Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS3393363 A
Publication typeGrant
Publication dateJul 16, 1968
Filing dateOct 7, 1963
Priority dateOct 7, 1963
Publication numberUS 3393363 A, US 3393363A, US-A-3393363, US3393363 A, US3393363A
InventorsForster Jr Harry D
Original AssigneeForster Ind Inc
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Amplifying means employing pulse width modulation
US 3393363 A
Abstract  available in
Images(7)
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

y 16, 5 H. D. FORSTER. JR 3,393,363

AMPLIFYING MEANS EMILOYING PULSE WITH MODULATION Filed Oct. 7, 1963 7 Sheets- Sheet 3 3 J: cr --7.

INVENTOR. #444? 0. Faesrae, JP.

July 16, 1968 H. o. FORSTER. JR 3,393,363

RMPLIFYING MEANS EMI'LOI'IHU PULSE WlTH MODULATION Filed Oct. 7, 1963 7 Sheets-Sheet INVENTOR. #416? Y 0 Fdi fif, Jf.

July 16, 1968 H. D. FORSTER. JR

AMPLIFYING MEANS EMPLOYING PULSE WITH MODULATION 7 Sheets- Sheet 5 Filed Oct. 7.

mm I .w a 5 AW 5 i m f I a Y A56 It July 16, 1968 FORSTER. JR 3,393,363

.LZPLIFYINC- JEANS EZ- FLCTIZQB PULSE WITH MODULATION Filed Oct. L963 7 Sheets- Sheet 6 n t e SCAN I W -wv- W INVENT R I A li/e71 #2693 9, We

/rr:eaame, 5ee42zg $6M July 16, 1968 H. o. FORSTER. JR 3,393,353

IMP LIFYING MEANS EMPLOYING PULSE WITH MODULATION Filed on. 1963 'TSheets-Sheet 7 P T ml l-uh dxrnaawr, 5544, 4:4; 5 few IrWA IYJ' United States Patent 3,393,363 AMPLIFYING MEANS EMPLOYING PULSE WIDTH MODULATION Harry D. Forster, Jr., Roslyn Heights, N.Y., assignor to Forster Industries Incorporated, Roslyn Heights, N.Y., a corporation of New York Filed Oct. 7, 1963, Ser. No. 314,193 15 Claims. (Cl. 325-38) ABSTRACT OF THE DISCLOSURE Solid state circuitry is disclosed for amplifying input signals in a linear fashion. Input signals are mixed or summed with a periodic signal such as, for example, a saw tooth or triangle-shaped wave-form. The resulting signal is applied to a binary circuit providing a first binary output on a predetermined threshold is achieved and a second binary output when the predetermined threshold level is not achieved. The binary output signal is then amplified and filtered to reproduce an amplified replica of the original input signal; a feedback path is provided for compensating for any drift in the circuit. Additional means may be provided to effect D.C. isolation between input and output circuits. This system may be employed as a communication means with the binary output being utilized to modulate the carrier such as a light source; receiver may be provided with light-sensitive means which then drives a transducing means converting the carrier pulses into the common form of intelligence such as, for example, audio intelligence.

The instant invention relates to electronic amplifying means, and more particularly, to amplifying means employing non-linear elements, thereby utilizing a binary concept to provide the desired amplification.

The basic approach to amplification of input signals in order to obtain a desired gain lies in the approach of using amplifier elements, such as vacuum tubes, transistors, and so forth, and operating them in the linear portions of their operation characteristic curves so as to provide an output signal having a waveshape substantially identical to the input signal such that the output signal is amplified in accordance with the gain requirements for the circuit. Such prior art approaches are not completely successful in this regard due to the fact that these devices, if they have any linear operating ranges at all, have extremely narrow linear operating ranges and attendant high power dissipation, so that a significant amount of distortion will nevertheless appear in the output waveshape, as compared with the input signal which it is desired to amplify.

The instant invention employs a concept which is totally different from such prior art approaches in that a deliberate operation of non-linearity is performed upon the input signal which it is desired to amplify.

The instant invention is comprised of electronic summing means upon which are imposed a scan signal and an input signal, thereby providing a resultant waveform having a waveshape which is non-linear relative to the input signal. The resultant waveshape is impressed upon encoder means which means has a binary operating characteristic in that it is either in a first state (i.e., otf) or a second state (i.e., on) at any given time, and is basically incapable of being in any state other than the on or off stages.

Thus, the resulting output is comprised of substantially square pulses whose width or time duration are a function of both the scan signal and the input signal impressed upon the summing circuit. Since the significant information of such pulses reside in their pulse width or, duration, and not in their amplitude, it is then possible to amplify such pulses to any desired gain until the desired output power level is obtained. This is done by impressing the output of the encoder means upon amplification means having a suitable gain characteristic.

The amplified input signal may now be recovered by suitable filter means having a frequency response characteristic capable of passing all frequencies of the input signal, while providing cut-off of the frequencies contained in the scan signal. It has been found that the repetition rate of the scan signal need not be greater than the frequencies contained in the input signal. Tests on the system and calculations made to date indicate that the system is usable over a range considerably greater than that heretofore believed possible by such systems. The reason for this increased bandwidth is the scan signal and general system concept. Conventional limitations were based on sampling theorems using rigid sampling periods. This circuit permits a more random sampling of the input and hence gives better performance. The exact limits are not definitely established and appear to be dependent upon the application of the system.

In numerous applications, it becomes advantageous to obtain an output signal or waveshape which bears a nonlinear relationship to the input signal. This is provided for through the employment of scan signal generating means having the capability of providing a scan signal which is designed to provide an output signal which bears a nonlinear relationship to the input signal as one example, and which may further serve to cause the encoder means to fail to produce predetermined portions of the input signal waveshape for a variety of different purposes. As one typical example, it may be desired to enhance the signalto-noise ratio of an incoming input waveform. This may readily be done by providing a scan signal having a requisite slope so as to provide amplification of all portions of the signal above a particular threshold level, while completely cancelling or attenuating all portions of the input signal below said threshold level.

In still a further example, the inverse relationship may be obtained whereby all portions of the input signal above the threshold level are completely attenuated, while those portions of the input signal below said threshold level are amplified to the desired gain level.

One preferred manner for extracting the amplified input signal is the provision of oscillating means which is controlled by the binary output signals from the encoder means. The output of the oscillation means is then detected by suitable detector means which may then be filtered in order to extract the amplified input signal. The provision of such oscillation means, however, provides suitable isolation between input and output of the amplifying circuitry which is an undesirable characteristic, and is to be avoided whenever possible.

