Vacuum Tube (Valve) Radio and Audio Repair Guide: Tools for servicing radio receivers

As regard small tools, anyone who is already something of a handyman already will have a selection of screwdrivers with different blades, various pliers, both blunt and snipe-nosed, side- cutters, a hand drill and a good selection of bits, various files and so on. It is very useful, too, to have at hand a selection of nutdrivers from 8BA up to 0BA and open-ended spanners to cover the same range. You will need at least one good electric soldering iron rated at about 25 W with a fine bit for getting into awkward places; if possible, acquire a larger iron of 60W or more rating for the odd times when you will have to solder direct to a steel receiver chassis.

Solder ‘guns’

Ordinary soldering irons have heating elements for mains supplies, wound with resistance wire and insulated in mica and steel cases. They take a few minutes to warm up, then have to be left switched on in order to be ready for instant use in the workshop. When the trigger switch of a solder ‘gun’ is depressed its element reaches working temperature in a few seconds, so it consumes current only when in use. This is achieved by replacing the usual heating element by a mains transformer with a low voltage secondary rated typically at about 1V, 80A. This is made up of a turn or two of thick copper strip around the primary, to which are attached a pair or heavy- duty terminals. The bit, which is fashioned from thick copper alloy wire, is attached to these terminals and when the trigger is pulled a heavy current flows through it, causing it to heat up. At the business end the element is flattened over about a quarter of an inch at which point the greatest heat occurs. The better makes have a tapped primary with two-position trigger, so that full power (full depression) is used to gain operating temperature and reduced power (part depression) to maintain it. Be warned that guns without this feature may carry on getting hotter and hotter all the time the trigger is depressed, so one has to make a joint, etc., quickly before damage is done to heat-vulnerable components. In this context, take care with any type of imported soldering iron, especially from the Far East because their elements may overheat on UK mains voltage. Irons from the USA and intended for 110/120 V supplies will be marked so; they are usually very well made and worth acquiring to be used with a suitable step- down transformer.

Irons with 12 V elements are obtainable for use on automobile batteries or low voltage trans formers. Some have very small tips that are useful for working in confined spaces.

Solder

To go with the irons you will need some good quality resin cored solder, preferably about 18—20 gauge. Steer clear of thick industrial types and very thin types meant for printed board connections.

What about instruments?

Most professional service engineers will be able to tell you that in the final analysis a very high proportion of repair jobs may be completed successfully with no more than a good multimeter, which in vintage radio terms is virtually synonymous with the name AVO, the firm which has been making the best on the market since the mid 1920s. The AVO meter has all the attributes that make up a first-class tool for radio servicing, including a good selection of voltage, current and resistance ranges, a large clear dial and a smooth needle movement of the type known as ‘dead beat’, which means that it comes immediately to rest on the dial reading to be indicated without any preliminary swinging about.

What ranges are required?

To be able to check anything from a miniature battery portable set up to a massive luxury mains radio-gramophone you will need a multimeter that will read up to about 1000V both AC and DC — although in theory this ought to be possible with a single 0—1000 V scale for either AC or DC in practice it would be impossible to obtain accurate readings at the low end of the markings. Thus the complete coverage needs to be broken down into four or more sections, say 0—1OV, 0—100V, 0—500V and 0—1000V. For resistance checks 0—1000 ohm, 0—100k-ohm and 0—10 M-ohm (note: the unlinear nature of the dial markings on resistance ranges means that roughly the lower half of any range actually spreads over about four-fifths of the dial with the rest compressed at the one end; therefore a resistor of say 750 ohm is much more easily tested on a 0—100k-ohm range than on a 0—1000 ohm range).

Fortunately, multimeters made in the vintage radio days were designed specifically to cater for these kinds of requirements, so in this respect a good old AVO meter may be a far better practical proposition than a brand new one designed for far different purposes.

