Can i do 2.5 to 5GHz frequency doubling using full-wave rectifier?

[Terminator3]

Is it good idea? What should i do else? Short-circuit stubs at f0 on output and 2f0 at input to filter out?

Thanks!

 

[raiya_23]

I think its perfectly possible theoretically

 

[jiripolivka]

I think its perfectly possible theoretically

There are many commercial doublers using this principle. Check those by MiniCircuits. Such doublers need typically +17...+20 dBm to saturate the diode pair or quad, output power can be +4...+8 dBm over wider bnd.

 

[biff44]

a full wave rectifier will double the frequency. And if balanced, it will not leak out too much fundamental frequency (a good thing if you need a broad frequency range of coverage). But they are lossy...you should expect at least 8 dB of conversion loss.

You can get maybe 1 dB of conversion loss if you use a properly designed varactor X2. And active multiplier can have conversion gain, but watch what it does to the noise floor!

 

[jiripolivka]

a full wave rectifier will double the frequency. And if balanced, it will not leak out too much fundamental frequency (a good thing if you need a broad frequency range of coverage). But they are lossy...you should expect at least 8 dB of conversion loss.

You can get maybe 1 dB of conversion loss if you use a properly designed varactor X2. And active multiplier can have conversion gain, but watch what it does to the noise floor!

I have designed varactor multipliers long ago. They usually work at one frequency only, with a limited bandwidth. They require a high driving input and are very sensitive to any mismatch. I would not recommend anyone to try designing a good wideband doubler with a varactor.

Passive diode doublers only require good input/output match and a good drive power. Then use a band-pass filter and amplifier to get enough output power.