It is therefore one object of the instant invention to provide means for amplifying an input signal which employs novel encoder means to transform the input waveshape into binary information.

Another object of the instant invention is to provide means for amplifying an input signal which is comprised of novel binary encoder means for encoding the input signal which is summed with a suitable scan signal.

Still another object of the instant invention is to provide means for amplifying an input signal which is comprised of novel binary encoder means for encoding the input signal which is summed with a suitable scan signal 3 where the output of the encoder circuit is related to both the scan signal and the input signal.

Still another object of the instant invention is to provide means for amplifying input signals which is comprised of novel encoding means for transforming input waveshapes into binary information, which binary signals may be suitably amplified to any desired power output level.

Another object of the instant invention is to provide means for amplifying input signals which is comprised of novel encoding means for transforming input waveshapes into binary information, which binary signals may be suitably amplified to any desired power output level and wherein the amplified binary signals are decoded by suitable filter means to obtain the amplified input signal.

These and other objects of the instant invention will become apparent when reading the accompanying description and drawings in which:

FIGURE 1 is a block diagram showing an amplifier means designed in accordance with the principles of the instant invention.

FIGURE 2 is a schematic diagram of the nonlinear amplifying means of FIGURE 1.

FIGURE 2a is a plot showing a curve relating the input signal to the output signal of the encoder means of FIGURES 1 and 2.

FIGURE 2b shows a plurality of waveforms employed to facilitate the description of the instant invention.

FIGURE 2c shows an alternative embodiment for the circuit of FIGURES 1 and 2.

FIGURE 3 is a block diagram of an alternative embodiment of the scan generating means shown in FIG- URES 1 and 2.

FIGURE 4 is a plot of a waveform of one typical input signal which may be applied to the amplifying means of FIGURES l or 2.

FIGURE 5 shows an alternative scan signal which may be employed in the amplifying means of FIGURE 1.

FIGURES 6 and 7 show two more alternative scan signals which may be employed in the amplifying means of FIGURE 1 for additional applications thereof.

FIGURE 8 is a schematic diagram showing one preferred output stage which may be employed in the amplifying means of FIGURE 1.

FIGURES 9, 9a and 10-13 show other alternative embodiments which may be employed in the output section of the amplifying means of FIGURE 1.

FIGURES 14, and 15a show alternative embodiments which may be employed in the amplifying section of the non-linear amplifier means of FIGURE 1.

FIGURE 16 is a waveform showing the diode characteristics of the diodes employed in FIGURES 15 and 15a.

FIGURES 17, 18 and 19 are schematic diagrams showing alternative embodiments of circuits for providing isolation between input and output of the non-linear amplifying means of FIGURE 1.

FIGURES 20 and 21 are schematic diagrams showing two alternative arrangements for the feedback loop of the non-linear amplifying means of FIGURE 1.

FIGURE 22 is a block diagram showing one application of the amplifying circuit of the instant invention.

FIGURES 22a shows the manner in which spacemen may converse by utilizing the scheme of FIGURE 22.

Referring now to the drawings, FIGURE 1 is a block diagram showing the amplifier means 10 of the instant invention which is comprised of an input signal source 11 which may be any suitable input signal such as, for example, input signals in the audio range, radio frequency range, or video range, as some typical examples. A summing means 13 is provided for accepting the input signal at its input terminal 14, a scan signal at its input terminal 15 and a feedback signal at its input terminal 16. Summing means 13 then suitably sums all of these signals to provide a single output which is impressed upon the encoder means 17.

The scan signal source 12 may be any suitable scan signal generating means capable of providing the desired binary output information at encoder means 17 in a manner to be more fully described. Typical examples of the scan signal are a sawtooth waveform which is the waveform 2 and a triangle waveform such as the waveform 3.

It is possible to generate a scan signal by causing the device itself to act as an oscillator. An equivalent manner of analysis is to say that the amplifier is a non-linear, astable device with controlled limit cycles. The pulse width is controlled by the control of the limit cycle. An example is shown in FIGURE 2c. Suppose that Q of FIGURE 2c is on, then so is Q The voltage v is therefore going toward ground potential. This reduces the current fed back through r and hence tends to turn Q off thereby yielding the opposite effect. If the input signal is also applied to the base of Q then the points where Q turns on and off are controlled by the input thus developing pulse width information.

The summation of the inputs to summing .circuit 13 (FIGURE 1) are impressed upon a non-linear encoder circuit 17 having a characteristic substantially as that shown in FIGURE 2a such that its output is at a voltage level d when the input signal is below a first threshold level r and further so that the output level is at d when the input signal is greater in magnitude than the threshold level I;,. By using a scan signal such as, for example, the scan signal 2 or 3 of FIGURE 1, the width of the output signal derived from encoder means 17 may be controlled within any desired limits so that the amplitude of the input signals from signal source 11 are converted into pulse width information at the output of encoder means 17 With the information in pulse width as opposed to signal amplitude form, the output signals may then be amplified by the amplifying means 18 which need not be perfectly linear since the only portion which need be preserved in the output signal of encoder means 17 is the width of the pulse and not the height. After amplifying to a suitable power output level, the amplified binary information is then impressed upon an output section 19 which is utilized to filter out from the binary information the desired portion, or portions, of the input signal from signal source 11. It should further be understood that the filter means may not be required in many applications. In order to provide relatively constant gain for the non-linear amplifying means of FIGURE 1, a feedback loop 21 is provided so as to control variations in gain or attenuation which may occur between summing means 13 and output terminal 20. In addition, the gain can be used to control D.C. level and to help in filtering the output, and to do frequency response shaping or any other function generally performed by a feedback circuit.

Theory of operation The objective of the circuit 10 of FIGURE 1 is to transform an input signal 2 into a binary signal and then to ultimately recover the desired signal from the transformed binary information. The functional dependence of the binary signal upon the input is determined by the scan signal developed by scan signal generating means 12.

The device which converts the continuous signal derived from input signal source 11 to the binary signal is the encoder means 17 which has the transfer function, as shown in FIGURE 2a. As the input signal impressed upon the encoder means increases from a minus quantity to zero and then from zero to a plus quantity, the output for encoder means 17 changes from d to d It will be noted that at the threshold level 1 the output of encoder 17 moves along positive slope S until threshold level 1 is attained, at which time the output remains constant at d While it is somewhat advantageous to have the slope of the portion S of the waveform shown in FIGURE 2a as being perfectly vertical, it is not necessary that the encoder have a perfect binary transfer function, and it may be imperfect, as is shown in FIGURE 20.

