CH656270A5 - CIRCUIT ARRANGEMENT FOR MODIFYING THE DYNAMIC RANGE. - Google Patents

Description

The invention relates to a circuit arrangement across the entire dynamic range;

according to the preamble of claim 1. (b) circuit arrangements with a characteristic that

Compressors and complementary expanders are made up of two or more parts, only one of which is lienary ("used frequently (a compander system) to create a linear system").

Achieve noise reduction; the signal will be before 60

Transmission or recording compressed and according to the A circuit arrangement with bilinear characteristics

Reception or playback from the transmission channel has special advantages and is widespread. The threshold is expanding. However, compressors can also be used alone above the input noise level or the noise level.

used to reduce the dynamic range, e.g. The transmission channel must be set in order to match the likelihood of control of the circuit by adapting it to the capacity of a transmission channel without subsequently excluding an expansion of pre-noise. Due to the high level part is taken along when the compressed signal for the substantially constant gain is a non-linear

desired purpose is appropriate. In addition, linear treatment of high-level signals are avoided

656270

otherwise would cause distortion. In the case of an audio signal for which the circuit must be syllabic, the high level portion also provides a range within which to handle the overshoots that occur with a syllabic circuit when the signal level increases abruptly. The overshoots are suppressed by clipper diodes or similar devices. This combination of advantages can only be achieved with bilinear characteristics.

Known single-stage, bilinear-level circuits, such as those used in home audio products today, allow 10 dB of compression and expansion, which is appropriate for many purposes. However, some noise remains for some listeners, and more compression and expansion, for example 20 dB, is desirable in order to achieve the best possible fidelity. A new circuit is already being used in home audio products, which is described in Belgian patents 889 428,889 427,889 426, in "Audio", May 1981, pp. 20-26 and in a publication J-6 and a preprint. presented at the Audio Engineering Society Congress in New York in November 1981.

Before the circuits mentioned above, circuits were known and commercially available which enabled compression or expansion of 20 dB and even more; but these were mostly logarithmic circuit arrangements with constant inclination, in which the gain changes continuously in the entire dynamic range or almost in the entire dynamic range. Such circuits suffer more from distortion and signal tracking problems at very low and very high signal levels than the bilinear circuits, in which the change in gain is limited to an intermediate range of characteristics, and overshoot problems occur more than with arrangements with bilinear characteristics. In known companders with constant inclination, the compression ratios are in the range from 1.5: 1.2: 1 and 3: 1, with 2: 1 being the most common ratio.

With regard to the staggering of bilinear circuits in accordance with Belgian patent specification 889 428, a first circuit which has a bilinear input-output characteristic is followed by one or more further circuits which likewise have bilinear characteristics at any frequency within a circuit with a common frequency range. The thresholds and dynamic ranges of the circuits are set to different values to stagger the intermediate or mid-level portions of the characteristics of the circuits so that gain change is achieved over a wider range of mid-input levels than for each of the circuits alone, and between the gains at low and high input level a greater difference is obtained. However, the maximum compression or expansion ratio is essentially not greater than the maximum compression ratio of an individual circuit alone, due to the staggering.

If the circuits in audio circuits have elements for suppressing (limiting) overshoots, their thresholds can likewise be staggered together with the staggering of the syllabic thresholds. The overshoots of the circuits or stages with the lower level are reduced accordingly, with a minimal overall overshoot of the various stages. This is in contrast to known logarithmic compressors, which naturally generate strong overshoots.

Each of the circuits can introduce a change in the spectral content of the signal, e.g. an elevation at low level in the case of a compressor. Each subsequent stage can be actuated by a signal with a progressively changing spectral content. In the case of complex signals, this has the consequence that the chances of error in the decoding function are spread spectrally. In the case of a tape recorder with uneven frequency response characteristics, e.g. io the tendency to spectral shift the total errors of dynamics and frequency response in the decoded result.

The possibility of staggering bilinear levels gives the designer an additional means to optimize an overall circuit. The shapes of the compression characteristics of individual levels can be interpreted in particular with regard to the staggering. The transition characteristics of the circuits are also taken into account, and preference is given to the possibility of staggering the overshoot suppression thresholds in audio-20 compressors and expanders to achieve a minimal overall overshoot.

A circuit commonly known as a "sliding band circuit" that can be used as the first and second circuits produces the specific, desirable characteristic for the 2s case of high frequency audio compression or expansion by applying high frequency boost in the case of compression or cut in the case of the Expansion using a high-pass filter with a variable lower cut-off frequency. To the extent that the signal level rises in the high-frequency band, the cut-off frequency of the filter slides upwards, narrowing the raised or lowered band and excluding the useful signal from the increase or decrease. Examples of such circuits can be found in US Pat. Re 28,426, US Pat. No. 3,757,254,4,072,914,

35 3 934 190 and Japanese Patent Application 55 529/71.

