AnsweredAssumed Answered

Labview Measurement Approach Query

Question asked by w3kq on Feb 4, 2012
Latest reply on Feb 7, 2012 by odanzy
Hi All,

I am working on an Labview application to collect data from a 4 port, 20Ghz ENA series analyzer.  The vast majority of the DUT's will be RF filters and related items (i.e. duplexer/triplexers), so basics such as RL, IL, group delay, etc. must be collected for almost everything. 

Some basic assumptions will be made (i.e. analyzer is properly calibrated.) moving forward.  Some of the approaches we're considering include:

1. Set up the ENA, and use the analyzer's built in processing functions for data measurement.  I.e. set markers, etc.  We would be using the processing hardware and algorithms developed by Agilent for measuring the specific data points, or parameters (i.e.BW, pk-pk, etc). I've found many examples of this approach on the web. 

2. Set up the ENA and use one of the various data export formats to collect the corrected data and post process using Labview created algorithms.  I've not found much on this approach (not on the post processing point, but there are examples for the data collection process) using Labview.  I am aware of one application that was written using Visual Basic.  The writer collects the data using the ASCII character format, then post processes using straight line (linear) interpolation. He insists his data measurements (i.e RL, IL, Group Delay, phase match, etc) are just as accurate as the data that would have been collected using approach #1 above.

3. Collect the S Parameter data from the analyzer and use Labview to post process, as opposed to number 2 above. Customers request SnP files for their parts on a regular basis, so we would have to collect this data using 1 or 2 above on a regular basis.

This goal is to create  a generic program that can be set up for many different filters and some simple active RF parts (i.e. switches), so we're prepared to put more effort up front in the program creation.  Our priorities are accuracy, ease of use by a test operator, and throughput speed - in that order.
I would greatly appreciate some feedback on any/all of the approaches, as well as additional thoughts, suggestions.

Thanks for your replies,

R  

Outcomes