AnsweredAssumed Answered

4-port S-parameter meas & uncertainty calculation

Question asked by tomoey on Jan 17, 2008
I am trying to come up a way to calculate the combined uncertainty of a set of 4-port S-parameter data that I acquired from PNA E8364B / N4421B.
Particularly, I am interested in calculated the uncertainty  of S31 and S32 in real & imag (or mag [dB] and phase [deg],  in the end).

First, I am trying to separately estimate

1. Type A uncertainty due to my meas. errors in my DUTs, and

2. Type B uncertainty due to the systematic error;
    * PNA E8364B / N4421
    * Non-agilent CAL kit
    * SOLR method with a local delta match

For Type A, I am going to estimate a 32 x 32 (=16 S parameters * 2 R/I parts) covariance matrix from my repeated measurements, and use that to run a Monte Carlo simulation under the assumption of multivariate normal distribution.

For Type B, I am having a problem and hope that some experts will provide me some guidance.  I looked at PNA-HELP document to find a set of the residual error estimates for 4-port measurement with SOLR CAL.  I could not find any documentation under SOLR CAL or Multiport test set configuration.  Furthermore, I realized that I used Maury's 2.92mm CAL kit, even if I found data, they would not be directly applicable to my situation. 

In this case, do I need to measure the residual errors with the precision airline, at a pair of ports at a time?

I would appreciate any input for the matter of uncertainty calculation in multiport measurement setting.

Thank you for your help.