OVSA Calibration procedures

--------------------------------------------------
step 1: Antenna pointing (important for TP & mapping)
--------------------------------------------------
	different for small and large ants
	For two 27m ants, fix one and adjust the other to match
	(called interferometric pointing)
	For small ants, interferometric pointing not possible.

For 27m dishes:
IDL> wpoint
	choose a file
	click "analyze"
	Gaussian fit is done, and offsets from the center
	recorded in a file as a function of HA and declination
	This file is called "pointing" for 27m dishes #.

	HAoffset as a fcn of HA

For 2m dishes:
IDL> wacal
	Files called "ANT POINT CAL" for 2m dishes #.
	
--------------------------------------------------
step 2: Delay scan (important for interferrometer)
--------------------------------------------------

use ARs to calculate the delay and send the delay to the interferrometer

proper delay (delay peak) brings the maximum signal
can be done in real obs. of the sun in "schedule" window
caommand line, type
> do-dlascan
file section suffixed "delayscan"
get a plot of phase & amplitude as fcn of delay (HA)
	This is equlivalent to F.T. of bandpass


--------------------------------------------------
step 3: Reference calibration (to get amp & phase)
--------------------------------------------------

- use strongest calibration (3C84) - once a month
- do 3 ants at a time --> two large ones + one small
	(one cannot use small-small)

--------------------------------------------------
step 4: Analysis of Reference calibration 
--------------------------------------------------
idl> wanalyze
	--> create a file called "headrec.sav"

--------------------------------------------------
step 5: amphit
--------------------------------------------------

idl> wamphit
	--> fitting the amplitude and phase (noisy)

amplitude & phase calibration
e.g. 1,2,4
	a_4 = sqrt(a_14 a_24 / a_12 )
For triad ijk
	phase closure: phi_ij + phi_jk + phi_ki = 0
In OVSA convention we use phi_14 instead of phi_41 (=-phi_14) so
	phase closure: phi_12 + phi_24 - phi_14 = 0
But for a hardwired reason the sum is not 0 but 135 deg. Finally
	phase closure: phi_12 + phi_24 - phi_14 = 135 is right!

Ant.-based phase: 
Define p1=0, then p14 = p4 - p1 = p4, and
p24 = 135 + p14 - p12 
    = 135 + p4 - p2
So we have all amp & phzs for the antenna system 124.

Next night do the same for 1-2-5
Next night do the same for 1-2-6
Next night do the same for 1-2-7
After 4 nights, we have 12 constratints (too much for the unknowns?)
(ant 1 & 4 are measured four times redundantly)

--------------------------------------------------
step 6:  overwrite old amp.calibration file
--------------------------------------------------

idl> apcal_add_2mant

say "yes"
"test124.sav"
*.rec (-->overwrites te current)


overwrite old amp.calibration file
amp.rec & phz.rec with newly set 124 calibration

--------------------------------------------------
step 7:  insert calibration records into general files
--------------------------------------------------

idl> insert_rec 

run for ampcal & phzcal separately
order does not matter.

--------------------------------------------------
step 8:   apcal_add_27mant
--------------------------------------------------
a difference from  apcal_add_2mant is that
there are two polarizartion states, R, L, I

--------------------------------------------------
step 8:   daily calibration - every day
--------------------------------------------------
idl> ref=daily(/interact, /noprint)

Satellite experiement: closure phase deviate from 135-->
has something to do with the DLCAL.
(Note when the src is the sun, a broad spectrum so the
phase slpoe determination error  may avaerage out. 
Fot satellite a sharp frequency spec no way to average out)

Newscan include reference and daily calibration 
to generate refcal structure
then applycal does apply the calibration factors.