AnsweredAssumed Answered

TRL calibration in microstrip with 8510C problems

Question asked by Amiet on May 24, 2010
Latest reply on Jun 3, 2010 by Amiet
Hi there, I am having a problem setting up the TRL calibration for microstrip line on my 8510C. I've gone through the set up a number of times and still cannot get the system to work properly. The problem is this:

We have created our 50 Ohm calibration standards for a short, a 10.1mm long line used for the thru, and an 18.06mm long line to be used as the line standard. The microstrip in on 0.502mm thick Rogers 5880 board with a permittivity of 2.2, effective permittivity 1.88. The frequency range is 4.5 - 9 GHz, corresponding to a 60 to 120 degree phase shift for the line standard. I used a response transmission cal to physically measure the time delay and it came out around 11.0 mm, well in line with expected. I used this figure for the line delay in the calibration setup.

Everything else was pretty standard, the Offset Z0 was 50 Ohm, the L values for the short were all 0, I used the coaxial dispersion for the cal standards. I used a velocity factor of 0.73 and 1.0, but it didn't seem to make a difference.

The main result I found with the S11 measurements I susbsequently took had a large phase shift at around 5.3 GHz. This was not a simple -180 to +180 wrap around, it was a proper phase shift. After calibration, remeasuring the shorts gave the same phase shift, and the phase was not equal to -180 degrees, rather it moved around the smith chart with frequency. We measured the short with a multimeter and it seemed to be fine, but the microwave result was just crazy.

The system works on a different off-site VNA (Wiltron 360), but I would like for it to work on our 8510C too.

We are using software version 7.16 (Dec 12 2000), could this be a source of the problem?

Any help appreciated, we are tearing our hair out here.

Thanks, Andrew.