The effect of the scan signal which is summed with the input signal is to cause the summed signal to scan across the threshold levels of the encoder. One example is shown in FIGURE 2b for a triangular scan signal. A change in the level or amplitude of the input signal causes a corresponding change in the level of the output by changing the level produced by the pulses. The amount by which the output changes as a function of the input varies as a function of the slope of the scan signal. Referring more particularly to FIGURE 2b, the first plot 30a shows a triangular waveform summed with an input signal at a time when the input signal voltage level is zero volts. At this time, the triangle waveform rests directly upon the time axis which is a zero volt reference level. At time t the triangular waveform has a value of zero volts, but as time increases the value of waveform 30a increases in a positive direction. At time 2 the value of the triangular waveform output reaches the threshold level of the encoder means 17 at which time the output of the recoder 17 changes from the voltage level d to the voltage level d as shown by waveform 30b. Between the times 1 and t the value of the triangular waveform is greater than the threshold level 33 so that the output of encoder .17 remains at the voltage level d At time t the voltage level of triangular waveform 30a decreases below the threshold level 33, causing the encoder output 17 to drop to the voltage level d as shown by waveform 30b. Between the times and t the voltage level of the triangular waveform 30a remans below the threshold level 33, causing the output of encoder 17 to remain at voltage level d In a like manner, between times t and t the voltage level of triangular waveform 30a exceeds the threshold level 33, causing the output of encoder 17 to be at the level d during this time period.

Considering the waveforms 31a and 31b, it can be seen that the triangle waveform 31a has been lifted, due to the summation of the triangle waveform with a substantially constant voltage (relative to the scan signal) of a magintude V. In this particular instance, the encoder will generate an output, as shown by waveforms 31b at time 2 wherein time t; occurs almost immediately after time t due to the fact that at t the summation of the triangle waveform 31a with the constant voltage V, exceeds the threshold level 33. Thus, the time t in the waveform 31a occurs substantially before the time t in the waveform 30a. In a like manner, the triangular waveform does not go below the threshold level 33 until time t which can be seen to occur after the time 1 of the waveform 30a. This forms the square pulse having a time duration t t as shown by waveform 31b, which is substantially greater than the time duration of the first square pulse shown in waveform 30b. In a like manner, the second square pulse is formed during the times t; to t.,, and is substantially identical to the first time pulse. Since the square pulses of waveforms 30b and 31b have substantially the same magnitudes, the information which they carry is contained within their pulse widths, or time duration, and it is the time duration which is employed as the means for relating the binary pulses to the amplitude of the input signal source. While the waveforms of FIG- URE 2b show a scan signal of a frequency greater than the frequency or repetition rate of the input signal it should be understood that in many applications the scan signal may be chosen to have a repetition rate which is equal to or less than the repetition rate of the input signal.

As still a further example, the waveform 32a which is another triangle waveform, is shown summed with a negative voltage, so that is forms the two square pulses shown in waveform 32b which are substantially narrower than both the pulses shown by waveform 30b and waveform 31b. This is due to the fact that the summation of the constant voltage with the triangle waveform 32a does not exceed the threshold levels 33 until the time t, which occurs substantially later than the times 1 in the examples of 30a and 31a, and further, the resultant waveform of 32a goes below the threshold level of time t and further the resultant waveform 32a drops below the threshold level 33 at a time 1 which occurs substantially before the times of the waveforms 30a and 31a, causing the two square pulses of waveform 32b to be formed which are substantially narrower in pulse width (i.e., time duration) than the square pulses or binary pulses of wave forms 30b and 31b.

T he scan signal The scan signal is the primary determinant of the system, and controls the application thereof. Its form, therefore, ascertains the use of the system. The uses can generally be classified as:

(l) Linear amplitude operations.

(2) Non-Linear amplitude operations.

(3) Scan fundamental operations.

A linear operation is one in which the output desired is a linear function of the input signal. A non-linear operation is one in which the desired output amplitude is a non linear function of the input signal amplitude.

In some applicatios such as, for example, the driving of an A-C servo motor, the primary objective is to maintain a fundamental frequency component in the output signal and to operate upon this fundamental frequency component so as have complete control of the driven device (i.e., the servo motor). One method of doing this is to utilize the frequency of the scan signal as the desired output frequency fundamental and to operate upon this fundamental frequency so as to control the output thereof. This is defined as a scan fundamental operation.

Each of the following applications utilize a different form of scan signal generator:

Linear amplitude operation A scan signal having a waveform that is linear with time may be used as the scan signal in an application where it is desired to have an output whose amplitude is a linear function of the input amplitude. The two simplest linear forms are the ramp and the triangle. These signals may be generated by any conventional means such as, for example, a sawtooth generator, 21 tri angle generator, etc.

A typical circuit which may be employed for the block diagram of FIGURE 1 is shown in schematic form in FIGURE 2. The scan signal generator which, in this case, is a sawtooth generator, is comprised of a unijunction transistor Q having tow base electrodes b and b connected through respective resistances 22a and 22b to a positive 10 volt source and ground potential respectively. The emitter electrode of the unijunction transistor Q is connected to the common terminal between series connected resistor and capacitor elements 24 and 23 respectively, the opposite terminals of which are connected to the positive 10 volt source and ground potential. The operation is such that when the emitter of unijunction transistor Q; has a very small current, capacitor 23 begins charging thereby forming the positive going ramp of the sawtooth scan signal (see segment 22a of waveform 22 in FIGURE 1). After a predetermined time, capacitor 23 charges so that the common terminal between capaci tor 23 and resistor 24 reaches a voltage level causing the emitter to b resistance to drop rapidly. At this time, transistor Q causes capacitor 23 to discharge through unijunction transistor Q and resistor 22, thereby forming the negative going portion of the sawtooth waveform (see portion 22b of waveform 22 in FIGURE 1). This sawtooth waveform, which is the scan signal, is passed through series connected capacitor 25 and resistor r to the base electrode of transistor Q Also connected in common with the base of transistor Q; is a resistor r which receives the input signal from the input signal source, such as the source 11, as shown in FIGURE 1.

The arrangement of resistors 26 and 27 connected in common with the base electrode of transistor Q provides the summation circuit 13 of FIGURE 1. The resultant waveform is, therefore, used to drive the transistor Q in a bi-stable manner (transistor Q serving as the encoder means 17) so as to form the square pulses 29a at the collector electrode transistor Q These pulses are passed through resistor 28 to the base electrode of transistor Q which suitably amplifies these binary pulses shown at 290 which appear at the output terminal 29b. The purpose of FIGURE 2 is to show the use of the unijunction scan generator in a typical circuit and its intergration with the remainder of the nonlinear amplifying means circuitry.

