This is an excerpt from a detailed description of a solid state communications receiver based on the Wadley loop principle. It was published as a series of articles in Electronics Australia from January 1971 to May 1971. The author of the construction article was the late Ian Pogson. Part one of the series is presented here because I think it explains the Wadley loop very well. You will note a strong similarity to the RACAL block diagram. Applying this description to the diagram in the RACAL manual should not prove too difficult.
"When the writer was in the United Kingdom around 1959, a receiver made by the "Racal” organisation was brought to his attention at an exhibition. In common with many other receivers, this one is based on the familiar super heterodyne principle. However the actual system employed is a development of the basic superhet principle, and is so ingenious that full marks for originality must be awarded to the originator, Dr Wadley, of South Africa. The principle is now known as that of the "Wadley Loop". The Racal receiver covers continuously from 1MHz (it is possible to go lower) to 30MHz, in steps of 1MHz, with remarkable frequency stability. Yet it uses only one crystal. In effect, the Wadley Loop principle produces the equivalent of 29 crystal-locked converters, which may be accommodated in little more space than is normally occupied by one conventional crystal-locked converter.
The system functions in the following manner, which may be understood by reference to the block diagram below.
Incoming signals, in the range 1-30 MHz are selected by the RF amplifier, which is switched and tuned in the same way as in a conventional full coverage receiver, The RF amplifier is designed so as to have as uniform gain as possible over the complete range to be covered. However, while the stage is normally peaked on the desired signal, adjacent signals will be passed to an extent, depending upon its selectivity.
After amplification by the RF stage, the signals are passed to the first mixer. The output from the mixer feeds into a band-pass filter having a flat top centred on 40MHz. The bandwidth of the filter is ±650KHz, the reason for which should become apparent later. The signal which passes through this filter may be considered to be the first IF of the receiver.
The other signal fed to the first mixer is the output from the VFO, The frequency of the VFO is tuneable over the range which extends from 1 MHz above the lower limit of the band-pass filter, to 30MHz above this frequency: from 40.5MHz to 69.5MHz. It is the VFO frequency which therefore, by heterodyne action, determines which particular 1 MHz band of input RF signals is "translated" into the (40MHz ±500KHz) IF range accepted by the band‑pass filter. The VFO tuning control thus becomes the 1MHz selector of the receiver.
After passing through the band-pass filter, the IF signals derived from the chosen 1MHz band is fed to one input of the second mixer. The function of this mixer will be examined in a moment. At this stage it should be noted that the operation is very similar to that of a normal super heterodyne receiver, with some change in the IF and its bandwidth.
In another part of the' receiver, however, signals are produced by a 1MHz crystal oscillator and harmonic generator, and fed through a low‑pass filter with cut-off at 32MHz. The resultant series of harmonics are fed to a third mixer.
The output of the VFO is also fed to this mixer and heterodynes with the harmonies from the crystal oscillator. The heterodyne, in which interest centres, is that which happens to fall, in the course of VFO setting, within the pass-band of a second substantially flat topped band-pass filter on 37.5MHz ±150KHz.
It is the amplified signal on 37.5MHz which becomes the injection for the second mixer. This frequency, when heterodyned against the IF signals passed through the 40MHz band-pass filter, results in an output on 2.5 MHz ±500KHz, or in other words 3-2MHz.
This becomes the tuneable IF, which is passed on to the rest of the receiving set-up. It is in the following tuneable IF section that the individually desired signals are selected, the preceding circuitry having served only to change the incoming signals from their original frequency to within the 2-3MHz range.
And here an important point should be made in relation to the frequency stability of the system.
Any frequency drift in the VFO will cause a change in the frequency of signals passing through the 40MHz band-pass filter. Simultaneously, there will be a deviation from the nominal 37.5MHz signal which is injected into the second mixer. These differences exactly cancel at the second mixer and consequently the tuneable IF is not changed. In fact, no adverse effects occur provided that the VFO resultants stay within the pass band of the two filters.
In effect, the only independent frequency determining device in the front end is the 1MHz crystal oscillator. The net result is a system with a very high degree of stability. In a complete receiver, the overall frequency stability is virtually determined by the local oscillator used in conjunction with the tuneable IF. And with careful design, it is possible to obtain good stability here as well.
