AnsweredAssumed Answered

Interaction of Cal Sets loaded into memory with AdaptorChar

Question asked by BrianAllred on Dec 22, 2011
Latest reply on May 7, 2012 by daras
Hopefully I've got the right sub-forum.

Some background information first.
We're using a PNA-X (N5242A), with firmware 09.22.18
The PNA-X is mounted in a rack as part of an automated test system.  The test system can route port 1 of the PNA-X to one of 30 inputs, and route port 2 of the PNA-X to one of 30 outputs (in reference to the DUT).
Each one of these 30 in/out paths is slightly different, and requires a full two port calibration.
As this would take an exceptionally long time to fully calibrate 900 paths, we were advised by local Agilent Field Representatives to use the Adaptor Characterization macro (or, as this is an automated routine, the Adaptor Characterization Console version) as follows:
  Perform a 2port calibration on one path pair (called reference)
  Perform a 1port calibration of each of the inputs/outputs
  Run AdaptorCharConsole to generate a s2p file containing the delta between reference and each of the input/outputs
  Deembed the s2p file from the reference calibration, and save as a new calibration set

(Please note that we opted to deembed the s2p file at calibration time, as doing so at test time would cost us more test time)

The issue that we're having is that as the final calibrations get created from the deembedding, the AdaptorCharConsole macro begins to slow down.  Initially (with a few cal sets in memory), it takes about 30-60 seconds to complete.  Once we get closer to completion (around 500 calibration sets), it begins to run closer to 10 minutes.  Is there anything that we can do to help speed this up, or do we need to set our timeout to an arbitrarily large number?

Any advice is greatly appreciated.