[evlatests] switched power check at X band.

[Vivek Dhawan]

I had a small data set lying on disk, from April 2nd, so I compared
the standard calibration on 3C286 to the TYAPL route.

The sequence of operations is BDF2AIPS, TYAPL, SETJY, FRING, BPASS,
CALIB.

The last step gives the residual gains after switched power calibration,
if they are 1.0 the mean flux is good. I get within +_3% overall, in each
polarization and subband; no opacity correction was done. It IS true, as
others have noted, that there are individual discrepant antenna/IFs.

I looked at the worst deviants, (this was X band) and found some were from
old receivers, some not. So I'm NOT sure blaming the T_cals is the answer.

--------

Details:

* ea01 is a brand new wideband Rx. ea01B shows a factor 2 lower flux than it
   should after TYAPL, though its T_cal, T_Rx, T_sys all looke quite reasonable.
   Not sure what the problem really is.

* ea02 has an old receiver (T_cals measured in 1996) but calibrates out fine.

* ea04 also dates from 1996, is quite discrepant.

* ea19A had very ratty psum,pdif, I do not know why, but perhaps a station
   board  issue. (Stuartt Corder also had reported ea19 problem at X band).
   Pdif etc from ea19BCD were good. Visibility data from 19A were good.

* ea22 dates from 1987 and is not especially bad.

-------

  SETJY2: / Flux calculated using known spectrum
  SETJY2: BIF =  1 EIF =  2 /Range of IFs
  SETJY2: Using Perley-Butler (2010) coefficients
  SETJY2: FREQ( 1) =   8.223000 GHz
  SETJY2: FREQ( 2) =   8.823000 GHz
  SETJY2: '1331+305        ' IF =  1 flux = 5.3079 (Jy calcd)
  SETJY2: '1331+305        ' IF =  2 flux = 5.0586 (Jy calcd)

  CALIB2: Writing SN table    2
  CALIB2: RPOL, IF= 1 The average gain over these antennas is  9.849E-01
  CALIB2: RPOL, IF= 2 The average gain over these antennas is  1.024E+00
  CALIB2: LPOL, IF= 1 The average gain over these antennas is  9.867E-01
  CALIB2: LPOL, IF= 2 The average gain over these antennas is  9.736E-01
  CALIB2: Found           96 good solutions