Beam1a Tsys 19oct11

22oct11

The galfa data from 14oct11 was used to measure the ripple in the spectrum caused by the fiber optics. On 19oct11 the fiber connectors were cleaned up stairs and downstairs.
The galfacts data taken 19oct11 after the cleaning showed that beam1a Tsys was 150 kelvins (.ps) (.pdf)

Black beam1a, red beam1b.
The computation used: calValAvgK * calOff/(calon-caloff). I used the cal value at 1450 MHz for the entire 172 MHz band centered at 1450 MHz.
The scalloping of beam1b is the Tsys variation with za. The dome was moving 2 to 19.5 deg za.

On 20oct11 the fibers for beam 1a were switched to:

Upstairs:

beam1a cable -> beam6b fiber input
beam6b cable -> beam1a fiber input (beam6b is not working)

Downstairs:

beam1a fiber receiver output -> beam6b cable
beam6b fiber receiver output -> beam1a cable

After the fiber switch the Tsys for beam1a dropped backed to around 30 Kelvin.
Looking at the cal deflection plots for a2130 from 20oct11 beam1a cal deflection now has a +/- 2% ripple that varies with za.

this was not present when fiber 1a was used.
a2130 is using a winking cal. For the caldeflection to change implies that the gain is changing (assuming tsys is constant over the .2 second winking of the cal).
At 2 and 20 deg za the turret floor is tilted by +/- 10 degs .I wonder if there is something that is shifting with the tilt of the floor causing the gain to change?

processing: x101/111019/a2130tsys.pro

home_~phil