Need help for the receiver front-end measurement, please.
I have a question about my receiver measurement. I have the receiver front-end (up to the based-band amplifiers) with differential output and I want to characterize it for gain, noise figure, IIP2, IIP3, P1dB. My question is what I should do to connect my circuit with the measuring equipment. Since the circuits have differential output from the based-band amplifier and the instruments have single-end input. I'm considering about:
1. Can I use a balun to convert to single-end output and using a high-impedance probe to measure it. (My based-band signal is 0-2MHz and I found a low frequency balun from MiniCircuits).
2. I saw some note suggesting about an instrument amplifier as a output buffer. Is there any pros and cons for this method? Is is a suitable way for other measurement such as noise figure, IIP2, IIP3?
3. Can I just directly measure at only one-end of the differential output by using an active probe connected with the spectrum analyzer? With this way, I am not sure how to calculate my results correctly. Are there any literature or guideline for this method?
Please give me some advice about this measurement. I am new to these measurements and want to learn an appropriate way that many people do. Thank you so much in advance for any comments.
I think You have answered yourself.
For IIP2, IIP3, the active prove or a more cheaper loop connectet to the SA is a good method.
This is because the IIP2 and IIP3 measurements need the knowing only the difference of the OUT signals and the knowing of the absolute Input power.
Even Pi1 may be measured in this way.
Of course, be sure to pick-up the signal coming from the last stage of the chain.
Po1 may be later calculated as Pi1+Gain.
The Gain measurement may require a balun . It depends on the model of VNA You'll use and calibration method.
The Noise Figure Measurement need a balun if You want use a commercial instrument (I.e. Agilent or Maury) , otherwise , if Your back-ends circuitry may give a numeric total power output, You may perform manually the Y factor measurement without any balun.
Even if theoretically a balun may be not necessary, I strongly suggest to buy pairs of balun that cover Your whole bandwidth. The pair is necessary in order to calibrate each one.
You mentioned that your circuit have differential output from the baseband output, that means this is a direct conversion receiver and there is a mixer inside.
In this case you cannot use a balun, or at least doesn't help for your measurement. Also you have to take into considerations the difference in noise figure between DSB and SSB noise figure.
Just search the net how to measure NF of a direct conversion receiver and you get a lot of answers.