DRV8353R: DRV8353RS Startup/SPI Issues

Jack,

This is impressive for a high school project! I wish I knew anything about PCB design or MCUs in high school.

Have you checked the power supply voltages at the pins of the DRV when everything is running?

Is the nFAULT low (indicating a fault such as undervoltage or similar)?

When you mention the device "won't enable", please explain what you mean. Do you see any increase in supply current when you pull enable high? Have you checked the ENABLE signal at the DRV pin to make sure it is coming up?

Regards,

-Adam

When I output to EN_GATE, with 3.3V (checked), nFault goes high. Supply voltage is 48V, 54.8v max, from a battery.(More in attached pdf) Output of integrated buck converter is ~4.8v. (Works as intended). 

When I say "wont enable", I mean that the SPI will not respond, or work in any meaningful way, and outputs to the gate driver for mosfet control, from the microcontroller, are rejected by the chip (no output, bootstrap not working, no actual driving functions are working as intended).

I have not checked input supply current, but I am confident there is some due to the integrated buck converter, and nFault LED being on.

Resetting EN_GATE momentarily turns off the nFault LED, but it swiftly turns back on. (Again, can't check errors with SPI).

Thanks,

Jack

Jack,

"Bootstrap not working": Are you talking about the charge pump of the DRV? Or the BST of the built-in buck?

Please check the supply current. If the supply current stays the same when you flip ENABLE OFF or ON then the chip is not enabling properly.

On my EVM for example, I see an increase of ~20mA when I enable the DRV.

The SPI is harder for me to debug. The circuity looks correct but I'm not sure of the implementation in code and on the teensy.

Regards,

-Adam

Sorry for being unclear there. There is no voltage across any of the charge pump capacitors on the DRV. (More evidence that I believe the driver is not properly enabling).

Here is the code for the Teesny. Lines 96-155 involve the SPI communications. github.com/.../smv_BE_2018-2019.ino

I can test the actual boards supply current sometime next week. That might be a little difficult with the PCB setup I have, I really didn't make it easy to debug or probe for myself.

In the meantime, I would assume the DRV is not properly enabling. Would replacing the chip be a valid solution for this? This is one of the few boards that I haven't hand soldered, and used a professional reflow oven at a local company. I don't imagine that being the problem, but is that a likely possibility? Is it just that the chip is bad?

Sorry for not being ready to test the board, I expected a much longer response time than I got. (Not a bad thing, for sure!)

Thanks again,
Jack

Jack,

I cannot debug the code unfortunately. If you suspect the chip is damaged, please try another board or swap the chip.

The supply current will tell us a lot once you can check it. Even monitoring the main input supply current would be helpful.

Regards,

-Adam

Hi Adam,

Sorry for the delay. I tested the supply current, and it goes to 270 mA on startup for 50 ns, and then goes to 50mA for a ~200 ns longer amount of time, before going back to zero. I removed main input capacitors and probe the main line, with a supply input of 12v(Don't have 48V power supply).

Edit: here is the scope from another attempt, 1mV/10mA probe. (this was on the enabling of ENGATE)

Thanks,
Jack

Jack,

Can you summarize for me, what is the steady state current before you trigger the ENABLE of the DRV and after you trigger the ENABLE?

Regards,

-Adam

Sorry. Before ENABLE is triggered, it hangs around 10mA, and after it jumps around(image of scope above) it drops back down to around 10mA. Inductor was removed, and 5V power was supplied from USB for the microcontroller.

Thanks,

Jack

Jack,

Are you checking for a fault? if the current is increasing and then dying back down to the 10mA level after ENABLE is triggered, the part may be going into fault condition.

Did you ever try this on a second device?

Regards,

-Adam

Yes, it is high on nFault. It has never stopped. I have not tried the second chip yet, I still have to figure out how to hand solder VQFN-48.

Thanks,
Jack

Hi Adam,

Today I replaced the driver. The issue persists, and is actually somehow worse. The driver now no longer gives a 5v output for buck, and shows no functionality at all. I'm not sure what exactly the problem is here. It must be somewhere else on the board, I just don't exactly know where.

Thanks,
Jack

Jack,

Did you figure anything out here? Sorry for the delay.

Regards,

-Adam

Hi Adam,

No worries. I haven't done much on it since then, I have had other projects to do. I'm honestly not even sure how to troubleshoot this issue, as it is inconsistent. 

Jack