Facts about "A" Battery Eliminators

Radio Record, Volume II, Issue 12, 5 October 1928, Page 26

Facts about "A" Battery Eliminators

Several Systems Outlined

By

Megohm

| won HIS is a brief review of ode) Bais several systems of A bataay fee tery elimination that may be successfully adopted provided that essentials are supplied in a suitable way. At ihe outset it as well to explain why the problem of eliminating the A battery is a greater obstacle to overcome thin is the elimination of the B pattery. In the case of the latter, the current to be supplied to the average set is about 30 milliamperes or 03 ampere (three hundredths of an ampere). A comparatively sniall current such as this, though of a voltage near 200, is easily smoothed by suitable chokes and condensers of «©comparatively small capacity. The consumption of plate current in a receiver cannot be great, owing to the high internal resistance of the valves caused by the space between plate and filament which has to be traversed by the current. HEN we come to filament supply the question is very different, as the amount of current to be smoothed is many times that required in the plate circuit. The 201A valve requires a quarter-ampere for its filament, so that a receiver with five such valves requires 14 amp. at 6 volts to operate its filament. The minimum amount of current available, allowing a margin, should therefore be 14 amperes. Many valves are now in use that require only 1 amp. or .06 amp. for the filament. Five of the former would only require a total of half an ampere, whilst five of the .06 type would only consume one-third ampere. The foregoing figures are for the usual method of connecting up the filaments in parallel. Filaments Run from B Eliminator. A SYSTEM of running both A and B supply (and sometimes C also) from. a B eliminator containing a 125

milliampere Raytheon tube has been successfully developed by elploying only .06 valves and connecting all the filaments in series. In such a case the amperage of every valve must be the siume, and the amperes econstuned will be that of one valve, in this case .06

amp. Whilst the voltage will be that of all the valves added together. The UNX199 is a suitable valve, and there are algo a number of Philips’ 06 valves requiring 4 volts at 06 amperes. The voltage of several valves placed in series need not be the same for all, so eA

long as a voltage equal to the total is applied. There is some disadvantage in this system owing to the restriction placed upon all but the last yulye. A power valve of any filament rating may be used in the last stage by heating its filament with raw A.C.

current from a filament winding on the eliminator transformer. Arrangements are made in the eliminator whereby the 15 volts or so required for the filaments is drawn off separately through suitable resistances. Suitable resist‘ances are placed across each filament to pass the added plate current when this is suflicient to take into account. Smoothing Arrangements.

AS’ mentioned above, a current of at least one ampere, and probably more, has to be filtered and smoothed for use with parallel connected fila-. meuts, this arrangement obviating any alteration of the wiring in the receiver. Condensers of two and four microfarads capacity are totally inadequate to handle the required current. and the advent of the improved electrolytic condenser wakes A elimination a much easier problem, and allows of even 2 sinall balancing accumulator being dispensed with. There are now on sale .. the United States, "filter blocks" specially made for A elimination, and these contain two electrolytic condenvers and suitable choke coils (Tobe Co.). Abox.. Aerovox, Dubilir, Elkon and Tobe Companies also make suitable electrolytic condensers, some being practically dry. Some of the most recent type of condenser make use of the oxide film which forms on aluminium, as the sole dielectric and means being taken to prevent the formation of the gas dielectric film, a large increase in capacity: for given area is the result.

Such a condenser, in two sections, Igy have a capacity of 1800 microfarady a one section and 3800 in the other, or a total of 5600 microfarads in a space of 5 x 5 %x 14 inches! The breakdown ~alue of such a condenser is 50 volts, so it is well suited to smooth out filament current. Whilst a total of about 16 microfaruds may form the condenser outfit of a B eliminator, a capacity of some thousands of microfarads is necessary to smooth an A supply. In early attempts to make high‘apacity condensers, lead sheets were separated by wood-pulp paper kept moist with acid solution, but as the capacity of a condenser raries inversely as the thickness of the dielectric and directly as the area of the plates, efforts were made to reduce the separation to a minimum, because ellarging the size of plates meant great bulk and extra cost. Thus the ultimate form of condenser was the electrolytic type, in which a very thin film of gas on the condenser plates acts as the dielectric to separate them from the liquid which acts as the opposite set of plates. Condensers of this type are now manufactured, varying in detail, but having a capacity

of thousands of microrarads. Kither one or two chokes must be wtilised in addition to the smoothing condensers in order to suppress any ripple that may be present in the current. Great care must be taken in designing such chokes, as no great voltnge drop is permissible across them. The d.c. resistance must therefore we kept low, and this is effected by employing 2 heavy gauge of wire. As the inductanee of such a choke need only be a quarter henry or less, the number of turns of wire need not be very great, and will not exceed a few hundred. By not unduly reducing the cross-section of the core, the number of turns, and thus the d.c. resistance, Can be kept low. In addition to low (.c¢. resistance, the wire must be capable of carrying two amperes, but this does not present any difficulty. .