Correspondingly, both the first and the second circuit can be such a “sliding band” circuit. In principle, the quiescent cutoff frequencies of the two sliding band circuits can be different, and this can be used to obtain a degree of compression or expansion that is greater in one part of the frequency band treated than in another. However, in a modified embodiment, the cutoff frequencies are chosen to be essentially identical. This has the advantage of greater discrimination between the frequency range in which an increase or decrease is applied and the range in which this does not occur, and accordingly sharper discrimination between the range in which noise reduction no longer takes place, because a significant useful signal occurs and the area in which the noise reduction remains effective.

On the other hand, circuits are also generally known in which the frequency spectrum is divided into a plurality of bands by corresponding bandpass filters, the compression or expansion in each band in the case of a compressor by means of a gain control device (either an automatically responsive, diode-like limiter or a controlled limiter) and 60 any reciprocal or complementary circuits are provided in the case of an expander. Examples of this type of circuitry can be found in U.S. Patent No. 3,846,719. These split-band or multi-band circuits have the advantage that they operate independently in the different frequency bands, and if this property is desired, such circuits may be the first, second or further stage can be used in the row arrangements.

It is known bilinear compressors and expanders,

656270

to build up both the “sliding band” and the split band type using only one signal path. Overall, however, a construction of such devices is preferred which has a main signal circuit which is linear in terms of dynamics, with a combination circuit in the main circuit, and a further circuit which derives its input or output from the further circuit and whose output is coupled to the combination circuit . The further circuit includes a (self-acting or controlled) limiter, and in the case of compression, the limited signal of the further circuit amplifies or increases the signal of the main circuit in the combination circuit, while the signal of the main circuit is counteracted in the case of expansion. The limited signal of the further path is smaller than the signal of the main path in the upper part of the input dynamic range. It is most advantageous and expedient if the main circuit and the further circuit are separately identifiable signal paths.

Known compressors and expanders of this type are particularly advantageous because they make it possible to achieve the desired type of transmission characteristic in an exact manner without the difficulties of high level distortion. The low level portion with substantially constant gain is obtained by giving the path a threshold above the noise level; below this threshold the further path is linear. The intermediate level part is created by the area in which the limiting effect of the further path becomes partially effective, and the high level part with essentially constant amplification results after the limiter has become fully effective, so that the signal of the further path no longer increases and in Compared to the signal of the main path becomes negligible. In the highest part of the input dynamic range, the output of the circuit arrangement is actually only the signal that has been passed through the linear main path, i.e. linear in terms of dynamic range. With dual-path audio circuits, it is particularly convenient to provide overshoot suppression.

Examples of these known circuits can be found in US Pat. No. 3,846,719.3,903,485 and US Pat. Re 28,426. Analog circuits are also known with which similar results can be achieved, but in which the further route has characteristics which increase Limiter characteristics are inverse, and in which the output of the further path counteracts the signal of the main path for compression and amplifies or increases the signal of the main path for expansion. (U.S. Patent Nos. 3,828,280 and 3,875,537).

Each of these known bilinear circuits can consequently be used as the first and second circuits of the series circuit arrangements according to the invention.

As already mentioned, it is not absolutely important to generate the desired type of bilinear characteristic using such “dual patch” techniques. There are also alternatives that work with individual paths, such as in U.S. Patent 3,757,254.3,967,219.4,072,914.3,909,733 and Japanese Patent Application 55,529 / 71. With these alternative circuits it is usually not possible to achieve as good results as with "dual path" circuits or they can be less practical and therefore less economical; but they can produce equivalent results overall. Consequently, these known circuits can also be used as one or more of the circuits of a series circuit arrangement according to the invention. If necessary, one of the circuits, the first or the second, can be a “dual path” circuit and the other can be a one-way circuit.

In a known arrangement of two bilinear series circuits developed by the applicant and already used in commercial circuits, which is the subject of US Patent Application No. 325,529 dated December 1, 1981, a crosswise coupling is provided in each device between the separate control circuits. A control signal component from the low level circuit is supplied to the control circuit of the high level processor via a coupling network.

The object of the invention is to create additional techniques for coupling bilinear, staggered series connections. These are intended to a) help make the system immune to control by undesired signals, b) help to reduce the effects of noise modulation, c) suppress wild responses, d) allow a reduction in individual circuit requirements, e) compression or increase expansion without side effects, f) reduce the complexity and cost, etc. of the overall circuit.

