Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Cree MOSFET gate driver card wierd waveforms

Status
Not open for further replies.

Sidd59

Newbie level 3
Joined
Feb 7, 2020
Messages
3
Helped
1
Reputation
2
Reaction score
1
Trophy points
3
Activity points
38
I am testing a Commercial SiC gate driver CGD15SG00D2. We will find the data sheet here:

The driver is designed for SiC Mosfets and is able to switch on and off the FETs with +15V and -3.3V respectively.

To test this gate driver card , I am sending PWM of 3.3V from a micro-controller and measuring the output of the gate-driver ( gate-source pins) with a differential probe. Additionally, I am powering up the gate driver with 12V laboratory power supply.
My test-setup looks like this:
testsetup.jpg

The waveform of the gate-source pin when no when no MOSFET is connected looks like this: As you can see there is huge positive and negative spikes during turn-on and off.

scope_21.png

I also tried to connect the corresponding terminals to the gate and source pin of single isolated SiC Fet, just to charge and discharge the gate capacitor to see if the spikes goes away. However only upper peak was reduced by one Volt but negative spike remained the same.
What am I missing? I am doing something stupid during the measurement? I twisted the cable of the differential probe as well with no effect.,

Thanks in advance and happy weekend!

best,
Sidd
 

You didn't tell what the differential probe is and how you have connected it. It's easy to get huge overshoot with inappropriate probing.

In addition, it's not suitable to drive SiC-MOS through 20 cm of twisted pair. The overshoot will be so large that you may achieve partial gate breakdown of Cree transistors, showing up by increased gate leakage current. Follow Cree layout suggestions for gate drive circuit.
 
You didn't tell what the differential probe is and how you have connected it. It's easy to get huge overshoot with inappropriate probing.

In addition, it's not suitable to drive SiC-MOS through 20 cm of twisted pair. The overshoot will be so large that you may achieve partial gate breakdown of Cree transistors, showing up by increased gate leakage current. Follow Cree layout suggestions for gate drive circuit.


Thank you for your reply. I am using this probe: Testec TT-SI 9002
https://www.batronix.com/files/Testec/TT-SI-9001_Manual_EN.pdf

I connected them to the gate source terminal of the driver card. Yes connecting MOSFET like that was a bad idea. But the spikes are there when there is no connection.

Can you please elucidate what you meant by the wrong probing? I believe my probing setup is at fault. 100kHz or more switching frequency may cause the probe reading to go heywire. Furthermore false or incorrect waveforms further complicate the design process :-(

What kind of measuring instrument should I buy to get better measurement results? Since people hardly talk about how they test SiC or GaN devices. For me it is imperative to correctly read gate source and drain source voltage.
Thank you in advance!


Best,
Sidd
 

I don't expect to read output of a fast gate driver correctly with a 25 MHz probe. It's too slow and the probe cables are causing signal reflections and ringing with low impedance sources.
 
  • Like
Reactions: Sidd59

    Sidd59

    Points: 2
    Helpful Answer Positive Rating
you need at least 150MHz probes with very short exposed probe tip <10mm and direct earth connection to probe braid also < 10mm

else you will get very misleading results.

You can simulate the gate with 1 ohm and 1nF - as FvM says long leads from the gate driver will alter performance and overshoot ...
 
  • Like
Reactions: Sidd59

    Sidd59

    Points: 2
    Helpful Answer Positive Rating
Hi,

In addition the the posts above:

A 25MHz probe is useful for 25MHz sinewave, not 25MHz square wave.

An ideal square wave includes all odd numbers of overtones up to infinite frequency.
If you are interested in "almost clean" readings of 25MHz square wave then a 150MHz probe is the lower limit ....

Klaus
 
  • Like
Reactions: Sidd59

    Sidd59

    Points: 2
    Helpful Answer Positive Rating
Thank you for your reply. I now measured with a new passive probe with a pig tail ground loop and waveforms now looks good. I will buy a bumblebee PMK for future measurements. Optical ones are out of my budget :)
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top