Radio Broadcast (May 1928-Apr 1929)

AUGUST, 1928 ADDING REGENERATION TO ANY SET 211 +90 FIG. 2 The same detector circuit as shown in Fig. 1 , hut drawn differently. The bypass condenser, C\, connects directly from the bottom of the primary to the filament and hence to the bottom of the secondary winding. Radio frequency currents are bypassed from the plate to the filament directly is often as good for weak signals as two or more stages of radio-frequency amplification. How can we make our neutrodyne detector, Fig. I, look like Fig. 3? Add a tickler coil? No, says Mr. King, there is a simpler way. Let us look at Fig. 4. This is exactly the same as Fig. 3 but drawn differently — the tickler coil is at the bottom of S instead of on top of it. It works just the same. Thf lext step, shown in Fig. 5, is more complicated. Here is where the trick begins. In Fig. 5 we have added a radio-frequency choke and regeneration condenser, C3, and placed the audio system as a shunt path, instead of a series arrangement, to the regeneration circuit composed of T, G> and C3. The choke keeps the r.f. currents out of the audio system; C3, which is about 0.0001 mfd. keeps the audio currents out of the tickler circuit. We have provided an effective filter system, keeping the two frequencies in their respective channels. The tickler coil, T, is now fixed in place and regeneration secured by varying C3. THE R. F. PRIMARY AS A TICKLER HERE is where Mr. King steps in. He notices that the primary coil, P, is attached to the secondary by Q, just as tickler T is, and says why not use P as the tickler coil as well as the primary of the r. f. transformer which connects the r. f. amplifier to the detector? The result is Fig. 6. If applied to a neutrodyne or t. r. f. set of the conventional type it will be necessary to adjust R.F.Choke FIG. 5 Instead of a series tickler and audio system arrangement, we can use this arrangement, which shunts the audio input across the regenerative apparatus. Variable regeneration is obtained by the condenser, C3 the regeneration condenser at each frequency, since more regeneration is needed at the longer waves. But if applied to an Equamatic, and Mr. King had this receiver in mind when he worked out the trick, C3 can be placed inside the cabinet, adjusted, and let alone. The photograph at that head of the article gives a good view of the installation. It can be fixed so that the detector oscillates all the time, or nearly oscillates, or so that the additional pep is just enough to give one the dx required. This simplicity is due to the variable coupling between P and S, which is the heart of the Equamatic system and which automatically increases the coupling between T and S at the longer waves where more coupling and more regeneration are needed. Sometimes if the impedance of the audio frequency channel is sufficient at radio frequencies to keep r.f. currents from flowing through the winding and to force them through the shunt circuit, the choke may be omitted. It will be necessary to remove the condenser across the audio input, and if the regeneration condenser, which may be 0.0001 mfd. as a starter, is not sufficient to cause oscillations at the longest wavelength to' be received, the choke may be added. Suppose, then, we have an old-style neutro FIG. 4 The same regenerative detector drawn differently . The tickler coil is at the filament end of the secondary instead of near the grid end It works equally well in the two cases dyne. We get into the box, attach our regeneration condenser from the plate of the detector to the plate of the previous r.f. tube, and listen-in. If nothing happens we look for the bypass condenser across the audio transformer primary and remove it. The detector should now oscillate at the shorter wavelengths and with about 0.0001 mfd. regeneration capacity. If the detector does not oscillate, the primary of the transformer, which is also being used as a tickler coil, must be reversed. If the detector oscillates on the longer waves as well as on the short, all well and good. The next step depends upon the type of set to which this trick is being applied. If it is an Equamatic receiver, in which the primary automatically moves as the tuning condenser capacity changes, nothing need be done but to determine the best setting for the regeneration condenser. This is the point at which regneration occurs at all wavelengths, but actual oscillations do not occur at any. Some receivers have decided bumps and hollows in the gainfrequency curve, so that a setting of the regeneration condenser which does not cause the FIG. 3 A rengerative detector. The additional coil marked T is the tickler. There is no more sensitive or efficient single circuit than this detector to break into oscillation at one frequency will cause decided oscillations at another. The trick is to find the position of this condenser which causes oscillation at no setting of the tuning dials. If the receiver has fixed primaries, as in the neutrodyne or the t.r.f. sets, the regeneration condenser had better be on the panel so that the amount of regeneration is under control at all times. With such an additional control it is always possible to make the detector circuit oscillate, and to tune in stations by the familiar "squeal" method. If oscillations do not occur in the longer wavelengths, the choke will be necessary. In the latest Equamatic, in which a stabilizing condenser is used as shown in the photograph at the head of the article, the procedure is as follows. Open up the stabilizing condensers until the receiver oscillates. Then screw them down slowly until the set is under control at all frequencies. Then add the regeneration condenser, and choke, if necessary, and ascertain the combination of stabilizing capacity and regeneration capacity that is best — it must be determined by experiment alone. The result of such a procedure is to make the receiver, be it an old-style t.r.f. set that has little or no amplification at the lower radio frequencies, or a modern high-gain set, more sensitive and naturally more selective. The parts necessary for these changes are but two, the 0.0001 mfd. variable condenser and the r.f. choke — which may not be necessary. FIG. 6 Instead of adding a tickler, we may use the primary of the interstage transformer, not only as a primary, but as a tickler as well. The r. f. choke may not be necessary, provided the condenser across the audio primary is removed. Condenser C3 provides the variable regeneration necessary to cover all frequencies