US2093882A - Electron tube generator - Google Patents

Description

Sept-21, 1937. H. DIRKs 2,093,882

`ELE@TRON TUBE GENERATOR Filed Feb. 5, 1955 Herma/WMM;

fi'ajizzey l Patented Sept. 21, 1937 irs sra maar ELECTRON TUBE GENERATOR Hermann Dirks, Berlin-Tempelhof, Germany, assignor to C. Lorenz Aktiengesellschaft, Berlin- Tempelhof, Germany, a company Application February 5, 1935, Serial N0. 5,154 In Germany February 5, 1934 1 Claim.

The invention has for its object to create an electron tube generator wherein load variations shall be avoided under any circumstances. 'Ihis problem must be considered especially if the op- 5 eration is required to be most constant in frequency even if mechanically-oscillating controlling means such as quartzes are not employed. This requirement is now growing more and more important, since quartz only admits of operating on a predetermined wavelength whilst there is a necessity for working on any desired wavelengths.

The invention is adapted for use with simple transmitting arrangements, i. e. arrangements merely having one self-excited stage, as Well as with transmitters having several stages, but it is possible also to employ it with receivers. The inventive idea is explained hereafter as employed in a two-stage transmitter controlled from a separate oscillator, the drawing being a diagrammatic illustration of this embodiment of the invention.

The controlling transmitter is designated R1,

the main transmitter designated R2. Transmitter R1 operates on the principle of self-excitation in a well-known manner. Disposed between the anode and grid of the transmitter R1 is an oscillatory circuit containing the split or tapped condenser C1 and the self-induction coil L1. In the arrangements hitherto in use the grid is connected to a suitable tapping of the condenser C1. There is thus a capacitive potentiometer-connection for effecting the back-coupling. Condenser C1 is earthed at a suitable point E. Moreover, transmitter R2 is connected to the condenser C1 and is through a variable condenser N2 of the transmitter R2 neutralized or balanced out from the anode in a Well-known manner.

This customary arrangement has the following disadvantages, which are simplest to explain in connection with the usual adjusting of the transmitter. For setting the neutralization condenser N2 the anode potential is disconnected from R2 whilst N2 is Varied until the oscillatory circuit connected to R2 shall be free of current. In this way the stage is neutralized. When now the anode potential is connected again, the inconvenient phenomenon results that at the same time the frequency of the transmitter R1 is changed, although this transmitter has not been subjected intentionally to any influence. The act of connecting the anode Voltage is identical with the keying effected in practice. Closer re- 5 flections on this state show that the variation in frequency is attributable to dephased currents in ltransmitter R1 which are duev to the detuning ofl transmitter R2 and at the same time act to'l change the phase of the back-coupling voltage and thereby the frequency.

According to the invention therefore it is pro- 5 posed to effect the back-coupling oi the transmitter R1 by purely inductive means and to so arrange this transmitter that the controlling voltage supplied to its grid is in unkeyed and keyed condition merely a voltage out of 10 phase over the anode alternating potential. Transmitter R1 is neutralized itself. It is known per se to neutralize stages located behind the controlling transmitter, but it is not known to neuralize a controlling transmitter, this transl5 mitter not being able to oscillate owing to such neutralization. Its oscillations are merely attributable to the inductive back-coupling.

For eiecting the inventive idea, the back-coupling of the transmitter R1 is not capacitive but 2O inductive, namely through coil L2. In addition there is provided a variable condenser N1 connected after the manner of a neutralizing condenser (one may perhaps consider such arrangement tobe a compensation device) and 25 having for its object to compensate for dephased currents.

From the customary structure, wherein the grid is connected to a tap on the main coupling condenser whilst Lz and N1 are missing, the 30 following results:

If transmitter R2 is not keyed and not well neutralized, then the frequency of the transmitter R1 can of course not be influenced by transmitter R2. If, however, transmitter R2 is keyed, 35 i. e. detuned, then capacitive currents I, 2 flow over the grid conductor and cathode conductor. These currents split up differently. Current I splits into a component 3, that ows to earth, and a component 4, whereas current 2 because 40 of the less resistance mainly passes to earth. In condenser C1 a current 5 flows at a definite moment, this current arriving from transmitter R1.

It has been assumed that all these currents are cophasal. Already on this assumption it will be 45 seen that owing to the presence of currents I and 2 the back-coupling potential in the grid of the transmitter R1 is changed, for in condenser C1 the currents are added in part, for example 3 and 5, and in part subtracted, for instance 4 and 5, or 2 and 5. This involves that the back-coupling potential derived at the point S, i. e. the potential by which the frequency of the transmitter R1 is determined decisively, depends upon the currents I, 2, that is, upon the state of operation of the transmitter R2. In addition, however, phase rotations will occur, because connected in parallel with condenser C1 is the self-induction coil L1, and current 4 flows not only over the internal tube capacity of the transmitter R1 but in part also through the self-induction coil L1 which for this current is connected in parallel with the internal tube capacity.

The described disadvantages are avoided by the invention according to which, as stated, the backcoupling is not effected capacitively but by purely inductive means, that is, with the aid of coil L2, for example. Owing to the transmitter R1 being neutralized, that is, located in a bridge arrangement, the capacitive currents cannot bring about an action on it, provided that their coaction takes place symmetrically, that is, the capacitive currents I and 2 for instance should not act to produce a drop of potential in the bridge diagonal, i. e. in the coil L1. It will be easily seen that this can be performed readily, because the neutralization acts to balance the bridge, no matter from Which side the currents are supplied to the bridge. Hence there remains connected to the coil L1 a purely inductive potential only which is conducted to the grid of the tube R1 over coil L1 in a state of correct phase, that is to say, the back-coupling Will be always of the correct phase coincidence,

no matter Whether transmitter R2 is loaded or not loaded.

Experiments have shown that it is possible in this way to avoid the said disadvantages and to attain that a small number of stages shall Abe sucient. It is to be understood that still other measures than those described may be adopted for effecting that the back-coupling of the controlling transmitter or the tube generator is in general performed by a single potential which is 180 out of phase, and that the other potential influences, due to loading or unloading, are compensated.

What is claimed is:

In an electrical system, an oscillator comprising a space discharge device having input and output circuits coupled by inductance coils, one of which is inthe output circuit and with a condenser forms the frequency determining circuit of the oscillator, a variably loaded space discharge device having its input circuit capacitively coupled to the oscillator by said condenser, and means connecting the inductance coil of the frequency determining circuit across one diagonal of a balanced Wheatstone bridge with the output terminals of the second discharge device connected across the other diagonal thereof.

HERMANN DIRKS.

Images