AnsweredAssumed Answered

Newbie Balun measurement question

Question asked by johndp on Aug 8, 2007
Latest reply on Aug 9, 2007 by johndp
I measured the balun as a three port network, port 1 is the unbalanced input, port 2 is the 'through' connection and port 3 is the indirect
connection. I only have a two port VNA so the unused ports were terminated  with 50 Ohms during the measurements

Port 1         -----OOOOOOOOO--------o Port 2
                     |                         |
                     |                         |
                    ---                        |
                    GND                     |
                                                ----o Port 3

If I generate S-parameters in touchstone format as DB-phase and plot the phase of port  1 to 2 (S21) and port 1 to 3 (S31) on the same graph I see the 180 degree (approx) phase shift I expect to see, the phase balance is +-3 degrees up to about 2GHz and then it goes off, but this is expected.

If I generate the RI touchstone files and then manually convert the phase using phase (degrees) = (180/pi) * arctan (im/re), the phase for Port 2 is correct but the phase for Port 3 is inverted and has approximately the same phase as Port 2.

Is it normal to loose phase information in RI format ? or do I just need to correct my phase calculation?. Obviously I could correct the DB-phase in the conversion, but this does not fix RI data.

I need to use the RI format because I am manually de-embedding the test jig from the data, so I would like to 'fix' this data if possible.

As an aside I measured the balun in a 50 Ohms system, I know that I can correct this using some math to re-normalises Port 2 and Port 3 to 25 Ohms so i can directly compare the manufacturers data to our measure data, do you have a reference on how to do this ?