A bad cable was found in wapp 4 brd 1 channel B on 07mar07.
The cable went from the A/D digitizer box up to the filter box.
Wiggling the cable caused the attenuation to change by many db.
Inspecting the cable, the connector at the end would spin around (rather
than being rigid).
What did it affect:
For alfa observations this was beam 6B (counting from zero). For these
users, the 6B data is duplicated in the last band 7B ( alfa has 7 beams
and the wapps return 8 band passes). The 7B data did not have any jumps
since the data did not go through the bad cable. You can replace
the 6B data with the data found in 7B.
For single pixel observers, wapp 4 polB will be jump around. There is no
duplicate data to replace (as in the alfa case).
The worse day was 19feb07 where polB was changing by over 10 % of Tsys.
When did it occur.
Looking at the a2010
alfalfa total power ratio PolA/polB (check out the feb07
.pdf file in the column Strips: badRatios rms A/B). Look at the
second page at the bottom).
The ratio polA/polB was stable for jan07
The ratio started to jump around at the start of feb07.
The worse day was 19feb07. The severity decreased on 20feb07. By 23feb07
it did not jump very much.
The attenuation of the cable must have been changing when the temperature
variation in the room caused the bad connector to expand/contract.
The Fix:
The cable was replaced on 07mar07 after it was isolated.