UK Vintage Radio Repair and Restoration

This discussion will be based solely on super-heterodyne (superhet) circuits. Earlier sets used TRF (Tuned Radio Frequency) circuits, with reaction to increase the gain and selectivity. For information on this type of circuit click here.

AM stands for Amplitude Modulation, and this is the method used for MW, LW and some SW transmission. The carrier is a fixed frequency (the frequency you tune the set to), and the amplitude (voltage level) of this is varied by the audio signal being transmitted.

The MW, LW and SW dials of virtually all British valve radios will be marked in wavelengths (metres) rather than frequencies. To convert from one to the other, divide 300,000 by the known figure (this conversion works both ways). The result should be rounded up or down to the closest whole number.

Thus 300,000 divided by 1215KHz gives 247 Metres. Conversely, 300,000 divided by 247 Metres gives 1215KHz. This confirms that Virgin 1215 is received at the 247 Metre point on the dial (the former position of the Light Programme on MW).

As noted in the table, the wavelengths used by current UK broadcasts are the same as those allocated after the war (of course the stations are different now) so the stations still line up with indicated stations on the dial.

This sort of information can be useful for checking that the tuning is reasonably accurate. It is worth calculating and noting down the wavelengths of the stronger stations in your area for this purpose.

This table gives the main frequencies and wavelengths of the national stations broadcasting in the UK, on MW and LW. Some of these stations may have additional relay transmitters on different frequencies in some areas.

The Light Programme on LW used to be at 200kHz or 1500 Metres. This was changed when the wavebands were revised in the 1970s to bring it into line with the 9kHz spacing between allocated wavebands on MW and LW. The difference is small, and BBC Radio 4 will still appear at the Light position on the LW dial.

Please note that kHz is a modern term, meaning kilo-Hertz. Period literature and service sheets may use the term kcs, meaning kilocycles per second. Both values are identical - 1 Hertz is 1 cycle per second. Kilo-Hertz means one thousand Hertz. "Hertz" is used as a measurement of frequency in honour of Heinrich Rudolf Hertz, whose experiments with electromagnetic waves led to the development of radio.

The IF and RF circuit of the Philips B2G81U (Cossor CR1201U) radio is shown here. This low-cost set has MW and LW bands only. The various sections of the waveband switch are marked m (MW) or l (LW) and are closed when the set is switched to the band indicated.

L1 and L2 are the coils on the ferrite rod aerial. S1 and S2 select the appropriate one for the waveband selected. The aerial is tuned by C4, with C3 being a trimming component. The received signal passes to the control grid of V1b.

V1a is the local oscillator, which is tuned by L5, L6 and C10 to C14. C14 is ganged with C4, so the oscillator frequency varies as the tuning is adjusted. S3 and S4 are the wavechange switch. The oscillator frequency is (in this case) 470KHz above the tuned frequency.

V1b is the mixer stage (the type of combined valve used for V1 is often referred to as a "mixer-oscillator" or a "frequency changer"). This mixes the oscillator signal with the received signal and produces four output signals on the anode. These are the two original input signals, the oscillator frequency plus the received frequency, and the oscillator frequency minus the received frequency. Modulation on the received signal is also present on these last two outputs.

Since the oscillator is always 470KHz above the received signal frequency, the oscillator minus received signal will be at 470KHz, regardless of the tuning setting. This is the IF (Intermediate Frequency).

The IF varies between different sets, but 465KHz is the most common value. 455KHz and 470KHz are used sometimes, and are fairly common on transistor equipment. The service sheet for the set will give the frequency used, but you only need to know it if you intend to realign the set.

If noise is heard which alters in note and volume as the set is tuned across the band, with the increase coming towards the low frequency (high wavelength) end of MW and the high frequency (low wavelength) end of LW, this indicates that the local oscillator is probably not working.

Measure the voltage on the anode of the oscillator section. If the oscillator is running the voltage will be between about 50 and 150V. If it is not oscillating, faulty the voltage will be much lower, probably around 10V.

