lbw receiver status jun05-sep05. 1400Mhz resonance and diode 1 dies
01sep05
The lband wide receiver has had a number of things
happen to it between jun05 and sep05.
What has happened:
-
In 2004 tsys for polB had been drifting. On 01jun05 the dewar
was warmed up and a bad cal cable in polB was found and repaired.
-
After the jun05 cool down a resonance was seen in polA at 1400 Mhz (see
01jun05
resonance in polA). This affected the daily Tsys measurements since
they are made at 1400 Mhz with a 1 Mhz bandwidth (07mar06: the warmup
actually created resonances in polA and polB).
-
On 22jul05 diode 1 failed (hcorcal). Users were switched to use diode 2
(h90cal). People using the circular option (hybrid after the dewar) got
no cal in one polarization because of the 90 deg phase shift in cal going
to polA and polB in combination with the hybrid used to create the circular
polarization.
-
On 25aug05 the observatory lost power and lbw warmed up again. After
the cool down, Tsys for lbw polA remained about 3K higher than it
was before the warm-up.
Tsys april through sept.
The daily
tsys measurements between may05 and sept05 (.ps) (.pdf)
were plotted using diode 2 (the h90cal that was not replaced). Black
is polA and red is polB. Epochs where things occurred are flagged in color:
-
Green. This is after the jun05 warm-up and cal cable repair. The the new
cal values were measured and then installed on 21jun05. The jump
in polA Tsys at 01jun05 is because there was now a resonance at 1400 Mhz
where the tsys was being taken.
-
Blue: This is when diode 1 failed. The users were switched over to
use diode 2.
-
purple: 24aug05 when the dewar warmed up. polB eventually returned to the
pre warm-up value. polA stayed 3 to 4 K warmer. This may be do to the strength
of the resonance changing after the warm-up.
01sep05 looking at the 1400 Mhz resonance
On 01sep05, cal onoff spectra were taken around
1400 Mhz with a 25 Mhz bandwidth centered on 1400 Mhz. This
was sent to 4 separate boards of the interim correlator. The lbw
spectra show the resonance
at 1394 and 1400 Mhz (.ps) (.pdf):
-
Page 1: The cal on and cal off spectra. Black is polA while red is polB.
The upper traces of each color has the cal on.
-
Page 2: CalOn - calOff spectra (with no bandpass correction). You do not
see the bump in the cal deflection so it is not coming from the cal
diode.
-
Page 3: (Calon-calOff)/normalized(caloff). This bandpass corrects the cal
deflection. It also produces the dip at 1394 and 1400 because of the resonance
in the calOff. The dip in polA is stronger than polB. The different frequencies
may be do to the physical separation of the probes.
Summary:
-
There is a resonance in polA at 1400 Mhz that is increasing the daily Tsys
measurement by up to 6 kelvins. Pol B shows a resonance at 1394 but it
is not as strong. The resonance appeared after the jun05 dewar warm-up.
It may have gotten worse after the aug05 warm up. I should check the "known"
resonances in lbw to see if the polarization's are offset in frequency
(the last test added the two pols together since a polarized source was
used).
-
Diode 1 died on 22jul05. Diode 2 (the h90cal) is being used in its place.
There is a 90 degree phase shift in the values that go to polA and polB
for this cal. Users can create circular polarization's using a hybrid that
sits after the dewar. When the h90cal value is passed through this
second hybrid, no cal will end up in one of the circular channels (90 degree
phase shift + 90 more from the 2nd hybrid cancels the signal). So users
should not use the circular polarization option for lbw until the new hcorcal
diode is reinstalled and measured.
processing:x101/050901/chkcal.pro
home_~phil