• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 29 subscribers
  • Views 1020 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.

DRV8303 many times damaged

N.AB
Intellectual 390 points
Other Parts Discussed in Thread: DRV8303, DRV8305

Hello,

I have a motor control design with a DRV8303 that I had duplicated 16 times. Unfortunately for 3 times the DRV had been damaged. And I am wondering about the reason.

My design is joint. As could be seen, the shunts values aloows up to 16A measurement. Unfortunately the DRV damages much lower. the internal gain is set to 10.

here is my register config:

//	DRV8303_cntrl_reg1.bit.GATE_CURRENT = 0;		// full current 1.7A
//	DRV8303_cntrl_reg1.bit.GATE_CURRENT = 1;		// med current 0.7A
	DRV8303_cntrl_reg1.bit.GATE_CURRENT = 2;		// min current 0.25A
	DRV8303_cntrl_reg1.bit.GATE_RESET = 0;			// Normal Mode
	DRV8303_cntrl_reg1.bit.PWM_MODE = 0;			// six independant PWMs
//	DRV8303_cntrl_reg1.bit.OC_MODE = 0;				// current limiting when OC detected
//	DRV8303_cntrl_reg1.bit.OC_MODE = 1;				// latched OC shutdown
//	DRV8303_cntrl_reg1.bit.OC_MODE = 2;				// Report on OCTWn pin and SPI reg only, no shut-down
	DRV8303_cntrl_reg1.bit.OC_MODE = 3;				// OC protection disabled
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 0;			// OC Vds=0.060V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 4;			// OC Vds=0.097V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 6;			// OC Vds=0.123V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 9;			// OC Vds=0.175V
	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 15;			// OC Vds=0.358V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 16;			// OC Vds=0.403V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 17;			// OC Vds=0.454V
//	DRV8303_cntrl_reg1.bit.OC_ADJ_SET = 18;			// OC Vds=0.511V
	DRV8303_cntrl_reg1.bit.Reserved = 0;

//	DRV8303_cntrl_reg2.bit.OCTW_SET = 0;			// report OT and OC
	DRV8303_cntrl_reg2.bit.OCTW_SET = 1;			// report OT only
	DRV8303_cntrl_reg2.bit.GAIN = 0;				// CS amplifier gain = 10
//	DRV8303_cntrl_reg2.bit.GAIN = 1;				// CS amplifier gain = 20
//	DRV8303_cntrl_reg2.bit.GAIN = 2;				// CS amplifier gain = 40
//	DRV8303_cntrl_reg2.bit.GAIN = 3;				// CS amplifier gain = 80
	DRV8303_cntrl_reg2.bit.DC_CAL_CH1 = 0;			// not in CS calibrate mode
	DRV8303_cntrl_reg2.bit.DC_CAL_CH2 = 0;			// not in CS calibrate mode
	DRV8303_cntrl_reg2.bit.OC_TOFF = 0;				// normal mode
	DRV8303_cntrl_reg2.bit.Reserved = 0;

I'm wondering about the possible reasons of this damage and mitigation

Thanks

  • 0
    Intellectual 390 points

    I forgot to join the designDRV.pdf

  • 0 in reply to N.AB
    TI__Mastermind 26065 points
    Hi N.AB

    Note that you are using high voltage >60V in another post, is that the same case here?

    Normally DRV8303 will not die because of over current as it is a pre-driver. But the voltage >60V will be the No.1 killer to the device.

    Even you lower the system voltage < 60V but not enough margin included, the motor is most likely to generate VM pumping voltage when it is in the deceleration state.

    Also in your schematic, only see 2.2uF to PVDD and all the half bridge top as the bulk cap. This is not enough for most BLDC applications.

    Best regards
  • 0 in reply to Wilson Zuo
    Intellectual 390 points
    Hi,
    The two posts are actually related.
    Howevers according to the voltage, I don't had used the board more than 52V, and actually the damage occures at 25V.
  • 0 in reply to N.AB
    TI__Mastermind 26065 points
    Hi N.AB,

    Could you have detailed running state before the damage happens? If motor quickly speeded down at 25V, the real PVDD can go up to 2x or more, depends on the deceleration rate, especailly when you don't have the big bulk caps. You can redo the tests with 1000uF cap placed close to the half bridage tops. And monitor the PVDD voltage through the pins if any damage happened. A safe way to investigate the issue is to apply a slower deceleration first, example, from 100% speed to 80% speed and monitor the PVDD change during this speed down.

    You can do the test with and without the big bulk caps installed. By this way, we can identify whether there is directly relationship of the root cause and high pump voltage.

    Besides, DRV8303 must be well soldered down with its PowerPAD connected to the GND. This is very important to make it function correctly. Many customers have the issue when testing in lab with hand soldering.

    Best regards,
  • 0 in reply to Wilson Zuo
    Intellectual 390 points
    Hi,
    Actually in addition to the 2.2uF, we have a chemical capacitor of 68uF.
    Before the damage, the pwm of U phase was at 95%, and the pwm of V and W pahses was at 5%. the supply volatage was 25V.
    Without shutdowning the logical part of the board (the uC), the power part was powered of then powered on. Just after the DRV has beeen damaged
  • 0 in reply to N.AB
    Intellectual 390 points

    Hello,

    Is there any extra info needed to diagnoze the issue?

    thanks

  • 0 in reply to N.AB
    Mastermind 9125 points
    Hi,

    I don't think 68uF is enough. Recently, I am also designing my motor control project using DRV8305. The suggested bulk cap is 470uF*2 = 940uF. According to my understanding, if the bulk cap is too small, the induced voltage will cause high voltage peak which will damage the driver.

    If you can, please use oscilloscope to measure the voltage waveform. This is be the way to see what's happened.