36 GHz sensor - use VCO F/2 output for Lo drive of I/Q mixer

For this particular application I have a VCO which outputs Fo and Fo/2 and a I/Q demodulator which has an active x2 multiplier on the LO line.

Normally, if I had the I/Q demod with no multiplier, I'd have the VCO output going through a coupler to the antenna and the coupled port I would amplify and input as the LO drive, so the downconverted IF would be at DC...

Would it be possible to get rid of the coupler and have the VCO Fo/2 output amplified and connected to the x2 multiplier of the I/Q downconverter to give me the same IF at DC?

What would be some pitfalls of this strategy?

Thanks!

The potential pitfalls revolve around the phases getting screwed up, and the counter waveform not being good for the mixer LO. But since you are only dividing by 2 and then multiplying by 2, you luck out. Even if the counter slips an input cycle, the output phase X2 is still in synch, so you do not get a big phase rotation in the I/Q plane. (if you were dividing by 3, and then multiplying by 3, you would not be so lucky)

The output from the divide by two can have some odd harmonics in it. You might want to experiment around with the divide by 2 driving the LO port, and then adding a 18 Ghz lowpass before the multiplier and seeing if there is any big change in the I/Q outputs. You also might want to try a 31 GHz highpass right at the LO port, to see if any 16 Ghz feedthru is screwing up the I/Q outputs.

Of course, the cost of adding a 32 Ghz active multipler will be significant.

Rich

Thanks Rich I appreciate it...

I'll post follow-ups if I have questions.

Best regards

BGW, just curious, who's chips are you using?

I'm looking to build a 24 and 36 GHz FMCW-type sensor system based on Hittite chips and microstrip patch antennas.

Do you have any experience and/or recommendations with these frequencies?

Initially I thought of the architecture as a VCO with PLL that outputs Fo into a coupler; the through would go to a circulator before the antenna (either amplified before the circulator or not), the coupled port would be amplified and used as the LO for the mixer/down-converter. The third circulator port would be the output to mixer (maybe with a power detector hanging off of it? before the mixer) which would be amplified with a LNA.

Now I'm thinking of using the Fo/2 VCO output as a LO drive (which is x2 in the chip I'm looking at), and hence eliminating the coupler.

Thoughts?

Thanks for all the suggestions and help btw.