DRV8350: motor phase current

Thanks for your reply. 

We use MOSFET of SQJQ510ER, and the working voltage is 72V, and maximum phase current is 400A peak.

Currently when we debug with DRV8350RH, and when run with pahse current which rise up to ~120A and will cause gate driver unit damage.

 And especially we find the pin of GL and SL will have big oscillation and will extend +/-1V, and this may the reason of damage the gate driver.

Could you help to check if this gate driver unit which can be used with big current application or do you have any design examples with big current? Thanks a lot.

Hello,

What is the IDRIVE setting being used? I just checked the MOSFET being used and the Qgd is extremely low. You mentioned seeing oscillation on GL and SL, but did you see any oscillation on GHx and SHx at the same time? Because of the low Qgd on the MOSFET the IDRIVE setting selected could have caused this ringing severe enough to cause a shoot through event.

I should also mention that using a working voltage of 72V is also quite high for this device and doesn't leave a lot of room before the abs max.

Please provide:

1. Waveforms for GHX to SHx, SHx to GND, GLx to SLx, and SLx to GND 

2. What IDRIVE setting is being used

Regards,

Yara

2. Idrive setting is configurated with Vl2(18kΩ) and current is 100/200mA.

Waveforms as below:

Hello,

What point are you measuring at? near the driver or the MOSFET?

Can you try to lower the IDRIVE and setting to the lowest possible setting to see if this improves the ringing you are seeing?

Regards,

Yara

Hello,

GND means measure close to the gate driver, PGND means measure close to the MOSFET.

Yes, we also try to set the lowest Idrive current, but when run with high phase current still have ringing and also damage DRV8350RH.

Do you have any application with high phase current or example?

Hello,

I think the MOSFET being used might have a Qgd that is too low.

This FAQ goes in depth on how Qgd of the MOSFET relates to the IDRIVE setting that should be selected:

https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/796378/faq-selecting-the-best-idrive-setting-and-why-this-is-essential

From my calculations, even the lowest IDRIVE setting may be too high for the MOSFET.

Have you ever tested using a different MOSFET?

Are you able to share the schematic for a review?

Regards,

Yara

Hello,

We didn't tested with different MOSFET. 

The below is the schematic of DRV8350RH. As for the setting of Idrive, we already adjusted.

Hello,

I checked out the schematic and it looks good but here are a few note:

1. The BST to SW cap should be 0.01uF not 2.2nF, Is there a reason for this?

2. Is there a pull-up resistor for SDO?

Regards,

Yara

Hello,

1. The BST to SW cap should be 0.01uF, it's a mistake.

2.We use DRV8350RH, there is no pin of SDO.

Any other suggestions about current design? Because GDU is easily damaged with high current. 

Hello,

So was the previous waveforms captured while using a 2.2nF or 0.01uF? 

I'll be meeting with my team tomorrow to discuss your issue, I will provide an update then.

Regards,

Yara

Hello,

Actually this DC-DC function we didn't use and we removed all these components.

From my current testing, I think mainly damage is due to the oscillation of SL-GND, below picture channel 7 is SL-GND.  Could you help to check? Thanks.

Hello,

I think the issue really is the Qgd of the MOSFET, increasing Qgd would help but not completely eliminate the issue. I think a layout review should be done since layout can heavily impact the performance of such a high power system. 

Regards,

Yara

Hello,

What value of increasing Qgd do you recommend?

Below picture is the waveform of SL-GND before damage.

Hello,

It would be ideal to have your Qgd at a minimum be able to handle the minimum setting of our IDRIVE settings.

From my understanding with information being handled via email, three MOSFETs are being used in parallel? This changes things just a little bit since that does increase the overall Qgd.

The Qgd stated in the datasheet is 4 nC, if three MOSFETs are in parallel then that would be the total Qgd is actually 12nC, if that is the case then 12nC adding an additional 3nC in total to the Qgd would allow you to use the lowest IDRIVE setting in the datasheet.

However, this is probably not going to completely resolve the severe ringing that is being seen considering the this is a high current application and changing the direction of this current could be causing transients that are violating the abs max of this device.

Regards,

Yara

Hello,

Yes,  three MOSFETs are being used in parallel.

Actually with the lowest IDRIVE setting cannot generate the PWM waveforms, so we set Vl2(18kΩ) and current is 100/200mA.

So do you have any suggestions about the ringing?

Hello,

The best course of action right now would be to do a layout review to ensure the layout is as optimal as it could be for this high curr4ent application. My understanding is layout may have already been shared via email with someone else on the BLDC team?

Regards,

Yara

Hello,

Could you help to give your email address? And can keep you in loop.