Return Loss of a Mixer
I intend to use a Network analyzer to do the same..My question is that shall i provide the LO drive input with a signal while making these measurements.or shall i just terminate it with a 50Ohms impedance?
If i need to pump the LO then why exactly its so?
Thanks & Regards
Because the LO level is not very high (~ 0dBm) probably won't change too much the impedances, but for accuracy is good to have the LO ON.
Usually when the LO level is very high, could produce an internal self-bias inside the active mixer, and varying in this way the input/output impedances.
A passive mixer (I do not know your device to be tested) requires to receive a nominal power level to its LO input. By this LO power, mixer diodes or transistor switches present a matched impedance to its RF and IF ports.
Without LO input power, the RF and IF ports are not matched.
You should see the difference on your VNA when you measure S11 and S22 of the mixer.
"Active" mixers which have RF or IF amplifiers integrated, usually do not change their RF and IF impedances.
RF and IF impedance matching in mixers is important whenever you connect RF and/or IF filters to those ports. Impedance mismatch also affects filter response.
Does that means for Active mixers the LO need not be powered to measure RF & IF return losses?I think HMC483 is an active mixer.
One more question to you guys,
Measuring S11 of RF & IF ports , with Powering LO, does it need a 4 port VNA? or just a 2 port VNA (with external Signal Generator to LO)?
I think to measure return loss(S11) of IF & RF a 2 port VNA will suffice(May be simultaneous measurement is not possible,i don't know) isn't that right?
Thanks & Regards
RF and IF Port impedances are almost independent from LO drive amplitude so you can measure directly these ports by applying supply voltage.
While you're measuring the ports, 2-prt VNA will be sufficient.Measure simply S11 and S22.
The Hittite HMC483 is an "active" mixer only in that it has a LO buffer amplifier integrated.
In its data sheet you can see the return loss (s11) of LO and IF ports but NOT of the most important RF port.
The advices of my colleagues above can be used to measure S11, the question is, how exactly you need to know it?
If your application is using the mixer with a RF amplifier before it and/or the IF amplifier following it, the simplest solution of handling mixer mismatch is to use e.g. 3-dB attenuators at its RF and/or IF ports.
Mixer mismatch can affect a filter response; if your application is narrow-band, you can add tuning trimmers to sensitive places and tune the system using a sweep oscillator, a detector and oscilloscope (scalar network analyzer if you wish).
Hello. It is important not to starve your Mixer with LO port, being it in actual operating enviroment (in your PCB or module implementation) or during lab characterisation using VNA. Give it the desired LO, and you will get the right VSWR. The published resturn loss measurement of all mixers has this underlying condition.
Resistive termination of Mixer IF & RF ports are done so because we want to prevent reflection of some problematic spurs from components such as filters which are highly reflective in their stopbands. These spurs, if re-enter the mixer, can cause re-mixing, worsen the intermodulation or conversion loss ripple of the mixer.
You can refer to very good Mixer application note from Marki Microwaves.
Regards
lguancho
short answer is yes. You need the right LO to pump the mixer up.
With the conversion loss in the spec it appears to be a passive mixer which absolutely requires LO for port impedance measurements.
It has an L.O. amp only.
If you look at conversion loss versus L.O. input power you can get an idea where impedance will start to change. This will happen then L.O. drops to a point where conversion loss starts to rise.