popular: The Filadyne detector by G. V. Dowding

Extracted from Popular Wireless and Wireless Review, July 10th, 1926. Pages 671, 672

THE FILADYNE CIRCUIT. A Novel Design, for the Experimenter. By G. V. DOWDING, Grad.LE.E. (Technical Editor.)

Once upon a time I drew a diagram in great haste to illustrate something or other and omitted to include phones. " That won't work," declared one of my friends with bland finality. This was in the early days of radio telephony, some ten years ago, and, although some of our " hook-ups " were temperamental to the extreme, the particular circuit in question was a straight Det., 3 LF, all transformer coupled. Therefore, I was rash enough to rush into a wager, and then I saw my mistake!

But I strung together a number of components exactly as shown in my diagram, including with great cunning in the last LF stage a transformer privately labeled by me as " the talker."

Its peculiarity was that it had badly assembled laminations, and, given sufficient input, it would " loud speak," to the complete mystification of all listeners.

Now, I don't want to justify my early artfulness here, but the incident serves to point out a moral, and, according to my interpretation, it is to the effect that no one knows how far an electron will go until they take it there and that no one knows what it will do until it has been made to do it!

The Three-Electrode Valve.

To treat the matter more scientifically, let us take as an example the three-electrode valve. Here we have a filament that when heated emits electrons, an anode or plate to which a proportion of such electrons are attracted by virtue of its positive potential, and a grid that influences this flow of electrons by virtue of either a negative or positive potential, according to circumstances.

The three electrodes are shown in the above order 1, 2, and 3, in Fig. 1 A. That in a nutshell, or, rather, an evacuated glass bulb!- is the theory of the operation of the valve as used in ordinary present-day circuits.

Fig. 1 shows a straightforward detector valve circuit, the reaction being eliminated for the sake of simplicity. Impulses received from the aerial circuit are passed on to the grid, which, by influencing the flow of electrons from the filament to the plate or anode, causes variation in the flow of current from the high-tension battery through the telephones and valve, for the stream of electrons forms, as it were, a bridge over what would otherwise be a break in the HT telephones circuit.

First Steps.

But all the electrons emitted from the filament do not reach the anode, and that is why I employed the reservation "a' proportion."

A great number tends to form what is known as a " space charge," and, by gathering together in a cloud materially resists the progress of others, for an electron is negative in sign and will repel the approach of another electron.

Readers will remember that the inner grid of a four-electrode valve is employed in the Unidyne for the very purpose of overcoming the effects of the space charge.

Briefly, then, in connection with the ordinary three-electrode valve, it seemed to me that it should be possible to do more with the emitted electrons than is done at present. Obviously, the greater the number that is allowed to bridge the HT battery - 'phones circuit, and the more they are influenced by the incoming impulses, the more sensitive will be the receiver.

Instead of influencing the electron stream in the middle, as it were, it occurred to me that a far more effective scheme would be to influence them at their source and to adopt measures to reduce, if not entirely abolish, the space charge.
For many years we have regarded the valve as a sort of concrete base upon which to build our circuits. Seemingly, the commencement of every " hook-up " is " take a valve " and then connect to its grid the input, and so on. Now and then, it must be admitted, the beaten track is departed from, and anode inputs are employed, although for what reason it is difficult to see. There is far more diversity in transmitting than in receiving circuits, although in practice few of the startling diversions are of any real value!

But I propose to describe a system of filament input which, with all due modesty, I claim to be worthy of the attention of amateurs as a system that is more logical than grid or plate inputs. I have styled it the Filadyne, and, as dyne means force or power, the name should be quite appropriate.

Filament input, or the control of the electron emission from the filament at the filament by incoming impulses, it might be thought, could be carried out merely by reversing the grid and filament connections from the aerial tuning circuit to the valve, but in practice, it could not. In any case, in evolving the Filadyne one of my endeavors was, as previously stated, to reduce the effects of the space charge. 

This was almost a starting point. and was achieved by giving the nearest electrode to the filament, i.e., the grid, a very high positive potential by connecting the positive of the HT to it through the phones. Fig. 2 illustrates this, the first step. But as a circuit, I did not expect results at this stage. 

The filament is, of course, connected to the LT battery, and HF charges could be expected to spread right through this system, and, indeed, pass to earth capacitively or directly. Therefore, I isolated the filament by introducing two H.F. chokes of low ohmic resistance, as at A and B in Isolating the Filament. 

From this point it was a simple step to introduce reaction, as in Fig. 4. Results then were excellent, and both in loudness and clarity of signals and in DX capabilities it was found to be better than an ordinary straight circuit.

