DRV8412DDWEVM: PWM input 100%，output motor will unstable

Hi Kevin,

Please, go to chapter 7.3.2.1 of below datasheet

https://www.ti.com/lit/ds/symlink/drv8412.pdf?ts=1666599582444&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FDRV8412

I think it may be the problem.

Regards,

Grzegorz

Hi Grzegorz,

Please suggest Please suggest the correct BST capacitors.

Since the input frequency is 3.6KHz, change the 4 BST capacitors (C15, 18, 20, 23) were changed from 100nF to 220nF, and the situation improved, but sometimes it still unstable.

And change the 4 BST capacitors from 220nF to 470nF, the DRV8412 burns out.
(Connect a 4.7Ω resistor in series between POWER SUPPLY +12V and EVM GVDD.)

Hi Kevin,

I think the main problem is that you try to run DRV8412 at PWM 100%, I guess 220nF or 470nF should work but you will need to increase minimum OFF time for PWM as well. Instead of 50ns (chapter 6.3) LS FETs will have to be ON for a longer time. I am not sure if you will be able to place series resistors between GVDD_X pins and decoupling caps C8, C9, C13 and C14 on EVM board. It would be easier to place them in series with bootstrap caps C15, C18, C20 and C23 (in this case switching times will be a bit longer, so I would stay between 5-10 Ohm).

I would try 220nF without series resistors and keep PWM lets say below 99.9% and if necessary increased PWM frequency to 5kHz. If I needed PWM at 100% I would choose another driver with charge pump instead of bootstrap caps.

Regards,

Grzegorz

Hi Pablo,

Is DRV8412 and DRV8962 the same EVM? Or does DRV8962 have a dedicated EVM?

Kevin,

Both devices are dual H-bridges but the DRV8962 has a different control interface and has a charge pump. 

There is a DRV8962EVM but for some reason it is not showing up on ti.com. We are working on fixing this problem. Once it is resolve, you can order the EVM.

Regards,

Pablo Armet

Kevin Huang8 said:

When PWM input 100%，output motor will unstable

The input pwm input cannot be 100% as this gives zero time for the low FET turn on to charge the BST cap. The chip requires a minimum 50ns per pwm cycle. To avoid to run at 100%, you might want to increase supply voltage higher than 12v.

The output is turned off (unstable) because the BST voltage is under voltage threshold and the driver chip turned off the high FET and turned on the low FET a short time to charge the BST cap.

Kevin Huang8 said:

And change the 4 BST capacitors from 220nF to 470nF, the DRV8412 burns out.

The internal diode burned out because the datasheet states that 220nF is max. When using more capacitance for BST, you might want to add a Schottky diode between GVDDx to BSTx pins.

If you need to have pwm=100% then choose other drivers that can run at 100% as they have different high side gate drive supply circuit.

Regards,

Brian