It should also be noted at this point that it is not necessary to provide a scan generator for each amplifying means. It is possible to use a single scan generator .for driving many amplifier circuits in common. In such situations, the scan signal can be viewed as another input signal to the amplifying means.

To obtain a linear scan, it is also possible to employ a triangle waveform scan signal rather than the ramp or sawtooth (see waveforms 2 and 3 of FIGURE 1). FIG- URE 3 shows a block diagram embodiment 30 for obtaining a triangle waveform scan signal. The arrangement is comprised of limiting means 32 for receiving an A-C signal 31 at its input terminal. and providing the square pulses 33 at its output terminal due to the limiting action. The square pulses 33 are impressed upon an integrating circuit 34 which, due to its inherent characteristics, forms the triangle waveform 35 at its output terminal 36. Thus, output terminal 36 may be connected to the base electrode of transistor Q as a substitute for the sawtooth waveform scan generating means comprising unijunction transistor Q of FIGURE 2. The circuitry of FIGURE 3 is of a highly advantageous arrangment, since the input signal in the limiting means 32 is usually available from a variety of locations such as any suitable nearby power supply source, etc.

N on-Iinear amplitude operation While the following examples are by no means intended to be an exhaustive treatment of the variety of non-linear applications, it is felt that the following examples will'suffice in order to obtain an understanding of non-linear operation for the amplifying means of FIG- URES 1 and 2.

In the section on the theory of operation, the effect of the scan signal slope upon the system gain was considered. The general concept for non-linear operation is to have a scan signal whose slope is non-linear with respect to time. The output is then a function of the amplitude of the input signal to the encoder. Consider the following examples:

Let it be assumed that an input signal shown by the waveform 41 in the plot 40 of FIGURE 4 is comprised of information pulses such as, for example, the pulses 42, and is further comprised of the low level noise components 44, wherein a high content of such low level noise is present.

In order to minimize the effect of the noise level upon the information signal, it is possible to employ a scan signal, such as the scan signal waveform 51, shown in the plot 50 of FIGURE 5. The scan signal waveform 51 is comprised of two portions during each cycle wherein a first portion 51a which occurs between the time Ot and 2 -2 has a substantially steep slope. The portion 51b occurs within the times t t and t t of the two cycles of the scan signal, shown in FIGURE 5. It can be seen that the portion 5112 has a smaller slope which is significantly less than the slope of portion 51a. The breakpoint between the two slopes 51a and 51b occurs at times t and t and is designated by the gain transition level phantom line 53. The input waveform 41 and the scan signal waveform 51 are impressed upon the amplifier of the instant invention in such a manner that the scan signal has a reverse polarity from that shown in the waveform of FIGURE 5 so that it is subtracted from the input waveform 41. Consider first the noise segments, 44, of the input waveform 41. The combination of the input, 41, and the scan waveform, 51, is such that it never causes the gain transition level 53 to pass the threshold of the encoder. The effect is therefore that of a substantially steep slope scan signal 51a. thereby giving little gain to the input. Consider now inputs that are sufiiciently large so that their combination with the scan signal will cause the portion 51b of the scan to be employed. This will occur for all inputs in excess of level 43. When such portions of input 41 as the pulses 42 occur they exceed level 43 and are effectively scanned by segments 51b and therefore have a higher gain. Whereas the scan signal waveform 51 is shown to occur substantially in synchronism with the input waveform 41, such that its steep slope portions 510 are in alignment with the noise level portions 44 and its relatively small slope portions 51b are in alignment with the signal evel portions 42, it should be understood that when operating the amplifier circuit in this manner it is not at all necessary that synchronism occur between the scan signal waveform 51 and the input waveform 41 since the scan signal acts to effectively reduce the noise level present in the input waveform, regardless of both the phase and frequency relationship between the scan signal and the input signal waveforms.

The reverse situation which may occur from that given above is that wherein the desired signal is contained in the low level information while the high level information is that which is undesirable or unwanted. Thus, considcring the plot of FIGURE 4, in this example, the information desired to be derived from the input waveform 40 is the low level signal ponions 44 whereas the high level portions 42 in this particular example are either undesired or unwanted.

In this situation, the scan signal waveform 61 of the plot 60, shown in FIGURE 6 is employed. The scan signal Waveform 61 is comprised of two waveform portions 61a and 61b wherein the portion 61:: which occurs between time O-t t -I and t -t and has a substantially steep slope. The portion 61b occurs between the time t -t r -r and has a substantially small slope. In this instant the scan and information signal waveform are impressed upon the amplifier circuit in an additive fashion. Considering the low lever portion 44 of FIGURE 4 and the scan signal portion 61a of FIGURE 6, it can be seen that the resultant waveform will be substantially greater in magnitude than the low level portion 44 and will lie well above the cut-off level 63 of the amplifier. Considering the high level portion 42 of FIGURE 4 and the scan signal portion 61b of FIGURE 6, it can be seen that the resultant signal occurring due to a summation of these two portions will be increased only slightly above the original high level signal portion 42. Thus the resultant waveform permits the user a vastly improved opportunity for observing and examining the low level portion 44 of the input signal waveform 41. Again, it should be noted, as was stated previously, with respect to the scan signal 51 of FIGURE 5, that phase and frequency synchronism is neither required nor desired as between the scan signal waveform 61 and the input waveform 41 of FIGURES 6 and 4, respectively.

In the selection of the slope of the scan signal portion 61:: and 61b of scan signal 61, the objective is to increase the gain of the amplifier to as high a level as possible without saturating the amplifier, the effect of which would be to cancel or diminish the function for which the scan signal 61 is employed.

In addition to the amplifying function recited above where it may be desired to either increase or diminish the effect of low level noise, the encoding capabilities of the amplifier which is comprised of a waveform having square pulses of equal amplitude and varying pulse 9 widths. Such output waveforms would be similar to those shown in the plots 30b, 31b and 32b of FIGURE 2b described previously.

The scan signals 51 and 61, shown in FIGURES and 6, may be generated by any suitable scan signal gen erator, such as, for example, the scan signal generator identified by the name waveform synthesizer and manufactured by the Exact Electronics Corporation of Hillsborough, Oreg. It should be understood that any other suitable scan signal generator or waveform synthesizer may be employed since the specific generator employed lends no novelty to the circuitry of the instant invention.

