9 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

Vibrators and vibrator receivers - 1

There are many vibrator powered radios that find their way into swap meets and auction marts. There is a very good reason why they are still to be found in such numbers: they were very practical, for many years...

T HAS BEEN mentioned here before In this context a vibrator is an electro- in electric sets were likewise used because of that reticulated 240V AC power did not mechanical device, which when used with the convenience of a 6.0V heater supply. I extend beyond the cities and the major an appropriately designed , In an all-battery powered radio, a bank of country towns of Australia until about the formed either a DC to DC converter or a three 45V dry batteries would last from four middle 1960s. By implication, until then DC to square wave inverter. The former to six months, depending on current drain and most farmers had to rely upon other sources were referred to as a `synchronous type' hours of use, while the 2V accumulator lasted of power — notably the 32V home lighting vibrator, while the latter were the `non-syn- about 6-8 weeks. Vibrators eliminated the plant and the 6V vibrator powered radio. chronous type'. As a rule for radio work, need for costly dry B batteries, and changing Although radios were produced for 32V the synchronous types supplied 130V DC to 6V avoided the need for special charging DC, six volt vibrator-powered radios were far or so from a 6V accumulator, as a substitute arrangements for the 2V accumulator. and away the more numerous, and were real- for the expensive B batteries. The one drawback was that the 6V accu- ly only superseded by the advent of reticulat- Six-volt accumulators were chosen for mulator only lasted only two to three weeks ed electricity and the emergence of the new three good reasons. Firstly, most cars up to between charging. However, as we have cheaper Japanese radio, which the FJ Holden, in addition to farm machin- seen, charging the radio battery was essen- made the purchase of a new accumulator or ery, used a 6V system. Charging the battery tially at no cost, to the enterprising farmer! the repair of the faithful old set uneconomical. was merely a case of swapping the radio battery over with the car battery, and by the The basic principle Why 6V, and vibrators? time the `cocky' had made a couple of trips Now let's have a look at the way vibrators Firstly, now is as good a time as any to lay to to town, the battery was charged. work. Fig.1(a) shows the circuit of a com- rest an unfortunate misconception. The term Secondly, in the case of the 2.0V series of mon electric bell or buzzer. The make-and- `vibrator' was a wholesome and proper part of valves, it was possible to design a very con- break action of the buzzer spring (hereafter a radio man's terminology and was freely dis- venient series-parallel network for supplying called the `reed') and the adjusting screw, cussed by Catholic and Protestant radio men the valve filaments. And thirdly, in the case of causes a pulsating or interrupted DC to flow alike. There was no hint of unseemliness. car radios, the same valves which were used through the armature coil.

(TOUCHING

ARMATURE R MAGNET

Fig.1: In (a) we see the principle of an electromechanical buzzer, where an electromagnet interrupts its own current. In (b) we see how a vibrator is essentially just the buzzer in a different physical form, while (c) shows the first practical attempt at using a half-wave non-synchronous vibrator to generate the HT in early auto receivers.

72 ELECTRONICS Australia, May 1999 ••••••••••••••••• ••••••• •• tio ••••••... by Roger Johnson

Fig.2: A closeup of a vibrator cartridge removed from its protective shielding can. The array of adjustments is quite complex, to be discussed next time.

