LF-HF Reflection Coefficient Bridge
This circuit is based on the design by Hans Nussbaum DJ1UGA translated in Rad Com 9/2000. The bridge can be used to measure the reflection coefficient of an unknown load. The usual 'unknown load' is an antenna. This version, with increased LF coverage, has been tested from 73kHz to 30MHz.
Circuit
The circuit below comprises of the bridge - 3 resistors and the balun
transformer and a wide bandwidth amplifier built around a monolithic microwave
integrated circuit (MMIC) The amplifier is needed in my case as the drive source
- The SM6LKM DDS source must be amplified to about 3V P-P. It is necessary to
drive the bridge with at least 20mW to mask any effect from signals that are
picked up in the antenna from high powered transmitters. A VFO with 20mW output
would also be suitable. The output from the SM6LKM DDS covers 1Hz to 6MHz. The
IC is a class A amplifier. It has a flat response from DC to over 500MHz and can
provide 50mW output up to 1000 MHz.
The balanced to unbalanced transformer is a small 15mm 3C85 core
with 23 bifilar turns of 0.4mm enameled copper wire.
The 82 Ohm resistor on the input to the amplifier is only necessary to attenuate
the output of my DDS to prevent overloading the MMIC. The 110R resistor shown is
selected to drop the supply voltage to 5.0V at the supply pin of the I.C. If you
have a higher supply voltage.....Use a higher value resistor.
Operation.
The bridge circuit has an input connected to the DDS or VCO/VFO and an output to
an oscilloscope or a diode probe and moving coil meter.
The third port is for the unknown load.
When the load is a resistive 50 ohms the bridge will be balanced and the output
to the meter / scope will be close to zero. If the load is above or below 50
Ohms the bridge will be unbalanced and there will be an output to the
meter/scope.
The ability to give a deep null with a perfect 50 Ohm load is dependant on the quality of the bridge. This will depend on the accuracy of the resistors, the construction of the transformer, and its losses due to stray and interwinding capacitance etc. Quality of the bridge is measured by the difference between the output voltage when the unknown port is terminated in 50 Ohms and when it is open or short circuit.
For the circuit constructed as described the following measurements were taken
73
kHz 20dB
100kHz 23dB
137kHz 26dB
200kHz 28dB
500kHz 37dB
1 MHz 42dB
5 MHz 46dB
30 MHz 29dB
The unit is particularly useful when first setting up an LF antenna. With the
antenna and matching network attached to the unknown port, the frequency is
swept across the range of interest and any dips in the output noted. By adding
or removing loading inductance the dip can moved to the correct frequency. The
matching can then be optimised by watching for the deepest dip in output as you
change tapping points or transformer ratios etc. With this instrument it's
possible to adjust a new antenna in a few minutes whereas using a transmitter
and VSWR bridge may not show any indication of resonance until you have the
matching very close to optimum.
Using the unit on HF is also very interesting. A Cushcraft R5
vertical was swept with a VCO covering 10 - 30MHz. The VCO voltage was generated
from the timebase of an oscilloscope while the bridge output was connected to
the X input of the oscilloscope.
The display continuously showed the 5 dips corresponding to resonance on the HF
amateur bands. Degree of matching, frequency and bandwidth can be seen at
a glance!