Most of us in VK5 will be well aware that the Coastal Radio Station VIA at McLaren Vale was made redundant a few years ago and purchased by Harro Krause VK5HK as his home QTH. It is a magnificent elevated site for amateur radio, well away from the built up areas and with a large aerial field of masts and transmission lines, the envy of all red-blooded radio amateurs.
As part of the installation, Harro has a beautiful vertical mast 45.72 metres high, insulated at the base and surrounded by 120 earth mat radials each 225 metres long. The original installation also had 90 metre top loading wires to four other surrounding towers. These wires have been removed but, with a bit of work, could be restored. To add to it all, there is a 75 ohm coaxial line and a 600 ohm open wire line already installed back to the radio building.
With or without the top loading, we have the ideal site for some LF transmission experiments and, if we can give him the necessary help, Harro is keen to make the site available for that purpose. Right from the earlier months of 1997, we have been looking at avenues to get an LF licence and how we might set up gear to transmit LF from the site. Initial thoughts were to use the mast as it stands (no top loading, at least for the present).
So that we could correctly couple and load the antenna at its base, we needed to know the values of the resistive and reactive component it presented. Some constants can be calculated but, in the real world, it's nice to be able to measure precisely what one has to work with.
Neither Harro nor myself had gear which could measure the constants required. Some 'farmyard' methods of measurement were tried but results were not too convincing. Also, in making measurements on an antenna at this site, there is a problem of confusing readings caused by RF pick-up from high powered MF broadcast transmitters at nearby Pimpala. The problem of inadequate test gear led me to build the LF bridge which is the subject of this article.
The bridge is based on the usual HF noise bridge circuit, which is actually a type of capacitance bridge arranged to measure both positive capacitive reactance and the reverse of this which is interpreted as inductive reactance.
However, it is different from the usual noise bridge in that the bridge components are selected specifically for the LF frequency range and the in-built signal source is a single frequency and not noise.
The single frequency is much more satisfactory to get a bridge balance. As a matter of fact, I often use a fixed frequency with my HF noise bridge for the same reason (Reference 2).
The unit was designed to operate within the range of 160 to 200 kHz, which covers frequencies used by several amateurs operating with Scientific Licences in Australia and the 160 to 190 kHz band allocated to New Zealand amateur stations. At the time of building the unit it was not known that the WIA might make application for an amateur band extending below 160 kHz to include the 135.7 to 137.8 kHz range now available to European amateurs under the CEPT agreement and later to be made available to UK amateurs.
An LF receiver is used as the null detector in conjunction with the LF bridge as is normally done at HF with the HF noise bridge. However, the output to the receiver is fed via a 160 to 200 kHz bandpass filter to protect against cross modulation at the receiver input from the broadcast stations mentioned previously and from VLF noise.
The unit can be used to measure the antenna resistance and
reactance constants to assist in the design of the coupling system.
It can also be used on the transmitter side of the coupling system
so that the system can be checked and adjusted, as necessary for
the desired reflected load. This is done without putting any signal
to air from the transmitter and, of course, this as how most people
make use of the noise bridge at HF.
| N2 - LF353
T1 - 20 tri-filar turns on a 16 mm ferrite core
with an Al factor of 1125 mH/1000T.
C2 - Adjust value for frequency range.
L1 to L6 - Miniature RF chokes (off shelf).
The LF signal is generated by a 555 timer circuit NI which is adjustable over the frequency range of 160 to 200 kHz. It is inclined to drift a bit when first turned on but it does the job. Its waveform is shaped by components L1 and C4, and drive level is set by RVI.
The bridge coupling transformer is fed via balanced driver LF353, N2. The bridge components are the 560 pF capacitor C9 in series with the unknown, 3 x 20 to 460 pF variable gang capacitor C10 with sections paralleled, and either of the potentiometers RV2 (250 ohms), or RV3 (50 ohms).
As with a lot of my projects, the unit was made from gear accumulated in my own junk box but I had some trouble finding suitable potentiometers with low enough inductance for use in RV1 and RV2. Carbon pots are really called for but in low resistance values, they are a bit hard to find. I didn't have any and I couldn't find any in the catalogues of the local electronics shops.
At these low frequencies a small amount of inductance can be tolerated and several wire-wound pots were checked out for inductance. Most of those measured did not make the grade but I found two which just did the job. The 250 ohm Colvern pot selected measured 7 uH at 250 ohms and 1.4 uH at 50 ohms. This corresponds to a reactance at 200 kHz of 7 ohms and 1.5 ohms respectively. The 50 ohm A G Naunton pot selected had an inductive reactance lower than I could resolve at 200kHz.