To confirm that the local oscillator is at fault, tune the receiver to a position on the dial where you would expect to receive a strong station (such as Radio 4 on LW or a local station on MW). Connect a signal generator to the control grid of the mixer-oscillator valve, and tune the generator across a band of frequencies around 400KHz to 500KHz higher than the station frequency (for example 600KHz to 700KHz for Radio 4). This simulates the action of the local oscillator, and if a station is heard it proves that the local oscillator is indeed faulty.

To obtain good selectivity and sensitivity, several stages of IF tuning and amplification are required. Coupling between the IF amplification stages is by IF transformers. These are tuned on the primary and secondary windings, giving two stages of tuning.

In this set, the first IF transformer is L3/L4 which is tuned by C6 and C7. V2 is (amongst other things) the IF amplifier, which is followed by the second IF transformer L7/L8 tuned by C17 and C18.

One stage of IF amplification and four stages of tuning is adequate for good reception in most cases, and is the arrangement used in most sets. Additional stages can be added, and will result in improved performance and sound quality, particularly when trying to pick out a weaker station which is close to a strong signal. Some higher cost receivers, and many Bush AM/FM sets, were fitted with two IF amplifier stages, and hence had three transformers giving six stages of tuning.

Voltage checks are the most useful method of checking the IF stages. The anode of the IF amplifier valve will be fairly high, generally only a few volts lower than the HT rail feeding that part of the circuit. Applying the meter should give a crackle from the speaker if the stages following this valve are in order.

The voltage on the screen grid varies widely in different designs, but anything below about 60V is cause for suspicion. The control grid should be between 1V and 3V negative relative to the cathode with no signal present. This may be derived by various means but the most common is a cathode resistor.

The voltages on the mixer section of the mixer-oscillator should be similar to those mentioned above. If a crackle is heard when measuring the mixer anode voltage, the IF amplifier is probably OK.

The advantage of superhet circuits over earlier TRF arrangements is that the signal passes through several stages of very sharp tuning, which gives superb selectivity and sensitivity. Since the IF is at a fixed frequency, the IF tuning is fixed. This is much more straightforward and reliable than attempting to keep four or more variable tuning stages in line with each other across the waveband.

The function of the detector is to extract the audio signal from the IF signal (a). This is carried out by a diode, which strips off the bottom half of the IF signal (b). The result is then passed through a low pass filter to remove the IF, leaving the audio intact (c).

This passes to the volume control, and then on to the amplifier which has been described previously.

In the Philips circuit (repeated here), the detector diode is contained in V2. The anode of the diode is shown to the right of the shared cathode, which is at the bottom centre.

The IF is filtered out by C19, leaving the audio signal present across the volume control (not shown, to the right of R10). There will generally be additional IF filtering components in the audio amplifier stages.

The second diode in V2 is unused in this set, so the relevant pin is connected to the cathode. In better quality sets it would be used as a separate AGC diode as detailed below.

The purpose of the automatic gain control (sometimes called automatic volume control, or AVC) is to ensure that all stations, regardless of the signal strength, give a similar output level. Without AGC, the stronger local transmissions would come through much louder than programmes that are more distant. Fading at night would also be more pronounced.

In the Philips set shown above, the AGC voltage is developed by the same diode in V2 as is used for detection. Since the cathode is at 0V potential, the positive peaks of the IF will be held at this level due to the diode action. The negative peaks will be at some level below 0V, depending on the signal strength. This is smoothed to an average DC level by R9 and C2. Therefore as the signal level increases, the negative DC potential on the junction of R9 and C2 increases.

The gain of V1a and V2 can be varied by altering the biasing voltage on the control grid. This sort of valve is often referred to as vari-mu, and is sometimes denoted by a diagonal arrow through the symbol (this indication is not universal though, and is not shown on this circuit).

The AGC voltage is used to set the control grid biasing of V1a and V2. Therefore, as the signal strength tends to increase, the negative AGC voltage increases. This increases the control grid biasing voltage on V1a and V2, reducing the gain, which compensates (largely) for the increases signal.

