I'm working with an DC power supply feeding an an H-bridge topology with some gate drivers. The output of the IGBTs are going to a transformer to step up the voltage to 250V and then goes to a rectifier bank and smoothing cap to get my VDC value. I was probing around to see what needed to be done for my current feedback circuit I'm wanting to implement, when I noticed these massive amperage spikes(Kiloamp range) on the main DC power supply.



The spikes line up with the Hi/LO pulses given to the gate driver so my first thought was I had some overlap in my IGBT firing. I did some looking and the Hi/LO firing aren't over lapping but the massive spikes are still in line with some of the pulses. Zooming to the spikes, the width of the spikes are on average of about 1ns in width. Just to be on the safe side, I adjusted my time delays thinking maybe its my tail currents getting me, but its the same thing.

Right now I'm digging into maybe the transformer is wrong. But besides that I'm kinda at a loss.

Thank you,

double

 

Leads back to my question on your previous thread about why you need a H bridge fo drive a transormer when your system block diagram indicates your system is all DC?

An H Bridge at the power levels you are trying to control may be fundamentally the wrong approach unless you need AC as they don't come without problems, especially with inductive loads such as a transformer.

Without even looking futher at your design. My advice would be to start by posting a question to the board at a block level to get some community advice for your whole approach i can't help but feel you may be starting on the wrong foot

 

Thank you for your advice Marc but I'm already committed to working this project through. Sometimes its good to learn via trail and error and failure.
But right now I'm focus'd on trying to find out as to why I have this issue.


double

 

Thank you for your advice Marc but I'm already committed to working this project through. Sometimes its good to learn via trail and error and failure.
But right now I'm focus'd on trying to find out as to why I have this issue.


double

Fair enough in that case this thread on inverters may help you https://forum.allaboutcircuits.com/threads/proteus-filter-design.169353/page-2#post-1507727

If you should change your mind here's a concept for your consideration.

Operation at 200khz approx 85% Duty Cycle with 48V input will give you 350+V and upto 800+Watt.

You may also be able to use a COTS inductor https://www.mouser.co.uk/datasheet/2/445/7443763540470-1721922.pdf

If you made the diode a synchronous rectifier you'll get better efficiencies than a diode at those currents.

Best of luck

 

Well that is considerable less components for the same output. My next project just got a whole lot simpler. But that thread did explain an a lot of issues. The IGBT module showed that there's a diode in parallel internally, there wasn't. That cleared up a massive amount of stray charge.
I'm going to look into the COTS and synchronous rectifier.

I appreciate it very much, thank you


double

 

Well that is considerable less components for the same output. My next project just got a whole lot simpler. But that thread did explain an a lot of issues. The IGBT module showed that there's a diode in parallel internally, there wasn't. That cleared up a massive amount of stray charge.
I'm going to look into the COTS and synchronous rectifier.

I appreciate it very much, thank you


double

Your welcome.