measuring impedance using spectrum analyser

i do not have a network analyzer, but i want to measure the input impedance at one of the sma connectors of my ic(supposed to have seen 50 ohm). i have seen some post which can give a basic idea about the matching but they all require tracking generator but is there any other way by which i can see whether the impedance is around 50 or not. (again it's basically a matching problem and i just want to check the effects of the matching network before the connector.)

Impedance is usually a frequency dependent complex quantity and you need to specify the frequency at which the impedance is being measured. The in-phase component (the real part) is commonly frequency independent but it need not be. It is common to plot the magnitude of the impedance as a function of frequency (and also the phase as a function of frequency) but in Bode plots, in which the real and imaginary parts of the impedance are plotted as a parametric function of frequency, are also quite popular.

In short, impedance is not a simple number, it is a complex function of frequency.

but how can i do this by using a spectrum analyzer. i am using 2437 MHz so the only way i am getting the idea about the impedance is that one of the ports where it is properly matched(50 ohm) i measured the power and then try to measure the power level at the ports where i need to get the matching. i know there will be some loss in power between these two points. one of the paper i saw that pcb loss isaround 0.3dbm/inch so the power diffrence btween these tw should be around 1 dbm.(tis is what i am supposed to get).

but i can notthink of any othr way except measuring thepower and compare them(based on intuition). is there a way to get the idea about impedance using only spectrum analyzer(i do no have a tracking generator).

there is no way to do it that I'm aware of, and that is what the others are also suggesting to you


Dave

you will need a directional coupler.
take a 10 or 20 db directional coupler. Orient it to show the power traveling BACK from the load.
Hook your spectrum analyzer to the coupled port, and leave the "load" port of the directional coupler open circuited
Sweep the frequency source, and use the spectrum analyzer "max hold" to capture the sweep. The trace stored on the face of the spectrum analyzer will be approximately the "0 dB return loss" trace.

THEN replace the open circuit with your device under test at the load port of the directional coupler. Once again, sweep the frequency and store the max hold again.

The difference between the two traces, in dB, will approximately be the return loss of the DUT.

Sir,

as i have already told that first i am measuring power from one of the matched ports and then measuring from the port where i am supposed to get the matching, i think this measurement is same as getting the return loss since the power from both the ports is supposed to be same (the power difference between these two by simply placing a o ohm in series and measuring the power from the second port) so why to use a directiona coupler ?

i also think this can not be done by just using spectrum analyzer. my bad



thanks for replying anyway

If you have a accurate a RF Signal Generator and a Directional Coupler, you may measure the magnitude of the Input Impedance but it's quite erroneous for small reflections.