s for air. The only effect to consider is the probe model accuracy of about 5%. ThusIn the case where PTFE is measured at 10 GHz using an air/short/water calibration, the total measurement error would be the sum of the errors due to probe model accuracy (typically about 2% to 5%), air, and water.
First compute s for air. The only effect to consider is the probe model accuracy of about 5%. Thus
sAIR = 0.05*magnitude of the permittivity of air = 0.05*1 = 0.05
Next compute s for water. The model is accurate to within about 1% and the probe model contributes another 3%. The permittivity of water at 10 GHz and 25 
C is 62.3 – j 30. 
rWATER is then the square root of the sum of the squares or 69.
sWATER = (0.05 + 0.01) x magnitude of the permittivity of water at 10 GHz = (0.05 + 0.01) x 69 = 4.14
Next find the sensitivities for the various permittivities at 10 GHz. The values for SSAIR, SSWATER , and SMTEFLON can be generated by first performing an air/short/water calibration. After the calibration, remeasure air and store the measurement data into a data file. Remeasure water and store it into a data file. Then measure teflon and store it into a data file.
Finally, refer to Importing Data into Other Programs, and use a spreadsheet program or notepad.exe to read the sensitivity numbers. Insert those sensitivity numbers in equations as shown below to calculate the uncertainty of the measured permittivity of the MUT.
Assume that the data files yield these sensitivity numbers:
SSAIR = 8.2
SSWATER = 223
SMTEFLON = 8.1
The worst case combination yields:
The RRS combination yields:
The error can occur in either of the orthogonal components ’ and ’’ such that
and