The AGC voltage is at a very high impedance. Any leakage in the AGC decoupling capacitors (C2) will reduce the AGC level. This will give excessive gain for the signal level received, resulting in possible distortion and instability. It is normally worth replacing all the AGC decoupling capacitors if the set is prove to noise and whistles, or the performance does not seem quite right.

The IF and detector stages of the Bush DAC90A are shown here. This uses separate diodes for the detection and AGC. Detection is carried out by the diode whose anode is shown to the left of the shared cathode of V3, and operates in the same manner as described above. R7 is the volume control and C13 is an IF filter. Note that V3 is also the first stage of AF amplification, while V2 is the IF amplifier.

The AGC voltage is developed across R11 by the other diode in V3 (with the anode shown to the right of the shared cathode). The IF is picked off from a tapping on L7, and coupled to the diode by C14.

The use of separate diodes for detection and AGC gives better performance, particularly when receiving weaker stations, than using a single diode for both. This arrangement was used on many AM sets for this reason. In most VHF (FM) sets the single diode arrangement is used on AM. However, the performance does not suffer as much as might be expected, possibly because the valves used are optimised for this method of operation. Also most people would use these sets mainly for listening on VHF, so the designers were probably not that concerned about getting the best possible reception on the other bands.

My previous description of AGC Delay was inaccurate and has been removed. Two contributors have described it properly for me!

The AGC delay is often obtained by placing a resistor (R6) in the cathode circuit of the AGC diode, giving a positive voltage on the cathode. Thus, the AGC voltage will remain at zero until sufficient signal is received to overcome the cathode offset voltage. Resistors are placed in the cathodes of the mixer-oscillator and IF amplifier valves to give correct biasing when the AGC voltage is at zero.

The majority of sets will have circuits similar to that outlined above. On sets having several wavebands and/or preset tuning, the circuit will appear to be much more complex. This is just due to the complex switching in the aerial and oscillator sections.

Earlier sets do not have ferrite rod aerials. Some used an internal frame aerial, but the majority relied on an external aerial. This is connected via a socket on the rear of the set, which is connected in a similar manner to the one on the Philips. An earth socket is also provided, which is connected to the chassis. On AC/DC sets the aerial and earth sockets are connected via capacitors, to isolate them from the mains voltages.

If you are repairing a set of this type, you will need to arrange a suitable aerial otherwise very little will be received. The signal level available now is much greater than that available when the set was made, due to the proliferation of local stations and relay transmitters. Consequently, a simple wire aerial will generally give good results. A three metre (ten foot) length of 5 amp wire connected to the aerial socket will generally be adequate, although you could arrange a longer and more permanent aerial if you are repairing sets regularly. Single-core or 7-strand cable seems to work better than multi-strand flexible wire. I have a length of wire going up through the workshop ceiling and tacked to a few beams in the loft, which works very well.

In some higher quality sets, there is an RF amplifier stage between the aerial and the mixer-oscillator. This is common on high quality SW receivers, and gives an improved signal-to-noise ratio on weak signals.

The RF and IF circuits are generally fairly reliable. As mentioned previously, leaky capacitors in the AGC circuit can give rise to instability.

On some sets the screen grid of the mixer-oscillator is driven via a resistor (R2 in the Philips) or potential divider, and these resistors sometimes go high or open circuit.

Decoupling capacitors are frequently fitted between the screen grids of the valves and ground. Leakage here can affect the biasing, and possibly damage the related resistors. The failure of any decoupling capacitor can cause instability or distortion.

If the tuning crackles or is dead towards the high wavelength (low frequency) end of the scale, the fixed and moving vanes of the tuning capacitor could be touching. This is often caused by the vanes being slightly bent or damaged, or sometimes by dirt between the vanes. Crackles can be caused by dirty contacts to the moving section, which can often be resolved by a little contact cleaner on the slip contacts and pivot points. Take great care not to get the cleaner between the moving and fixed vanes, as this will drastically alter the capacitance and is difficult to clean out.

The position of the moving vane assembly relative to the fixed vanes can be adjusted by means of a grub screw and lock nut, which are sometimes situated on the rear of the component. Do not adjust this unless it is absolutely necessary, as it is easy to make things worse! Seriously damaged tuning capacitors cannot be repaired, and should be replaced with a unit salvaged from another set.

