The "KESTREL THREE"

Radio Record, Volume V, Issue 2, 24 July 1931, Page 16

The "KESTREL THREE"

By

The Technical Editor.

. CONSTRUCTOR walked into + our Office last week and left a note-or rather an ulti-matum-for me, for at the time I was out. His remark was that unless the "Kestrel Three’ was published this " week he was going to sue me. Now, in face of that, the constructional details can be held back no longer. We are very sorry there has been delay with the publication of this set. It was due in the first place to the absence of differential condensers, then, on further experimenting with the set, some litile troubles came to light, and we had to eradicate them. This necessitated certain structural alterations, and alterations in the circuit, with the result that the "Iestrel Three" was pulled down and rebuilt. In order to refrain from publishing a circuit that was not going to give 100 per cent. satisfaction, we again withheld publication until it was going absolutely perfectly. Now we can offer it to readers knowing that it is the best "Three" yet; that for an all-wave set, if made to specifications, it will give absolute satisfaction. At the time of writing we are not quite satisfied with the results below 80° metres, and we are convinced that our coils can still be altered to good effect. However, we shall not hold back the description any longer, but next week, or at least when we are quite satisfied our coils are right, the complete specifications for short-wave coils will be published. The "Kestrel Three" embodies several different ideas-we shall not call them new-and constructors are asked to follow the details minutely.- Do not think because we say a_ connection should be made in such-and-such a way that another way is just as good. You can rest assured that it is not. Constructing. the Set. Now that we have told you all about what the set will do, we shall not bandy further words, but get right on to the job and tell you how to make it. In a separate panel you will find & shopping list. Get all the parts be-

fore you start and make sure the parts are as specified. There is no real reason why two .0005 variable condensers will not do just as well as a .00085. If you like to make

alterations in the coils they- will be all tight. Furthermore, a larger or a smaller differential condenser will also do, but you must be prepared if you alter the specifications in condensers to juggle with the number of turns on the coils to get them absolutely right. "We are publishing only one set of turns-those for the .00035 tuning condensers and for .00025 differential condensers. We know that they work, and work well. We know’ nothing about the others, although they will work equally as well, we feel sure.

For smooth operation we are introducing potentiometer control of grid returh, This to us, seems easily the best method of bringing the grid return to the filament. It is an idea that is employed in most high-class English sets, and’ we can thoroughly recommend it. A proper setting of the potentiometer will give a smooth control of reaction. There is no need to be bothered with ploppy oscillation, for it can be made as smooth and silky as desired merely by adjusting this resistance. See, too, that you get a good radio frequency choke. A poor choke igs one of the most common causes of . poor regeneration. It should have.a low self-capacity and a high inducfance. You should be able to get quite a good one for 6/- or. 6/6. Do not compromise on the grid condenser, as this has to act as a blocking condenser with regard to the high potential current that feeds the plate of the screengrid valve. If it is cheap and shoddy, it is likely to break down, with detriment to your batteries and valves. Apart from these the components are not very critical. A word here concerning the baseboard will not be out of place. It is unnecessary to use a special aluminium sub-panel for this set. If the shielding recommended is employed there is no need to go to the extra expense entailed in the aluminium panel or sub-panel. A piece of three-ply will do splendidly, but it must be kept off the ground with wooden | strips about din. x Zin. These are placed 2 along both long sides, and the panel’ can then be fixed to the plugs along the side that will be thé front. The size of the baseboard is Tin. x 14in. If, however, you are using biggish components, allow another inch or so in depth, Sin. or 8tin. x 14in.

