Actions
Engagement
  • Author Author: rscasny
  • Date Created: Date Created
  • Last Updated Last Updated:
  • Views 462 views
  • Likes 0 likes
  • Comments 30 comments
Related
Recommended

Why Haven't You Applied for RoadTest: Gate Driver EVM with Truly Differential Inputs?

I've been running this roadtest for about 2 weeks: Infineon Gate Driver with Truly Differential Inputs: Infineon Gate Driver with Truly Differential Input

 

Right now, I have 3 applicants, with 10  kits given to me for the RoadTest.

 

I'm trying to determine why there hasn't been more interest. Is it the messaging on the landing page that's confusing? Not interesting? You're too busy? Too complicated? No Test Equipment?

 

I'll agree this is not your typical maker dev board. The product being evaluated would be in a gate driver IC -- part of a larger design such as DC-DC converter, an inverter, or other power supply. It would be ideal for someone into power supply engineering.

 

A little bit about the gate driver IC: It's a new family of single-channel non-isolated gate-driver ICs from Infineon. Due to the unique fully differential  input circuitry with excellent common-mode rejection, the logic driver state is exclusively controlled by the voltage difference between the two inputs, completely independent of the driver’s reference (ground) potential. This eliminates the risk for wrong triggering and thus is a significant benefit in all applications exhibiting voltage differences between driver and controller ground, a problem typical for systems with 4-pin packages (Kelvin Source connection), high parasitic PCB inductances (long distances, single-layer PCB) or a bipolar gate drive. https://www.infineon.com/dgdl/Infineon-1EDN8550B-DS-v02_00-EN.pdf?fileId=5546d46262b31d2e01635da0c1c02652

 

So, could you take my poll and give me an idea why this roadtest is not enticing you to apply?

 

Thanks.

 

Randall Scasny

RoadTest Program Manager

  • I am really interested but I am already involved in two complex road tests (at least to do what I have in mind). Before the end of September it will not possible for me to start on this stuff so I resigned.

     

    Enrico

  • in reply to Jan Cumps

    It makes life easier if I'm selected but kind of makes my application lame...

    The inductor has fold-under 'J' leads so it would be reasonably easy to get it off the board, even using a soldering iron. That would leave you with a half bridge which you could do other things with if you wanted something more involved.

  • in reply to Jan Cumps

    ... but I may do a better soldering job than this image

     

    image

  • in reply to jc2048

      wrote:

     

    Their application note (see 1.2.4) includes installing an alternative low-side driver on the spare placement underneath. I think the idea is that you then knock off whichever gate resistors don't apply, though I wasn't entirely sure just skim-reading it.

     

     

    ...

    Yes, my idea is not unique after all - I thought I had this unique angle image. I checked the application note to see if the design could be modified and how they disturb the ground level - didn't read that part.

    It makes life easier if I'm selected but kind of makes my application lame...

  • There have been many other designs fit for projects. It will take more time and test shifting to new products.

    One only applies for Roadtest  that may used in further projects rather than research.

  • in reply to Jan Cumps

    Their application note (see 1.2.4) includes installing an alternative low-side driver on the spare placement underneath. I think the idea is that you then knock off whichever gate resistors don't apply, though I wasn't entirely sure just skim-reading it.

     

    It might be tactful to use the alternative driver they suggest there (one of their own) since they are sponsoring the roadtest.

     

    It will be fun to see if you can manage to do that testing without destroying the MOSFETs (you should be able to if you think about what it is that causes the damage, how it comes about, and then take some precautions).

  • in reply to Jan Cumps

    I submitted an entry.

     

    Here's what I wrote as test procedure:

    (d) What is your testing procedure (Be as specific as you can)?

     

    I will first review the behaviour of the test board with the on-board gate driver, with focus on the improved ground bounce handling that it promises to be good at in the documentation.

    I plan to use the test approach of the board's application note for that.

    I will then replace the low-side driver with a more traditional one and validate how the design is handling the DC offsets and AC distortion, using the same test procedure.

    I will look for a generic gate driver part as replacement- if Infineon wants me to use a specific alternative, I am open to that.

  • in reply to jc2048

    I’m thinking about applying for this road test, although I’ve been delaying my enrollment.

    It is a power FET related topic, something that interests me. But I thought I’d hold off on road tests for a while and enjoy following others doing some. But there are not so many takers...

     

    One way to test this one would be to first test the setup with the driver-under-test, then replacing that driver by a more traditional driver and see the difference when the rail lifts away from ground...

  • in reply to Jan Cumps

    For the PiCasso Design Challenge, given the nature of the prizes and their cost I wouldn't feel very good about participating and then blatently disregarding the T&Cs that I'd signed up to if I were to win one of them. That's the reason why I haven't entered the competition and will make sure I don't qualify (it's 9 Slices of Pi and not ten, ie 9 blogs and not the required 10 to qualify).

     

    The RoadTests are more complicated because of the wide range in value of the 'prizes'. If I were lucky enough to get one of the more expensive pieces of test equipment to test then I would feel an obligation that I wouldn't necessarily feel with a Microbit worth a tenner, but where should the cut-off come? At what point is it acceptable to play fast-and-loose with the agreement that you sign up to?

     

    I have to confess, though, that I'm not totally consistent because I did enter a Project14 competition and I think that has basically the same T&Cs.