If a secondary winding was wound Following on from this was an even worse around the armature coil, a voltage would contraption designed to eliminate the rectifi- be induced in the secondary as a result of er valve. A second set of contacts was placed the collapsing magnetic field when the reed on the reed — electrically quite separate was in the `break' or `off' period of the from the primaries of course — to enable cycle. The frequency of the vibrating reed rectification. This model developed the could be determined by the adjusting screw. same faults as the early non-synchronous The vibrator cartridge is simply a more type of Fig.1(c), but with worse results. The sophisticated and elaborate development of secondary contacts tended to arc and pit this simple buzzer principle. even more so than the primary contacts. Fig.1(b) shows a modified arrangement. Fuses and secondary delay relays were This time the armature coil is at the free sometimes incorporated to give the sec- end of the reed, so that the reed can more ondary contacts that fraction of a second easily oscillate, and with a greater ampli- delay, in order to get the reed moving tude. Not only that, the primary contact has before the load was placed through the sec- a certain resilience. This means that it can ondary contacts. accommodate variations of the amplitude of the reed vibration, so that the 'gap' Parallel coil vibrators remains more constant. This is important A big improvement came with the full wave for reasons that will become apparent soon. parallel coil vibrator, of the type seen in The system of Fig.1(b) also allows the Fig.3(a) which happens to be a synchro- addition of extra contact points. nous (i.e. rectifying) type. This is the type in As you can see, Fig.1(b) is simply a common use from about 1936 until the end rearrangement of Fig.1(a). The chief of the vibrator-powered car radio (1958 or advantage is that it permits additional flex- thereabouts) and many a receiver was kept in ibly mounted contacts on each side of the service until the mid 1960s. Not only that, vibrating reed. These in turn can control they were extensively used in military other electrical circuits, independent of the receivers (in particular in WW2) where driving armature coil. much development took place in insulating, This is indeed what happens in a vibrator moisture proofing, sound proofing, and most cartridge. The close-up photo in Fig.2 shows adjustable resilient contact, in which the important of all, providing reliable contacts. a complicated array of contacts as used in a energising coil was in series with the pri- With this type of vibrator the armature synchronous, or rectifying, cartridge. mary of the transformer. Half-wave rectifi- coil is wound with many turns of fine wire Like the ignition points in a car engine, cation only was possible, and was achieved the contacts of a vibrator cartridge open using a conventional valve. and close literally millions of times during Since the armature or magnetising (inter- their service life. A vibrator cartridge oper- changeable words) coil was in series with ating at 115Hz will have over one million the transformer primary, this meant that the contact movements every 2-1/2 hours. total primary current was drawn through Special electrical and mechanical engineer- the armature coil. It therefore had to be of a ing must be undertaken if any sort of very low DC resistance. Not only that, the longevity is to be anticipated. The engi- DC resistance of the transformer, and the neers of the day aimed at 1500 to 2000 imbalance of the reflected load during hours of operation. operation, meant that the operation of the reed could easily be disturbed unless the The early attempts load conditions, transformer primary and The development of the ultimate `modern' armature coil were carefully designed, vibrator cartridge came as a result of the matched and maintained. demand for automobile receivers in the early Furthermore, if the `points' (synony- 1930s. In this instance, the vibrators were mous with `contacts') became in any way used as shown in Fig.1(c). Until then, what pitted of corroded, there was an excellent car radios there were had to be supplied by chance that the reed would not commence separate HT or `B' batteries (see, for exam- vibrations at all. This is a classic example ple, Vintage Radio in Electronics Australia of a case where `giving it a slight whack' for June 1998). may have fixed the problem, by jolting the These early attempts comprised a single reed into operation!