The ohms-per-volt game

The sensitivity of a meter is expressed as an ‘ohms- per-volt’ figure, and the greater the number of ohms the better, at least in theory. What it actually means is this: the heart of any multimeter is a meter movement that measures small amounts of current, typically with a full-scale deflection (f.s.d.) of 1 mA. When you make a voltage reading with the multimeter the movement is responding to current flowing through it, the dial of the meter being calibrated to indicate the amount of voltage that corresponds to that current.

To give an example, Ohm’s law tells us that a voltage of 1V applied across 1000 ohm will cause a current of 1 mA to flow. Thus a multimeter with a basic movement of 1 mA would have a sensitivity of 1000 ohm per volt. The 10001 is made up of the internal resistance of the basic movement, commonly 50 ohm, plus a series resistor which would be of 950 ohm. For a 10V range the overall resistance would need to be 10,000 ohm, so the series resistance would be 9950 ohm. By this time the meter resistance is becoming such a small percentage of the overall that it may be ignored, so for a 100 V range the series resistance would be 100 k-ohm, and so on. Wire-wound resistors are used for the lower ranges and very high stability carbon types for the higher ranges.

What does ohms-per-volt mean in practice?

Suppose you wish to read the anode voltage of a tube (valve) which draws 3 mA anode current through a load resistance of 47 k-ohm from an HT line of 200 V. The resistor will drop 141 V at the rated anode current so the voltage should be a shade under 60 V. The AVO ‘Seven’ has a basic movement of 1 mA but it is shunted to make it draw 2 mA at f.s.d., so the sensitivity is reduced to 500 ohm per volt. Thus on its 100V range the overall resistance is 50000 ohm which, when applied to the tube (valve) anode, will draw around 1 mA extra current through the load resistor and the anode voltage will drop accordingly, giving a meter reading of perhaps no more than 20V even though there is nothing wrong with the set.

Now let’s look at the AVO ‘Eight’. The basic movement of this has an f.s.d. of only 50 uA, and to make it register 1 V the overall resistance has to be 20000 ohm. The resulting sensitivity of 20000 ohm per volt makes a great deal of difference to the reading than would be obtained using the same example as before. The tiny current of about 10 uA drawn by the meter would drop the anode voltage by less than half a volt, so a much more accurate reading would be obtained.

It would be easy to be misled by the above into believing that the old AVO ‘Seven’ has no place nowadays in radio servicing. Nothing could be further from the truth, because the ‘Seven’ was the industry standard through much of the vintage years until about 1955 and nearly all the radio manufacturers used it to produce the voltage check lists in their service manuals. Therefore the readings produced on a ‘Seven’, even if they are not totally accurate in actual voltage, do indicate what is right and what is wrong. Reading given by a meter of high sensitivity, although much more accurate, may in practice be misleading!

For the rest of the tube (valve) years the ‘Eight’ took over and most service data voltage readings were based on its use. What needs to be remembered is that when an ‘Eight’ is used to check voltages originally taken on a ‘Seven’ the readings usually will be higher, and this should be taken into account (the reverse is true when a ‘Seven’ is used to check figures taken with an ‘Eight’).

There is little doubt that the ‘Eight’ is now the most popular meter for vintage radio servicing, assisted by its ready availability on the Government surplus market at a fraction of the cost when new; thus does the taxpayer aid the vintage radio enthusiast. There are, of course, many other makes and types of multimeter, not all with conventional dials of the type now referred to as analogue but also with digital readouts and even simulated voice readouts. These last two impose virtually no load on a circuit and thus might be thought to be the best for servicing, but in fact they lack the ability of the analogue type to give the smooth indication of rising and falling voltage or current that is necessary for certain tests on tubes (valves) radio receivers.

The right way round for meters

I hope that experienced readers will forgive the inclusion of what they may consider to be an obvious point. However, at one of the Radiophile Workshops no fewer than eight participants arrived with beautiful new ex-Government AVO ‘Eights’ yet had no clear idea of how they should be used. For the benefit of the complete beginners, then.

