Measuring interferes with impedance matching

Thread Starter

Nailer

Joined May 11, 2021
7
Hello,

I have setup like shown below.
schematics.jpg
Notes: freq gen and rf amp are 50 ohm rated, when doing the measurements for impedance matching (with S11) VNA is placed between RF amp and RF transformer, RF transformer has differential output, RF transformer is 50 to 200 ohm, so that means L-Match circuit will be placed between RF amp and RF transformer, differential probe have 3Mohm input

I need to measure (or calculate) the output differential voltage amplitude before it goes into antenna.
But I found out that when differential probe is connected at the output of decoupling capacitors, I have different measurements of reflected power.
That means that differential probe is changing frequency characteristics of the system.
So when I am not having probe connected I have different power delivery therefor different amplitude that goes into antenna.
How can I know what is the amplitude of differential voltage at decoupling capacitors exit?
Can I do the impedance matching with differential probe connected, but does that mean that some of the power will be delivered to the probe instead of antenna?
 

peterdeco

Joined Oct 8, 2019
415
It's been a long time since I worked with RF so my advice might be a little rusty. To measure output voltage, I used to use a 50 ohm non-inductive load on the output (not an antenna) and measure the voltage with a high impedance RF probe. I would then tweak the output coupling (in my case I used pi-networks) for maximum voltage.
 

Danko

Joined Nov 22, 2017
1,189
Simple measure level of power, emitted by antenna, with help of some receiver.
Adjust circuit for maximum signal from receiver. That's it.
 

Thread Starter

Nailer

Joined May 11, 2021
7
It's been a long time since I worked with RF so my advice might be a little rusty. To measure output voltage, I used to use a 50 ohm non-inductive load on the output (not an antenna) and measure the voltage with a high impedance RF probe. I would then tweak the output coupling (in my case I used pi-networks) for maximum voltage.
Simple measure level of power, emitted by antenna, with help of some receiver.
Adjust circuit for maximum signal from receiver. That's it.
Hi guys thanks for the advice.
I have to point out that I am total newbie considering the RF. I just somehow understand a little bit of impedance matching.

Also I have to explain what can I do I why I am confused with measuring the differential voltage at output of RF transformator:
So, my RF amp can display forwarded power and reflected power on its screen (E&I 240L RF amp).
Now let say I just power up my freq gen (with some fixed freq and Vpp), rf amp and have setup as pictured without differential probe.
In that scenario my RF amp will display for example forwarded power of 12W and reflected power of 3W.
Then I connect differential probe as pictured on differential output of decoupling capacitors which is in parallel with RF carpet as well. And now my RF amp displays forwarded power of 24W and reflected power of 21W.

So conclusion is when I connect my differential probe my power delivery is different and I can understand that it is happening because of impedance mismatch due to differential probe being connected. Therefore measurement of amplitude is not the measurement of amplitude I would have if there is not probe connected.

Questions:

Can I do VNA impedance matching (at the place where RF amp is) with probe connected where it usually is, or that will mean that matched circuit will have split power delivery, therefore I will not have full power delivery to the antenna?

How can I avoid having differential probe connected but still know what is the differential output amplitude of voltage at that place where I am trying to measure it (output of decoupling capacitors)?
 

Tesla23

Joined May 10, 2009
488
Unless there is something strange in your networks, the 'usual' approach would be to assume that everything after the power amp is linear, so what you would do is disconnect your matching circuit from the RF amp and connect it to the VNA, and adjust the matching circuit until you have the match you want. This would also allow you to see the effect of connecting your probes on the impedance seen by the RF amp.

Once you have a good match, reconnect the RF amp and the forward and reflected powers should be what you expect. If they aren't, connect a dummy load to the amp and check that the power readings are correct.
 

Thread Starter

Nailer

Joined May 11, 2021
7
Unless there is something strange in your networks, the 'usual' approach would be to assume that everything after the power amp is linear, so what you would do is disconnect your matching circuit from the RF amp and connect it to the VNA, and adjust the matching circuit until you have the match you want. This would also allow you to see the effect of connecting your probes on the impedance seen by the RF amp.

Once you have a good match, reconnect the RF amp and the forward and reflected powers should be what you expect. If they aren't, connect a dummy load to the amp and check that the power readings are correct.
I already tested that, I dont need matching circuit to prove that measurements are different. As I explained in my latest reply above:
RF amp to non matched circuit without diff probes - 12W forwarded 3W reflected
RF amp to non matched circuit with diff probes - 24W forwarded 21W reflected

So it would be the same with matched circuit because differential probes are doing some kind of impedance missmatch and I confirmed that with VNA measurements. But how to avoid that, or how to know what is the differential voltage at place pictured in the setup? Should I do impedance matching with connected differential probes, wouldn't that mean that power would be split between antenna and diff probe?
 

Tesla23

Joined May 10, 2009
488
Then your differential probe is not a good way of measuring the voltage across the antenna. You will have to either improve it of come up with another way of measuring it, or of working it out. With the only description of your antenna as a literal black box, and no knowledge of your operating frequency, it is hard to advise further . You could possibly measure the antenna impedance, measure your matching circuit losses when working into that impedance, and calculate to power going to the antenna. Good luck.
 
Top