This is achieved according to the invention by the measure specified in the characterizing part of patent claim 1. Claims 2 and 3 describe preferred embodiments of the invention.

A feature of the staggered bilinear series connections is that they work with different operating thresholds and mostly also with different overshoot thresholds (at least in the case of audio devices). As a result, the circuits respond differently. The cross coupling of alternating current signals between or below the series connections gives the design engineer an additional design parameter that can be useful for optimally designing the operation of the system.

With bilinear dual path circuits, e.g. the signal output of the noise reduction path of a circuit with a higher threshold level is applied to the circuit with a lower threshold level and, with a suitable frequency response and phase modifications, is input into the fixed and variable filter signal circuits, so as to produce active filter effects which promote the sliding band effect and the subsequent noise reduction effect .

In addition, the circuit is less complicated and the cost is lower when using a single control circuit for two or more stages of bilinear circuits in series, the respective inputs and outputs of the single control circuit being cross-coupled between or below the series-connected stages.

The invention is explained in more detail below with further advantageous details on the basis of schematically illustrated exemplary embodiments. In the drawings:

1 is an exemplary set of curves illustrating complementary bilinear compression and expansion characteristics;

Figure 2 is a general block diagram of bilinear devices connected in series;

3 shows a circuit diagram of a known “sliding band” compressor;

4 shows a circuit diagram of a known “sliding band” expander;

5 shows a circuit diagram of a modification of the device according to FIGS. 3 and 4;

FIG. 6 shows a block diagram of a bilinear "dual path" - "sliding band" compressor as described in connection with FIG. 3 or FIG. 3 with the modification according to FIG. 5;

7 and 8 are block diagrams of a known fixed-band compressor and expander;

4th

5

10th

15

20th

25th

30th

35

40

45

50

55

60

65

9 is a general block diagram of the invention;

10 is a block diagram of the invention implemented in a bilinear two-stage compressor and expander;

FIG. 11 shows a more detailed block diagram of the exemplary embodiment according to FIG. 10;

FIG. 12 shows a circuit diagram of an exemplary cross-coupling network for use in the exemplary embodiment according to FIG. 11;

13 is a block diagram of the invention implemented in an arrangement with a common control circuit for bilinear sliding band devices connected in series;

Fig. 14 is a block diagram of the invention which is implemented in an arrangement which provides bilinear, in-line fixed-band devices a common control.

In Fig. 1, exemplary bilinear complementary compression and expansion transfer characteristics are shown at a certain frequency, which for the compression characteristic include the low level part with substantially constant gain, the threshold, the part where the dynamic action takes place, the end point and the high level part reveal essentially constant gain.

Figure 2 shows generally bilinear, series devices; a first bilinear compressor 2 receives input information and its output is connected to a second bilinear compressor 4, which is connected in series with it and whose output is connected to a noisy, information-transmitting channel N. A pair of bilinear expanders 6 and 8 connected in series receive input from channel N on expander 6 and provide a noise reduction system output at the output of expander 8. The ranges of the dynamic effect of the devices lying in series are separated or staggered in relation to one another within the frequency range common to the devices. Although two devices are provided on each side of the information channel N in the figure, not only two but also more devices can be used. The invention provides for the cross coupling of two or more bilinear compressors or expanders connected in series, as will be explained in detail below. When designed as a complementary noise reduction system, the same number of bilinear compressors and expanders are connected in series.

The sequence of stages selected in the compressor, which have a specific characteristic, is reversed in the expander. The last stage of the expander is e.g. Complementary in every respect to the first stage of the compressor - steady state and time-dependent dynamic response (frequency, phase and transition response behavior under all conditions of the signal level and the dynamics).

As already mentioned, the high level is usually the first in a compressor series and the low level is the last. However, a reverse arrangement is also possible. Conversely, the first stage control amplifier needs strong amplification to achieve the necessary low threshold. This low threshold then applies even in the presence of signals with a high level, which usually leads to poor behavior with regard to noise modulation of the overall system in the known sliding band systems. In this reverse arrangement, each stage must have sufficient control amplifier gain to be used for

5 656270

to reach this level required threshold. In addition, each threshold is essentially fixed and independent of the operation of the other stages. This is a consequence of the fact that the signal gain of each previous stage has decreased substantially to one when the threshold for the corresponding subsequent stage is reached.