For most DC voltage tests the negative terminal of the meter goes to chassis and the positive to the HT line, tubes (valves) electrodes, and anything else carrying HT voltage. Always start with the meter switched to a range higher than the maximum to be expected, e.g. if the HT voltage on a tube (valve) should be around 100v, don’t start with the 100V range but with the 400 V range on the AVO ‘Seven’ and the 250V or 300V range on the ‘Eight’. When negative voltage has to be checked (as, say, the voltage drop across a choke used for negative smoothing) connect the positive terminal of the meter to chassis and the negative to the choke.

For DC current tests always connect the meter in series with the device that is drawing current so that the positive goes to the positive side of the supply.

For most resistance tests it is immaterial how the leads are connected, except when checking metal rectifiers, which should be tried twice with the leads reversed the second time. There normally should be a much higher resistance one way than the other and similar results, either high or low, indicate trouble.

AC measurements may be made with the leads either way round but when voltage checks are being made with respect to chassis it is best to connect the negative terminal to the latter in readiness for further DC tests.

The signal generator

The one other instrument that is essential for repairing superhet receivers is the signal generator or service oscillator (strictly speaking the former is a highly accurate device more for the laboratory than the workshop, whilst the latter is a less expensive general purpose tool). It is used to align, i.e. to bring into correct tune, first the IFTs, then the RF and local oscillator circuits and needs to cover from about 100 kHz up to 100 mHz. Anyone reading certain articles in old radio magazines may well come across an assertion that it is possible for an amateur to align IFTs without a signal generator. Forget it; not even the most experienced engineer can do this properly, the best that can be hoped for is that if just one of the IFTs has been mis-adjusted it may be possible simply to peak up the other on a station.

Try to get hold of the best signal generator that you can, those made by Advance being particularly good. The writer has an Advance Model E2 bought new in 1952 which is still in regular use, so reliability is guaranteed. E2s appear regularly at vintage radio events and auctions, seldom costing more than a few pounds.

The most sophisticated and accurate signal generators were those built (erected might be a better term!) by Marconi’s Wireless Telegraph Co. through the 1940s, 1950s and 1960s, with model numbers commencing with the letters TE Some of these are enormously large and heavy and cost huge sums only affordable by large laboratories and the military (i.e. the taxpayer). For instance, the mighty TF86 measures about two feet six inches square and eighteen inches front to back, weighs over a hundredweight and cost new more than a contemporary family car. In recent years TFs have been appearing at radio events for only a few pounds, their sheer size daunting most potential buyers. However, if you do have the room to house one, you will have the satisfaction of owning one of the best generators available anywhere.

The resistance and capacity bridge

This enables the values of resistors and capacitors to be checked accurately, added to which bridges also incorporate a means of checking capacitors for good internal insulation by applying suitable voltages in ranges from about 25 V up to 300 V. This feature may also usefully be employed to ‘re-form’ electrolytic capacitors. It is certainly an instrument that will come in handy now and again but don’t go out your way to obtain one — one is sure to come along sooner or later at a bargain price.

What about an oscilloscope?

What indeed? You may read reams and reams of articles in old radio magazines about how wonderful this instrument is, but in practice it probably will prove to be a fairly expensive dust-gatherer. Place it low on your shopping list unless you are able also to obtain a wobbulator which is the common name for a frequency-modulated oscillator. When used to align IFTs the wobbulator supplies the 465 kHz or whatever, rapidly and constantly swept up and down for 25 kHz or more on either side. This produces at the detector a varying voltage which starts low at say 440 kHz, then climbs gradually to maximum at 465 kHz, only to start dropping again until 490 kHz is reached, whereupon the whole thing starts over again. If the probe of an oscilloscope is connected to the detector the varying voltage will be shown on the screen as a replica of the response curve of the IFTs, which may be adjusted until the correct shape is obtained. A control voltage produced in the wobbulator is applied to the ‘scope to keep it running in perfect step. For anyone engaged professionally in repairing high quality receivers or communication receivers the wobbulator/’scope set-up is a good investment but the hobbyist can afford to ignore it unless it comes at a non-refusable price.

