"Type-D ‘Phase Localized’ implies a single standard (either a line or a reflect/reflect pair) being used along with the assumption that the fixture is electrically long enough based on the frequency range being used and proving the bulk of the fixture mismatch is not too close to the inner plane. In this demonstration, a zero length line was chosen as the standard. The symmetry between the fixture halves is known for the insertion loss. The match is presumed to be perfect since we try to divide a single transmission line by two, representing a two halves fixture. Instructions: Make sure the appropriate 2-port calibration is active. Zero out match terms if desired. This sets all reflection terms to 0 and may be helpful if the fixture is extremely unrepeatable. Select the measurement type (using thru). Set the magnitude thru line to 1 which means 0 dB loss and the length is zero. Enter an estimate of the network's electrical delay. Entering 0 will activate the automatic length estimator. Enter the file names and path where the output .s2p files will be stored. The Quick Extract option can be used instead: time stamped files will be saved to a predetermined hard disk location and the de-embedding engine will automatically load those files. Remember to keep track of available disk space. When all fields have been entered and proper connection of the standard(s) has/have been made, click on 'Perform Network Extraction'. If successful, files will automatically be saved and a confirmation dialog will appear. If Quick Extract was not selected, after the .s2p files have been saved, go to Measurement/Edit Embed/De-embed/Edit Network configuration panel to recall the .s2p files and configure the network. Go to embedding/de-embedding menu, select port 1. Add .s2p file (the one generated by the port 1 in Type-D), click on Add Network, click Apply. Add .s2p file (the one generated by port 2 in Type-D), click on Add Network, click Apply. Turn on the de-embed, a 0 dB insertion loss is observed from the trace S12."