Blog Post Actions
Group Actions
Engagement
  • Author Author: Jan Cumps
  • Date Created: Date Created
  • Views 471 views
  • Likes 8 likes
  • Comments 6 comments
Related
Recommended

High Voltage GaN FETs - Part 1: Several 100 Volts

Jan Cumps

I'm a low power guy. And a low voltage one too. I don't have a power supply that delivers more than 33 V.

On the other hand, I'm interested in DC to DC conversion. I think this is an area in electronics where you can get a deep understanding of the electronics basics, for active and passive components.

If you fully understand a switching converter, you know voltage, current, inductors, capacitors, resistors, diodes and transistors (and varying wave forms).

That's about half of the "Art of Electronics" handbook.

image

The subject is un-hip in few people's minds. And that's OK. In my mind, it isn't. That's why I'm happy that I managed to lay my hands on a DC converter test motherboard.

I bet this test bed will help me to step up from low voltage DC conversion skills to high voltage conversion solutions.

 

High voltage is several 100 volts DC. And with several hundreds,  I mean so many volts that I had to buy another DMM to be able to do the measurements

 

Prologue

 

I'm following the Coursera "Power Electronics" course. For the fourth time now.

There's nothing wrong that that course. I think it's one of the best out there on the web. Grab it if you can.

But unlike other online university courses, I've never managed to pass that one. And I know why.

 

I like to blame my professional agenda. I have an unpredictable job. I go on business trips on short notice.

They always interfere with the course agenda.

But why did I pass the Coursera "Linear Circuits" course then?

That wasn't easy either. And I had my work interfering with that one too.

 

My personal analysis is: because the Power Electronics one requires good math skills,

and because the switching power converter designs require a more-than-in-depth understanding of electronics, beyond the ones we use in many other areas.

 

You can happily build a microcontroller based design if you know the basic electronics rules

and if you know that you need snubber diodes when powering a relais or motor.

That's about all the inductive knowledge you need.

 

Not so with DC to DC converters. And that's one of the reasons why I value them.

Power electronics is often looked down to as the stepchild of the business.

But the experts in that sector may be among the ones that have (and need) the best understanding of electronics - active and passive.

 

The Series

 

I'm inviting you to follow me in my investigation into high voltage switch-mode circuits.

 

image

 

 

I'm going to put the latest high voltage GaN FETs to the test. My story hook is that I'll validate this brand new technology.

In reality, it's an excuse for @self to get real with that "Power Electronics" course, and finally pass the exams.

 

Hooray for the  Power Electronics discipline

 

Part 1: Several 100 Volts
Part 2: Test Setup with LMG3410 Half-Bridge
Part 3: Probing the LMG3410 Half-Bridge
  • in reply to Jan Cumps

    If that test circuit was built using silicon MOSFETs, the upper FET would be operating simply as a commutating diode for the coil, using the intrinsic body diode to catch it just above the rail and with the transistor itself playing no part. GaN FETs don't have that body diode [which, apparently, is one of their strengths - no diode recovery time], so I wondered how that circuit worked to catch the coil. This application report explains it nicely:

     

    https://www.ti.com/lit/an/snoaa36/snoaa36.pdf

     

    The higher voltage than a silicon diode means you won't want to leave the coil current circulating for too long during the interval between the double pulses, as you'll drain the energy from the coil more quickly [and the current won't then resume at much the same level you interrupted it at].

     

    I find the symbol on the test circuit you've shown there quite confusing. Inside, at first glance, it looks like a cascode, with the bottom FET doing the switching and the top one shielding it from the high voltage. But, actually, it seems that the top FET is switching and the bottom FET is just there to disable it when the driver section isn't in a position to control it properly.

     

    On the datasheet I've just downloaded for the LMG3410R070 part, the symbol is drawn differently to the circuit you show, without the lower FET explicitly shown:

     

    image

  • I'm revisiting this GaN design. To see how they bias the upper half of the half-bridge, and to train my dc load.

    image

  • I'm ready to continue with the series.

    The GaN FETs that I was going to use in this exercise hadn't arrived @Cumps Lab yet (the mysteries of sales orders going into odd directions) .

     

    image

     

    I now have everything needed.

    • On the right: uncommitted motherboard without FETs. A (rather smart) configuration that lets you evaluate power transistors in several configurations.

    and arrived this morning, delivered by 's friends FedEX:

    • On top left: daughterboard with 2 GaN 600V FETs (technology preview material: LMG3410) in half-bridge configuration. This board plugs into the motherboard.
    • Left middle: 4 more fresh-from-the-FAB LMG3410s

     

    That's a decent lot of sparkling new technology to talk about.

    I have enrolled into the BK Precision electronic load RoadTest because I can very well use a programmable load in this series.

     

    No excuses anymore to procrastinate (except maybe that I still have to finish the SPI performance story - a blog series that I started because there was a delay in the FET delivery image )

  • Count me in.

     

    I like to see what you are doing and get an update on the latest technology available.

     

    DAB

  • Hi Jan,

    I too have been fascinated by DC to DC conversion and switching power supplies. It has been my dream to be able to actually fix a switching power supply that has failed. It hasn't happened yet but I haven't given up yet either. Every day I learn a little bit more and I am hoping to gain some new insights from your blog on this subject. Thank you for taking the time to include us in your exploration.

    John