US2173925A - Noise reduction circuits - Google Patents

Description

Sept. 26, 1939. Q T XEN 2,173,925 I NOISE REDUCTION CIRCUIT Filed March 15, 1937 3 Sheets-Sheet l LOWAF.

6 2 orro TUXEN 1 2 BY ygg ATTORNEY Sept. 26, 1939. oQTUxEN NOISE REDUCTION CIRCUIT 3 Sheets-Sheet 2 Filed March 15, 1937 INVENTOR 0T TUXEN BY 7% ATTORNEY Sept. 26, 1939. o. TUXEN 2,173,925

' NOISE REDUCTION CIRCUIT Filed March 15, 193'! 3 Sheets-Sheet 5 llll A'Al GI SZ ==-C To g 7UGE/DOF V T D A INVENTOR TTO TUXEN BY M-y-W ATTORNEY Patented Sept. 26, 1939 PATENT OFFICE NOISE REDUCTION CIRCUITS Otto 'Iiixen, Berlin, Germany, assignor to Telefnnken Gesellschaft fiir Drahtlose Telegraphic in. b. H., Berlin, Germany Application March 15,

Germany, a corporation of 1937, Serial No. 130,833

In Germany April 9, 1936 10 Claims.

The present invention relates to an arrangement for eliminating or reducing atmospheric disturbances in receivers for Wireless telephony, especially radio receivers.

It is known that the so-called atmospheric disturbances reveal themselves within the demodulated low-frequency spectrum as short voltage pulses whose harmonic analysis furnishes partial frequencies situated primarily in the higher part of the sound spectrum, hence above about 4000 cycles. There is based upon this fact the method of weakening the disturbances by cutting off the higher audio frequency range. However, this method has the disadvantage that at the same time the higher notes are obviously suppressed.

' It has been proposed to out off, through the insertion of elements limiting the amplitude, the voltage peaks produced by the atmospheric disturbances and the like, and which are superimposed on the remaining audio frequency mixture. If the maximum amplitude which the amplitude limiting element just still permits to pass is equal to the amplitude of the non-disturbed audio frequency, then the disturbance will be practically entirely out off if a disturbance occurs in the form of an additional voltage pulse at the moment in which the voltage curve of the non-disturbed audio frequency passes through the maximum,

and if the two directions of the voltages conform with each other. The actual disadvantage is seen in the fact that disturbances occurring in a moment in which the audio frequency voltage curve passes through zero, or in which its principal direction is opposite to the incidental direction of the voltage, are not weakened at all or only to a small degree. On the other hand, also the amplitude of the audio frequency, i, e., the volume, fluctuates within wide limits. In order to avoid distortions the limit voltage of the amplitude limiting element must lie above the natural audio frequency voltage appearing at maximum volume, whereby only very great disturbances can be weakened at all.

An arrangement has become known in which the limit voltage is automatically adapted to the respective volume through an automatic control device. However, there remains the disadvantage that only those disturbances are notably reduced which appear in the maximum of the voltage curve in the same direction of the voltage. Furthermore, distortions appear when the volume suddenly increases from low values to very high values.

The present invention avoids these disadvantages to a wide degree. It relates to an arrangement for eliminating or reducing disturbing impulse voltages at telephony reception such as are caused by atmospheric disturbances and the like, and in which the frequency ranges containing the principal frequencies of the disturbance noises, are filtered out by means of filters, then applied to the amplitude limiting elements, and thereafter again combined with the other frequency ranges. In particular in the audio frequency portion of a receiver the range of the higher notes will be filtered out from the audio frequency spectrum by means of electric filters, and passed across an amplitude limiting element, whereafter it is again added to the sound spectrum.

In the drawings:

Fig. 1 schematically illustrates the invention,

Figs. 2a and 2b graphically show the operation of the invention,

Figs. 4 to 8 inclusive show diiferent embodiments which may be employed in the invention.

