The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Mixer NF confusion
https://designers-guide.org/forum/YaBB.pl?num=1189208706

Message started by liu6747 on Sep 7^th, 2007, 4:45pm

Title: Mixer NF confusion
Post by liu6747 on Sep 7^th, 2007, 4:45pm

Hi,
I am designing a VHF band mixer, and I am confused about the result:
The apps eng use following eq to calculate the output noise for each stage:

sqrt(K*T*R_source) * voltage_gain_linear * sqrt(NF-1)

then rms w/ the product of previous stage's output noise and current stage gain to get overall noise
it makes sense when input source marchs Zin of mixer, because source noise sqrt(4KTR), V/sartHz, will make sqrt(KTR) at input of the mixer, but my mixer input is HiZ, then when spectre report NF (pnoise + pss), it is "marched NF" or not? in other word, when spectre calculate NF form output noise contributor %, it set noise voltage at mixer input to sqrt(4KTR) or sqrt(KTR)?

If it use sqrt(4KTR) at input, I guess it does because that is what the circuit is really doing then above eq to calculate output noise is not valid anymore?

Thanks
Jun

Title: Re: Mixer NF confusion
Post by Bill Toole on Sep 10^th, 2007, 4:35am

NF calculations get a bit tricky when you start considering mismatches. You will need to know what source impedance was used for determining NF of the Device Under Test (usually 50 Ohms but not always) and you need to know unloaded voltage gain, input and output impedences (or available power gain). I would recommend looking at section 2.3.2, pages 41 - 45 of Razavi's book "RF Microelectronics" He has a very good treatment of noise figure with arbitrary input and output impedences.

I don't mean to confuse the issue any more but with mixers you also have to consider if the NF was "Single-sideband" (SSB) or "Double-sideband" (DSB). Again Razavi has a good treatement in Sec 6.2 pages 183-185 and pages 200-203

Bill