Filament Current From Battery Charger. ERUAPS the simplest way of producing an A eliminator is to employ a valve rectifying eharges of suitable capacity, say 2 amperes, connecting this to electrolytic condensers, and one or two low inductance chokes. Nothing more is required except a 10 or 15-ohm power-type rheostat and a voltmeter reading to 8 or 10 volts. Two condensers are sufficient for the average receiver, but a good home-con-structed set with low-note amplification will probably require three. The cireuit is shown in a diagram. Another method, obviating the use of electrolytic condensers, is to place two wet batteries across in place of C1 and C2. Small dry-cells have even been used for this purpose, but are not so satisfactory as accumulators. /he operation of this system depends on the fact that, at voltages equal to the cell voltages, the resistance across the cell banks is very high, so that preety tically none of the current flo is through them, but continues to flow out through the filter circuit to the receiver. But for all current whose

voltage is higher or lower than the operating voltages, the resistance of the cell banks is very low, and these , currents are by-passed. For this reason rhe voltage remains constant, and a smooth current is delivered from the charger. For sets of more than five vagres dry batteries are not satisfactor¥. The accumulators used need be of only very small capacity, and for a 4-volt supply they would of course be of that voltage. Low Inductance Chokes. ILTER chokes for A supply must be specially designed for the current they have to carry, and as this current is many times that carried by a B eliminator choke, the wire must necessarily be of much heavier gauge. Such a choke cannot be of high inductance value if of convenient size, because of the heavy current which tends to cause saturation of the core, even when its cross-section is liberal, and turns of wire are kept to a minimum. By increasing the cross-section of the core, the inductance is proportionally increased, and this allows of a small (inversely as the square) reduction in the number of turns, still obtaining the same inductance. Not only on account of saturation, but to keep down the d.c. resistance of the choke, must the turns be limited, because when working from a battery charger the available volts will not give much surplus over the six volts required for working the receiver. On this account the d.e. resistance must on no account be more than three or four ohms, and may very well be less. One commercial choke is advertised as having an inductance of -l henry and a d.c. resistance of .3 ohm. No smaller gauge than 20’s s.w.g. enamelled wire should be used to carry 13 to 2 amperes continuously. Of this "Wire, one pound has a resistance of ‘Only 2 ohms, and less than a pound would be required on a stalloy core 1} inches square. If the core is nearing saturation its inductance value is automatically lowered, and in such a case the gaps may be slightly increased by trial, and although this lowers the inductance, is preferable to saturation. A Matter for Consideration. "THERE is one point that must receive careful consideration by anyone adopting an A_ eliminator. Such a piece of apparatus usually has "bad regulation," which means that when the supply to one stage or stages of the receiver is reduced, there is an increase in supply by that amount available, and this at once divides itself between the filaments of the other valves. This points to a different method of volume control being adopted in receiyers where the R.F. filaments are used for control. Otherwise there would be a tendency for the detector and audio valves to receive too great a filament voltage on the R.F. filaments being dimmed. Dry Rectifiers. HE Kuprox dry rectifier arranged for full-wave rectification forms a very suitable means of producing direct current from A.C. mains, but a transformer and smoothing arrangement Rre required just as when a valve rectifier is used. Dry rectifiers of any type should not be overloaded, and the manufacturers’ statements of carrying capacity should be taken as liberal if a reasonably long life is desired for the unit. In using any dry rectifier there is a point to be remembered. These recti-

fiers will not stand up to heavy backvoltage without injury, so that it is not safe to employ with them a trans. former secondary of much _ higher voltage than specified, reducing the voltage with a rheostat or resistance. because in such a case the back-voltage would be that of the transformer sec. ondary plus the battery voltage. This. however, only refers to half-wave rectification, which is not recommended for A eliminators on account of the greater difficulty of removing hum.