50 ohm measurements using 75 ohm vector network analyzer

[kada]

Is it possible to use 2-port 75 Ohm vector network analyzer to match 50 ohm transmission line to antenna of 10-100mW 2.4GHz radio modules (BTLE, WiFi, etc.)? The matching circuit topology is pi-network.

[tggzzz]

Is it possible to use 2-port 75 Ohm vector network analyzer to match 50 ohm transmission line to antenna of 10-100mW 2.4GHz radio modules (BTLE, WiFi, etc.)? The matching circuit topology is pi-network.

You will have to insert 50<->75 ohm matching networks, which will introduce a fixed loss. These can be a 2 resistor "L" topology rather than a 3 resistor pi topology, simply by setting one of the shunt resistors to a high value. Play around with http://chemandy.com/calculators/matching-pi-attenuator-calculator.htm

[Howardlong]

I don't know your exact setup, but I use and 8753A with an 85044B 75 ohm test set, it just happened to be the ebay deal I struck nearly 15 years ago. I simply use a 75 ohm to 50 ohm minimum loss pad (HP 11852B) with it and calibrate appropriately. You lose a few dB of sensitivity/dynamic range as a result but it's not really a problem for my uses, which is antenna matching and filter testing where stop band measurement down to -100dB or so is unnecessary.

Just take super super care NOT to plug in a 50 ohm N into a 75 ohm N socket or the socket will be ruined. The minimum loss pad is attached to my 75 ohm test set 24/7 to avoid any unnecessary upsets.

[kada]

You will have to insert 50<->75 ohm matching networks, which will introduce a fixed loss. These can be a 2 resistor "L" topology rather than a 3 resistor pi topology, simply by setting one of the shunt resistors to a high value. Play around with http://chemandy.com/calculators/matching-pi-attenuator-calculator.htm

Thank you for useful link. I suppose that this topology will work for bidirectional VNA port for S11 return loss and antenna impedance measurements? Ie this topology is actually a two-way and converts the impedance from 75 ohms to 50 ohms, and vice versa from 50 ohms to 75 ohms with the same attenuation?

I don't know your exact setup, but I use and 8753A with an 85044B 75 ohm test set, it just happened to be the ebay deal I struck nearly 15 years ago. I simply use a 75 ohm to 50 ohm minimum loss pad (HP 11852B) with it and calibrate appropriately. You lose a few dB of sensitivity/dynamic range as a result but it's not really a problem for my uses, which is antenna matching and filter testing where stop band measurement down to -100dB or so is unnecessary.

I plan to use VNA with 2 bidirectional ports like in the attached picture. I guess that in this case I do not need to use a test set and I can manage with one 75 ohm to 50 ohm minimum loss pad? A minimum loss pad consist of 2 or 3-resistors network?

In my case I have to spend OSL calibration before measurements with 50 ohm calibration kit? Or do I have to calibrate with 75 ohm set as well?
And the last question, do I should somehow change the impedance settings of VNA from 75 ohms to 50 ohms for the correct measurements?

Sorry for so many questions, I'm new in the microwave measurements. Your input are very helpful for me.

[Howardlong]

I think from memory that the 8753D has its own test set installed as default, so I assume if it's the 75 ohm option it will have two 75 ohm N ports. You can use two minimum loss pads, but of course you will lose more sensitivity and therefore dynamic range, each pad is an additional 5.7dB loss.

Always calibrate at the point where you plug in the DUT, so that will be a 50 ohm cal. This means you're only measuring the DUT, not the cables or indeed the minimum loss pads.

[tggzzz]

You will have to insert 50<->75 ohm matching networks, which will introduce a fixed loss. These can be a 2 resistor "L" topology rather than a 3 resistor pi topology, simply by setting one of the shunt resistors to a high value. Play around with http://chemandy.com/calculators/matching-pi-attenuator-calculator.htm

Thank you for useful link. I suppose that this topology will work for bidirectional VNA port for S11 return loss and antenna impedance measurements? Ie this topology is actually a two-way and converts the impedance from 75 ohms to 50 ohms, and vice versa from 50 ohms to 75 ohms with the same attenuation?

Yes. You should consider choosing real resistor values and suitable transmission line lengths, and doing a spice simulation (e.g. LTSpice) to see the magnitude of the resulting system's performance.

[KJDS]

In that block diagram the test set part is within the dotted lines. I think it's possible as an option to delete that section but it's usually installed.