[evlatests] 3-bit state counts, input power levels, & GSE settings

[Keith Morris]

>(Keith can provide
> details
> as to how "difficult" is defined).

I pulled receiver gain plots for receivers that were installed in the
3-bit antennas.  I found frequency regions in the receiver bands that
looked particularly in need of gain slope correction.  It may not be an
ideal criterion, but it was a place to start looking.  We also chose an
arbitrary region of K-band, whose inherent slope was not known.

Not to be nitpicky, but the range of equalizer adjustment is 0 - 15,
representing (nominally) a -15dB/2GHz to +15dB/2GHz slope adjustment
range.  Due to frequency roll-off in the T304, this tends to be more like
-20dB to +10dB per 2GHz.  The default position of "10" is the setting that
flattens the output spectrum of the T304 itself, when fed with a
relatively flat input signal.  As Michael mentioned, the majority of cases
used settings that were very close to this default position.


>
> Herewith the promised, more detailed report on 3-bit power levels and
> gain-slope equalizer (GSE) settings.
>
> We were able to set the attenuators so that the sampler rms was 1.6-1.7 in
> all cases. The resulting state counts looked remarkably similar for
> all the samplers, especially A1 (RCP) -- see attached plots (data path 0
> [upper] is LCP; data path 1 [lower] is RCP).
>
> However, this required a wide range of input powers to the samplers,
> as detailed below.  After setting those power levels we tweaked
> the gain slope equalizers to flatten the bandpass; again this required a
> range of settings.  A quick look showed no obvious spectral differences
> when the GSE setting was changed -- the slope across the band changed, but
> the smaller-scale features were constant.
>
> Quick summary of GSE results:
> - GSE needed for a given sampler was generally similar for all
>    tunings
> - Most samplers required GSE settings within a couple "steps" of the
>     default (10).  The clearest exceptions were ea12-A1 (even a GSE
>     setting of 0 left 12-15 dB residual slope across the band) and
>     ea22-A1 (GSE settings of 3-7).
>
> Here are the required settings for each sampler, for a number of
> different tunings, chosen by Keith as particularly difficult places in
> the bands -- the goal was to see whether the GSEs could indeed flatten
> the bandpass for even the most difficult tunings (Keith can provide
> details
> as to how "difficult" is defined).
>
> Key:
>     RMS --> sampler RMS
>     Ch1Power --> power out of T304, into the sampler
>     GSE setting --> setting which produced flat bandpass (0-31; 10
>       is the default), or residual slope across the 2 GHz band if
>       this failed (ea12-A1 only)
>     Frequency covered --> frequency range covered by that tuning
>
> Sampler   RMS    Ch1Power   GSE setting     Frequency covered
> -------   ---    --------   -----------     -----------------
> ea12-A1   1.7   -21.5 dBm   12dB res.slope     5- 7 GHz  (C  band)
>             1.7   -22.2 dBm   12dB res.slope    19-21 GHz  (K  band)
>             1.7   -22.2 dBm   15dB res.slope    44-46 GHz  (Q  band)
>             1.7   -24.6 dBm   15dB res.slope    35-37 GHz  (Ka band)
>             1.7   -22.1 dBm   13dB res.slope    38-40 GHz  (Ka band)
>
> ea12-C1   1.7   -41.9 dBm       9              5- 7 GHz  (C  band)
>             1.7   -41.1 dBm       9             19-21 GHz  (K  band)
>             1.7   -41.4 dBm       8             44-46 GHz  (Q  band)
>             1.7   -41.4 dBm       7             35-37 GHz  (Ka band)
>             1.7   -41.3 dBm       9             38-40 GHz  (Ka band)
>
> ea15-A1   1.7   -38.3 dBm      10              5- 7 GHz  (C  band)
>             1.7   -38.2 dBm       9             19-21 GHz  (K  band)
>             1.6   -38.8 dBm      10             44-46 GHz  (Q  band)
>             1.7   -38.6 dBm       9             35-37 GHz  (Ka band)
>             1.6   -38.9 dBm      11             38-40 GHz  (Ka band)
>
> ea15-C1   1.7   -39.7 dBm      10              5- 7 GHz  (C  band)
>             1.7   -39.5 dBm      10             19-21 GHz  (K  band)
>             1.6   -40.0 dBm      10             44-46 GHz  (Q  band)
>             1.7   -39.5 dBm       8             35-37 GHz  (Ka band)
>             1.6   -40.4 dBm      10             38-40 GHz  (Ka band)
>
> ea22-A1   1.6   -32.6 dBm       7              5- 7 GHz  (C  band)
>             1.7   -32.3 dBm       3             19-21 GHz  (K  band)
>             1.7   -30.2 dBm       6             44-46 GHz  (Q  band)
>             1.7   -32.2 dBm       4             35-37 GHz  (Ka band)
>             1.7   -30.2 dBm       7             38-40 GHz  (Ka band)
>
> ea22-C1   1.7   -35.0 dBm       9              5- 7 GHz  (C  band)
>             1.7   -35.5 dBm       7             19-21 GHz  (K  band)
>             1.7   -35.4 dBm       8             44-46 GHz  (Q  band)
>             1.7   -36.6 dBm       6             35-37 GHz  (Ka band)
>             1.7   -35.7 dBm       9             38-40 GHz  (Ka band)
>
> ea28-A1   1.7   -39.1 dBm      10              5- 7 GHz  (C  band)
>             1.8   -38.9 dBm       8             19-21 GHz  (K  band)
>             1.6   -39.5 dBm      10             44-46 GHz  (Q  band)
>             1.7   -39.6 dBm       7             35-37 GHz  (Ka band)
>             1.7   -39.3 dBm       9             38-40 GHz  (Ka band)
>
> ea28-C1   1.7   -34.6 dBm      11              5- 7 GHz  (C  band)
>             1.7   -34.1 dBm      10             19-21 GHz  (K  band)
>             1.7   -35.0 dBm      11             44-46 GHz  (Q  band)
>             1.6   -34.4 dBm       9             35-37 GHz  (Ka band)
>             1.6   -35.5 dBm      11             38-40 GHz  (Ka band)
>
>     -- Michael_______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>