ELECTRONICS Australia, May 1999 73 PR MARY CONTACTS

:ECNDARY.:. CONTACTS

PRI REED REED DRIVING REED, &CAN &CAN COIL, COIL, CAN

OFF-CENTRE 6 PIN BASE ITH LIX :BASE. SMALL 7 PIN PIN FERROCART SIDE PINS REMOVED BASE OAK

Fig.3: In (a) is shown the basic circuit for a synchronous (rectifying) vibrator supply, while (b) shows that for a non-synchronous vibrator, where a valve must be used. In (c) is shown how the uncommon split-reed synchronous vibrator was used. Representative vibrator cartridge base connections are also shown, for reference. to give a much higher DC resistance, suit- whilst the primary voltage is connected. ously be negative for the high tension. The able for parallel connection with one side Under these conditions, contact is made secondary centre tap is therefore positive. of the low resistance DC winding of the alternatively between the reed and the two When the reed travels to the top set of transformer primary. The magnetic energy `primary' points (contacts), and pulses of contacts, the same thing happens. The reed was provided by the large number of turns, DC pass through the respective halves of the is at earth, the top of the secondary winding with a small current flowing. It will be seen transformer winding in opposite directions. is negative and the centre tap is therefore that the contacts are open when the reed is The frequency of vibration is governed by positive HT. at rest, and that the armature coil is ener- the characteristic of the reed and the pull of Because LT negative is common with HT gised via one lead going directly to one side the magnetic circuit. Usually, a frequency of negative, with the standard synchronous of the DC source (negative), and the other between 115Hz and 120Hz is the standard. vibrator, there can be no attempt at supply- via one half of the transformer primary. The schematic of a vibrator circuit found ing back bias. However, a `split reed' vibra- Because of the higher voltage/low cur- in many circuits for a full wave non-syn- tor cartridge was developed, with a sec- rent operation, the problems with contact chronous (i.e., non rectifying) is shown in ondary reed. This was mechanically cou- resistance was virtually eliminated Fig.3(b). . pled to, but electrically isolated from the together with starting difficulties. primary reed, as shown in Fig.3 (c), and this At off, the reed is neutral to all Synchronous types did allow for the provision of back bias. contacts. Immediately a voltage is applied, The big advantage of synchronous or rectify- However, circuits and applications for these the armature coil is energised in the manner ing vibrators was elimination of the rectifier split reed cartridges are uncommon; the vast described above, and when the reed moves tube, thus saving its cost and also eliminat- majority of domestic radios used a six-pin to the bottom contact, it shorts out the ing 0.6A of current drain from the accumu- 'Oak' or five-pin `Ferrocart' (made by armature coil. This has two effects, one lator. In a car radio, where the car's accumu- Astor) synchronous vibrator. being that the bottom half of the trans- lator is constantly being charged, this does Vibrators and their associated trans- former is now connected to the DC source. not pose a problem. However, for a domestic formers were designed for 4V, 6V, 12V So the magnet is effectively switched off, radio, whose drain depends on the number of and 32V operation. 32 volt jobs were for and contact is broken. The spring inertia of valves and total HT current, the extra 0.6A home lighting plants, whilst the vast the reed moves over to the top contact. becomes quite significant and can shorten majority of domestic radios used the 6V Once the bottom contact is broken, the battery life by as much as one third. type with the principal exception of top half of the transformer primary is still Returning to Fig.3 (a), the secondaries of AWA who, for two or three years, pre- disconnected from its DC source, until con- the transformer are connected by the con- ferred 4V types. The 12 volt types were nection is made with the top points. As a tacts to the reed, which is always at negative used for car radios with 12V systems, and result of the magnet coil releasing the reed (ground) potential. Assume that at a given were mostly non-synchronous. and energising the top half of the primary, instant, the reed is touching the lower prima- In closing, I should mention that the the reed in the process of disconnecting ry contact. With the polarities as marked, the schematics of Fig.3 only illustrate the basic itself from the bottom contact allows the lower end of the primary will be negative concepts of vibrator operation. No mention armature to be energised again and the with respect to the centre tap. If the trans- has been made as yet of filtering and asso- armature draws the reed down to the bottom former secondaries are correctly phased, the ciated circuitry. We'll look at these aspects points; thus the whole process is repeated lower end of the secondary will simultane- in the near future. ❖

74 ELECTRONICS Australia, May 1999 • • • • • • * * • 0 4 0 0 • @ * 0 0 • 0 * • • • • • • • 0

Vibrator Power Supplies

Last month we discussed the general principles of vibrator cartridges and the way they work. This month we look at the peripheral circuitry and how to put theory into practice.

