As previously described, the compensation devices are test transformer samples that have had their tested winding pins shorted (short compensation) or had their windings removed from their pins (open). The compensation devices are placed in the test fixture, compensation is performed, and readings are taken afterward with the compensation parts in place (especially the short compensator) to verify low inductance (preferably below 1nH) and low resistance (preferably below 10 milliOhms). There is only one measurement fixture. It is moved from the Agilent to the VTi and it is used with the compensation devices described above. With dirty or non-coplanar DUT's, I may have to spend a minute or more to get some transformers to provide a realistic, stable reading, but I always can with the Agilent. I never have with the ATi. Note that I haven't done this test on the ATi before getting it back from this month's repair.
In the photos, you can see the board in a plastic case which is sitting on another 1" tall plastic piece on top of the DUT test area of the VTi. The four BNC's to the VTi are the only electrical connections from the fixture to the VTi. I do hold the transformers against the contacts, but I don't touch the contacts and the impedances are so low (around an Ohm), that noise pickup doesn't seem very likely. This is not an issue with the Agilent meter.
Series model R in the tests is around 0.1 to 0.2 Ohms. Reactance for a 60uH inductor at 1MHz is 0.38 Ohms. So the Q is around 2.
The ATi reported test currents are normally around 2mA (even though it shows a programmed current of 20mA). 2mA times 0.4 Ohms is 0.8mV measured voltage. What is the precision (not accuracy) of the ATi voltmeter?
I used 20mA because the Agilent likes that current for this measurement. The Agilent meter also lets me force a range for reactances (I use 10 Ohms full scale). I don't see a range set function for the VTi.
The Agilent meter likes shielded Coax (mine is RG-58-ish) and it likes the positive drive shield to be connected to the negative drive shield on fixtures as that meter pushes a complementary drive current equal to the measurement drive current through the shields to reduce stray inductance and magnetic susceptibility.