The Panel. {2 will ‘be seen from the diagram of the panel that the set when finished will present a quite neat appearance, and so the lay-out of the panel is quite an important part. Here are the directions. , . Lay the Tin. x 14in. panel flat, and mark the points for the tuning condensers. These are 4in. from the bottom and Sin. from the end. The switch on the left is 1}in. from the bottom and lin. from the left-hand side, while the phone jack on the other side is in exactly the same position, The rheostat is 2in. from the bottom of fhe panel and 6in. from the left-hand end. The differential reaction condenser 0occupies the same position on the right-

hand side of the panel, that is, it is 6in. from the right hand end and 2in. from the bottom, Be sure you get these two last-mentioned components in this position, otherwise you will have difficulty with the screen. If you are using a big rheostat it is better to move both theostat and differential condenser in to 6jin. or 6%in. from the end. This will, to some extent, spoil the look of the set, but better do this at the start than discover, after the hole is bored, that the condenser and rheostat have to be shifted and other holes drilled. It*must be borne in mind that the shield will come exactly five inches:

from the left-hand end. About a quarter of an inch at the bottom of the panel on the long side drill three holes to which the screws are to pass into the strip that is supporting the three-ply baseboard. Affix the components to the panel, noting that the tuning condensers are to be mounted at an angle of about 45 deg. with the baseboard. It is preferable to mount the rheostat with the terminals up. Now take your baseboard and lay out the components as shown in the diagram. Slight alterations in position may be necessary owing to slightly different sizes 6f various makes of components. Note particularly the way the valve-holders are turned round. If you do not get them round the right way, you will find that when the set is finished the Wiring will be long and unsightly. Generally speaking, the valves are toward the back edge and the coils are near the front of the set. It is well, when laying out the components, to put the shielding ia ees eee

position and lay the components out on it as well. You can then get an idea of the set as a whole. The shielding: consists of three sides of an aluminium box 44in. wide by Tin. long and Gin. high. Unless you have facilities for bending the aluminium, do not attempt to do it yourself, but order it from your dealer, bent to the shape indicated. A copper box would do quite as well. It fits'flush with the left-hand end with the metal end up against the panel. When it is fitted in position, hold it steady and mark the point where the r.f. condenser shaft is to go through. This can now be drilled and ringed out to the right size, and the condenser mounted. It will hold the shield in position, while the rest of the lay-out is completed. It might be as well to pierce the shield where the switch is to come through, and also to fasten it

in position: The shield will then be quite firmly held. When. the components have been fitted into position, take a seriber or similar sharp instrument and mark where the screws ‘are to go through them to the baseboard. Move the components one by one and make sure that the drilling points are marked before the component is removed. Apart from screw holes, the point where the baseboard must be drilled to allow a wire to go through to the other side has been clearly indieated. Before mounting the midget condenser that is to go in the aerial, make’ certain that no part of it will touch the shield. For this reason we recommend one of ‘the types with a bakelite base. This can then be 8cfewed down to the baseboard without any fear of its touching the screen. (To be continued. next week.)

No. of Turns for Broadcast Coils, .000385 Condenser. Qin. Former. Regeneration .00025 Differential. Radio and Detector: Secondary ...: 82 turns 26 d.s.c. * Detector Coil: Secondary .... 78 turns 26 d.s.e. Tickler .......... 50, 30 dis.e.

=-Pu@ Parts List of "Kestrel Three" Ebonite panel, 14in. x Tin. Three-ply baseboard, 14in. x. Tin., with quarter-inch slats to protect under baseboard wiring. Aluminium screening box, Jin. x 5in. x 6in., open on two sides and top. -00035 variable condensers. vernier dials, -00025 differential condenser, 30 ohms rheostat, 400 ohms baseboard mounting potentiometer, filament switch, -0001 condensers with grid-leak clips. -0002 condenser. 2 megs. grid-leak with holder. 10 mogs, grid-leak (for short-wave). px sockets or 3 UX and 2 Engish 5 mfd, by-pass condenser. .05 by-pass condenser. midget condenser. push-pull battery switch, transformer, ratio 34 or 4-1. single output jack. ‘ rf. choke, . . 2 terminals with short 4in. armoured cable, Special coils for broadcast or two four-inch lengths of two-inch former. 4lb. 26 d.s.c. wire. . 8 or 9 yards of 30 gauge d.s.ec. wire, 6 wire battery cable. paki CT OR RO