DRV8323R: 6-step commution with 6x PWM Mode

Hi,

The 6 Step commutation works in 1x PWM Mode. Please, verify you are in 1x PWM mode and have correct pin configurations, as stated in the highlighted text and table. 

Hi,

My developed board works well using 1xPWM mode.

And also 6xPWM mode works well using sinusoidal commutation.

So, the pin connection is right.

I tried to 6-step commutation with 6xPWM mode.

But the GHx signal is different according to INHx.

Hi,

I see you also have written in original post inverted gate signal values on Phase B when compared to PWM input signals

(INLB : !PWM, INHB:PWM >> GLB : PWM, GHB : !PWM)

Can you verify if you are measuring the voltages using the correct device pins? Thank you very much.

Additionally, please verify in firmware the value sent for INHA is written correctly. Thank you.

Sorry, thers is a mistake writing in original post.

I check the input signal, gate signal and voltage in MOSFET output.

I think gate signal is wrong.

INLA : low   , INHA:low

GLA : low, GHA : high(<-- wrong, it shoould be low)

(So high side of mosfet turns on. It should be turn off)

INLB : !PWM, INHB:PWM

GLB : !PWM, GHB : PWM

INLC : high, INHC:low

GLC : high, GHC:low

I can use 1xPWM mode. and it works well.
I know 1xPWM mode is simple with hall sensor.
But, I try to use 6xPWM mode with 6-step commutation

Hall sensor is used and the firmware was implemented.
I checked IHx/ILx signals related to hall sensor signals. it is right.
But the gate signal(GHx) is high, when both IHx and ILx are low. I think it is wrong.
Could you check the gate signals?

I will discuss with colleagues and/or experiment and get back to you with information and/or results.

Feedback:

Can you verify if your commutation tables follows the steps outlined in the table shown in the following blogs?

e2e.ti.com/.../generate-your-own-commutation-table-trapezoidal-control-3-phase-bldc-motors-using-hall-sensors

e2e.ti.com/.../generate-your-own-commutation-table-part-2

 a firmware is implemeted well according to communication table.

The problem is the gate signal when INLx and INHx signal is low.

Both gate signal(GHx, GLx) should be low, but GHx is high in 6xPWM mode.

This is unnatural.

thank you.

On my end, the signal is correct utilizing our BOOSTXL-DRV8323RX EVM GUI/FW.

Please, verify the signal integrity of the 6 step commutation by spinning motor using hand torque. Since you are using hall sensor position feedback updates, validate in firmware if the correct INHx/INLx signals are sent for the respective step commutation caused by your applied hand torque to the motor rotor.

I checked all INLx and INHx signals with oscilloscope.

Ths signal is correct.

I don't know why the GHx is high with low INLx and ILHx.

all signals are checked with oscilloscope.

INHA : low     INLA : low

INHB : pwm   INLB : !pwm

INHC : low     INLC : high

then gate signal is as follows.

GHA : high     GLA : low   --> GHA is wrong!

GHB : pwm     GLB : !pwm

GHC : low       GLC : high.

DRV8323RH is used.

mode pin is connected to GND. 

Is there a any possibility to make the gate signal?

Hi Kim,

Did you try to commutate using your the torque applied to your hand? I recommended this suggestion since it will give you a definite accurate commutation reading on your oscilloscope and make sure you are not skipping in logic table.

Dear Hector,

I can't apply the power with motor connection, because the high side FET which should be turned off is always ON.
the abnormal FET operation make the full-duty switching.
So I cheked the signals without motor connection. It means motor power lines are unplugged and hall sensors are connected.

I checked hall sensor and gate input signals with oscilloscope.
And I am sure the commutaion logic is right.

I wonder why both low gate input signals(INLA:low, INHA:low) make high and low gate output signal(GHA:high, GLA:low)
This is strange and it does not matter with commutation logic.

If the logic is programmed correctly, the device configuration is correct, and the INHx output is correct, then the layout is manipulating the GHA signal or the device you have is behaving incorrectly. Can you provide image of your PCB layers of the entire node of GHA and the signals beneath it?

If you need to share privately, please hover over my name and you can email me. To do so, you must friend request me first.