ERELY CONNECTING a vibrator,

accumulator and transformer together METAL BOX M and hoping for the 130-odd volts DC 50-10012 sufficient to operate a valve radio is a far cry THE 'CAN' 2 OnF FILTER from reality, as it happens. The circuit for a CHOKE vibrator power supply is shown in Fig.l, and RFC HT as you can see there's quite a bit more to it. F -130V There are some basic design considera- T tions to bear in mind. Firstly, the final out- 0.5uF put voltage is basically determined by the 0.5uF turns ratio of the transformer, not the vibra- L.T CHOKE — tor cartridge. A given cartridge can work ► 500uF 6V TO VALVE almost any vibrator transformer for a deter- 6V FILAMENTS + ► SHIELDED mined voltage, within certain limits. CABLE FUSE Secondly, there needs to be filtering in both the primary and secondary circuits, as you can see. This cannot be over empha- Fig.1 A typical vibrator power supply using a synchronous or self-rectifying vibrator sised, because a vibrator power supply is cartridge. This is the complete circuit showing the shielding of the battery leads. The essentially a switch-mode system — and numbers refer to the base connections of a typical cartridge. using mechanical switching contacts, so without filtering it generates huge amounts voltages and grid bias would ensure that Now the vibrator cartridge itself is only of RF interference. this figure would be kept down to about about 75% efficient at best, which means The third consideration is whether the sup- 14mA for the sake of battery economy. that an average of about 0.75A will flow ply is to use a valve rectifier, or a synchronous Therefore, the secondary load is 2.7 watts. through the primary contact points, to satis- vibrator (as shown in Fig.1) which provides its Given a transformer efficiency of 80%, fy the set's HT requirements. own rectification for the secondary side. the transformed secondary loading Of course the valve's heaters must also By the way for most purposes you can becomes 3.375VA in the primary circuit. be powered, and the ubiquitous 1 D4/ 1 L5 -G regard a vibrator power supply as a substi- As the primary voltage is 6V, the primary output valve drew 0.24A. The remaining tute for HT batteries, for radios which nor- current is therefore 0.5625 amp. four valves were most often types chosen mally would have used a bank of three 45V that drew 0.12A, so that they neatly formed dry batteries in series. a series-parallel network which provided each valve with its correct voltage and cur- In more detail rent requirements when it was connected Like an ordinary mains transformer, the across a 6V accumulator. entire load of a vibrator transformer is Hence the total current supplied by the ultimately transformed into the current accumulator would be very close to and voltage drawn through the primary. 1.0A (0.75 plus 0.24). In the majority of cases the primary volt- In theory, a 160AH (ampere-hour) age was 6V, so the HT loading becomes accumulator would provide 160 hours of transformed into a certain current at this operation at this drain, although it seldom voltage. was able to do so for radio work. Basically, Let's assume a typical five-valve receiv- listening to the radio for about five hours per er consisting of an RF amp, mixer, IF amp, day would give three weeks or so life from the detector/audio and output. Such a set oper- accumulator, while seven to eight hours per ating under maximum coalitions could day reduced that figure to a fortnight. These easily draw 20mA from the HT supply (say Fig.3: Underneath the sub-chassis of a figures seem to match the anecdotes provided 135V). For dry battery operation, screen typical vibrator power supply. by the old timers who used these sets. 32 ELECTRONICS Australia, June 1999 The transformer see, reduces the ten- The transformer used for vibrator sup- dency of the sec- plies was typically designed with a ondary contacts to liberal amount of iron and plenty of arc or spark. turns-per-volt, to ensure that the While the reed core did not approach saturation, swings across and even with a nse In primary volt- closes the other sec- age, such as may happen with an ondary contacts (the automobile radio. (The voltage opposite primaries across a 6V accumulator may having closed first rise to as much as 8V under full under no-load con- charge, particularly with some ditions), the charged of the primitive regulating sys- filter capacitor main- tems on the older cars.) Such pro- tains the secondary vision minimises magnetising cur- potential at the sec- rent and makes for greater efficiency. ondary centre tap. The transformer efficiency quoted above Fig.4: A complete mantel radio chassis We must remember that much the using a vibrator supply for HT. Most of same as an ordinary mains transformer, may therefore be a little conservative. the vibrator supply components are in the big metal box on the right. the ends of the secondaries are opposite The vibrator itself in polarity to the centre tap. With a prop- This is where the designers' persistence erly designed filter system with ade- paid dividends, because when properly under load, particularly when the ignition quate capacitance, the HT input filter dis- adjusted and initially set up, vibrators were timing is advanced by the vacuum advance. charges only fractionally during the entire designed for something like 1500 to 2000 The points in a vibrator cartridge must cycle. This means that the respective ends hours of service life, about the equivalent of therefore be as clean and flat as possible, of the half-secondaries are at zero poten- a radio valve. However, unlike a radio valve, with the contact surfaces as parallel as pos- tial, or very nearly so, at the instant just a vibrator cartridge is repairable. sible, and the gaps for the primary contacts before secondary contact closure. As a Of course if the magnet (or armature) as equal as possible. result the potential difference between the coil is burnt out, there is little that can be The gaps for the primary and sec- secondary contacts and the reed is also done. `Ferrocart' cartridges designed for ondary contacts of a synchronous vibra- 32V home lighting plants were said to have tor are different, by the way. The prima- given trouble in this regard. ry contacts are spaced a little closer There are five adjustments on most stan- together than the secondaries. Thus the dard 6V volt synchronous vibrators. Firstly, secondaries close after, and open before, there is the adjusting screw and locknut the corresponding primary contacts. This shown on the framework in Fig.2 last is done so that when the primaries close, month. This adjustment sets the frequency there is only the magnetising and core of vibration of the reed, and can have a saturating current that initially passes bearing on the output voltage. through the contacts, which is much less Also shown in the same photo are two pairs than the 0.75A that passes through the of contacts. In the photo as shown, these are primary contacts under full load. the primary contacts. Directly behind these, In other words, the primaries are closed and obscured by the primary contacts in the before full load is applied, and opened after photo, are the secondary contacts. full load has been removed. This markedly reduces the tendency to primary arcing. The `points' When the secondary contacts are closed, The condition of the contacts is vital if the input filter capacitor is almost fully `hash' is to be eliminated and correct excita- charged to the voltage which appears tion at switch-on is to occur. In this instance across the particular half-secondary wind- they are like the `points' in the older type of ing. This charged capacitor is vital to main- ignition system used in cars. If the points are tain the potential as nearly as possible pitted, the car is hard to start and will misfire across each of the contacts, and as we shall ELECTRONICS Australia, June 1999 33 Fig.2: The vibrator circuit of a Healing 408A receiver. Only the socket connections are shown in this case. The buffer capacitor is the 0.5uF unit connected across the primary, and also the two 1OnF units in series across the secondary.

