MORE ABOUT OSCILLATION.

Evening Post, Issue 88, 12 April 1924, Page 22

MORE ABOUT OSCILLATION.

AN IMPORTANT MATTER.

The writer of this column has received a number of bitter complaints about "squealing valves " backing-up entire ly his recant remarks on the subject, and asking him to say some rnqra about it. The unhappiest victims are people vho have tried to entertain their friends with a. wireless evening, arid have had their efforts spoiled by having to explain those funny noises. Those " funny noises " are quite enough to help some people to make up their minds that wireless may be all right for enthusiasts, but it is poor in the drawing-room. - Let all listeners-in who read these notes, then, concentrate, their attention ior ■a 1 "while upon a few questions that reed their .best efforts. >As surely as they. are considered rightly, they will improve their pastime; as surely as they are neglected, they will help to bring it into disrepute. , 1. Does your set'cause interference? .. 2. If so, why? 3. How can you stop it? Probably every listener-in knows the symptoms of interference caused by somebody else—the squeal, the: wliisle, buzz, or chirp that "butts in'" when one is listening to a transmission. But it is quite likely that a. good many do hot know the symptoms of interference caused by their-own apparatus. If when one is listening the squeal is heard, and is extremely loud, it is almost certainly due to one's own set. 'In such a case.it will be found ' that the squeal changes in note when the tuning altered. Similarly, if'the squeal be ever so feeble, it will, if caused by one's own set^ rise or fall in the scale if one oi the tuning dials is moved. On the contrary, if the interference: comes from some other receiving set, the moving of a dial will not change the note, but only the loudness of the interfering whistle. At the same time, it will bo noticed that a squeal due to one's own. set will almost prevent any music or voice being heard, whereas- - the foreign one will merely be a noise ■ imposed upon tha transmitted sounds.' What' always puzzles ,the victim who has to put up with interference duo to other sets in his neighbourhood is why their owners continue in their bad habits, for he knows perfectly well that the perpetrator is hearing nothing whatever but howls. THE CAUSE OF- THE TROUBLE. There is only one cause of this nuisance of the either, so far .as the receiving apparatus is concerned. It is that the set is," oscillating." It never occurs except with what are known as- " regenerative^' circuits, which necessarily employ1 valves for reception, though in a "small percentage of cases the sets may; oscillate without having been intentionally constructed to do so> owing to un-designed-regenerative effects. ''. All users of valves' should study at least the simpler aspects of the theory of their operation. There are always two circuits connected to a valve—the input circuit and the output circuit. One end of each of these is connected to the glowing filament of the valve. ;,The.other end of the input circuit, which receives the signal, inpulses from the aerial, is connected to the "grid," a fine spiral of wire in the vacuum tube; and the other end of the output circut is connected to the "plate ". or sheet of metal which surrounds the grid. * Now the nature of the valve is such that when the filament is glowing-it is possible under certain conditions (which are duly provided for in the construction of the set) for a 'small current to pass between the filament and the plate,' traversing the vacuous space, and, of course, passing through the grid. This current exists because the filament discharges electrons, which are regarded as actual particles of electricity. It is also in the nature of the valve that any change •in the ' electrical condition (the potential) of the grid exercises a powerful control over the number of electrons which pass through it, and the incoming signals, being conducted into the grid" change its potential condition and thus modify the plate current. Actually the changes in the plate current thus caused are many times greater than the changes in current affecting the potential of the grid, so that the plate current is in a sense a magnified copy of the signals. Thus the valve acts primarily as an amplifier. But even then tho signals heard by converting the plate-current variations into sound are not usually as loud as may be. desired. But if the plate current is passed through a coil placed close to the tuner coil which feeds the grid with its varying potential (or some other means is adopted for the same purpose) some of the plate-current energy is passed into the grid circuit and intensifies the changes of grid potential. This further, enlarges the plate-current variations, and the process obviously promises to go on indefinitely. That is just the trouble. .If the "feed-back" or "regenerative" action is carried beyond a certain point/ the effect-is that the valve begins to "oscillate" the circuit; it pays no further attention to. the incoming signals, byt proceeds to carry out a continuous generation of interrupted currents, of which some of the energy is passed back into the aerial and is dissipated in the form of ether waves. If these have a frequency approximating that of a wireless trans mitter, as they do when the set'is tuned to receive the signals, then listeners in the neighbourhood will experience interference because the carrier-wave of the transmitter and the waves from the oscillating set, both inaudible themselves, combine to produce an audible note. The note will vary in pitch according to the difference in frequencies, and changes in this difference are brought about through alterations of the turning of the set which is oscillating. That is why the interfering noises go "whee-oo" or vico-vorsa, or chirrup like dick-birds at bed-time, drat them. ■ SO MUCH FOR THE "WHY." The cure ia simple enough: avoid excessive regeneration. It will be noted that this oscillation only occurs "beyond ■A cortain point," and-it is easy to prevent passing that degree of regeneration. Feed-hack is' performed in two principal ways—by means of a plate-circuit coil placed'near the grid circuit coil; or by means of a variometer. or : other tuning device, in the plate circuit,/notplaced against a'grid coil.' Whichever of these is ; used, it is made adjustable for. the sole purpose of' controlling regeneration. Move the plate coil close to the grid coil, and regeneration increases ;at a certain adjustment the act will osc-llatet.' Move it away, and unless the coil is too big, oscillation will stop. If it does not stop, try a smaller coil in the plate-cir-cuit. If a'variometer is used, it will pass into and out of the oscillating position as il is turned from minimum towardo maximum. To «top oscillation,