Fig. 1 shows the principal circuit. The audio frequency alternating potential is divided into two portions by the filter F, among which the one may contain for instance the frequencies below 4500 cycles, while the other one consists of the higher audio frequencies. The latter are passed across the amplitude limiting element A by which all voltage peaks exceeding a certain value are damped, and then the two audio frequency spectra are again joined with each other in the element V either electrically or acoustically. For the favorable action of the arrangement according to the invention the fact is essential that in the natural sound spectrum the amplitudes of the notes above 4000-5000 cycles appearing in general only as weak upper notes are very much smaller than the amplitudes of the intermediate notes and in particular of the very low notes.

The working performance of the arrangement is seen from Figs. 2a and 273'. Fig. 2a shows two sinusoidal voltages superimposed whereby the one may correspond for instance to a note of 300 cycles, the other one to a note of 6000 cycles, and the figure also shows a series of aperiodic voltage pulses, such as may correspond to atmosphericdisturbanees and the like. A division of the frequency spectrum into a higher and a lower portion denotes in this case a separation of the non-disturbed 300 cycle frequency from the 6000 cycle frequency having the disturbances superimposed, such as shown in Fig. 2b. It is seen that the ratio between disturbance voltage amplitude and the amplitude of the useful voltage has become very much higher, and through a voltage limitation as indicated by the dash line w, it is. possible to weaken the disturbance voltages to a considerable degree, although the amplitude of the natural audio frequency may even increase to twice its value. The arrangement has the special advantage that for the case in which also the natural voltages of the high frequencies are to be limited and hence distorted at the greatest volume, the appearing upper oscillations lie in the frequency range above 9000 cycles, and, therefore, outside the range reproduced by the loudspeaker.

In Fig. 3, two transformers T1 and T2 are connected to the output of the audio frequency end tube E, whereby T1 serves for the transmission of the intermediate and low notes, while T2 serves for the transmission of the very high notes. The primary coil of T2 is connected to the anode of the tube E across a comparatively small condenser C1 thus obtaining a damped circuit tuned, for instance, to the frequency 6500 cycles, and representing a low impedance for the high audible frequencies, while the latter are prevented by the choke D to enter into the upper transmission branch. The voltage is transformed by T2 to a higher value so as to bring it into the vicinity of the ignition voltage of the glow tube G which responds at high voltage peaks, and prevents an increase of the voltage beyond a certain limit. The voltage is again transformed downwards by the transformer T3 such that the proper matching to the loudspeaker L is insured, the latter being connected in series to the two secondary windings of T1 and T3. The parallel condenser C2 shall be so dimensioned that it completely suppresses the high frequencies still transmitted across T1, while C3 is to reduce normally only the frequencies above 8500 cycles if these are still present, and eventually it may also be a variable condenser and thus act as tone control means. Furthermore, a choke (not shown in the figure) may also be connected in parallel to the secondary winding of T3, and which represents for the intermediate and low frequencies a short circuit, but for the higher frequencies a high resistance, and which prevents that the intermediate and low frequencies may reach the glow tube across the transformer T3.

A possibility of properly setting the amplitude limiting means resides in providing taps on the secondary winding of T2 and in a corresponding manner on the primary winding of T3 and which are connected to a double sliding contact S by means of which the glow tube can be connected at the same time to different tap points of the two windings corresponding with each other. In this manner the voltage at the glow tube can be increased or decreased while the entire transformation ratio of T2 and of T3 remains constant. In the position shown the transformation ratio upon the glow tube is a maximum thus obtaining the lowest voltage limit. As T1 and T2 transformers having a relatively low self induction may be used, whose stray factor should however be as low as possible so as to insure an unobjectionable transmission of the highest audio frequencies.