or two in total. Even one ohm will cause a voltage drop of 0.75V, which is enough to drop the primary voltage to 5:25 volts and reduce the secondary voltage by perhaps 15V. A resistance of one ohm in the leads to the filaments will produce a voltage drop of 0.25V, which is tolerable. HT filtering In Fig.1 you can see that there is also an RF very nearly zero, resulting in minimal cause the transformer to resonate at the same choke in series with the HT output from the sparking across the secondary points. frequency as the mechanical frequency of transformer secondary centre-tap. The vibra- the vibrating reed. When properly adjusted, tor and transformer are an excellent source The `buffer' capacitors the induced. EMF has reversed in polarity of RF hash, but by careful design it can be The buffer or timing capacitors are those again, and is approaching the terminal volt- almost eliminated. shown connected between the ends of the age of the battery when the points close. The buffer capacitors may reduce the secondary. In some circuits, each capacitor Variations in transformer characteristics, sparking, but do not completely eliminate it. is connected between one end of the sec- vibrator characteristics, primary and sec- Also, pitting and corrosion of the points due ondary and the centre tap, or only one capac- ondary voltages and loading can all affect to maladjustment or normal wear-and-tear itor is used, connected across the entire sec- the electrical resonance, and the values of can and will cause sparking. Not only that, the ondary. Occasionally, series resistors of buffer capacitors are best determined at the buffer capacitors do not completely eliminate between.. 50 and 100 ohms are placed workbench with the aid of a CRO to study damped oscillations, and even though spark- between the transformer secondary ends and the waveforms. The unsuppressed spikes ing across the points may not have occurred, the buffer capacitors, and a larger value can reach quite high voltages, and ordinary plenty of interference can make its way to the capacitor (20 to 50 times the value) is placed paper capacitors are not recommended. high tension output. The RF choke and a cou- directly across the primary. Mica types of 1 nF or 2nF value and 1000V ple of bypass capacitors of 1 or lOnF are nec- The functioning of the buffer capacitors rating were preferred. essary to provide adequate filtering. is quite complex, and is described by the The DC output from a vibrator supply is older texts as follows: Low tension filtering like the unfiltered output from a valve recti- When one of the set of primary contacts The continual make and break action of the fier, except that the waveform is of a differ- is closed, the battery is connected across points (carrying 0.75A) fractionally lowers ent shape. Instead of being a succession of that respective half-primary. The resultant the battery voltage when the points are adjoining positive portions of sine waves, current flow in the primary is opposed by closed. Hence the voltage measured at the the output waveform from a vibrator is a the so-called `counter EMF', which builds battery terminals is fluctuating at the vibra- succession of square waves. If both sides of up to a certain value during the brief inter- tor frequency. This causes ripple to appear the transformer were identical in every val that the contacts are closed. When the at the valve filaments, and can greatly exac- respect, and also the gapping of the points, contacts open, the field collapses, and a erbate hum just like an electric set with poor the output might be continuous. In reality, voltage builds up in opposite polarity to the filtering. Consequently, both air- and iron- there is likely to be transient negative `dips' original counter EMF. cored chokes of very low DC resistance from Vmax to zero, as well as fluctuations In a no-load position, i.e., when the reed were used for filtering, together with capac- in Vmax according to minor differences in is oscillating between the contacts, the itors as shown. the constants described above. building up and collapsing of the magnetic One, and sometimes two air-cored The normal filter circuit of two 10uF elec- field may in fact repeat itself and caused chokes (marked 'AFC' in Fig.1) were trolytics and an iron-cored choke of 10 or 12 damped oscillations within the transformer. placed in series with the primary, and at henries is designed to eliminate any HT rip- This can cause interference of its own least one iron-cored choke was placed in ple at the fundamental conversion frequency. accord, but more importantly, a situation series with the valve heaters as shown, with As you can see, another important ele- can arise when a portion of the damped a 500uF filter capacitor for further smooth- ment in containing the RF hash generated oscillation is opposite in polarity to the ing. Shielding the battery leads, down to by a vibrator supply was to enclose the applied voltage when the contacts close, but not including the negative battery clip, vibrator, transformer and RF filtering com- thus causing serious sparking, and greatly reduced the tendency for radiation of RF ponents in a metal shield box. reducing the service life of the contacts. hash which could otherwise be picked This and last month's columns have The situation can be corrected by placing up by the antenna. described how a vibrator works. Next time the buffer capacitors in place as already The DC resistance of these filter chokes we'll look at repairs to vibrators and vibrator described. The effect of these capacitors is to must be of necessity a no more than an ohm powered receivers in general. ❖ 34 ELECTRONICS Australia, June 1999