turn,_ it towards the minimum or zero position. " . ' ' This is important: if the set will not work without oscillating, have it .overhauled; there is something wrong with Make this experiment: With the reaction coil turned right back -to ■ "the panel, or the reaction variometer atzero., tune in'the local broadcast with the tuning .adjustments alone. Now•slowly and gently increase the reaction. • waking, if necessary, slight readjust-' ments of the tuning, dials.. You will find, that the music or speech will come in\ stronger and stronger,, up to a point just short of where oscillation begins. Go further, and you will find that the reception is not nearly so good, so "as you were." That's.the place to stop. REACTION BY TUNED PLATE. The writer was interested to notice in a prominent wireless'magazine a description of a receiver regarding' which the author remarked that its performance was remarkable because though there was no evidence of any arrangement capable of causing regeneration, the effects were .similar to those, of a.regenerative receiver. He was^ emphatic as to the absence of a regenerative'arrangement, and truth,to .tell,, the set had neither the plate-circuit coil coupled with the grid circuit, as.in the ordinary threecoil outfit, nor had it a«.varioineter in the plate circuit. ■' But it was regenerative none the less, because the plate circuit contained a.coil timed with a condenser.. It should .be obvious that a plate circuit of .that kind—such' as is used in a "tuned anode" radio-frequency amplifier—is regenerative in exactly the same way" as the . well-known variometer feed-back circuit. The regeneration is effected through the grid-to-plate capacity, owing to the high'potential variations impressed on the plate by the oscillation of the circuit inducing opposite charges in the grid, and so,increasing the variations in the plate current. In such a circuit the virtue does not lie in ;the fact that .a variometer is used, but in the fact that the circuit is tuned; and it is not even necessary to tune the plate circuit directly at all. Suppose, for example, that the plate current actually traverses a coil or 25 or 30 turns/ which is wound as the primary of a transformer, the secondary, consisting of 60 or 70 turns, may be tuned'with'a "variable condenser. It will be found that tuning this circuit will cause regeneration to the full extent, notwithstanding that the voltage changes in the actual plate circuit are much less than they would have been had the larger tuned coil been inserted in it. Such a ' arrangement is highly praised for ' use with amplifying valves, including those/of reflex circuits. The possibility of these circuits causing interference by oscillating the set must not be overlooked. .When a crystal detector is ' used, fhe" resistance of the . crystal may absorb sufficient energy to prevent oscillation; but if in the course of operation the contact of the crystal is broken, the set will certainly oscillate., strongly, and the fact will not necessarily be notified to the operator by the squealing of the carrier wave. . Special precautions Bliould therefore be taken to prevent inference from this cause.