In contrast to this reverse arrangement, in the preferred arrangement, where the high level is the first in the compressor chain and the low level is the last, there is a useful interaction between the level gains and thresholds. The thresholds of the downstream stages are partly determined by the signal gains of the preceding stages. In a two-stage system with a 10 dB low level gain per 15 stage, the required gain of the second stage control amplifier is reduced by 10 dB due to the low level signal gain of the first stage. When a high level signal occurs, the 10 dB gain of the first stage is switched off and the threshold of the low level 20 stage is effectively increased by 10 dB. In the case of sliding band com- panders, this improves the behavior of the noise reduction effect with regard to noise modulation.

In the preferred arrangement, the gains of all the preceding stages are fully effective up to the threshold of any 25 subsequent stages. In contrast to the system in reverse order described above, the prevailing signal amplifications of the individual stages are thus used in the preferred arrangement in the best possible way. That means:

30th

1. In signal conditions with a very low level (below the threshold), the gain of the control amplifier required in each stage is reduced by an amount which corresponds to the cumulative signal amplifications of all previously

35 walking levels.

2. A variable threshold effect which is dependent on the signal is achieved, as a result of which the noise modulation effects are reduced in the case of sliding band stages. The effective thresholds of the low level levels increase with increasing

40 signal levels progressively increased at a certain frequency. At high signal levels (in the linear high level part of the transmission characteristic) the effective threshold of the level with the lowest level is raised by a value which corresponds to all low level level gains (below -45 half the threshold) up to this point.

A known embodiment of bilinear processors connected in series works with "sliding band" devices connected in series: the compressors 2 and 4 and the expanders 6 and 8 according to FIG. 2 are "sliding band" devices as described in US Pat Re 28 426 are described, with modifications according to the Belgian patent specification 889 428. These modifications include that the syllabic and overshoot thresholds are staggered 55 and the filter cutoff frequency is changed.

Details of the basic circuit are shown in FIGS. 3, 4 and 5, which correspond to FIGS. 4, 5 and 10 of U.S. Patent Re 28,426, respectively, and further details of these circuits, their operation, and the theory on which they are based described there. The following description of FIGS. 3, 4 and 5 is largely taken from US Pat. Re 28,426.

The circuit according to FIG. 3 is especially designed for installation in the receiving channel of a home tape recorder, two such circuits being necessary for a stereo device. The input signal is applied at a connection 10 to an emitter follower stage 12, which supplies a low impedance signal. This signal is first of all

656 270 6

The main path, which consists of a resistor 14, to a signal at the limiter and thus at the resistor 18 can also be

Output connection 16 is applied and secondly short-circuited to earth via a switch 53,

teren way, the last element of which is a resistor 18 when the compressor is to be switched off,

also applied to connection 16. The resistors 14 The amplifier 30 is an NPN transistor with one and 18 add the outputs of the main and the further s emitter time constant network 52, which at high fre-

Way to achieve the required compression law results in a greater gain. Strong high frequency

lichen. sequences, e.g. a pelvis therefore lead to one

The further route consists of a fixed filter 20, a rapid narrowing of the band, in which a compression filter 22 with a variable cut-off frequency, including a sion, takes place in order to avoid signal distortion. FET 24 (these form the filter / limiter), and a ver. The amplifier is connected to the smoothing filter 32 via the stronger 26, the output of which is connected to a double diode rectifier diode 31. The filter has a Rei limiter or clipper 28 and is connected to the resistor 18, a resistor 54 and a shunt capacitor 56. The non-linear limiter suppresses over-. In parallel with the resistor 54, a silicon oscillation of the output signal with abruptly rising diode 58 is provided, which enables the condenser input signals to be charged quickly. The amplifier 26 amplifies the signal in the îs tors 56 for a rapid rise together with a good smoothing-further path to a level such that the characteristic curve enables a steady state. The voltage of the limiter or overshoot suppressor 28, voltage on the capacitor 56 is located directly at the gate of the FET of the silicon diodes, at the corresponding signal 24.

level is effective under transitional conditions. A complete circuit diagram of the complementary expansive threshold of the overshoot suppressor 20 is shown in FIG. 4. A full description is slightly above that of the syllabic filter / limiter. However, this is not necessary because the circuit is identical to resistors 14 and 18, so that the required one is shown in FIG. 3, so that values of the rural degree of compensation of the attenuation for the signal elements are largely shown in FIG. 4 are not shown.

is further available. The following differences exist between FIGS. 3 and 4:

The output of the amplifier 26 is also connected to an amplifier 25. According to FIG. 4, the further path leads its input from the amplifier 30, whose output is derived through a Germani output connection 16a, the amplifier 26a is inverted.

umdiode 31 rectified and rend through a smoothing filter, and those combined by resistors 14 and 18