The same remarks apply to the next two items, the AF signal generator and the ‘Q’ meter. The first, with usual coverage from about 10 Hz up to 30 kHz is useful if one needs to be able to plot the audio response curve of a receiver or amplifier. The job of the second, strictly speaking, is to test the ‘Q’ or goodness of tuning coils, which it does by applying a signal at the appropriate frequency and measuring the gain by means of special kind of detector connected to a milli-ammeter. It also enables one to check the frequency range of a coil placed in parallel with a variable capacitor, the setting of which can be altered from about 10 pfd to 750 uF This is only likely to be a useful facility for anyone engaged in rewinding tuning coils on a fairly regular basis.

The output meter

This is no more than an AC voltmeter which can be connected across the secondary of the speech coil in a receiver. With the steady 400 kHz AF modulation produced by the average signal generator applied to the set, the output meter registers a voltage dependent on the output from the receiver. It is thus an excellent indicator of when the IFTS and the RF and local oscillator coils are brought into correct alignment. If you have a multi meter you already have an output meter, since the same job may be done by a suitable AC voltage range connected across the speech coil.

Tube (valve) testers

Oddly enough, these are more likely to appeal to the person who does only the occasional radio repair than to someone with a busy workshop and a good spares section. Certainly in the radio trade, although every shop had its tubes (valves) tester, few engineers bothered to use them because it was quicker and more positive simply to try a new tubes (valves) in a set. There is also the argument that in a sense a set is itself a tube (valve) tester since, if all the supply requirements are in order, a few voltage tests will indicate the condition of its tubes (valves) with fair accuracy. However, for anyone lacking a large tubes (valves) cupboard and wishing for an independent means of checking suspect examples, a tester can be very useful. The best-known types are the various models made by AVO since about 1935, and by the Taylor Instrument Co. from about 1938—1958. The early AVOs and the Taylors worked by measuring the mutual conductance of triodes and other multi electrode tubes (valves), and the anode current of diodes and rectifiers at a given voltage. The AVOs were of twin-panel construction, one carrying all the volt age selector switches, the supply transformer and the actual meter movement, the other housing a wide selection of tubes (valves) holders and a multiway thumb switch enabling many combinations of connections to be made to the holders to suit almost any known valve. The Taylors used a single large panel carrying the meter, the power supply, the tube (valve) holders and three switches which worked in various combinations to give the correct base connections. Either type of tester is likely to be available at vintage radio events at seldom over about $75 and maybe a lot less.

The later AVO Tube (valve) Characteristic

Meters are far more sophisticated devices enabling very comprehensive and accurate checks to be made on various factors that affect a valve’s performance. The military used them in large quantities and it is those versions which are most likely to appear at radio events, although the civilian original does pop up now and again. Be prepared to pay up to $300 for a really good example but look out for the occasional bargain at a fraction of this.

Warning: whichever of the above tubes (valves) testers you may buy, make sure that it has the essential operating book with it, for if this is missing you may have to pay up to $40 to obtain a replacement (they are not infrequently more valuable than the testers themselves!).

The Mullard ‘high speed’ tubes (valves) tester

This offers the most picturesque way of testing tubes (valves) with its cathode ray tube display and ‘one armed bandit’ action. It is basically a large metal box with a few knobs on the front below a 3-inch c.r.t., and a large metal lever on the right-hand side. At the top rear is a slot into which is inserted a punched card made of paxolin, individual examples being supplied with the machine for all the popular tubes (valves) types. Pulling the lever causes contact springs to bear on the card, completing circuits according to the positions of the holes and thereby setting up all the tube (valve) heater and electrode voltages automatically. A spot on the c.r.t. moves to show, with the aid of a scale printed alongside, the state of health of the valve. The complete outfit consists of the tester itself and two large steel boxes containing the cards, all supposed to be mounted on a strong cradle which may or may not be present.

These testers were fabulously expensive back in the 1950s but can now be obtained at auction for around £50 complete with cards — but make sure that the popular tubes (valves) types are still represented and that the essential mains adjustment card also is there. Another tip: watch out for ex-Government versions for which all the cards were for military tubes (valves) with CV numbers. It is possible to cross reference a lot of them but it is a tedious job.