Intro

    Data for model3A was taken feb,mar16 (more info). The 1D gaussian fits to the az and elevation strips were used to compute the system performance. Data with poor fits (large pointing errors, were excluded).

    The gaussian fit used for each strip was:

• y=A0 +A1*exp(-(*x-A2)/A3)^2) + a4*x
• x is the great circle offset from the center of each strip.
• I used 8 sources. We had fluxed for 6 of them.
• the data was taken over 2 weeks.
• There was no cal available
• The data from each rise,set of a source was normalized to have a median value of 1. So the units of each data set should be close to tsys
• The electronic gain was necessarily the same for each source data sets (although the src and Tsys measurement for a single cross will have the same units..).
  

The results:

    The plots show the results from the fits (.ps) (.pdf)

• Page 1: Tsys and src strength
• This plots all of the data taken
• black has the azimuth strips, red has the elevation strips.
• Top: Fit to Tsys  vs cross sample number.
  
• the units are normalized to 1 for each data set (so all the data does not have the same electronic gain).
• 2nd: Fit to source strength vs cross sample number.
• The strongest source (Cas A) was about .25 to .3 of Tsys.
• Bottom: srcFit/Tsysfit
• this is source strength in units of Tsys (but the flux has not been included yet).
• Tsys can change because of the elevation, or the weather.
• Page 2: Compare data from az and el strips.
• Top: ( fit TsysAzstrip)/(  fit TsysElStrip).
• The variation is a fraction of 1%
• bottom: Ratio of (azFit srcStrength)/ (elFit srcStrength)
• The noisy ratios are the source with smaller flux  (< 1% of Tsys).
• Page 3: Tsy Fits vs elevation. plotting each source separately.
• Try to see if there are any strange variations in gain or noise.
  
• Top: the 3 strong sources
• green: 3C461.. one of the strips looks like it had a strong tsys or gain variation (rain?)
  
• Bottom: the 5 weaker sources
• black 3C123 had a large gain or tsys variation around el=60 ..
  
• Each source is plotted in a different color
• Page 4: TSource/Tsys vs elevation by source
• Top: 3 strong sources
• bottom: 5 weaker sources
• The large change in 3C461 remained with the ratio. So the change was probably a Tsys change.
• 3C123 had no jump around el=60.. so the Tsys jump was probably an electronic gain jump (it cancels in the ratio).
• Page 5,6 : SEFD vs elevation by source
• Each source is plotted in a separate frame.
• I corrected the source strength by the pointing error (although it was < 1%)
• No correction was done for the source being extended..
• The beam is 40 Amin, the largest sources were about 7'.
  
• Black is from the az strip, red if from the elevation strip..
• There doesn't seem to be a very large difference between the two.
• The median value for the az and el strips are also printed on the frame.
• The fluxes were taken from:
• baars et al.
• I corrected 3C461 from the 1980 fits to 2016 using:
• (.97 -.3*log(freqGhz)) % change per year
• Looks like the correction was a little to much.. the sefd came a little lower than the others
• Pictor_A flux comes from the 2nd attempt pictor_a fits by chris salter.

SUMMARY:

• The sefd was computed using the 1d fits to the az,el strips taken when making model 3A.
• The median sefds are:

• source	SEFD jy	Notes
  3C123	4711	1% of Tsys
  3C144	4807.	crab. extended
  3C274	4146	virgo A
  3C405	4429.	+40Dec Cygnus A
  Pictor A	4080	-45Dec. chris salters 2nd fit attempt
  3C461	3832	+58Dec Cas A (variable,extended) looks like time correction overdid it...

  
  
• Using the 1st 5 sources, the sefd is 4434Jy.  The measured rms is 7%.
• There is a 20% Tsys variation with el for sband.. (more info)