B Battery Eliminators for Small Sets

Radio Record, Volume I, Issue 36, 23 March 1928, Page 10

B Battery Eliminators for Small Sets

First {nstalment of Simplified Construction

By

Megohm

3H purpose of this article is to describe two differ‘ent methods of constructing a B battery eliminay ie tor, suitable for small sels tio of two and three valves, and possibly four, if the demand for current is not too great The chief idea in an outfit of this kind is to cut dowh expense as much as _ possible, without impairing efficiency. An eliminator of this description will cost about half the cost of constructing the full-wave unit described some time ago, but must not be expected to give mearly the same output. ‘Those who are tired of purchasing

ary B batteries for two-vaive Crystal amplifiers will find it worth while to construct a small eliminator, which will enable full power to be cbtained from the amplifier on all occasions, and with a probable i...provement in quality, as a power-valve usually gives best results when employing the highest plate voltage specified. The constructor may decide what voltage he will provide by putting on the suitable number of turns, the maximum of which will give a full 200 volts when rectified, and this is more than the average . power-valve will reanire. The power supply for which the eliminator is designed is 230 volts, 50 cycles. The first unit to be described is the one that is recommended, althongh it entails rather more work in the shape of constructing a small transformer. Yet the work will be well worth while, and well repaid in after use. And, after all, this column is to a great extent for those who get pleasure out of the making, not only for what the article will do, but for what is learnt jn tNe process of making. (in the battery charger article, third column, ‘thin card" is, owing to a misprint, made to read "thin cord.’’) THE TRANSFORMER. N the issue of March 2 the making of a fransformer was described for charging an A battery, and in this article the making of a spool to contain the windings was fully described, together with a winding jig, which makes the process easier. Constructors will there find how to make the spool, and must alter the sizes to those given here. The length inside is the same, 2} inches, and some strips oi paper this width will be required. The spool ends are 2.3 inches square, and the wooden former on which the spool is made is a bare 1.8 square and @bout 3 inches long, depending upon the exact thickness of the spool ends. THE PRIMARY WINDING. HIS consists of 2500 turns of 382’s &.w.g. enamelled wire, in 15 lavers, averaging 200 turns per layer, and the turns should be put on without spaces between in order to get the full number in, and still leave a simall space of at Jeast 1-16-inch at each end. When @il the primary turns have been put on, the end of the wire is passed out through a hole in the spool end, and the last layer covered with empire cloth cut to leave no space at the ends, and over this a layer of paper, after which , THE SECONDARY WINDING is mioceeded with. This may be varied

at the option of the constructor, according to the voltage decided upon, or the whole 2000 turns may be put on and taps provided as required. This -method of tapping is good in any case, but only one tap must be used at one time, s different voltages cannot be tapped off at this point. The one voltage must be taken and reduced by reBistatices ‘This winding 19 86’s s.w.g. enamcllcd wire averaging 300 turns per layer, and for 3000 turns 10 layers will be required, cach laver giving about 25 volts The following table shows voltayes obtainable bs tapping the end of every layer, though actually it would be sufficient to tap alternate layers, giving 100, 114, 200, 250. It must also be remembered that from these figures has to be deducted the drop, about 40 volts, caused by the rectitying valve.

Seven layers will suffice for most purposes, as that will give nearly 150 volts maxinium output, which will work at full volume any valves likely to be used in a two-valve amplifier. There

is provision for two voltages, so that the first valve may be worked with less plate voltage. With taps provided, a reduced yoltage may be obtained on both without the need of putting in a large amount of extra resistance. Those who have been working with, say, a W0-volt dry B battery will find even 100 yolts from on eliminator quite a good increase, because a 90-volt dry battery may only be giving 70 volts after a few weeks’ use and 50 when dis. carded, whereas the eliminator delivers the full voltage all the time. Valves ean be "paralysed" by getting too much plate voltage, so nothing is to be gained by greatly exceeding the maximum voltage recommended by the manufacturers. ‘Laps are to be brought out through the spool ends at the most convenient Se

