Really bad DVDD ripple voltage when EN_DRV is pulled HIGH on 6EDL7141

[ftl_12]

Hello,

I'm developing a prototype BLDC controller with MOTIX 6EDL7141.

I used the analog programming method recommended by the datasheet to set the DVDD_SETPT to 3.3V.

The IC's DVDD output looks pretty good on my scope when it's not in the 'ACTIVE' state. Everything works properly without any problem.

However, when I set EN_DRV pin to HIGH and start to control the MOSFETs with PWMs, the DVDD output has huge voltage ripples around the time when the MOSFETs are being turned on and turned off. The voltage can go to as low as 1.7V. After a short while (<1second generally), the IC appears to reset itself and the DVDD output is somehow set to 5V instead of 3.3V. What's causing this? Shouldn't the LDO always output 3.3V since it's set by analog programming?

The parts I chose for the integrated buck converter are identical to the parts used in the evaluation board design: https://www.infineon.com/cms/en/product/evaluation-boards/eval_6edl7141_trap_1sh/

The only difference is that my buck loop area is slightly larger than the reference design. The decoupling capacitors are also further away from the IC compared to the reference design.

I'm guessing the problem is my board design but I honestly didn't expect it to be so bad.

Any suggestions?

[SuhasBobade]

Hi,

Thanks for the post and sorry for the inconvenienced caused.

After reading your post, I could understand that something unusual happens to the board when the gate driver is enabled.

This might be due to the loading on the DVDD pin when gate driver is enabled. Kindly check DVDD_OCP  threshold. 

DVDD voltage can be used to supply a microcontroller (MCU) or additional elements in the circuit like Hall
sensors, LEDs. Here kindly check hall sensors supply polarity as well. 

Kindly check above points and revert back with your observations.

Thanking you!

[ftl_12]

Thank you for your reply.

1. The DVDD_OCP_CFG in the SUPPLY_CFG register is left as default: 450 mA.

2. I'm not using any hall sensors in my design.

I can't directly measure the current draw on DVDD but I am very certain that the MCU and other peripheral's current draw don't increase when EN_DRV is pulled high.

This is because the firmware on the MCU simply switches the EN_DRV pin to HIGH when the voltage reading of a potentiometer is lower than a predefined threshold and switches the pin to LOW when the voltage reading is higher than the threshold. I can confirm that the DVDD output is normal if I disable the switching of the EN_DRV pin in the firmware and keep everything else the same (the firmware will still detect voltage changes on the potentiometer and etc). 

Speaking of LEDs, it's noticeable that the LEDs powered by the MCU start to flash when voltage ripples appear. The MCU also restarts randomly since the power supply dips below the minimum voltage requirement.

[SuhasBobade · Accepted solution]

Hi,

Thank you for the reply.

As you mentioned, the IC appears to reset itself and the DVDD output is somehow set to 5V instead of 3.3V, this might be due to the resistor chosen, could be at the borderline of 3.3V or 5V setting. We suggest to choose values close to 2K for DVDD=3.3V, or 12K for DVDD=5V. Kindly check the output of the LDO at the 6EDL7141 and at load (MCU) or whatever load powered up with 3.3V and if required add some caps at the load DVDD pins.

You should always follow manufacture's design guidelines for guaranteed performance. If possible kindly share schematic and layout files. 

Kindly check above point and share your feedback.

Thank you!