28oct10 Tsys jump
09dec10
The lbw dewar was warmed up on 26oct10 because
of a ratcheting noise coming from the refrigerator. It was cooled back
down on 28oct10. The tsys
values jumped after the cool down (.ps) (.pdf)
- This is the daily Tsys monitoring of lbw for 2010
- Top is Tsys hcorcal. black is polA red is polb. polA jumped down
by about 6 kelvins.
- Bottom: TsysA - TsysB.
- Before the jump TsysB decreased a bit vs TsysA.
- After the jump it looks like TsysA continues to slowly decease
- The change in Tsys was 10 to 15% (depending on where you measure
it from).
The change in Tsys could be caused by:
- the system temperature actually changed.
- The amount of cal getting into the receiver has changed (either
the diode or the path into the dewar could change).
To see if the cals changed, the
x102 calibration data (lbw 1415 Mhz) jul10 thru 30nov10 was used
(.ps) (.pdf):
- All the x102 calibration runs jul10 thru 30nov10 at 1415 Mhz were
used.
- The data is plotted vs calibration measurement (so the x axis is
not equally spaced in time).
- Page 1: Tsys,Gain,SEFD stokes I
- The dashed green line shows data after the jump on 31oct10. the
data to the right of the jump was actually taken on 24nov10.
- Both Tsys and the gain went down.
- The SEFD was constant and then varied a bit.
- Page 2: Tsys,Gain,SEFD PolA, polB overplotted.
- Tsys PolA jumped a lot more the PolB
- Gain: The gain for points 75-95 behaves differently
- Page 3: Tsys,Gain,SEFD ratio polA/polB
- Tsys: the ratio decreased by 10%
- Gain:The ratio Decreased by 10% for the 4 points after the jump
and at the end.
- SEFD: Closed to unity after the jump, and then variation
75-95 followed by close to 1
- The dashed blue lines enclose the source 3C138. This has a
variable gain ratio.
- 3C138 is about 7% polarized so the polA/polB is going to
change as the parallactic angle changes.
- This source also comes close to the keyhole. The
calibration patterns are not very accurate when the azimuth is moving
rapidly during the pattern.
- Excluding 3C138, the SEFD was unchanged before, after the
jump. So the problem is with the cals, and not the Tsys (since the SEFD
does not depend on the cal values.
The x102 data on 24nov10
3C138 (B0518+165) was analyzed using the full stokes processing.
- this processing computes deltaG which is an error in the
pola,polB cal ratio. It was measured to be 11.6 %. So the cal values
really have changed.
Summary
- After the warmup cooldown sequence of 26oct10/28oct10, the cal
values for lbw changed.
- The change was 10 to 15%.
processing:
x101/101208/lbwTsysDrop.pro
page_up
home_~phil