X-band receiver for the Phase 3 D Satellite



Brief description of system


The X-band receiver is based on a Ku-band satellite TVRO low noise 
converter of commercial design. Much of it has changed though. Firstly
the DRO is heavily modified into a buffer amplifier. This was a major
modification ! The input to the buffer stage is via a male SMA connector.
the input is DC blocked with a 0.7 pF high Q chip capacitor to the gate
of the GaAs-FET. The gate is at 0 DC volts and the source is grounded 
via a source resistor. The drain voltage is 5 V.

I chose the IF to be 432 MHz instead of 144 MHz ( too low ) and 1296 MHz
( too complex as I would then need a second converter ).

The mixer and the image rejection filters are brought down in frequency to
cover the 10368 and 10451 MHz bands by simply loading them with pieces of
PTFE ( teflon ). The preamplifier stages are manually tweaked with thin
flakes of silver plated metal foil. The noise figure of the converter is 
around 1.5 dB. I am planning on replacing the first stage GaAs-FET with a
P-HEMT when time permits. 

The image rejection of the overall converter is over 35 dB which is ample.

Modification of the IF amplifier was also necessary as the original IF
strip had a very sharp cutoff below around 600 MHz. The IF strip has
four broadband transistor stages and the frequency response was achieved 
originally with the DC bias circuitry. By increasing the L I could decrease
the lower frequency cutoff. There also was a series resonant circuit 
between stages 1 and 2. I simply bypassed this with a 27 pF chip capacitor.
After modification the overall gain of the converter was in the order of
45 dB's at 432 MHz and peaking at over 50 dB's at 1296 MHz.

The IF amplifier operates reliably and is stable over a large variation
of temperature. The converter can be used for 10450 MHz satellite 
operation but also for 10368 MHz and others if a scanner is used at the IF.

The original DC feed via coax was modified so that the IF connector ( an
N connector ) is DC free and the 12 V supply is fed via a feedthrough 
capacitor of 1 nF. 

The local oscillator is a modified 6 GHz phase locked multiplier ( PLM ) 
from Microwave Associates. I changed the crystal to 104.3541667 MHz. This 
is a 60 degree C oven crystal and I am heating it with a PTC resistor that
stabilizes at this temperature with the -20 V supplied. The VCO is running
at 1669.6667 MHz as is locked at the 16th harmonic of the local oscillator.
The VCO frequency is multiplied by six with a step recovery diode multiplier
followed by a digital filter. I shortened the filter resonators in a milling
machine in order to get the filter to tune to 10018 MHz. Tuning is simple and
multiplication is reliable with no parametric oscillation observable. The
SRD is loaded via a small attenuator so it sees a fairly resistive load during
alignment. The multiplier output level is +13 dBm and is fed to the converter
buffer via a small attenuator.

The whole system is built into an Eddystone diecast box with a waveguide 
input on the end wall. The waveguide input to the converter is depolarized 
from circular to liner by a TVRO depolarizer ( this is also modified from 
LHCP to RHCP by rotating the Rexolite delay line in the circular waveguide
by 90 degrees ).

The diecast box is attached directly to the feedhorn of a 90 cm prime focus
dish and is light enough not to require any other support than the original
tripod mount of aluminium spokes.

Haven't measured the sky/ground noise ratio yet, because I still need to 
build a +12 V to -20 V inverter into the box to enable relaxed portable
operation with 12 V only...

(well, since writing this I have certainly built the converter and used
the receiver in a few experiments. The stability is quite good even without
a crystal oven and the sky/ground noise level is very easily heard with
my FRG9600 receiver and using the loudspeaker. Sun noise is several dB
with a 90 cm dish - easily copiable with the ear - 14.02.2000/MF)

Updated 13.04.1998