32 is integrated to the control voltage for the FET 24 signals are at the input (base) of the emitter follower stage 12

receive. whose output (emitter) is connected to connection 16a

Two simple RC filters are used, although 30 are also bound. To ensure a low driving impedance

equivalent LC or LCR filters are used, the input connection 10a is via an emitter follower

could. The fixed filter 20 has a cutoff frequency of 60 connected to the resistor 14. There must be

1700 Hz (now 1500 Hz), below which reduced measures are taken to prevent

pression takes place. The filter 22 has a series condenser that preload gets into the expander.

sator 34 and a shunt resistor 36, which a 35 The amplifier 26a is made inverting, followed by series resistor 38 and the FET 24, whereby that the output from the emitter instead of the collector of the source-drain path of this transistor as a shunt resistor second ( PNP) transistor is taken. For this change status is switched. In the idle state with a zero signal rank, a shift of the 10 k Q. Resistor 62 at the gate of the FET 24 is blocked and offers an im (FIG. 3) from the collector to the emitter (FIG. 3), which makes the essential auto-essential impedance; the presence of a suitable output impedance for driving the resistance of the resistor 38 can then be ignored. The delimiter is preserved. Therefore, the counter cut-off frequency of the filter 22 is missing in FIG. 4, so that 800 Hz (now 750 Hz) was 50.

which can be seen quite considerably below the cut-off frequency of the It should also be noted that it is in the adjustment of fixed filter 20. a complete noise reduction system is important

If the signal at the gate rises to such an extent that the opposite signal level at the emitters of the transistors 12

the FET was less than e.g. 1 k £ 2 drops, bypassed in both the compressor and expander. Of-

the resistor 38 is essentially effective the resistor half, as shown, with these emitters measuring terminals M

36, and the cutoff frequency increases, causing the pass. connected.

band of the filter is clearly narrowed. The increase in Fig. 5 shows a preferred circuit to replace the

Cutoff frequency is of course a progressive process. so circuit between points A, B and C in Fig. 3rd

The use of an FET is advisable because and 4. If the FET 24 is blocked, the second RC network is

such a device within a suitable, installed 22 out of operation, and the first RC network 20

restricted range of signal amplitudes then essentially determines the behavior of the further path. The Ver-

like a linear resistor (for signals of one or better circuit combines the phase advantages, the other polarity in it), the value of which consists of the control voltage, that in the idle state only a single RC section is determined at the gate. is present, while the presence of a signal

The resistor 36 and the FET have an adjustable damping characteristic of an RC filter with two seconds.

ten tap 46 returned in a potential divider, which is given with 12 dB per octave.

contains a germanium diode 48 for temperature compensation. In realizing the circuit using

The tap 46 allows the compression 60 of MPF104 FETs to be adjusted, the 39 k Q resistor 36a is necessary in order to obtain a finite source impedance in which the FET

The amplifier 26 has complementary transistors that can operate. In this way, the compression

which result in a high input impedance and a low endurance at all frequencies and levels at a maximum output impedance. Because the amplifier kept the diode loading of about 2. The 39 k Q resistor 36a has the same limiter 28 drives, a finite output impedance is necessary, 6s function of limiting the compression ratio at which a coupling resistor 50 provides. As already said of this improved circuit like the resistor 36 in the said, the diodes 28 are silicon diode circuits according to FIG. 6 or 7. In addition, this provides a sharp kink in the region of 1/2 V. The resistor has a low frequency path for the signal.

656270

3, 4 and 5 have been developed over the years and there are publications on more modern forms of the circuit which are known in the art. The reference to the special circuit according to US Pat. No. Re 28,426 is given for the sake of expediency in the present illustration.

In the Belgian patent specification 889 428, modifications of the circuit just described are disclosed, which are particularly intended for the purpose of operating two such circuits in series. These modifications include changing the frequencies of the filters 20 and 22, changing the overshoot suppression levels, and changing the syllabic threshold of one of the circuits by modifying the control amplifier 30. This is done by changing the pre-emphasis characteristics provided by the emitter time constant network 52 are controlled. Increasing the capacitor value in the emitter network of the control amplifier 30 increases the gain of the amplifier at a given frequency, thereby responding to lower level signals. As explained above and in US Pat. No. Re 28,426, the cutoff frequency of the variable RC filter 22 increases with increasing control voltage (from amplifier 30, rectifier 31 and smoothing filter 32). At higher capacitance values in network 52 (thereby lowering the control amplifier crossover frequency), the variable filter responds with an increase in its frequency compared to its idle value. The overshoot suppressor threshold is lowered by applying appropriate DC biases (forward) to diodes 28. According to an alternative, the gain of amplifier 26 (FIG. 3) can either be raised to the necessary level or brought to a high level and attenuation applied to match the signal level to the diodes.

