5) C94 is rated at 35 V. Is this sufficient in your view?

    Since Things are looking good. I will close this thread and have marked you solution as resolving this issue. If there are any other issues, I will open a new thread.

    Thanks again!

    Best Regards,

    John