Figure 4 shows a similar arrangement in which as amplitude control means two rectifiers G1 and G2, for instance diodes or also suitable dry contact rectifiers are used in parallel, but connected in the opposite sense, and negatively biased. The value of the bias potentials obtained from the two batteries B1 and B2 determines the limit voltage and can be set by means of the switch S. In the uppermost position the limiting element is completely disconnected. The circuit differs from that in Fig. 3 in that a condenser C is also placed in series to the secondary winding of the transformer T2 whereby intermediate and low frequencies which may eventually still be transmitted, can be further reduced, and, furthermore, this circuit differs from that in Fig. 3 in that for each of the two frequency ranges special loudspeakers L1 and L2 are provided. This signifies a more favorable decoupling, and as is known has also an advantage from a purely acoustic point of view.

A still more favorable decoupling of the two parallel branches serving for the two frequency ranges, and hence a still more exact separation of the two ranges can be achieved if, as shown in Figs. 5 and 6, the separation is already carried out ahead of the output, for instance in front of the end tube.

In Fig. 5 the transformer T transmits the frequencies below 4-5 kilocycles across the filter chains R, C and R C to the grids of two pushpull end tubes operating upon the loudspeaker L1 across an output transformer. The filter circuit C1D1 receiving the necessary damping from the resistor R1 takes care of the high frequencies, which it passes across the decoupling resistor R2 and the second filter circuit C2, D2 to the amplitude limiting element and to the grid of a further tube which in turn acts upon the loudspeaker L2.

The circuit according to Fig. 6 differs from that in Fig. 5 in that in the branch for the high frequencies, transformers T5 and T6 are provided at the input and output side, and that the amplitude limiting element lies in the anode circuit. The filter circuit D3, C3 connected to the output in both circuits is intended to withhold from the loudspeaker L2 the frequencies above 8500 cycles still present or again produced. The condenser C3 may also be a variable condenser and serves for the tone control. The switch S provided in all circuits herein shown may be operable from the outside, But it may also be coupled with the loudspeaker knob such that at greater volume the limit voltage will be higher. The bias potential batteries B1 and B2 shown may also be replaced by ohmic resistors at which a direct cur-- rent derived for instance from the battery eliminator, produces voltage drops.

It is, also, possible to provide an automatic control of the bias potential so that thereby the limit voltage will be automatically controlled as a function of the audio frequency volume. Still more suited is an automatic control of the bias potential as a function of the high frequency carrier amplitude. An arrangement embodying this feature is shown in Fig. '7. In the line for the high audio frequency transmitted by the transformer T2, the two rectifiers G1 and G2 are inserted across the condensers C1 and C2. These condensers have a bias potential since they are in parallel to a respective part of the resistors R through which direct current passes which is proportional to the high frequency carrier amplitude, and which is produced by detection of the high frequency through the detector G3 and by smoothing (also of the superimposed. audio frequency oscillation) through the filter chain consisting of condenser C and chokes D. The bias potential of the rectifiers G1 and G2 and hence the limit voltage thus obtained are therefore proportional to the high frequency carrier amplitude. In addition, the simultaneous displacing of the two contacts of the double switch S as an additional manual control is possible.

Fig. 8 shows an arrangement permitting of an amplitude limitation of the high frequencies filtered from the entire audio frequency spectrum by simple means. The circuit differs from an ordinary resistance coupling only in that a condenser C1 is connected in parallel to the resistor R, and a choke D is connected in series to said resistor, while in parallel to said choke D a series combination consisting of a condenser 02 and amplitude limiting element A is connected. The values of the circuit elements are to be so apportioned that for the high frequencies the condensers C1 and 02 represent practically short circuit, while D represents a high impedance, whereas for the middle and lower frequencies the reverse is true. The voltages of the middle and low frequencies therefore are at the resistor R, and that of the high frequency at the choke D and amplitude limiting element A. The combination of the frequencies is obtained in that the two voltages are in series between the grid and the cathode of the following tube.