Some practical examples will now be given to illustrate the foregoing description of operation. Consider that it is desired to receive a signal on exactly 21MHz. The RF stage will be set to pass this frequency. The VFO is set to 61.5MHz and this will beat with the signal on 21MHz to give a difference of 40.5MHz, which passes through to the second mixer.
The 61.5MHz from the VFO will also beat in the third mixer, with the 24th harmonic from the crystal to give a difference of 37.5MHz. This is passed through the 37.5MHz amplifier-filter, to the second mixer. The 40.5MHz and 37.5MHz produce in the output of the second mixer, a difference of 3MHz. This is acceptable to the tuneable IF at one end of its range.
Now, taking a second look, assume that the same 21MHz signal is to be received, but this time, with the VFO set to 60.5MHz. Obviously enough, this time a difference frequency of 39.5MHz will be produced, which is still passed by the filter to the second mixer.
The 60.5MHz from the VFO will now beat in the third mixer, with the 23rd harmonic from the crystal to give a difference of 37.5MHz and so to the second mixer. The 39.5MHz and 37.5MHz produce, in the output of the second mixer, a difference of 2MHz.This is also acceptable to the tuneable IF, at the other end of its range.
From these two examples, it is evident that each integral megacycle point is tuneable at both ends of the scale, which end being determined by the setting of the VFO. The exceptions to this are those frequencies at the start and finish of the overall coverage, normally 1MHz and 30MHz.
The dial of the VFO, which is the "MHz" selector, is not calibrated according to its actual frequency but rather to that band of frequencies which the receiver actually tunes. For example, 41.5 is calibrated as 1MHz, when I-2MHz is covered; 42.5 as 2MHz, when 2-3MHz is covered, and so on.
To illustrate the effect of slight mis-setting, or drift of the VFO, the first example will be referred to again. The 21MHz signal is to be received, but the VFO is either set to or has drifted to 61.6MHz. This gives a difference frequency of 40.6MHz from the first mixer. Now, although the band-pass filter is only required to pass 40MHz ± 500 KHz, it may be recalled that it has been made slightly wider, ie, ±650KHz. This provision allows the 40.6MHz signal to pass to the second mixer.
The 61.6MHz from the VFO beats in the third mixer, with the 24th harmonic from the crystal oscillator to produce 37.6MHz. It has been pointed out already that the amplifier-filter is on 37.5MHz ± 150 KHz. Therefore, the 37.6MHz signal will still be passed to the second mixer. The 40.6MHz and 37.6MHz signals are heterodyned together to give a difference of 3MHz at the output of the second mixer. This is exactly the same as the resultant frequency at this point as given in the first example, demonstrating the way in which drift in the VFO is automatically cancelled.
Should the VFO drift by an amount which runs the signals concerned down the skirts of the two filters, the frequency of the received signal will not be changed, but the signal strength will drop due to attenuation in the filters. This must be avoided for obvious reasons. In practice, the VFO can be made adequately stable and no such trouble is experienced. From the foregoing details, this system may be seen to offer many desirable features, which include: ‑
Continuous frequency coverage from 1-30 MHz.
Constant tuning rate.
Reasonably good band spread, ie, I MHz per sweep of the dial.
High order of frequency stability.
Excellent repeatability due to electronic rather than mechanical band switching.
Many other features could be added to the list. However, another point is worthy of note. Given a high grade dial on the tuneable IF, which has only to cover 1000 KHz and which effectively: "interpolates" between the settings of the MHz dial, the frequency of the received signal may be read off quite accurately. This is a very good feature where it is necessary to pre-set the receiver in readiness for an expected signal. As may be expected, a price must be paid for all the advantages which this system provides. Efficient low-pass and band-pass filters are required, in order to obtain correct operation. In addition, considerable care must be taken to shield various sections from each other, to prevent the harmonics from the 1MHz crystal from being injected and received as incoming signals. Power supply leads must also be filtered, where necessary, for the same reason.
Be this as it may, it is a small price to pay for the facilities which are available. No amount of writing can illustrate the delightful ease with which tuning can be carried out. A few minutes at the controls are all that is necessary to convince the operator. "