if i simulate the lna and mixer separately using Pnoise analysis, but the cascade NF is not simple from Friis equation, which is because of noise folding effects from sidebands. so, how to calculate cascade NF with a mixer?

thank you,pancho_hideboo

i calculate the NF(F) as follows:

first, i simulate LNA with matched input and load of mixer input impedance,  the ouput noise at mixer input is calculate by F and GT of phased noise analysis.
second,  the mixer is simulated with 50ohm port and 50ohm matching resistor without noise. so the output noise at mixer output is calculate in the same way as LNA
last,  F is calculate at mixer output as
F=(Noise from LNA and input 50ohm port + Noise from mixer)/(KTB*RS*GT^2)
where B=1HZ, RS=50Ohm GT is the transducer gain from LNA input to Mixer output.  LNA and Mixer are on-chip connected. Mixer output is also on-chip connected with next stage.

what wrong with above method?
the cacade F is smaller than that of direct combination of LNA and Mixer. but if i cascade two LNA. the cascade F of two LNA is well match with the result of  direct simulaiton.

mydreamhouse wrote on May 9^th, 2009, 1:03am:
What do you mean by "matched input impedance" ?
What is your definition of Gain for LNA ?

mydreamhouse wrote on May 9^th, 2009, 1:03am:
mydreamhouse wrote on May 9^th, 2009, 1:04am:
What is "phased noise" ?
Even if you mean "phase noise", why do you need "phase noise" in evaluation of output noise for LNA ?

mydreamhouse wrote on May 9^th, 2009, 1:03am:
What do you mean by "50ohm matching resistor" ?

mydreamhouse wrote on May 9^th, 2009, 1:03am:
How do you calculate "Noise from LNA and input 50ohm port" ?
Show me equations you used.

How do you calculate "Noise from mixer" ?
What values of Gain for "RF->IF" and "Image->IF" are ?, here I assume spurious receiving point is only primary image.
Show me equations you used.

Your denominator is for a definition of conventional SSB_NF.
Do you understand SSB_NF, DSB_NF and ASB_NF correctly ?
See http://www.designers-guide.org/Forum/YaBB.pl?num=1239925339

mydreamhouse wrote on May 9^th, 2009, 1:03am:
Nothing is correct.

mydreamhouse wrote on May 9^th, 2009, 1:03am:
First what NF do you get for "LNA+Mixer" in simulation, SSB_NF or DSB_NF ?

Do you understand various voltage gain definitions correctly.
  Conventional Voltage Gain(=Vout/Vin).  Gain by Direct Plot of PSP is this definition.
  Gain definition of Direct Plot is same for PAC, PXF and Pnoise, which is not Vout/Vin.
  S21 is another Voltage Gain.
  Sqrt(Transducer Gain)

See http://www.designers-guide.org/Forum/YaBB.pl?num=1237656418
See http://www.designers-guide.org/Forum/YaBB.pl?num=1215827287/3#3

Again, do you understand SSB_NF, DSB_NF and ASB_NF correctly ?

mydreamhouse wrote on May 9^th, 2009, 4:51am:
"Transfer Function" in "Direct Plot Form for Pnoise of Cadence Spectre" is not "Transducer Gain".

Do you understand "Transducer Gain" correctly ?
And you can't define SSB_NF for LNA.

However these are not so serious misunderstanding.

mydreamhouse wrote on May 9^th, 2009, 4:51am:
Again, nothing is correct very completely.

mydreamhouse wrote on May 9^th, 2009, 1:03am:
Again, do you understand "Transducer Gain" correctly ?

Generally GT_total is not equal to a simple multiplication of GT₁ and GT₂.
Here I define like followings.
  GT₁ is a transducer gain of LNA with Z_S1 and Z_L1 as source and load impedances.
  GT₂ is a transducer gain of Mixer with Z_S2 and Z_L2 as source and load impedances.
  GT_total is a transducer gain of LNA+Mixer with Z_S and Z_L as source and load impedances.

If the following conditions are satisfied,  GT_total will be equal to a simple multiplication of GT₁ and GT₂.
     Z_S=Z_S1,      Z_L=Z_L2
     Z_L1=Z_in2,      Z_S2=Z_out1
Here Z_in2 is an input impedance of Mixer, Z_out1 is an output impedance of LNA.

Do you satisfy this condition ?
However this is not so serious misunderstanding.

Your most serious misunderstanding are calculations of "Noise from LNA and input 50ohm port" and "Noise from mixer".

(1) Do you understand "Spurious Receiving of Mixer" ?

(2) What values of Gain for "RF->IF" and "Image->IF"  are ?
   Here I assume spurious receiving point is only primary image.

(3) Does your mixer have image rejection ability ?

Did you surely and correctly evaluate your Mixer by using PSS/PXF Analysis of Cadence Spectre ?
http://www.designers-guide.org/Forum/YaBB.pl?num=1236016393

All incomplete knowledges are always useless.

I'm not kind and patient enough for correcting all your misunderstandings which are too many.
Other people who are very kind and patient might help you.