A FULL-WAVE B BATTERY ELIMINATOR FOR HOME CONSTRUCTION

Radio Record, Volume I, Issue 15, 28 October 1927, Page 14

A FULL-WAVE B BATTERY ELIMINATOR FOR HOME CONSTRUCTION

A CONSTANT AND PLENTIFUL SUPPLY OF HIGH TENSION AT LOW RUNNING COST

' (Third Instalment.) This 1s the third instalmert of the B eliminator construction article. At the commencement it was statcd tliat the eliminator would run a five or six-valye get, but this statement was too consetvative, for it will run an eight-valve | super-het., and probably a ten-valve set, as the double-wave gives a good supply | of current. In cas¢ of running a large | set it might be necessary to add an extra 2md condenser each side of the choke. It is always good to have plenty &{ reserve of smoothing power, then all chance of lum is cut out. The condenser holder was originally designed to hold seven fixed condensers ef Qinfds. each, and it will work well with these on a four or five-valve set, but the extra condenser was added between the transtormer L/l centre tap and HT negative lead in order. to cnsure an overplus of smoothing capacity. That this capacity is provided is shown by the fact that the set will continue working for a second or so after the 230-volt switch is cut off, The cost of parts and materials required will fotal £7 odd, according to how the parts are purchased. ‘This cost is not great for a double-waye eliminator, as nearly twice the price is asked for the factory-made article. The valves used are two single Clearton rectrons, costing 12s. 6d. each, and with the two working together on wt

double wave, an output »f 100 milliamps may be obtained, As the average set with five valves takes about $v mifliamps, including» a super-power valve in the last stage, it will be seen that there is plenty to spare for the demands of a iarger scé. Raytheon atid other gas-filled doublewave rectifying valves are now appearing on the market. hese valves rectify without the assistance of a heating filament, and thus there is a saving in current consumption, The voltage drop in these rectitiers is apparently lugher than in the filament-heated type, so that more turns would be necessary on the hiigh-tension secondary. ‘the omission of the filament windinz would give room on the spool for the necessary extra turns of hizh-tension secondary. Experiments will sliertly be carried out to find the best method of making the necessary changes, and the result will be published,

THE SMOOTHING CONDENSERS. ‘The seven fixed condensers of 2 mids. capacity each are held in an upright receptacle made of tin, as shown, the outside finished with black cycle enamel, as used for the outside case, It is as well to include the extra condenser and place it at the outside of the case, where it will be held by the connecting wires. Of course, the 4 and 6 mfds. may each be bought in one unit of that value if preferred. Connections of the condensers are shown in the diagram. One side of every condenser is connected. to thie high tension negative, and other sides as shown. The holder shown above is to take the Duhilier make of condenser, but if the T.C.C. type is to be used, a different method of holding will be necessary. A strip of thin board 7in. long by 2}in. wide can be fastened in an upright position and the T.C.C. condensers can then be screwed to ‘it, one above the other. These con‘densers are exactly the same _ thickness as the Dubilier, so they will occupy the same space as_ regards height. Connections to the bank of conden-. sers are clearly shown in the diagram. ‘The one marked "Choke and arms" goes to one end of the main choke, and also to the arms of resistances, all three of which are connected together. THE FILAMENT PANEL, The small panel to which the filament taps of the transformer are attached is shown. The taps run

straight from the end of the transformer to the respective bolts in the panel, To make connection easier, it 1s as well to solder a connecting tag to each tap, and through the hole in this the panel bolt will pass. Near the centre of this panel there are two terminals, below which are two bolts. These are provided so that a short piece of resistance wire may be included in each filament circult to suit the current as exactly as possible to each valve so that the one rheostat will give an equal control of each valve. ‘The two terminals are connected together behind the panel, and the 18’s wire is continued round to the frout, where it is attached to the lefthand bolt giving the required voltage. The corresponding bolt on the righthand is connected by a piece of 18’s to the single bolt at bottom of panel, and from the hack of this a wire runs to the rheostat. The two bolts to which the lower ends of the resistance wires are attached are connected to a fila-

ment terminal of each valve respectively. The panel is drilled for two screws at the right-hand side, and is screwed to & narrow strip of wood, which is first screwed to the partition. This eliminator ‘was designed by ‘‘Megohm"’ for experimental work, and for that purpose this panel was included, as voltage can be obtained for any filament likely to be used. The eliminator will work just as well without the panel, but there is some satisfaction in turning out a well-finished piece of apparatus that is convenient to use. A half-scale drawing of the transformer fuse panel is also shown. The wiring under the baseboard is given, but will be described later. When taking this wiring from the eliminator it was necessary to refer to the published circuit diagram, and it was found not to agree with the actual article. ‘The discovery was then made that the original diagram made twelve months ago before the construction of the eliminator had been used from which to make the drawing. ‘The actual circuit was taken from a well known English radio paper, and before commencing the construction it was found that there was an error in the English diagram, and this was altered | on another working drawing that had heen prepared. The incorrec* diagram appeared sunerficially to be correct and was used, and in the hurry for publica-; tion got hv unnoticed, Next week we shall publish the corrected cirenit, and meanwhile ask readers to take no notice of the one already published.