A still different concept of non-linear scan signals can be obtained when considering the time perspective. Suppose, for example, that periodic data of the form shown in FIGURE 4 is being imposed upon the non-linear amplifying circuit of FIGURE 1, and that only a certain time portion of the incoming waveform is to be amplified. A scan signal of the form 71, shown in the plot 70 of FIGURE 7 may be employed, which scan signal is synchronized with the clock frequency rate of the input signals. If the lower levels 74 of the scan signal are sufficient to keep the encoder means off, then the only portions of the input signal amplified will be those portions where the normal scan signal appears. In this man ner, portions of any predetermined length may be deleted (i.e., attenuated) from the incoming data signals. Those portions which are to be amplified, however, may be either non-linear, linear, or a combination thereof.

Output stages The output stage 19 of FIGURE 1 is a design which depends upon the intended use of the amplifying means 10. The most fundamental amplifying means is that shown in FIGURE 8, which amplifying means 80 is comprised of a transistor Q receiving the output from the encoder means at its base electrode 83, having its collector and emitter electrodes connected through respective resistors R and R to the voltage source terminals 81 and 82, respectively. The output is taken at point 84 from the collector electrode. The embodiment 80 of FIGURE 8 assumes that either the load filters out any unwanted signal introduced by the scan signal, or that the spurious signals are used by the load as in the scan fundamental operating mode.

An alternative embodiment 90 is shown in FIGURE 9, and is employed primarily as means for compensating for a D-C level which may be introduced into the load. With the arrangement 90 of FIGURE 9 the transistor Q has its collector and emitter electrodes connected to the resistors R and R respectively. The terminals 92 and 91 of the voltage source are connected to the resistors R and R respectively, with suitable diode means 93 being connected between the terminal 92 and the resistor R The second voltage source is connected at terminals 96 and 97 through a diode 94 and a resistor 95. This power source is connected and polarized in such a manner so as to counteract a D-C level established by the fundamental circuit. The output as seen by the load can now go positive or negative.

A modified embodiment 90a of the circuit 90, shown in FIGURE 9 is shown in FIGURE 9a and is comprised of transistors Q and Q wherein transistor Q is connected to the first voltage source via terminals 92a and 91a, and the second voltage source is connected at terminal 97 which is in electrical series with the emitter electrode of transistor Q the collector electrode of which is connected to the terminal 92a through resistor Y. The base electrode of transistor Q is connected to the collector electrode of transistor Q and operates to control the power supply connected to terminals 97 and 98 in such a way that when transistor Q is conducting, transistor Q, is off, and vice versa. This arrangement also enables the output as seen by the load to go either positive or negative. Assuming that R is driven by two sources v and v the current through R is i 4 by superposition the sum of the currents due to v and v the objective of FIGURES 9 and 9a is to show that this current can be controlled by diodes D and D In FIGURE 9 when, D is closed; with D open the current is i When Q conducts, the current is i i If i i then the effect of being conductive Q is to reverse the direction of i In FIGURE 9a, both Q and Q are used. They are connected so that when Q conducts Q is in cut-off and i =i and when Q conducts Q; is cut-off and i =i thereby effecting a change in direction of i;,.

Bridge circuits can also be utilized for the bi-directional outputs. In some circuits, it may be desirable, however, to remove the high frequency components before they are admitted to the load. In such situations, a low pass filter may be added so that only the desired frequencies go to the load means. FIGURE 10 shoWS an embodiment 100 which may be employed with the bidirectional cir cuit of FIGURE 9 and which is comprised of a transistor Q which contains, in addition to the resistor R in its load circuit, a low pass filter means comprised of the resistor and capacitor elements 103 and 104, respectively, which elements cooperate to form a low pass filter means to filter out all unwanted high frequencies. It should be understood that other situations may exist where filters of different pass bands may be desired other than the low pass filter of FIGURE 10, and the output circuit or stage 100 of FIGURE 10 may employ filters of other pass band characteristics other than the low pass band filter shown therein.

Power amplifier stage The amplifying stage or section 19 of FIGURE 1 may utilize the circuit arrangement 110, as shown in FIGURE 11 of the drawings. In this arrangement, the binary pulses from the encoder means 17 are impressed upon the base electrode 111 of a transistor Q The level of the pulse is amplified and applied to the base of transistor Q through the resistance means 114 which, in turn, provides a second stage of amplification amplifying the already amplified signal output from transistor Q and providing this second stage of amplification at its output terminal 117. It has been found to be advantageous, but not absolutely necessary, to utilize the bi-level aspect of the transformed signal to drive the individual stages into or out of saturation, or nearly so. Operating in this manner causes less power to be dissipated in the active transistor devices Q and Q It also eliminates D-C biasing problems common to conventional amplifiers. The combination of this type of operation and the scan signal concept permits the use of a D-C amplifier with directly coupled stages, thereby eliminating the problem of D-C shift. The stability of the amplifier, therefore, is determined primarily by the scan signal and, hence, is an A-C and not a D-C problem.

The amplifier stages do not have to appear as shown in FIGURE 11. An alternative arrangement such as the schematic embodiment shown in FIGURE 12 may be employed. This arrangement employs NPN and PNP transistors Q and Q -Q respectively, such that the collector electrodes of the PNP transistors Q and Q, are connected to the negative voltage bus 129b, while the collector electrode of the NPN transistor Q is connected to the positive voltage bus 1290. This complementary symmetry arrangement has been found to be highly advantageous.

The circuitry 130 of FIGURE 13 teaches the employment of unijunction transistors, such as the transistors Q and Q to provide a unijunction operation. The output of the encoder means 17 is impressed upon the input terminal 131 which is connected through the voltage divider circuit com-prised of resistors 133 and 132 to the control electrode of unijunction transistor Q. As the input voltage is raised with the control electrode of unijunction transistor Q it is triggered and fired, thereby raising the voltage at the base electrode b of transistor Q due to the increased voltage drop across resistor 134. This causes the voltage level at the control electrode of unijunction transistor Q to be raised sufliciently enabling transistor Q to fire, thereby increasing the voltage drop across the load resistor R By dropping the input voltage at input terminal 131, this causes the reverse action to take place. The same general concept of operation can be obtained with any device capable of bi-level operations. Other such typical devices are relays, silicon controlled rectifiers, trigistors, and diodes, to name just a few.

Variations in the system become obvious when it is observed that the significant information of the transformed signal is in the time of transition. Hence only this transition time, or its equivalent, need be transmitted and amplified.

