Bandpass ripple in gregorian 430Mhz system
dec,2001
Some time in dec01 a large ripple appeared in
the gregorian dome 430 Mhz data. It did not look like interference because
it was relatively stable. While debugging the problem, it was discovered
that turning on the gregorian dome motors caused the ripple to appear.
Turning off the dome motors converted the ripple into a large tone at
435.75 with a width of about 200 Khz. The
figures show the birdie and the ripple.
Top plot, dome motors off. Birdie at 435 is about 10 times tsys in polA
and 3 times Tsys in polB.
Middle plot dome motors are on (but telescope is not moving). Ripples appear.
Bottom plot. 1 second dumps of the bandpass with .1 offsets added between
each strip. Black lines have dome motor off. Blue lines have dome motor
on.
When the azimuth or carriage house motors were turned
on, the narrow birdie spread out a little (say 400 khz) but no where near
the modulation when the dome motors were on. We tried turning off 1 dome
motor at a time but the modulation did not go away. With all of the power
in the vertex drive shelter turned off, the large narrow spike remained.
We worked on the shielding of the cables that ran from the vertex drive
shelter (the motor amplifiers) down to the motors, but the problem persisted.
And then we lucked out... The dome motors were off
and we were monitoring the large spike at 435 Mhz with 1 second dumps.
All off a sudden the narrow line changed to the modulated form for a few
seconds and then hopped back to the large narrow birdie. After a little
investigation we found that the operator had typed in the command to read
the dewar temperatures and this caused the modulation to appear!!!
Each dewar has a lakeshore temperature sensor that
can read the dewar temperature. Cables run from each dewar to a multiplexor
box in a separate rack situated on the rotary floor. This multiplexor can
connect these signals one at a time to a digital volt meter that reads
the voltage (temperature). Turning the receiver monitor multiplexor
box off and then on caused the birdie and modulation to go away. After
cycling the power we could not make the tone or the modulation appear.
It's still a bit of a mystery why the dome motors were able to cause
the birdie generated by this box to switch from a tone to a modulated signal.
Turning the motors on may have caused some vibrations (although you didn't
have to move the dome for the modulation to appear) or the motors
turning on could have caused the voltage to sag a bit causing the tone
to switch to a modulated wave form.
Whatever the reason, the problem has gone away but
has not been fixed. If a birdie at 435.75 Mhz appears or you see large
modulation in the dome 430 Mhz receiver, check out the receiver monitor
multiplexor box.
processing: x101/011226/doit.pro
home_~phil