DRV8353: motor drive issue

Cherry Zhou

Part Number: DRV8353

Hi Team,

The customer wants to drive a brushless DC motor using the 1×PWM mode of the DRV8353H and below is a part of the schematic:

The customer is going to use R14 to sample the total current, R30, R29, and C52 are filtering.

R31 and R32 connect a 680K resistor to ground respectively, R34 is not soldered and R33 is shorted directly to turn off the VDS monitoring function. R35 is not soldered and R36 connects a 680K resistor to ground.

During the test phase, the customer did not solder C52, and a 10k resistor was soldered at R30 and R29. The nFault pin output is low after power-up and the motor does not spin. It was found that the lower arm of the MOS tube was all open and the upper arm was all closed. The voltage on R29 was measured to be 4V on one side near the sampling resistor by a voltmeter. Then short each end of R29 and R30, repower-up and the motor starts to spin.

The power down and modify the program to change the direction of rotation of the motor, the motor will spin slower after repower up and nFault is pulled low again. The SPA will see a 7-V level when the Enable pin is enabled, even if R29 is not connected.

Could you help check this case? Thanks.

Best Regards,

Cherry

over 1 year ago

0 Cherry Zhou over 1 year ago

TI__Mastermind 22235 points

Hi,

Giving some updates:

If R29 is not populated, that is, the SPA, SPB, and SPC pins are shorted to float and at this time you can see the waveform of the shorting point as shown in the following figure, it is a triangle wave of large magnitude:

Please see the links below for PCB and schematic:

https://e2echina.ti.com/cfs-file/__key/communityserver-discussions-components-files/121/SCH_5F003575C178D26300955F00_2023_2D00_10_2D00_08.pdf

https://e2echina.ti.com/cfs-file/__key/communityserver-discussions-components-files/121/PCB_5F00_PCB4_5F00_2023_2D00_10_2D00_08.pdf

Thanks and regards,

Cherry

0 Akshay Rajeev Menon over 1 year ago in reply to Cherry Zhou

TI__Mastermind 23446 points

Hey Cherry,

Please find my schematic review attached. I suggest reviewing the feedback and then retesting to see if this improves the situation.

_SCH_电磁插销_2023-10-08-TI Reviewed.pdf

Hope it helps.

Best,
Akshay

0 Cherry Zhou over 1 year ago in reply to Akshay Rajeev Menon

TI__Mastermind 22235 points

Hi Akshay,

Thank you for the support!

R14 is used to sample the total current and has a resistor value of 400 mΩ. The BLDC motor used is rated at 1.7 a and will not trigger VSENSE overcurrent.
Leave SPA, SPB, and SPC all floating and power on and find all three lower bridges of the MOSFET open, all three upper bridges are shut down. The gate drive voltage for each of the three MOSFETs of the lower leg is 7.3 V. The SPA, SPB, and SPC pins are all at 7.3 V. The motor is not connected and the nFault pin outputs low. R33, R34, R35, R36 in the schematic are provided to allow for placement for debugging.

Cherry Zhou said:
R31 and R32 connect a 680K resistor to ground respectively, R34 is not soldered and R33 is shorted directly to turn off the VDS monitoring function. R35 is not soldered and R36 connects a 680K resistor to ground.

The correct practice should be to short unused SPX and SNX to ground. But there should not be a 7.3V even floating. According to page 36 of the manual, the internal structure of the drive has a 150-kΩ resistor as shown in the following figure:

Does SPX and GLX have a voltage of 7.3V at the same time due to this resistor?

The input INAH has a PWM frequency of 20kHz and a duty cycle of 32%.

Thanks and regards,

Cherry

0 Akshay Rajeev Menon over 1 year ago in reply to Cherry Zhou

TI__Mastermind 23446 points

Hey Cherry,

I will look into this and aim to provide feedback.

Best,
Akshay

0 Akshay Rajeev Menon over 1 year ago in reply to Akshay Rajeev Menon

TI__Mastermind 23446 points

Hey Cherry,

Please look at the following FAQ section 6 on unused CSA for our recommendation.

https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/1176539/faq-quick-guide-to-debugging-common-issues-in-bldc-motor-drivers

If all the HS FETs are off and all LS FETs are on when it seems like they are doing LS braking. Have they measured the brake pin in 1x pwm mode to ensure that this is not the reason of the observed issue?

Best,
Akshay

0 Cherry Zhou over 1 year ago in reply to Akshay Rajeev Menon

TI__Mastermind 22235 points

Hi Akshay,

The motor can now spin after the customer has resoldered a circuit board because the SPX pin of the chip was burned out due to previous misoperation.

The current behavior is that the forward and reverse speed is different after the motor is turned.

The PWM frequency and duty cycle given to INHA are the same, only the INHC level is changed to change the direction of rotation. However, the speed of one direction of the motor is slower than the other direction, and the motor stalls during rotation, the nFault pin is still high and no fault is reported. The motor shaft needs to be manually actuated before it can continue to spin.

They've tried on another drive and the speed of the motor is the same, so the issue should not be related to the motor. Could you help check what could be the possible cause of this phenomenon?

Thanks and regards,

Cherry

0 Akshay Rajeev Menon over 1 year ago in reply to Cherry Zhou

TI__Mastermind 23446 points

Hey Cherry,

Glad to hear that the motor is now spinning.

Hmm is there a load difference between the clockwise/counterclockwise direction?

Also I am suspecting that this might be due to hall signals being misaligned.

Motor drivers

Motor drivers forum

DRV8353: motor drive issue