FIGURE 14 shows a schematic diagram 140 which exemplifies this concept. The bi-level signal which is obtained as the output signal from the encoder means 17 of FIGURE 1 is impressed upon the base electrode 147 of transistor Q The signal which is suitably amplified and shown at 143 then appears at the output or collector electrode of transistor Q The signal is then impressed upon a differentiation circuit comprised of the capacitor and resistor elements C and R to produce the output pulses 144 and 145. Amplification or transmission of these pulses is sufiicient for maintenance of all the desired information of the signal. It is not necessary to produce the pulses in the specific manner shown in FIG- URE 14, nor need they be in the opposite directions, as shown therein. Only their position in time relation need be maintained.

A direct continuation of the concept employed in FIG- URE 14 is shown in the embodiment 150 of FIGURE 15 which employs diode amplifying means. In the circuit 150, the tunnel diodes D, and D (or their equivalent) are used for amplification purposes.

FIGURE 16 shows a plot 160 which presents the curve 161 representing the tunnel diode current-voltage characteristic. Let it be assumed that the diodes D and D are in their lower voltage state v A positive pulse 151 is then applied through resistor R to the anode electrode of diode D This drives the diode D through the negative resistance region to its second stable operating point at voltage level v The voltage across diode D thereby appears as the waveform 153 at the anode electrode of diode D due to the bi-stable characteristic of diode D This output voltage is suitably differentiated by the differentiation network comprised of capacitor and resistor C and R respectively so as to form the positive pulse 154 which it should be noted has a greater magnitude than the pulse 151. The pulse 154 is then employed to switch the following stage containing tunnel diode D to generate the output 156. The negative pulses 152 and 155 are employed to reverse the procedure so as to form the trailing edges of the square pulses 153 and 156 respectively. Power gain is obtained by the increased voltage and current levels in the successive diode stages.

FIGURE 15a shows an alternative embodiment 150a which may be substituted for the embodiment 150 of FIGURE 15. In the circuit 150a, a transformer means T is provided which receives an input waveform 153 which is differentiated to form pulses 154 and 155 by the differentiation circuit C -R thereby impressing these pulses upon the primary winding of transformer T. The secondary winding or transformer winding T is connected across the anode and cathode electrodes of the tunnel diode D which generates the output 156 is substantially the same manner as that described with respect to the diodes D and D of FIGURE 15. The transformer T thereby serves to provide a voltage gain in the embodiment of FIGURE 15a, as well as providing D-C isolation between the input and output stages of FIGURE 15a. An obvious extension of the embodiment 150a of FIGURE 15a would thereby be the use magnetic amplifiers having suitable characteristics to provide substantial gain in addition to providing the isolation function of the device of FIGURE 15a.

As another operating concept, let it be assumed that a bi-level signal is applied to a device such as a voltage tuned klystron whereby F.M. modulation results. Suppose, however, that the range of oscillation is between both levels. The resulting output is then a pulse of varying frequency marking the crossover. This is then all the information necessary for complete reconstruction of the signal. Such pulses are of a chirp nature which, according to the articles entitled:

(1) Klystron & Microwave Triodes; Rad Lab Series #7, McGraw-Hill, 1948. Chapter 16, page 441, shows the basic concept of the encoding process. FIGURE 16.1 in particular gives the output power and frequency spectrum for two types of input signals of interest.

(2) Radar Engineering; Fink, McGraw-Hill, 1947, page 424, FIGURES 305 and 306, shows a more specific form for accomplishing the above object. Any state of the art circuit is sufiicient.

The coding can also be in the form of modeing. That is, at both levels of the input signal, the klystron oscillates, but at two difierent modes.

Input-output isolation In the circuits discussed so far, it can be seen that a common continuous path occurs between input and output. Such a situation is not permissible in most circuit arrangements due to D-C level problems imposed, as well as other problems. Returning to a consideration of FIGURE 15a, this embodiment provides an indication as to one method of isolation between input and output. Another method which may be employed is that of triggering an oscillator on and off, in accordance with the binary level of the output derived from the encoder means 17 of FIGURE 1. Three such embodiments are set forth in FIGURES 17 through 19 respectively. Considering first FIGURE 17, the output of encoder means 17 of FIGURE 1 forms pulses such as, for example, the pulse 171 impressed upon the base electrode of transistor Q Transistor Q is arranged to operate as an amplifying means which impresses its output upon the control or trigger electrode of a unijunction transistor Q The operation of FIGURE 17 is essentially that of the unijunction relaxation oscillator. When Q is on, Q oscillates as normal. When Q turns off the oscillation of Q closes. The transformer T in the b lead of Q is used for coupling of the AC. components of the oscillation. FIGURES 17, 18 and 19 show that control of the oscillator can be had by placing Q in any of the three legs of Q. A detector circuit comprised of diode D and parallel connected capacitor and resistor C -R forms a square wave output pulse 173 which is basically the envelope of the waveshape 172. Employing this embodiment not only preserves the identity of the square wave 171 which can be seen is substantially identical to the output waveshape 173, but further provides isolation between input and output.

The embodiment of FIGURE 18 differs from the arrangement 170 of FIGURE 17 in that the control transistor Q is connected in series relationship with the unijunction transistor Q in such a manner that when the square pulse 171 is present, this completes the series circuit between positive and negative voltage busses and 186 respectively, enabling oscillation to occur in substantially the same manner as that described with reference to FIGURE 17 to provide an output 173 substantially identical to the output 173 contained in FIGURE 17.

The embodiment 190 of FIGURE 19 differs from the arrangement 180 of FIGURE 18 in that the transistor Q has its emitter and collector electrodes connected so that the transistor Q is in parallel with a resistor element R When the square pulse 171 is impressed upon the base electrode of transistor Q this effectively short-circuits resistor R to a degree sufficient to permit oscillations to occur. Any one of the embodiments 170, 180 or 190 may be employed as the isolating means for the non-linear amplifying circuitry of FIGURE 1.

Feedback Feedback around the entire circuit may be accomplished as in any other circuit arrangement. There are, however, some unique feedback characteristics which are inherent in the system of the instant invention. FIG- URE 2O demonstrates one of these for a simplified version of the non-linear amplifying means. FIGURE 20 shows an embodiment 200 having a summing circuit which is comprised of resistors R R R and R The encoder is comprised of transistor Q and the amplifier stage is comprised of transistor Q with the load appearing as resistor R The summed, or resultant, voltage that is fed into the base electrode of the encoder transistor Q, is developed across resistor R and is due to the currents present in resistors R R and R Suppose, for example, that the supply voltage V diminishes somewhat. It would be expected that the voltage drop across R would also decrease. The drop in supply voltage V however, reduces the current in resistor R This lowers the net voltage at the base of transistor Q Hence transistor Q will be cut off for a longer portion of the scan period. This thereby causes transistor Q to be on for a longer portion of the scan period, causing an increase in the D-C level of the output. This makes apparent the ability of resistor R to control the D-C level of the output signal.