place, but not where the laminations come, which position is marked by the cross-piece put on temporarily as directed. When the secondary winding is finished the usual covering of paper, empire cloth, and another layer of paper is puf on ready for the FILAMENT WINDING. (PHS consists of 76 turns of 22’s s.w.g. double cotton-covered wire, tapped at 5, 13, 88, 68, and 71 turns, Commencing at one end, the end of the wire is put through a hole in spool end. five turns wound on, a tap solder-

ed on, then eight more turns and tap, then twenty-five turms more arrives at the 38th turn, which is the centre tap, after which twenty-five more make the 68rd, then eight the 7lst, and five more completes. It will be found that the centre tap comes past the centre, and that when the 68rd turn is reached the layer is full, and this tap is at the end. ‘The winding so far in place must now be well shellaced and a strip of paper an inch wide wound twice round over the last completed end of the Jayer. ‘The remaining turns, 18, are now put on over the paper, the end cut and left free, wire shellaced and left to dry, when a suitable covering may be put on. A diagram shows the transformer with filament winding uncovered in order to give an idea of the arrangement of taps. The filament winding gives 6} volts over the full coil, five turns less at each end gives 5{ volts, and eight turns less again on each side, 4 volts, These voltages will suit any rectifying valve ‘likely to be used. A rheostat is put into the circuit for filament control, so that the taps to be used are the two giving the voltage slightly above that actually required, the reduction being — made with the rheostat. The centre | tap is the B positive output, which may at first appear strange, but as the curtent to be rectified enters the valve at the plate, and can only travel from plate to filament, the rectified current must be drawn from the filament, and this is done at the centre of the filament winding, which connects through the two chokes to the output termiuals. LAMINATIONS. "THESE are cut from ordinary sheetiron not exceeding 1-64th of an inch in thickness. A sheet-metal worker will supply about 52 pieces cut 5} by 8 18 for 2s. 6d. When the winding is completed it will be possible to see the exact width required for the ‘"‘window" through which the turns of wire pass. This should not exceed 5-8-inch, and way be less. It should not be made larger than necessary. The length of opening is 3 inches. The ‘ron. mav be marked by cutting a

xquare of card the same size, cuiting ont the window, and two notches on the edges where the iron is cut through into two pieces. This card can then be laid on each piece and scratched round or stencilled with white paint. If the two halves of cach piece are numbered 1-1, 2--2, and so on with white paint, they may be assembled in pairs end will fit, however irregularly the dividing cut has been miade. Now each piece of iron is to be divided into two pieces by cutting across the dotted lines shown on diagram, these being 2 inches from the nearest end. Next, a cut is made for each side of the window, and the cut across the short end made by bending down the right-hand end as shown in the diagram. Make a good clean bend of this- it will easily straighten out. Ordinary snips will do the cutting. The bends are now to be taken out roughly by bending back bv hand. ‘The iron is then ready to be flattened by tapping with a hammer on an iron sur-face-the face of an old flat-iron is suitable. The next process is to paint over both sides of the iron with thin

shellac, applied with a brush. The shellac varnish is made by dissolving common shellac in methylated spirits. This soon dries and the assembling may be proceeded with. The wood clamps are dtin. long, drilled 3}in. centres for $-16in. brass bolts 2in. long. The clamps are lin, wide, just under #in. thick to take the brass: bolts easily. Instruciions regarding stalloy strips and packing tight apply here, but the assembling is easier. One side of the iron is to be cut exactly 1jin. as shown, and it is this side that goes inside the winding in every case. ‘The joins are placed at alternate ends so that two

do not come together in adjacent layers. The clamps are put on and carefully screwed up tight, adjusting the position of laminations at the same time. If there is room, adhesive tape should be passed round the outside iron against spool, and the whole bound as tightly | as possible to ensure that there can_ be no looseness to allow of the iron setting up a mechanical hum, which would be detrimental. THE CONDENSERS. WE now come to the smoothing equipment which is to fill in the gaps left by the suppressed half-cycle of current, and turn it into a smooth, unbroken, direct current. The cotdenser capacity totals 8 microfarads, the actual capacities required are 4 microfarads and two of 2 mfd. each, but the 4 may be made up of twe 2’5, as shown in the shaded blocks on plan. Either the Dubilier or T.C.C. makes are suitable in the low-priced quality. A wire connects one end of each condenser and then continues to negative terminal. THE R.F. CHOKE. YHTIS is a useful addition to an eliminator. It is made by cutting two dises of cardboard 2in. in diameter. To the centre of one of these is glued a small section of a thin wire-spool core sawn a little over jin. thick. The diagram shows the R.F. choke spool with one side removed. -- "