The most common problems are with wavechange switches. These can suffer from dirty, bent or even broken contacts, which will render the set either very noisy or silent on one or more wavebands. Many problems can be solved with a little contact cleaner in the right place and possibly some gentle prodding to tighten the contacts.

Problems that are more serious can often be overcome by removing and dismantling the switch and cleaning the parts thoroughly with isopropyl. Before disconnecting or dismantling anything make some notes or sketches showing how it all goes back together.

On some sets with a gramophone input, a section of the switch is used to remove the HT supply to the RF and IF section. This switch section is prone to tracking due to the voltages involved. The effect is a continuous crackling or rustling sound, and the appropriate section can sometimes be seen to be arcing. The wires can be disconnected from the switch and permanently joined together, but some radio breakthrough may occur when the set is switched to the gramophone setting (since the gramophone input is unlikely to be used this is unlikely to cause problems in practice).

In the event of more serious switch failure, the only solution may be a complete replacement. This is a major undertaking however, and should only be considered when all the alternatives have been ruled out. Rotary switches can sometimes be replaced with a maka-switch type component, where the mechanism and wafers are purchased separately and assembled as required. With other types of switches, the only option may be to obtain a similar second-hand component from another set or a dealer.

You may be able to attach a modern push-button switch to the existing mechanism, and transfer the appropriate connections to this. If all else fails, you may be able to get the set working on one waveband by disconnecting the defective sections, and permanently wiring the circuit as required. This is definitely a last resort however!

VHF transmissions use Frequency Modulation (FM). The carrier frequency has a base value, and this is varied by the audio signal. The amplitude remains constant.

Because of this, the detector circuit is arranged differently. The tuning arrangements are also different to accommodate the much higher frequencies, and the IF is increased to (usually) 10.7MHz.

The circuit (up to the volume control) of the Ekco C273 (also A274 and A277) radio is shown here. This particular circuit was chosen because the set is FM only, so the complications of including the AM bands and the waveband switching are avoided (for now).

V1a is an RF amplifier. Its purpose is to increase the weak signals received by the aerial to a level suitable for the mixer-oscillator. V1a is configured in common grid mode, with the input to the cathode and the output from the anode. This gives a low input impedance, to match the 80R nominal impedance of the aerial. The anode load is tuned by means of variable inductor L3.

V1b is the mixer-oscillator. L4 and C7 set the oscillator frequency. L3 and L4 are variable and form the tuning control. C5 and C6 pass the RF signal to the mixer-oscillator. The anode load of V1b is the primary of the first IF transformer (L6).

The IF is almost invariably 10.7MHz, although different values were used on a few earlier VHF sets (particularly those made before VHF broadcasts started). The figure will be given on the service sheet for the set in question.

In some sets, the tuning is adjusted by variable capacitors rather than variable inductors. The VHF tuner section (everything up to and including L6) is normally contained in a separate screened casing. This is sometimes referred to as the FM front-end or FM tuner-head.

V2 and V3 are the IF amplifiers, which operate in the same manner as that described for AM above. The IF transformers are designed for the higher IF frequency, and are generally tuned to give a slighter flatter response than those used on AM.

On sets having the AM bands in addition to FM, there will be two IF transformers between each stage of IF amplification. The primaries and secondaries of the two transformers will be connected in series, and contacts on the waveband switch are generally used to link out the unused transformer, particularly in the earlier stages. The oscillator section of the AM mixer-oscillator will be disabled, and the mixer section used as the first FM IF amplifier. This is generally followed by one further IF stage before detection, although some sets (in particular many Bush sets) will have two.

The FM detector is more complicated than its AM counterpart. The secondary of the final IF transformer (L11) is centre tapped. The two ends are connected to diodes (V4a and V4b), which point in opposite directions. The operation of the ratio discriminator is fairly complex, and I have attempted to simplify it in the description below. Fortunately, a detailed understanding is not necessary to repair sets!

