How to measure input/output return loss of a mixer using VNA

[baltoudi]

mixer measurement using 3 mixer

currently I am doing 5-GHz mixer measurement for its return loss using 2-network analyzer. Since this measurments must be taken in the dynamic conditions, then I am in doubt if pumping the LO signal to the mixer while RF and IF ports are respectively connected to port-1 and 2 of VNA will destroy the 2-port VNA. Anyone has comments how to do it correctly ?

 

[marcomdd]

using vna & sig gen to measure mixer

See the picture of the setup measurement.

Bye

 

[sergio mariotti]

return loss in mixer

My friend and colleague marcomdd showed a good method to measure in dynamic mode the reflection coefficent.
My experience with mixers measurements is the following:
1) generally the Network analyzers can accept relatively high power without demage, at least 100 mW for SNA and more for VNA.
2) If you have a VNA, because it's tuned, you may measure directly the sii at the R port and I port. The tuned circuitry cut away the interference caused by LO leakege.
3) you may measure sii at L port if the out power of VNA is strong enough.
4) Of course close any unused port into matched load.
5) more difficult is the measurement of sij , read Agilent AN's and PN's

 

[baltoudi]

ussing the vna

i need to know as well when u measured Sii at R and I port using VNA, did u pump the mixer with LO signal @ L port too or connect only with a 50 ohm termination ? because I did not pump it with LO signal generator, but with a 50 ohm termination @ L port, resulting a very very bad return loss (RF power injected=-20 to 0 dBm) ! which one is correct?

 

[xinxin]

return loss measurement using vna

Baltoudi , here is doubt what you said above , it's just how to pump it with a 50ohm termination , but not connect SG? i need your help.

 

[marcomdd]

lo port termination

i need to know as well when u measured Sii at R and I port using VNA, did u pump the mixer with LO signal @ L port too or connect only with a 50 ohm termination ? because I did not pump it with LO signal generator, but with a 50 ohm termination @ L port, resulting a very very bad return loss (RF power injected=-20 to 0 dBm) ! which one is correct?

You must pump the LO port !

 

[marcomdd]

Re: How to measure input/output return loss of a mixer using

Baltoudi , here is doubt what you said above , it's just how to pump it with a 50ohm termination , but not connect SG? i need your help.

What do you mean?

 

[xinxin]

But 50 ohm termination is for?

 

[marcomdd]

Re: How to measure input/output return loss of a mixer using

When you have to measure the RL at the LO port you must pump the LO port.

Usually a VNA is not able to provide the right power level , so it is better to use a measurement setup equal to the picture above.

 

[baltoudi]

Re: How to measure input/output return loss of a mixer using

marcomdd, you mean in order to measure all return loss for R, L, and I port in the dynamic mode using VNA, the mixer must be pumped with an external LO signal generator at L port, right ?

The reason why i insist on using VNA (rather than SA) is that the VNA measurements can save more time when sweeping towards frequency and power level and give relatively more accurate results.

 

[marcomdd]

Re: How to measure input/output return loss of a mixer using

marcomdd, you mean in order to measure all return loss for R, L, and I port in the dynamic mode using VNA, the mixer must be pumped with an external LO signal generator at L port, right ?

Right !
For RF and IF ports RL measurements you may use the VNA.

But, I don't think your VNA may pump the LO port with the right power. So, for the LO RL measurement only it is better to use the SA setup.

 

[sergio mariotti]

Drive the L port with appropriate power level during the measurements of R & I ports!
Typical values for L power are +10 dBm.
Of course use CW signal, and prefere a spurious "clean" generator. Harmonics are not important, since they are generated in any manner inside the mixer.
If you want measure sii at L port, check, with a power meter, if your VNA is able to deliver enough power, some agilent 872x and Anritsu 37xxx can do it. Then, close the other 2 port into matched load and

A little suggestion. When measure sii at R & I ports, be sure that the generator is well matched, or place a ferrite isolator or an attenuator across the L port.

Enjoy with the measurement rather than chat