Test issue for Marchand Balun

[supersage]

Dear all,

I designed a Marchand balun operating at 5.5GHz for IEEE 802.11. The input impedance is 50 ohm at port 1, and output impedances are 100 ohm at port 2 and port 3, respectively.

As we know, 50 ohm is the standard impedance of regular network analyzer. The issue comes as: how can we use the regular 50ohm 4-port network analyzer to measure such balun?

Thanks a lot in advance.

 

[kspalla]

Yes, You can configure the NA for differential mode too.
Check the manuals of NA.

 

[supersage]

Thanks, Kspalla.

In my case, the differential output impedance is 200ohm, right? If so, how can regular 50ohm network analyzer measure it?

Btw, if I try to do 3-port measurement, how to do measure such balun with 100ohm output impedances at port 2 and port 3, respectively?

Thanks again.

 

[kspalla]

There is scaling for impedance too.
Few of the NA supports impedance scaling to 75 ohms and 100 ohms.
If you can configure the NA job is done.
Check with the NA vendor for scaling the impedance.

 

[kayaoo]

I sometimes meausred the 3-port s-parameter in pure 50ohm (3 s2p files were measured with 2-ports VNA and combined them) and import it into @ds where the port impedance can be set freely.

 

[supersage]

Thanks. But I want to see how to do such measurement by the regular 50ohm VNA in our lab.

There is scaling for impedance too.
Few of the NA supports impedance scaling to 75 ohms and 100 ohms.
If you can configure the NA job is done.
Check with the NA vendor for scaling the impedance.

---------- Post added at 00:32 ---------- Previous post was at 00:27 ----------

Thanks, Kayaoo.

I also know this method. But I am a little bit confused about this method, because the converted results (setting 50ohm to 100ohm for termination impedances at port 2 & 3) are not the real measured results... Is there any issue with way? Thanks.

I sometimes meausred the 3-port s-parameter in pure 50ohm (3 s2p files were measured with 2-ports VNA and combined them) and import it into @ds where the port impedance can be set freely.

 

[kayaoo]

I also know this method. But I am a little bit confused about this method, because the converted results (setting 50ohm to 100ohm for termination impedances at port 2 & 3) are not the real measured results... Is there any issue with way? Thanks.

I think it is perfect in theory unless nonlinearity is concerned. And all you get from VNA are not real but after calibration.

 

[supersage]

Thanks lot, Kayaoo. I will try to see this way for our work.

I think it is perfect in theory unless nonlinearity is concerned. And all you get from VNA are not real but after calibration.