History: (top)

• 19jan15: had left off refraction correction (see main history). corrected 19jan15
  
• 27jan15: 17:00 model 2A installed.
  
• 28jan15: 325 crosses on casA using model2A to check the model.

28jan15: model2A

Making the model

    model2A was made with model1A installed.

• model1A:
  
•  1A was a primitive model with only 2 or 3 parameters that were set by a patriot survey when the model was installed.
  
• the pointing error was .2 degrees or more in az and el.. I compensated for this by inserting great circle offsets during the crosses to keep the source partly centers on the strip.
• On 08dec14 i found a bug in the application of the refraction correction. It was being computed ok, but the offsets that were recorded in the header were left out. This will give a .05 - .1 degree error in elevation (varying with elevation)
• The sources used were:

• source	coord	flux at Sband
  3C273	J1229+02	39
  3C274 (M87)	J1230+12	122
  3C353	J1720-01	36
  3C405 (CygA)	J1959+30	785
  3C461 (CasA)	J2323+59	1495

•  
• Because of the poor pointing of model1A, the weaker sources had larger errors in their fitting.
• ... todo  add the plots showing the model fit
  

verifying the model

• 28jan15: 325 crosses were done on 3C461 (casA: 2323+58) with model2A installed
• 1d gaussians were fit to each az and elevation strips (+/- 3deg  at .1 deg/sec).
• The plots show the model2A  pointing error on casA (.ps) (.pdf)
• Top: az,el error for each cross. black azErr, red: elErr. units are great circle arcminutes
• 2nd: az,el error vs azimuth
• bottom:az,el error vs elevation
• the overall errors are:
  

• model2A errors using CasA

  set	mean Amin	rms Amin
  azErr	-.090	.72
  elErr	-.067	.91
  totErr		1.16

•  
• The next model will have the refraction correctly implemented so the el err should come down.
• The beam at sband is about 36 arcMinutes (12cm/12Meter) =1/100 of a radian..
• So an error of 1 aminute should not affect the gain much..
  

<- page up
 home_~phil