Dear Katik Sharma,

I'm curious...

You're basically asking "How can I do an invention" and in case I'd knew, I'd apply for a patent before sharing the answer publicly.

Depending on power levels, there are several topologies well established to act as DC-DC-Converters.

So what is your application?

DC-DC-converter for EV-Charging with 1000V/500A output?

DC-DC-converter for USB-devices 5V/2A?

DC-DC-Converter in a MPP-Tracker in windmill application, channeling a few Megawatts?

Super-precise DC-DC-converter to generate a reference voltage in measurement equipment?

If you know - then think about requirements and whether you consider it possible to outperform an established solution on at least one major parameter like efficiency, accuracy, cost, cooling, size, reliability...

And only if you suspect you can - go ahead and try finding a better solution than the ones available.

Cheers - Martin

Thank you Martin sir for your response. Actually I am asking about the process or the method by which I can learn how to design a new topology? How the researchers proceed when they are at the initial stage of such problems?

Dear Kartik Sharma,

in case you find a new topology or an idea how things could work, two options arise after developing a solid plan first:

1) Set it up, turn it on, see if it works as predicted. Typically, in this approach you learn a lot but also destroy some parts in the process.

Gaining hardware experience comes with this sort of expanses :-)

2) Grab a simulation tool, try your best on software and once you reached a point you're confident with, return to step 1).

This can help to eliminate the most basic mistakes but is not a warrantee for success.

Keep in mind that there is the BIBO-principle for simulations:

Bullshit in - bullshit out :-)

A well made simulation will help a lot sometimes.

Sometimes, a misinterpreted simulation is worse than a few burnt components...

Cheers - Martin