The IF transformer is tuned to the centre frequency, and C28 will be charged by the rectified signals from the diodes to a mean level. As the frequency varies due to modulation, it moves above and below the centre frequency of the transformer. The IF level to the diodes drops as the frequency moves further from the centre frequency. Since the level on C53 will not respond to these rapid variations, the diodes will pass less current and the loading on the transformer will be reduced.

This varying load will be reflected by a varying level (the audio signal) on the additional winding (L12). Since this is connected to the centre tap of the transformer secondary, IF signals will be cancelled out to a great extent. C24 filters out any remaining IF before the signal passes to the de-emphasis circuit.

Correct operation of the circuit relies on precise adjustment of the final IF transformer. If the adjustment of any of the IF transformers has drifted or been altered the sound will be distorted.

The de-emphasis circuit comprises R10 and C25. Its purpose is to correct the high frequency boost given to the signal before transmission. (It also suppresses the 19KHz pilot tone that is transmitted to indicate a stereo broadcast, although this would not have been considered at the time). The signal then passes to the volume control (R15) via C26.

The mean level developed across C28 is used as the AGC control voltage. This varies the gain of V3 only, by varying the potential on the suppresser grid. The gain does not need to be controlled to the same extent as with AM, because it is frequency variation, and not voltage level that affects the output level.

Many VHF sets have circuit arrangements very similar to the Ekco, and use a similar valve line-up. Earlier VHF sets used a different circuit arrangement and valve types. The dual triode (ECC85 or UCC85) in the tuner assembly was replaced with two RF pentodes (EF80 or similar), with the first RF stage being configured as a common cathode arrangement. The IF amplifier used an EF86 or similar pentode valve. This valve gave lower gain than the later EF89 at the FM IF frequency, so a different value cathode resistor was often switched in to compensate. In a very few cases (such as the Bush VHF41), the IF was higher, 19.5MHz being a figure I have seen.

The detection diodes are normally included in the same valve as the audio amplifier stage. This valve (UABC80 or EABC80) contains three diodes (one is used for AM detection on AM/FM sets) and one triode amplifier. Some sets use a separate double diode valve (e.g. 10D2) or metal rectifiers. This is followed either by a combined triode amplifier and pentode output valve (in cheaper FM only sets), or by a double diode and single triode valve, with the two diodes used for separate detection and AGC on AM.

Apart from the faults detailed earlier, the main problem again comes down to capacitors. If some of the decoupling capacitors are suspect, it is possible for a set having AM and FM to work fine on AM but be unstable on FM.

The comments about arcing on the gramophone switch also apply to the sections that remove the HT from the tuner unit when the set is switched to AM. This cannot be linked out as AM reception will be seriously impaired, so one of the approaches given previously should be considered.

Since the tuner unit is a separate assembly, sometimes on rubber mounts to reduce the effects of vibration, it is possible for one of the connections between this and the main chassis to be broken. This would result in the set being totally dead on VHF.

Valve condition has a greater impact on VHF reception than on AM. If in doubt, get all the valves tested. The types of valve used in V1 position (normally ECC85 or UCC85) are sometimes noted for their relatively short life!

Valve VHF sets are not as sensitive as modern equipment, so you will only be able to receive the stronger stations clearly unless you use an external VHF aerial. You should be able to get good results on the BBC national stations in most areas, and reasonable reception of the BBC and ILR local stations. Stations that are more distant will give poor results. Note that the VHF scale only extends to 100MHz, so you may not be able to receive Classic FM and some local stations. Sometimes it is possible to adjust the alignment slightly to bring in Classic FM at the top of the band without losing BBC Radio 2 off the bottom, but the calibration will then be inaccurate.

In some cases, significant distortion will be heard on some stations, particularly the local commercial programmes, even though the set is correctly aligned. The reason for this is that these transmissions use a greater amount of modulation than the earlier transmissions. The increased use of audio compression also means that the sound is louder and the modulation reaches the limits much more often. The problem will not occur on the BBC stations, apart from possibly BBC Radio 1, because they use higher quality transmitters and less compression, and care more about the transmission quality. This fact allows you to establish that you do indeed have this problem by checking the reception on BBC Radio 2.