The reason for using the two pots switchable via SW2 was to get a better resolution at values below 50 ohms. Resolution of the larger wire-wound pot at low values is somewhat limited by the spacing between the wire turns and its residual minimum resistance value.
The bandpass filter is made up of a 160 kHz fifth order Chebychev high-pass filter combined with a 200 kHz fifth order Chebychev low-pass filter. They are designed for a circuit impedance of 50 ohms and are terminated in R11, assuming high impedance receiver input.
The unit is powered from a 12 V bank of size AA cells mounted in an eight cell battery holder.
There is one thing about working at these low frequencies. We can select integrated circuit packages which just wouldn't work at HF. Typically, the 555 timer and the LF353 amplifier have been selected.
Potentiometers RVI and RV4 are carbon types. I would have preferred that RV2 and RV3 also be carbon types but, as discussed previously, low value carbon pots were a bit hard to find.
Inductors L1 to L6 are "off the shelf' miniature RF chokes normally available from electronics stores such as. Dick Smith Electronics.
Transformer TI was made up with 20 trifilar turns wound on a 16 mm ferrite toroidal core recovered from some old gear. The core measured an Al factor of 1125 mH/1000T. From this, we deduce, that the primary inductance is 450 uH giving a primary reactance of 453 ohms at 160 kHz and adequate for the application (as a rule of thumb, the primary reactance should be at least three times the circuit impedance at the lowest operating frequency).
Of course a maker's type number cannot be quoted for the core but, if anyone is interested, Amidon types FT50A-72, FT50B-43 and FT82-72 have suitable ferrite and Al factors close to that given above.
The three gang 20 to 460 pF variable capacitor C10 is an old receiver type often picked up at amateur 'buy and sell' marts.
Most of the other components are non critical. Good silver mica capacitors are used for oscillator timing (C2) and in the bridge (C9).
The aluminium case was re-cycled from some other old gear which had been dismantled.
Calibration and Measurement Range
To be of use, the resistance dials coupled to RV2 and RV3, and the capacitance dial coupled to C10, had to be calibrated against known values of resistance and capacitance.
RV2 is calibrated in a range of 20 to 250 ohms. RV3 is calibrated in a range of 0 to 50 ohms. Positive capacitance of Cl0 is calibrated in a range of 70 to 3000 pF.
Reverse capacitance is calibrated in terms of inductive reactance at 180 kHz over a range of 0 to 1000 ohms.
Using this type of bridge (as used in the normal HF noise bridge), derivation of capacitive or inductive reactance at a given frequency from the scale as read can be a bit tricky.
For capacitive reactance, simply use the normal reactance formula, 1/2pfC where C is the capacitance read and f is the frequency.
For inductive reactance, divide the inductive reactance read by the ratio f/180kHz.
As with the noise bridge, there is a calibration where reactance is virtual zero; that is, when the capacitance side approaches infinity and the inductive reactance side approaches zero.
As this virtual centre is approached, resolution of precise values becomes cramped; the highest capacitance calibration is marked at 3000 pF and the lowest inductive reactance calibration is marked at 100 ohms.
I have described an instrument which can be used to measure reactance and resistance components in the LF range of 160 to 200 kHz. Whilst it is not a precision instrument, it can be carefully calibrated to provide useful readings of the antenna constants and be used to assist in adjusting matching to the antenna.
At the time of writing, the February 1998 issue of Amateur Radio had just been released. In that issue, a press release by the Federal Media Liaison Officer announced the intention by the WIA to revive a submission for an LF band allocation (Reference 3).
As the press release indicated an intention to include application for frequencies below 160 kHz, I must now consider the possibility of extending operation of the instrument down to lower frequencies.
The bridge would work OK at the lower frequencies but the oscillator frequency range would have to be expanded downwards and the cut-off frequency of the high-pass filter section would have to be lowered.
As the main function of the complete filter is to remove MW broadcast station interference, it is also probable that it might be satisfactory just to disconnect the 160kHz high-pass section leaving only the 200 kHz low-pass section in circuit.
1. Bob Slutzkin VK3SK - Series of articles on the noise bridge - Amateur Radio, March, April, May 1981.
2. Lloyd Butler VK5BR -Another tip for Using the Noise Bridge - Amateur Radio, December 1994.
3. WIA News, WIA Revives Submission for LF Band Allocation, Amateur Radio, February 1998.