AnsweredAssumed Answered

Sweeping + triggering + repeatability

Question asked by b.d.hall on Mar 15, 2011
Latest reply on Mar 16, 2011 by b.d.hall
I would be grateful for an opinion as to whether the following is a correct way of triggering the PNA (E8363B).

I am collecting S-parameter data using linear sweeps. Averaging may be on or off.

In both cases I have set the trigger source  to 'immediate'.

When averaging is on, I then send 'SENSE1:SWEEP:GROUPS:COUNT n' (where n is my number) followed by 'SENSE:SWEEP:MODE GROUPS'. When the operation is complete, I read the data.

When averaging is off, I send 'SENSE:SWEEP:MODE SINGLE' and then wait for the operation to complete before I read the data.

I can't be sure from the manual whether this is correct. It seems to work, but I have some concerns about the data I am getting.

The repeatability of the first point in the sweep is much worse that all following points. By this I mean that the standard deviation of the data for the first frequency  (from 50 consecutive sweeps) is an order of magnitude worse than all the other points of the sweep.

So I am wondering if the way that I am using the PNA could be incorrect.

It does not seem to matter what the frequency of the first point is. Nor does the problem improve by throwing away a few initial sweeps and then looking at the stability of the remaining data. So, there seems to be something about the PNA measurement of the first point in every sweep that is not working as well as the others.