7 shows a block diagram of a bilinear “dual path” compressor and expander design with a fixed band.

The basic features of this system are disclosed in U.S. Patent 3,846,719,3,903,485 and in the Journal of the Audio Engineering Society, Volume 15, No. 4, October 1967, pp. 383-388.

In the known exemplary embodiment according to FIG. 7, four bands are created in the further way by networks 250.

Bands 1,3 and 4 have conventional 12 dB / octave input filters: an 80 Hz low pass filter 252 at the input of band 1, a 3 kHz high pass filter 254 at the input of band 3 and a 9 kHz high pass filter 256 at the input of band 4 Each is followed by an emitter follower isolation stage 258. Band 2 has a frequency response that is complementary to bands 1 and 3. Such behavior is derived by adding the outputs of the emitter follower stages 258 in bands 1 and 3 (in an addition stage 260) and subtracting this sum from the total input signal (in a subtraction stage 262). The output of emitter follower stage 258 in each band and the output of subtraction stage 262 is applied to corresponding limiters 264 and 264 '. The delimiters 264 and 264 'are identical except that the delimiters 264' in bands 1 and 2 have time constants which are twice as large as in bands 3 and 4. The outputs of bands 1-4 are combined in a combination stage 266 with the main path signal. The compressor output is applied to a noisy channel to be passed to the complementary expander, in which the output of the identical networks is further subtracted from the input signal to obtain the complementary expansion characteristic.

8 shows further details of limiters 264 and 264 ', each of which includes an FET damper 270, which operates in response to a control signal. The damper output is amplified by a signal amplifier 272, the gain of which is set to achieve the desired low level signal amplification. The outputs of all bands are combined with the main signal so

that a low level output is generated by the compressor, which is up to approx. 5 kHz evenly 10 dB higher than the io input signal, while above this value the level increase increases up to 15 dB at 15 kHz.

The FET attenuator is controlled by a control signal subcircuit that provides a compression threshold of 40 dB below a peak operating level. The control subcircuit includes a control signal amplifier 276 followed by a phase divider 278 which drives a full wave rectifier 280. The resulting DC signal is applied to a smoothing network 282, the output of which represents the control signal. Network 282 includes an RC-20 pre-integration stage, an emitter follower, and a final RC integration stage that cooperate with diodes so that both the pre-integration stage and the final integration stage have non-linear characteristics generated by the diodes. Rapid, large changes in the signal amplitude are passed on quickly, while small changes are passed on slowly. This dynamic smoothing effect produces optimal results in terms of modulation effects, low frequency distortion and distortion components generated by the control signal 30. The circuit enables quick recovery and low signal distortion.

Fig. 9 shows in general the possible cross coupling between two bilinear devices connected in series. If more than two devices are operated in row 35, the number of possible cross-coupling configurations increases. For example, the first device may be cross-coupled to the third device, etc. FIG. 9 shows “n” possible cross-couplings from the compressor 2 to the compressor 4, the respective transfer functions fi (s), f2 (s) to fn (s) to have. There are also shown “n” possible cross-couplings from the compressor 4 to the compressor 2, which have corresponding transfer functions gi (s), g2 (s), to gn (s). In the complementary expander arrangement of the expanders 6 and 8, the cross-45 couplings are reversed, so that the gi (s), g2 (s) and gn (s) cross-couplings from the expander 6 to the expander 8 and the fi (s), g2 ( s) and gn (s) cross couplings from expander 8 to expander 6. Overall, one or more cross-couplings can be provided either forward or backward-50 - f (s) or g (s) - and the cross-coupling (s) can only take place in one direction, e.g. either the f (s) or the g (s) directions are missing, or alternatively in both directions via a single coupling device.

The transfer functions fi (s), f2 (s), gi (s) etc. can be realized by various active or passive devices, which can contain frequency and / or level-dependent elements. The input and output connections of the cross-coupling paths can contain suitable points for deriving from or coupling to one of the following signal paths 60: input signal path, output signal path, signal path “main path” (in a bilinear “dual path” device), signal path “further Path (in a bilinear dual path device) and the AC input signal path to the control circuit (s).