The other side is then glued on and 1000 turns of 36’s s.w.g. enamelled wire are put on — THE SMOOTHING CHOKE. THIS can be made on the same model as a transformer, but smaller, and laminations are all assembled to bring joints together and a thin piece of card is put between the butting ends. If you have an audio transformer in your set that is of ancient pedigree and rather small in size .it ‘would be a good idea to take it down, make a new spool, wind on as much 86’s enamelled wire as there is room for, then assemble core with joins all together, slip a bit of eard between joing and clamp up. You can then invest in an up-to-date — transformer for the set and benefit by the improvement. In any case, try to obtain a burnt-out transformer to rewind for the choke. THE RECTIFYING VALVE. A PM4 power-valve can be used as the rectifier, but there are a numiber of speically-made rectifying valves on the market. Mullard DU10 half wave, 2.7 volts, .77 amps. ; Philips 873 half-wave, 2 to 3.5 volts, .6 to 9 amps., are two good valves for the purpose. The UX216B requires a filament voltaee of 74, which would necessitate about 18 extra turns on the winding. When an ordinarv valve such as the PM4 is used, the grid and plate are both to be connected together, the most conyenient way being to continue the connecting wire from plate to grid terminal on the holder. A fullwave valve may have both plates connected in the same way to use for halfwave, but full-vave valves cost more than half-waye. RESISTANCES, A ROYALTY type B variable resistance, 1500 to 100,000 ohms is the best to use in the "low’’ or detector circuit. ‘The two cartridge resistances shown are 100,000 ohms each, but are not likely to be necessary, but are provided for if required. In wiring up, for the first trial, wire across 8S, leaving out the brass clips and resistances. If a lower voltage is required, one or both may be added. ‘heir position is shown by an X on the theoretical diagram. BASEBOARD, PANEL, AND COVER. IIE haseboard is 84 inclies wide and 8} inches deep, and this size will be found to fit into a standard square biscuit-tin, which would make a cheap ready-made cover, stripped of paper and finished with black cycle enamel, All that has to be done to the tin is to punch a few holes at the top and along the back bottom edge for ventilation. The panel may be of ebonito. three-ply. or, better still, tin, also

enamelled black. The Royalty and rheostat will attach without. insulation, so -a small strip of chonite could be holted in to take the three terminals. If tin is used, the sides should be bent back about lin. and cut sloping to nothing at the top. The bottom end can then be serewed to side of baseboard to act as a sirut to stiffen panel and hold it uptight. Light battens should be screwed under base to prevent warping. GENERAL HINTS. A® American valve holder should be uscd unless the required valve cannot be obtained with an American base. In order to protect the secondary Wingines from damace in case of a short uo MATT

circuit in the set, and also to prutect valve filaments {rom damage from the same cause, it is wise to include a tinfoil fuse in the negative B. ‘This fuse (Continued on page 11.)

QUERIES BY CORRESPONDENCE. {. Every communication enclosing queries is to be addressed to ‘Meg: ohm,'’+ Box 1032, Wellington, and must be accompanied by a stamped addressed envelope for reply by post. 2. Questions must be written so that a space is left in which the reply may be added. 3. No charge is made for replies, BEN HUTH ATMA TAIT HI TINT M HAITIAN HTA AAT RATA HUAT ATT ItyT i dtd

Lays. Turns. Volts. 10 3000. 250 9 2700 220 8 2400 200 7 4100 170 6 1806 145 5 1500 120 4 1200 100 8 900 75 2 600 50