AnsweredAssumed Answered

4-port non-insertable CAL with mechanical CAL kit

Question asked by tomoey on Nov 27, 2007
Latest reply on Nov 28, 2007 by daveb
With PNA E8364B / external test set, I have been doing 4-port calibration with the same connector gender (2.92mm Female @ DUT, i.e., Male @ cable ends), by using a mechanical CAL kit.  I found that PNA's "Smart CAL" was less desirable in the following reasons:

1). 3 full-set of SOLT with adapter removal was too much work.

2). PNA only asks me to do the reduced THRU measurement (3), but I want to do all 6 THRU to enhance the accuracy, especially for my very short / low-loss devices.

I am thinking of writing a simple program to improve this situation, and I would like to get some ideas from the expert if what I am thinking is OK to do.  Here is my proposed CAL procedure:

Step 0:
In 2-port mode, perform SOLR and fully characterize an adapter with female connectors.  Save the 2-port S-parameter data.

Step 1:
At each port, perform 1-port CAL to get ED (directivity), ES (source match), ER (refl. tracking) terms.

Step 2:
Measure uncorrected 6 THRU path with the characterized adapter inserted between each pair of port.

Step 3:
Compute 12-term error terms using the data acquired from Step 0 - 2.
I am intending to do SOLT with "KNOWN" THRU, externally.  Then, upload all the errors to PNA.

This way, I do not need to worry about Swith error terms.  I do not know any way to perform "Global Delta Match" CAL on my PNA.  Is it possible to do so?

In addition, I could not figure out a way to define my adapter as a DATABASE standard by entering 2-port S-parameter as its definition.

I would appreciate any of your critique for my proposed CAL method.
Thanks for your time.