10 shows an example of a signal path cross-coupling arrangement between the further paths of bilinear “dual path” compressors and “expanders. The in line

656270

switched devices are arranged so that the syl labian threshold of the first compressor circuit is at a higher level than the second compressor circuit. The order in the series-connected expander circuits is reversed for complementarity. With block Ni and N2 the circuits of the other ways are designated. In the arrangement according to FIG. 10, the output of the further path of the high level stage 280 is input from Ni through a coupling circuit with a transfer or transfer function f (s) into the "further path" circuit, N2 of the low level stage 282. The transfer function f (s) can have suitable frequency and phase characteristics to promote the limiting effect of the low level compressor stage and consequently the noise reduction effect. For example, when the bilinear devices are sliding band devices, the signal of the high level stage is input to the filter circuit of the low level stage to promote the sliding band effect. In the complementary expander, the output of the circuit Ni in the high level expander stage is input via the same transmission characteristic f (s) into the «further path» circuit N2 of the low level stage.

A specific embodiment of the arrangement shown generally in FIG. 10 is shown in FIGS. 11 and 12. Figure 11 shows the compressor circuits 280 and 282 connected in series, showing the input and output connection points for the cross coupling which contains the transfer function f (s). FIG. 12 shows the details of the network of the transfer function f (s) and its connection to the filter circuit device of the compressor 282. For reasons of expediency, it is assumed that the filter circuit of the compressor 282 corresponds to that described in connection with FIG. 5.

12 includes a high-frequency boost network 284, which brings about a 10 dB boost at high frequencies and whose cutoff frequency corresponds to the quiescent cutoff frequency of the filter of the circuit 280. The output of network 284 is split in two ways and input to fixed filter 20 and variable filter 22 via an adjustable gain device. One end of the 3.3 k £ 2 resistor in the fixed filter 20 is disconnected from the earth and the signal derived via a potentiometer 286 and an amplifier 288 is then applied. The end of the 39 k £ 2 resistor 36a, which was previously connected to the junction between the 0.033 ji.F capacitor and the 3.3 k £ 2 resistor, is disconnected from this junction and is derived via a potentiometer 290 and an amplifier 292 Signal applied to it. Amplifier 288 has a gain of approximately 3/4 and amplifier 292 has gain one.

In operation, the high level circuit 280 would not yet function at low levels. Under this condition, voltage V2 corresponds to voltage Vt because voltage V3 contains a low-level, high-frequency boost (resulting from the gain below the threshold of high-level circuit 280), which is mimicked by boost network 284. The signal levels applied to the fixed filter 20 and the variable filter 22 can be adjusted so that the best results are achieved. When approximately 3/4 of the output of network 284 reaches the 3.3 k £ 2 resistor, its effective resistance becomes approximately 13 k £ 2. When the high level circuit reaches its threshold, both filters 20 and 22 of the low level circuit 282 would already have a band shift effect that overall improves noise modulation behavior without degrading mid band modulation effects,

i.e. Exaggerated band displacement is reduced to a minimum.

FIG. 13 shows a further exemplary embodiment of the invention for cross coupling of “sliding band dual path” devices connected in series. The components already contained in the exemplary embodiment according to FIG. 11 are identified by the same reference symbols. In the embodiment of FIG. 13, a single control circuit (blocks 30, 31, 32) is supplied by adder resistors 304 and 306 from the outputs of the further paths in each of the compressors 280 and 282 connected in series. The output of the control circuit is applied to the variable filter 22 of each of the series compressors via corresponding level adjusters 308 and 310 (if needed). The values of the adder resistors 304 and 306 can be chosen such that the control effect of one compressor is weighted in comparison with that of another. This arrangement enables circuit components to be saved, but has performance. which is largely similar to embodiments each having a control circuit for each of the series devices. A single control circuit can also be provided in the case of three or more compressors or expanders connected in series. In this case, the output of each further path is applied to the control circuit input via a summing device, the output of which is applied to the corresponding changeable filter via appropriate level setting devices (if required).

A single control circuit arrangement is also possible with fixed-band devices connected in series, as the exemplary embodiment according to FIG. 14 shows, in which fixed-band compressors have a single further path. The arrangement is also suitable for fixed-band compressors and expanders connected in series, each of which contains a large number of other paths. In this case, of course, the common control circuit is connected between the other paths that work in the same frequency band.

14, the input signal in the compressor 326 is divided into a main path, which is applied to a combining network 316, and a further path, which contains a fixed filter 312 and a voltage-controlled amplifier VCA 314. The VCA may be an FET attenuator followed by an amplifier as described in connection with FIG. 8 (blocks 270 and 272). In the second compressor 328, the threshold of a VCA 314 'is staggered with respect to the VCA 314 of the first compressor 326, as described in Belgian patent 889,428. A common control circuit (blocks 276, 278, 280 and 282, as described in connection with FIG. 8) is supplied by summing resistors 318 and 320 from the output of the further route in a manner described with reference to the exemplary embodiment according to FIG. 13. The output of the control circuit is applied to the two VCA 314 and 314 'via level adjusters 322 and 324. Just as mentioned in connection with the exemplary embodiment according to FIG. 13, this arrangement can also be used for three or more fixed-band devices connected in series.