What is claimed is:

1. In a system provided with a plurality of transmission channels, the method of transmitting electrical waves comprising a wide band of audio frequencies which consists in transmitting in separate channels the low frequency waves and the higher frequency waves, limiting the amplitude of only the waves of the high frequency channel to a predetermined magnitude, and utilizing the entire band of transmitted frequencies including the limited high frequency waves and the waves in the unaffected low frequency channel.

2. In a system provided with a plurality of transmission channels, the method of transmitting electrical waves comprising a wide band of audio frequencies which consists in transmitting in separate channels the low frequency waves and the higher frequency waves, limiting the amplitude of only the waves of the high frequency channel to a predetermined magnitude, and combining the limited waves derived from the high frequency channel and the waves derived from the unaffected low frequency channel.

3. In a system provided with a plurality of transmission channels, the method of transmitting electrical waves comprising a wide band of audio frequencies which consists in transmitting in separate channels the low frequency waves and the higher frequency waves, limiting the amplitude of only the waves of the high frequency channel to a predetermined magnitude, separately amplifying the limited waves in the high frequency channel and the waves in the unaffected low frequency channel, and combining the separately amplified waves.

4. A method of increasing the signal to noise ratio, of audio Waves having noise components in the high frequency end of the audio band, which includes dividing the waves into a group of higher audio frequency waves and a group of lower frequency waves, limiting to a predetermined value the amplitude of the waves of the higher frequency group which exceed such predetermined value, and combining the limited waves with the waves of the lower frequency group.

5. A method of increasing the signal to noise ratio of audio waves having noise components in the high frequency end of the audio band, which includes dividing the waves into a group of higher audio frequency waves and a group of lower frequency waves, limiting to a predetermined value the amplitude of the waves of the higher frequency group which exceed such predetermined value, combining the limited waves with the waves of the lower frequency group, and controlling the amplitude limiting in a predetermined manner.

6. A method of increasing the signal to noise ratio of audio waves having noise components in the high frequency end of the audio band, which includes dividing the waves into a group of higher audio frequency waves and a group of lower frequency waves, limiting to a predetermined value the amplitude of the waves of the higher frequency group which exceed such predetermined value, combining the limited waves with the waves of the lower frequency group, and reproducing the combined waves.

7. In a wave transmission system, at least two wave transmission channels of different frequency-response characteristics, means impress-' ing waves composed of a wide band of frequencies upon said channels whereby said waves are transmitted in twofrequency groups, means in only one of said channels for effecting transmission of waves whose amplitudes are below a predetermined value and for effecting the suppression of the peaked portions of waves which exceed said predetermined value, and means for utilizing the wave output of both channels.

8. In a wave transmission system, at least two wave transmission channels of different frequency-response characteristics, means impressing waves composed of a wide band of frequencies upon said channels whereby said waves are transmitted in two frequency groups, means in only one of said channels for effecting transmission of waves whose amplitudes are below a predetermined value and for effecting the suppression of the peaked portions of waves which exceed said predetermined value, means for utilizing the wave output of both channels, and additional means for automatically regulating the action of said limiting means.

9. In an audio wave transmission system, a source of audio waves, at least two audio transmission channels coupled to said source, one channel being constructed to transmit low and middle audio frequencies irrespective of wave amplitude, the second channel being constructed to transmit the high audio frequencies, means in said second channel for limiting to a predetermined value the amplitude of the high audio frequency waves which exceed such predetermined value, said limiting means including a pair of rectifiers in polarity opposed relation.

10. In an audio wave transmission system, a source of audio waves, at least two audio transmission channels coupled tosaid source, one channel being constructed to transmit low and middle audio frequencies irrespective of wave amplitude, the second channel being constructed to transmit the high audio frequencies, means in said second channel for limiting to a predetermined value the amplitude of the high audio frequency waves which exceed such predetermined value, said limiting means including a pair of rectifiers in polarity opposed relation, and means for applyinga variable bias to said rectifiers.

OTTO T'UXEN.

Images