Ripple in the lbw rcvr near 1400 Mhz
aug 04
Problem:
There has been ripple in the lband wide bandpass near
1400 Mhz during the a1861 run of may04 and during the a1914 run in aug04.
Looking at the dynamic spectra of the 1 second records, there is a broad
change in power over a few mhz for up to 300 seconds. It is not a
narrow feature sweeping across the band. Both experiments used the interim
correlator.
Data taking setups:
The a1861 data did on,off position switching using 300
1 second records. The first two sbc were polA, polB at 25Mhz while the
last two sbc were polA,polB at 12.5 Mhz. When they observed above 1370
Mhz, the 1320 hipass and 1370 hipass wave guide filters were installed.
When they were below 1370 and toward the ends of the runs, the 1370 hipass
was removed.
Project a1914 used the 1st two sbc for polA and polB
at 25 Mhz. The last two sbc were set to 12.5 Mhz bandwidth. On,off position
switching was done with 300 1 second records. For 01aug04 and 02aug04 the
1370 hipass and the 1320 hipass wave guide filters were installed.
Plots of the data:
a1861
may04 on/off-1
-
Fig 1. (on/off-1) degK vs frequency with offsets for plotting. The
top plot is polA and the bottom plot is pol b. Each color is a different
on/off-1. The bumps at 1381 Mhz are gps L3. The ripples occur at 1390 Mhz
and around 1400 Mhz. All of these on/offs have the 1370 and 1320 mhz hipass
filter installed.
-
Fig 2. The same as fig 1 except that polA and polB are now plotted on top
of each other. The ripple is in both polarizations but it is not necessarily
at the same frequency/amplitude.
a1914
: 01,02 aug04 (on/off-1)
-
Fig 1. (on/off-1) degK vs frequency with offsets for plotting. Each color
is a different on/off. The top plot is polA while the bottom plot is polB.
The ripple in pol B appears larger. All of these spectra had the 1320 and
1370 hipass filters installed.
-
Fig 2. Same as figure 1 but pola, polb are over plotted. The ripples in
the two polarizations are offset in frequency.
-
Fig 3. This is correlator boards 3,4 using the same center frequency and
12.5 Mhz bandwidth rather than 25 Mhz. The ripples are on the left edge
of this band. For a given polarization , the ripples in the same in the
two correlator boards. This says that the problem must be coming at of
before the 2nd mixer. It is not in 2nd IF distribution amplifiers or in
the correlator.
Conclusions:
-
There are ripples in the lbw system around 1390,1400 Mhz.
-
The ripples differs in phase and amplitude in the two polabrizations.
-
The same polarization going into two different correlator boards
is the same. So the problem is before the 2nd IF mixer.
-
Looking at the dynamic spectra of the ons or the offs,
-
in may04 it appeared that not using the 1370 hipass filter made the problem
go away.. this needs to be verified.
-
I've called them ripples in the bandpass, but there is nothing yet saying
that they could not be coming from some external noise source.
processing:usr/a1861/baselineinstabl.pro,/usr/a1915/blripple.pro
home_~phil