DRV8350R: Question of GHx signal Quailty

Hey Johnson,

1. The VGS (GHx) waveform of the high-side MOSFET appears to be a square wave, but the positive edge section shows a slight slew rate. However, the VGS (GLx) of the low-side MOSFET is a perfect square wave with no slew rate. We would like to know if it is normal for only the high-side MOSFET’s VGS (GHx) waveform to have a slight slew rate in the positive edge section, and why this slew region occurs.

A - Both High side and low side VGS voltage will have a slew during turn on and turn off. Is highside being measured from GHx-SHx and low side from GLx-GND? The time scale might be affecting how much the slew is being visible. The gate current will determine how fast the MOSFET slews up.

What is the rise/fall time for VDS? 

2. When zooming in on the positive edge section of the high-side MOSFET’s VGS (GHx) waveform (which has the slew rate), we observe a small spark-like noise.
We would like to know possible methods to eliminate this phenomenon, the reasons why it occurs, and what side effects it may cause if it persists. (IDRIVE PIN : SHORT TO GND (Minimum setting))

A - This looks like noise around the miller region of the MOSFET. This is the region where the MOSFET VDS is slewing. Can you take waveforms of GHx-GND, SHx-GND, and VGS for a rise and fall? There might be switching noise on SHX, which could be coupling into the VGS voltage.

I am curious, what is the battery voltage and the application for this project?

Best,
Akshay

Dear Akshay

1. I measured VGS as GHX-SHX and GLX-GND.

but, I don’t quite understand the meaning of the phrase: "The time scale might be affecting how much the slew is being visible."
Could you please explain it in simpler terms?

2. I will let you know soon regarding waveforms of GHx-GND, SHx-GND, and VGS for a rise and fall.

3. battery voltage : 48V 

Application is  Actuator (motor) in Robot (upper 7axis)

Hey Johnson,

1) If the waveform was too zoomed out the small time for falling might not show the slew. Also is this a simulation?
I  would like to see the waveforms from the actual hardware if possible.

2) Sounds good, thanks.

3) Thanks for the background.

Best,

Akshay

Dear. Akshay,

1. This waveform is a hand-drawn representation of the actual waveform. Due to the firewall, I am unable to attach the real captured waveform. I tried to draw it as closely as possible, and I kindly ask for your understanding on this.
   Our control frequency is about 20 kHz, and we are performing current control.
   We measured GHx–SHx and GLx–GND on the oscilloscope with the same time scale, and only GHx–SHx shows the slew rate, similar to the waveform I previously shared.
   GLx–GND, on the other hand, is almost a perfect square wave.

2. Please refer to the attached waveform diagram.

3. I also have an additional question.
   Our input voltage is 48 V, and when we drive the motor and measure as below:

• The GHx–GND voltage level of waveform shows about 60 V,

• The SHx–GND voltage level of waveform shows about 51 V,

• The GHx–SHx voltage level of waveform shows about 11 V.

It seems the reason each waveform has a different voltage level is because the reference points are different. However, since the input voltage is 48 V, I expected GHx–GND and SHx–GND to both measure around 48 V. I would like to understand why they are measured in this way.

< GHx - SHx Details>

Hey Johnson,

I will review and aim to provide feedback tomorrow.

Thanks,

AKshay