How can PC PSUs use such small transformers and have a BIG output?

[tom66]

Re: How can PC PSUs use such small transformers and have a BIG output?

No switching regulators work at 1 GHz.

Two problems with such a high frequency:
- Very high gate charge losses in MOSFETs and turn-off losses in BJTs. This can be observed by the fact that RF transmitters have poor efficiency. My 100mW video transmitter, for example, eats close to 400mW and does get pretty hot.
- Inductors of such high frequencies would self-resonate and have very high core losses. Efficiency would be poor.
- Designing a discrete driver circuit for 1 GHz+? Forget it. Must be ASIC. That means the MOSFET must be on the ASIC. And MOSFETs on dies have high on resistance (hundreds of milliohms) and high gate charge - low efficiency.

Bottom line is you would probably have a bigger heatsink on it than other components.

The highest I've seen is the LM2734Z, which operates at 3 MHz.

[PCBONEZ]

Re: How can PC PSUs use such small transformers and have a BIG output?

Oh, so the 2.4Ghz in my microwave oven isn't regulated at all.
Nor is the 5Ghz in my older WiFi.
Nor is the 6Ghz in the 1960's [or older] Radar unit I worked on in ET school.
- It's just NOT NEW.

Who said anything about using MOSFETs for the regulators?

Do you honestly think something that is designed for the temp extremes of outer space, to work in a vacuum and sustain a nuclear blast is going to look remotely like what you can buy at Digikey?
.