Best core material for QCW Buck modulator ?

[Simranjit]

Hi everone,
In the link below T300-2D is used for buck modulator but there is also recommendation for kool-mu.
https://www.loneoceans.com/labs/qcw/

Digi-key (kool-mu core with similar dimensions as listed on above website):
https://www.digikey.ca/en/products/detail/magnetics-a-division-of-spang-co/0077908A7/18626637

I'm guessing the main concern is able to handle peak currents. I'm also trying to future proof the buck modulator so that I don't have to build it again if I use more powerful IGBT modules in future.
Digi-key has kool-mu cores. Any other suggestions or feedback ?

Thank you.

[flyingperson23]

Yes, the main concern is getting an inductor that can handle enough current at a suitable inductance

[Anders Mikkelsen]

For a QCW buck modulator I would look at a SiFe type powder material, in permeability 40 - 60 range. These materials have better inductance retention under bias compared to Sendust (FeSiAl) type materials. Losses are typically higher (though not for some of the newest generation materials), and cost is similar or lower. Every manufacturer has their own name for these formulations, from Mag Inc this would be XFlux.

https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078912A7/18626799
https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078617A7/18626674
https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078072A7/18626830

For example, a pair of the 0078912A7 wound with 32 turns would give you 116 µH at zero bias and just above 40 µH at 150 A. Even 30 turns on a 0078617A7 might work well enough up to 150 A, though core losses might be excessive depending on buck switching frequency and input voltage.

T

[Simranjit]

For a QCW buck modulator I would look at a SiFe type powder material, in permeability 40 - 60 range. These materials have better inductance retention under bias compared to Sendust (FeSiAl) type materials. Losses are typically higher (though not for some of the newest generation materials), and cost is similar or lower. Every manufacturer has their own name for these formulations, from Mag Inc this would be XFlux.

https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078912A7/18626799
https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078617A7/18626674
https://www.digikey.com/en/products/detail/magnetics-a-division-of-spang-co/0078072A7/18626830

For example, a pair of the 0078912A7 wound with 32 turns would give you 116 µH at zero bias and just above 40 µH at 150 A. Even 30 turns on a 0078617A7 might work well enough up to 150 A, though core losses might be excessive depending on buck switching frequency and input voltage.

T

Thank you very much for your reply. I also saw your other post. Now I at least know start, what to look for and how to do calculations. I really appreciate it.