The importance of the isolation of the filament was such that passing any one of the chokes by means of a condenser caused complete silence.

Clearly proving that the circuit is abnormal in operation, it was discovered that a grid leak and condenser in any position was not only unnecessary but could not be used.

Interesting Points.

 Whether the HF impulses can actually affect the emitting properties of the filament of a valve, or whether the shape or disposition of the electrodes affects results, the fact remains that some valves do not operate at all in the Filadyne.

The B5 (B.T.H.) appears to be eminently suitable, and so do one or two others, but not many. Now, the B5 is a " lively " valve, but in the Filadyne 100 or so volts HT can be used with excellent stability of control, although actually between 50 and 60 only are required.

 It is interesting to note, too, that the grid and plate connections in the Filadyne cannot be reversed, nor can the LT connections, and that the disconnection of the plate causes complete cessation of signals. But the plate is at earth potential!

 The aerial circuit is - filament - plate and the " anode " circuit - grid - plate.

 In tuned anode circuits a 75-turn coil will cover practically all the ordinary, broadcast wavelengths, while a No. 250 will cover the wavelengths of 5XX and Radio Paris.

 If we were asked to specify the minimum number of coils for ordinary broadcast work with an HF-Det-LF three-valve set employing a tuned anode circuit and reaction, we would include the followings 40, 50, 75, 150, 200, 250. With these, we would hope to be able to get almost any station up to and including 5XX.

 The foregoing three paragraphs at first sight appear to form a confusing jumble of observations and surmises, but, nevertheless, every statement is highly significant.

For the benefit of amateurs who wish to carry out experiments with Filadyne circuits, it might be as well to mention that the HF chokes used consisted each of 250 turns of 24 S.W.G., wound honeycomb, coil fashion. They must not couple, that is, they should be well separated and mounted at right angles to each other. The total ohmic resistance of both chokes together is but a matter of an ohm or two, so, using a four-volt accumulator with a B5, a filament resistance with a maximum of at least 15 ohms must be used. Everything else is quite normal.

 Volume is excellent, and the purity of reproduction is exceptional. Filadyne receivers are remarkably selective, and yet are as easy to handle as ordinary sets.

 Finally, it is interesting to note what isolation is provided by an efficient HF choke. The filament resistance, as will be seen in Fig. 4*, is separated from the filament by a choke.

 The resistance can be touched with the fingers without affecting reception, but not the other side of the choke. Touching this causes a complete cessation. This is more interesting in that it proves that the circuit operates as I intended that it should do and that the filament is indeed, the governing element.

* Fig 5 mentioned in the original. No Fig.5 exists, assumed Fig 4.

 

To thank the Author because you find the post helpful or well done.

Hello Radiofreaks

Thanks to Gary, who introduced the „Filadyne“ to us.

Because it was unknown to me, I was encouraged to build something like this in a simple mock-up chiefly for study. Concernig the used tube, the TELEFUNKEN RE084 is also a Dull-Emitter and was made in the 20s.

During experimenting with proposal Fig.3 it became quite clear, the original filament voltage must be reduced 2...2.5 volts! Also surprisingly was the fact that the 'B' Battery voltage is of less influence and can be lowered to almost starving level! Even down to 4.5 volts, thus only a single 'A' Battery will do.....

Next step was to bring in regeneration with an additional tickler coil as recommended in Fig.4.
It was rather disappointing that no improvement could be seen. So I switched to higher supply voltages but no reaction at all....

When I compare circuitry Fig.3 with Fig.1, the difference is a lot. At first, the well known Grid Leak Detector (Fig.1) doesn't work with the drastically reduced voltages. On the other hand, when the HT is 16 volts or more, it comes to live! Not only the sensitivity is 15 times better also regeneration can be achieved too

Conclusions

The „Filadyne“ as shown above is a nice beginners project.
The lack or regeneration could be solved with a more suitable tube like the proposed B5 (B.T.H.), I think.

Kind regards

To thank the Author because you find the post helpful or well done.

To make things more understandable I made another drawing.

Basicly it's a so-called "Inverted Triode" circuitry. The steering voltage is connected between the anode and the cathode (here the filament). At the grid we have the outgoing signal.

Not new but seldom used because of some disadvantages especially with the correct biasing.

It becomes obvious that the HT source between the anode and grid must not exceed a certain level. Therefore a small amount was sufficient in my trial!

Kind regards

 

To thank the Author because you find the post helpful or well done.