Hello,

I have some questions about how to use the CharacterizeAdaptor Macro:

- I would like to perform the following: Given a 2-port DUT surrounded by a 'left-fixture' and a 'right-fixture' - I would like to solve for the left-fixture.s2p and the right-fixture.s2p

- I currently perform this operation with custom-software, the general procedure is: perform a first-tier calibration at the first-tier-VNA-cable-reference plane (using TRL) and then I perform a 2nd-tier-on-wafer-calibration - again with TRL (with the previous-first-tier calibration turned-on) at the DUT-reference-plane (the left-vna-cable is connected to the left-side of the left-fixture and the right-vna-cable is connected to the right-side of the right-fixture --- this yields enough information to solve for the left and right fixturers' S-parameters.

- However, we are now interested in using the on-board macro of the pna-x since I would like to explore using the NIST-MultilineTRL method for solving for the fixture halves.

finally, here are the two-questions:

- when executing the CharacterizeAdaptor Macro's step#2: "2. Perform a full 2-port CAL 2 where the DUT is inserted (reference plane). Save to MyCalSet2." - taken from: http://na.tm.agilent.com/pna/help/latest/Applications/CharacterizeAdaptor_Macro.htm --- is the first-tier calibration supposed to be on or off during the 2nd-tier calibration ? (For my own software/algorithm, the first-tier-cal must be on during the 2nd-tier-calibration --- I only ask since during a recent converstation with Dr. Joel at our lab I thought he mentioned that the first-tier-calibration should be off during the 2nd-tier for the PNA-X macro and that did not make sense to me or I must have misunderstood...).

- is it necessary to estimate the electrical-length of any of the fixture-halves to correctly solve for their S-Paramaters, or does the algorithm make use of the fact/knowledge that at DC, phase=0 ?

thank you,

Nick

- when the PNA-X software solves for the fixture-halves, does it need some type of estimate for the electrical-length of a fixture-half ?

Edited by: nickmmw on Jan 24, 2013 2:53 PM

Edited by: nickmmw on Jan 24, 2013 2:54 PM

Edited by: nickmmw on Jan 24, 2013 2:57 PM

Edited by: nickmmw on Jan 24, 2013 2:58 PM

I have some questions about how to use the CharacterizeAdaptor Macro:

- I would like to perform the following: Given a 2-port DUT surrounded by a 'left-fixture' and a 'right-fixture' - I would like to solve for the left-fixture.s2p and the right-fixture.s2p

- I currently perform this operation with custom-software, the general procedure is: perform a first-tier calibration at the first-tier-VNA-cable-reference plane (using TRL) and then I perform a 2nd-tier-on-wafer-calibration - again with TRL (with the previous-first-tier calibration turned-on) at the DUT-reference-plane (the left-vna-cable is connected to the left-side of the left-fixture and the right-vna-cable is connected to the right-side of the right-fixture --- this yields enough information to solve for the left and right fixturers' S-parameters.

- However, we are now interested in using the on-board macro of the pna-x since I would like to explore using the NIST-MultilineTRL method for solving for the fixture halves.

finally, here are the two-questions:

- when executing the CharacterizeAdaptor Macro's step#2: "2. Perform a full 2-port CAL 2 where the DUT is inserted (reference plane). Save to MyCalSet2." - taken from: http://na.tm.agilent.com/pna/help/latest/Applications/CharacterizeAdaptor_Macro.htm --- is the first-tier calibration supposed to be on or off during the 2nd-tier calibration ? (For my own software/algorithm, the first-tier-cal must be on during the 2nd-tier-calibration --- I only ask since during a recent converstation with Dr. Joel at our lab I thought he mentioned that the first-tier-calibration should be off during the 2nd-tier for the PNA-X macro and that did not make sense to me or I must have misunderstood...).

- is it necessary to estimate the electrical-length of any of the fixture-halves to correctly solve for their S-Paramaters, or does the algorithm make use of the fact/knowledge that at DC, phase=0 ?

thank you,

Nick

- when the PNA-X software solves for the fixture-halves, does it need some type of estimate for the electrical-length of a fixture-half ?

Edited by: nickmmw on Jan 24, 2013 2:53 PM

Edited by: nickmmw on Jan 24, 2013 2:54 PM

Edited by: nickmmw on Jan 24, 2013 2:57 PM

Edited by: nickmmw on Jan 24, 2013 2:58 PM

> Hello,

> I have some questions about how to use the CharacterizeAdaptor Macro:

> - when executing the CharacterizeAdaptor Macro's step#2: "2. Perform a full 2-port CAL 2 where the DUT is inserted (reference plane). Save to MyCalSet2." - taken from: http://na.tm.agilent.com/pna/help/latest/Applications/CharacterizeAdaptor_Macro.htm --- is the first-tier calibration supposed to be on or off during the 2nd-tier calibration ? (For my own software/algorithm, the first-tier-cal must be on during the 2nd-tier-calibration --- I only ask since during a recent converstation with Dr. Joel at our lab I thought he mentioned that the first-tier-calibration should be off during the 2nd-tier for the PNA-X macro and that did not make sense to me or I must have misunderstood...).

>

NO, and in fact, you cannot. If you use the PNA cal engine, it always spins up a new channel, without cal, does the cal, and deletes the cal channel. So you don't have the choice to do cal with cal on. But, it is not necessary to have the cal on, you can compute the adpater by comparing the 1 port error terms from two separate calibrations. The math details are in my book.

> - is it necessary to estimate the electrical-length of any of the fixture-halves to correctly solve for their S-Paramaters, or does the algorithm make use of the fact/knowledge that at DC, phase=0 ?

>

It estimates the delay from the slope of the line, and extropolated to DC to choose the proper root. An example of this is also in my book for reference.

> thank you,

>

> Nick

Note, if you undersample the fixture (measure with points spacing such that the phase change in the filter is more than 180 degrees per point) the delay estimate will be in error.