If you want to use the set to receive the programmes that give this problem, it is necessary to widen and flatten the response of the IF stages. This can be achieved by connecting a resistor in parallel with the primaries of the VHF IF transformers. You may need to experiment with the resistor values, but generally 18K across the final IF primary and 27K across the others is a good starting point. Note that the lowering the resistor value will flatten the IF response but the gain will also be reduced. Depending on the design of the set, you may not be able to totally remove the distortion without adversely affecting the reception of other stations. The advantage with this sort of modification is that it can easily be reversed in the future if necessary.

TI is a magic eye tuning indicator, and is driven by a proportion of the AGC voltage (tapped off by R11 and R12). This contains a triode amplifier, R14 being its anode load. This drives the indicator section. As the AGC voltage becomes more negative (stronger signal), the voltage on the triode anode becomes more positive. This increases the beam current in the indicator section, causing less of the visible area to be shadowed.

On AM/FM sets, the signal to the triode grid may be switched to the appropriate AGC line by contacts in the wavechange switch. Alternatively, the two signals may be combined on the grid by resistors.

After considerable use, the tuning indicator will become dimmer, and will eventually reach a point when it does not glow at all. The only solution is a replacement.

The tuning indicator is often mounted remote from the chassis, and may be attached to the speaker board of the set with a spring stretched across behind it. Take care when withdrawing the tuning indicator, as the spring could release abruptly breaking the top pip from the indicator.

This circuit section (Bush VHF91) shows a fairly typical AM/FM IF stage, complete with magic eye. At first sight it may look fairly horrific, but upon further examination we will see that it contains everything we have discussed previously. Unfortunately the diagram is not particularly clear due to needing to keep it small enough to fit a browser window on an 800 x 600 screen.

V2 is the mixer-oscillator on AM (MW and LW) and the first IF stage for FM (VHF). Switch S1f (top left corner) connects the HT to either the oscillator section of V2 in AM (as shown) or to the FM tuner head. S1a (left of V2) connects either the AM aerial circuits or the output of the FM tuner head (via screened cable) to the control grid of V2.

The two separate IF transformers at each stage can be seen. The FM IF transformers can be identified by having the lower value parallel capacitors, thus IFT3 and IFT4 are used on FM and the other two on AM. The primary of IFT3 is bypassed by S1e when the set is switched to AM. V3 is the first IF on AM and the second IF on FM.

The FM detector (two diodes in V4) is clearly visible to the right of IFT4 and is the same arrangement as discussed earlier. The AM detector diode is in the section of V4 at the bottom right. S1d (between V3 and V4) switches the appropriate detector audio output onto the top of the volume control (RV1) and on to the triode amplifier in V4 via C40.

The AM AGC is derived from the same diode and is filtered by R14 and C32 and applied to the control grid of V1. The FM AGC from C41 is applied to the suppresser grid of V3. Either the AM AGC signal or the FM AGC signal is applied to the magic eye tuning indicator (V5) by S1b.

The following diagram, taken from Newnes 'Radio and Television Servicing' 1955-56 edition, clearly illustrates the valve functions in a typical AM/FM receiver.

This website, including all text and images not otherwise credited, is copyright © 1997 - 2006 Paul Stenning.
No part of this website may be reproduced in any form without prior written permission from Paul Stenning.
All details are believed to be accurate, but no liability can be accepted for any errors.
The types of equipment discussed on this website may contain high voltages and/or operate at high temperatures.
Appropriate precautions must always be taken to minimise the risk of accidents.
Last updated 14th April 2006.

MW	Virgin 1215	1215 kHz	247 Metres	Light
	Talk Sport	1089 kHz 1053 kHz	275 Metres 285 Metres	Midland West
	BBC Five Live	909 kHz 693 kHz	330 Metres 433 Metres	London Northern
	BBC World Service	648 kHz	463 Metres	Third
LW	(currently unused)	252 kHz	1190 Metres
LW	BBC Radio 4 and BBC World Service	198 kHz	1515 Metres	Light