The exemplary embodiments according to FIGS. 13 and 14 are both based on the observation that in the case of such compressors and expanders connected in series, the dynamic effect suddenly occurs primarily in one step. Starting from input signals of low level, for example, the stage with the lowest level generates the largest proportion of the combined control signal. To the extent that

8th

s

10th

IS

20th

25th

30th

35

40

45

50

55

60

65

9 65627®

the input signal level rises, the level of the low and the next higher threshold level becomes active and contributes the highest level from the dynamic effect and the level contributes the most to the combined control signal.

B

7 sheets of drawings

Claims (3)

656270 2 PATENT CLAIMS Compressors used only for certain products, in particular 1. Circuit arrangement for modifying the dynamics, particularly in the case of audio products, which transmit only a compressed area of input information signals with a first radio broadcast or a circuit to be recorded in advance with a bilinear characteristic, which are to transmit signals from a signal. Expander low-level parts with essentially constant gain s are also used alone in certain products, especially in the case up to a threshold, an intermediate level part above that of audio products, which only have already compressed round thresholds with variable gain, which receive a maximum of radio transmissions or in provides the previously recorded compression ratio or expansion ratio, are intended to reproduce signals. With certain products, in and a high-level part with a different from the amplification of the special product for sound recording and again low-level part, essentially konxo there is often an individually provided device constant composition, at least one switchable, so that it as a compressor for recording a second circuit following the first circuit, which has signals and as an expander for playback, likewise has a bilinear characteristic within a frequency range common to the radio broadcasts or previously recorded circuits, the signals operating. Intermediate level parts of the characteristics of the circuits is the extent of the compression or expansion can be expressed in dB within a frequency common to the circuits. A compression of 10 dB means e.g. are so staggered that a change in gain compresses an input dynamic range from N dB to an out in a wider range of intermediate input levels than the output range from (N-10) dB. With one for each of the circuits alone and an enlarged differential noise reduction system, one speaks of a compression between the amplifications at low and high 20 sion of 10 dB, which is followed by a complementary expansion of the input level but at maximum compression or 10 dB, of 10 dB noise reduction. Expansion ratio can be achieved, which is essentially circuit arrangements for modifying the dynamics - not greater than that of any of the circuit areas of an input signal, series circuits can be characterized solely because of the staggering, which each have a bilinear characteristic that at least one coupling circuit information - 25 (where "linear" in this context means a constant signal component in one of the circuits with a gain) which is composed of the following, connecting signal path in another of the circuits to: to modify the action of said other circuit depending on the state of the one circuit. 1.) a linear low level part up to a threshold, 2. Circuit arrangement according to claim 1, wherein each of the 30 2.) a non-linear middle or intermediate level part with a coupling circuit (variable gain) connected circuits above the threshold up to a main signal path, which has a linear characteristic in an end point, the one has a predetermined maximum com-reference to the dynamic range, and provides a secondary expression or expansion ratio, and has a signal path which has a non-linear characteristic in 3.) a linear high-level part, the gain of which relates to the dynamic range and whose input is 35 from differs from that of the low level part. the input or output of the main signal path and its output with combining devices in the characteristic is therefore called bilinear, Main signal path is connected, which is the signal of the secondary because two parts with essentially constant gain additive or subtractive with the signal of the main power. Combine the path, characterized by the fact that the coupling 40 In practice, the threshold and the end point of the dual circuit have an input which, with the output of the level part, often does not form a very precisely defined side path or a further path of the one mentioned « Point". The two transition areas, where the intermediate circuit is connected, and an output, which passes to the level part in the linear low level part and in the linear part of a signal in the secondary path or in another high level part, can each have a variable path of the other circuit mentioned connected is. 45 Form from a smooth to a sharply curved curve 3. Circuit arrangement according to claim 1, characterized, depending on the control characteristic of the compressor and characterized in that a common control circuit pre-expander. seen that controls the dynamic effect of the circuits It should also be noted that switching and the said coupling circuit represents and signaling arrangements with bilinear characteristics are derived from components of the circuits and a signal thus two other known classes of circuit arrangement to control the dynamic effect of the circuits. differentiate, namely (a) logarithmic or non-linear circuitry with either a fixed or changing slope ss and without a linear part; the gain changes over the