An alternative form of overall feedback is shown in the schematic circuit 210 of FIGURE 21, which arrangement will compensate for any type of drift. To outline this operation, suppose, for example, that the output across the load resistor R becomes too positive. This means that the voltage across capacitor C will also go positive. The current fed back through the resistor R; is thereby reduced. This reduces the voltage drop across resistor R thereby causing transistor Q, to be off for a longer period of the scan cycle, and likewise causing transistor Q to be off for a longer portion of the cycle, thereby causing the level across the load resistor R to go more negative. It should be noted that the polarity of the feedback depends on the phase of the stage fed back relative to the encoder means. For example, the encoder means being transistor Q, is amplified by an amplifier stage comprised of first transistor Q and then a second amplifier stage comprised of transistor Q The first amplifier stage has as an inherent characteristic the phase inversion of the signal impressed upon its base electrode which inversion is the 180 phase inversion. By placing this completely inverted signal through a second stage of amplification, a second phase inversion of 180 occurs, thereby shifting the ultimate output signal after its original phase condition. By adding a third stage, or by removing one of the two amplifying stages, this arrangement may be employed to change the polarity of the feedback.

One scheme of application for the instant invention is of the arrangement 250 shown in FIGURE 22. The arrangement 250 is comprised of a first location 260 which utilizes the amplifier circuit of the instant invention which is substantially similar to the schematic arrangement 10 of FIGURE 1. The block 251 represents a suitable microphone or speaker means capable of converting voice into electrical signals. The electrical signals are impressed upon the amplifier 10 and are employed as the input waveforms. As previously described, the amplifier 10 generates square pulses of substantially equal magnitude, but of varying pulse widths. These signals are impressed upon a light source 252 and act to modulate light signals so that the light source 252 generates light signals of equal intensity but of varying time duration. These light signals represented by the line 253 are directed to a remote location 260 which is provided with a suitable light or photo-sensitive pickup means 254. The photo-sensitive pickup means 254 is capable of converting the light pulses or flashes which impinge on the photosensitive means and converting them to electrical signal pulses of substantially identical wavesbape to the output of amplifier 10 at location 260. These square pulses are impressed upon suitable speaker means, such as, for example, ear phones which are capable of converting the square pulses into sound. The speaker means is identified by the block designated 255. There is no necessity to decode the equal magnitude square pulses of varying time duration since the ear phones coupled to the light detection means 254 load the light detection means 254 sufiiciently to act as a decoding means, thus avoiding the need for any filter means to decode the incoming signals.

Remote location 260 may be provided with a transmission circuit comprised of a microphone or speaker 251, amplifier circuit 10' and a light generating means 252' substantially identical to the elements 251, 10 and 252 respectively, of location 260. Likewise, location 260 may be provided with suitable light sensitive pickup means 254 and ear phone means 255' substantially identical to the elements 254 and 255 of location 260'.

The arrangement of FIGURE 22 may be advantageously employed in the field of space exploration. Let it be assumed that spacemen are deployed to regions of the universe beyond the atmospheric layers of the earth or other planets and well into space. It will be under such conditions substantially impossible for two or more such spacemen to communicate to one another due to the fact that the space is substantially devoid of air or atmosphere so that there is no suitable means which exists for carrying voice signals from one location to another. In order to enable such voice communication to be carried out, each spaceman, such as, for example, the spaceman 270 and 271 of FIGURE 22a, is provided with a facility 260 or 260' substantially identical to those shown in FIGURE 22. For voice communication to take place, spaceman 270 merely speaks into the microphone 251, which signals are coded by the amplifier 10, the output of which is used to modulate the light source 252. The modulated light 253 is then received by the photo-sensitive pickup 254 provided in the facility 260' of spaceman 271. The modulated light is converted into electrical signals which are in turn converted by the ear phones 255 into suitable sound signals. Spaceman 271 may converse with spaceman 270 in substantially the same manner by by ear phones 255. This arrangement thus makes it quite.

simple for individuals to converse with one another in environments where the amount of air present to propogate voice signals is extremely small. It should be understood that the above arrangement is merely exemplary and that the invention is not limited to voice amplications.

It can therefore be seen that the instant invention provides an amplifying means for suitably amplifying input signals through the employment of a non-linear or binary coding concept, which, even through introducing nonlinetan'ty to the input signals by transducing them into binary signals, nevertheless, is capable of producing an output signal which is substantially an exact replica of the input signal amplified to a suitable level.

Although this invention has been described with re spect to preferred embodiments thereof, it should be understood that many variations and modifications will now be obvious to those skilled in the art, and it is preferred, therefore, that the scope of this invention be limited not by the specific disclosure herein, but only by the appended claims.

The embodiments of the invention in which an exclusive privilege or property is claimed are defined as follows:

1. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means having three input terminals for summing the instantaneous output signals of said scan signal generating means applied at a first input terminal and said input signal source applied at a second input terminal; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said first amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; feedback means connected between said filter means output and the third input terminal of said summing means.

2. Communication means for transmitting information between two or more remote locations wherein each location is provided with first means for generating electrical signals representative of the information to be transmitted, said first means comprising means for converting audio information into electrical signals;

amplifying means of the type described in claim 1 for encoding said electrical signal;

transmitting means operated by said amplifying means for generating light pulses;

receiver means at each location comprising light-sensitive means for converting light pulses depending upon said light sensitive means into electrical signals; and means for converting said electrical pulses into audio information.

3. The amplifying means of claim 1 further comprising first and second bias means; first and second diode means coupling said first and second bias means respectively, to said first amplifying means being connected with their polarities arranged to permit generation of either positive or negative polarity output signals.

4. The amplifying means of claim 1 further comprising first and second bias means coupled to said first amplifying means;

transistor means coupled between said first amplifying means and said first bias means;

impedance means coupled between said second bias means and said first amplifying means;

second impedance means coupled between said first amplifying means and said transistor means selectively operating said transistor means to permit generation of output signals of either positive or negative polarity.

5. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said first amplifying means; said scan signal generating means including means for generating an output waveform adapted to prevent the magnitude of said summing means resultant output signal from achieving said threshold level for selecting only portions of said input signal to be amplified.

6. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein the said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said first amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; said first amplifying means comprising a first differentiation circuit for converting said encoding means output signal into impulses spaced at predetermined intervals and means for converting said impulses into an amplified replica of the output signal of said encoding means.

7. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshhold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; said first amplifying means comprising a transistor amplifier stage for amplifying the output of said encoding means; a differentiation circuit connected to said transistor amplifier stage for converting the amplified binary output signals into impulses spaced at predetermined time intervals, the length of said time intervals being substantially equal ot the width of said binary signals.

8. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a secondary binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; said first amplifying means comprising a first differentiation circuit connected to the output of said encoder means for converting the binary output signal into impulses spaced at predetermined time intervals; 21 first amplifier stage comprised of tunnel diode means for converting said impulses into binary signals; the binary output signals of said first amplifier stage being amplified replicas of said encoding means binary output signals.

9. Amplifiying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means connected to the output of said amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; said first amplifying means comprising a first differentiation circuit connected to the output of said encoder means for cnoverting the binary output signals into impulses spaced at predetermined time intervals; a first amplifier stage comprised of tunnel diode means for converting said impulses into binary signals; the binary output signals of said first amplifier stage being amplified replicas of said encoding means binary output signals; a second differentiation circuit connected to the output of said first amplifier stage for converting said binary output signals into impulses spaced at predetermined intervals; a second amplifier stage comprising tunnel diode means for converting said impulses into binary signals, the binary output signals of said second amplifier stage being amplified replicas of the binary output signals of said first amplifier stage.

10. Amplifying means comprising an imput signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; filter means for averaging the output of said amplifying means, the output of said filter means being an amplified replica of the output of said input signal source; input-output isolation means connected between said first amplifying means and said filter means comprising a control stage having its input connected to the output of said first amplifying means; oscillator means; said control stage being connected to said oscillator means for enabling operation thereof upon receipt of a first binary signal from said first amplifying means; detector means for converting said oscillator output signals into binary signals; transformer means inductively coupling said oscillator means to said detector means; the output of said detector means generating binary signals which are amplified replicas of the binary output signals of said first amplifying means; said transformer means being adapted to isolate the amplifier means output from its input.

11. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; said scan signal generating means including means for generating a non-linear output Waveform having a portion of each cycle of said output waveform adapted to prevent the magnitude of said summing means resultant output signal from achieving said threshold level for selecting only portions of said input signal to be amplified.

12. Amplifying means comprising an input signal source; scan signal generating means for generating a scan signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoding means; said first amplifying means comprising a transistor amplifier stage for amplifying the output of said encoding means; a differentiation circuit connected to said transistor amplifier stage for converting the amplified binary output signals into impulses spaced at predetermined time intervals, the length of said time intervals being substantially equal to the width of said binary signals; tunnel diode means for amplifying and converting said impulses into binary signals.

13. Amplifying means for reducing noise level in an input signal comprising first means for receiving said input signal; scan signal means for generating a repetitive waveform, said scan signal waveform having a substantially steep slope during a first portion of each cycle of said repetitive waveform; and having a slope substantially less than the slope of said first portion during the remaining portion of each cycle of said repetitive waveform; and means for subtracting said scan signal from said input signal to generate an output signal in which the noise level content is substantially reduced.

14. Means for converting signal amplitude information into pulse width information comprising amplifier means; summing means having first and second inputs, said first input receiving output signals containing said signal amplitude information the output of said summing means being amplified by said amplifier means; feedback means connecting the output of said amplifier means to the second input of said summing means to reduce the effect of the input signal upon said amplifier means by providing negative feedback; said amplifier means comprising a transistor connected in common emitter fashion and having series connected resistor and capacitor means in series with the collector of said transistor; said feedback means being connected between said resistor and capacitor and said summing means second input.

15. Amplifying means comprising an input signal source; scan signal generating means for generating a periodic signal; summing means for summing the instantaneous output signals of said scan signal generating means and said input signal source; encoding means for encoding the resultant instantaneous output signal of said summing means into binary signals wherein said encoding means generates a first binary condition when said summing means instantaneous output signal exceeds a predetermined threshold level and a second binary condition when said summing means instantaneous output signal does not exceed said predetermined threshold level; first means for amplifying the output of said encoder means; filter means; input-output isolation means connected between said first amplifying means and said filter means comprising a control stage having its input connected to the output of said first amplifying means; oscillator means; said control means being connected to said oscillator means for enabling operation thereof upon receipt of a binary signal from said first amplifying means; detector means for converting said oscillator output signals into binary signals; means capacirtively coupling said oscillator means to said detector means; the output of said detector means generating binary signals which are amplified replicas of the binary output signals of said first amplifying means; said coupling means being adapted to isolate the amplifier means output from its input.

(References on following page) References Cited UNITED STATES PATENTS Fredendall 325-38 X Posthumus 332-15 Thompson 332-9 Farkas 332-15 Sharma 332-15 X Thompson 332-15 X Johannessen 332-15 X Gregory 332-15 X Staufenberg et a1. 250-199 Chasek 307-885 ROBERT L. GRIFFIN, Primary Examiner.

JCHN W. CALDWELL, Examiner.

J. T. STRATMAN, Assisrant Examiner.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US2391776 *May 29, 1943Dec 25, 1945Rca CorpIntelligence transmission system
US2471168 *Nov 22, 1946May 24, 1949Hartford Nat Bank & Trust CoDevice for converting a signal of variable amplitude into pulses of constant frequency and variable duration
US2990516 *May 29, 1956Jun 27, 1961John C Simons JrPulse-width modulated amplifier and method
US3090929 *Dec 18, 1959May 21, 1963Westinghouse Electric CorpController circuitry with pulse width modulator
US3149293 *Oct 19, 1960Sep 15, 1964Farkas Thomas PElectronic timer or pulse time modulator
US3167720 *Feb 10, 1961Jan 26, 1965Transis Tronics IncPower amplification means
US3170125 *Dec 18, 1959Feb 16, 1965Westinghouse Electric CorpController circuitry
US3187273 *Dec 30, 1960Jun 1, 1965Bell Telephone Labor IncZero axis crossing pulse modulator
US3260912 *Jun 19, 1963Jul 12, 1966Gen Motors CorpPower amplifier employing pulse duration modulation
US3290503 *May 6, 1963Dec 6, 1966Raytheon CoLight communication system
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US3493685 *Sep 28, 1966Feb 3, 1970Philips CorpPower supply regulator and amplifier circuits
US5070292 *Nov 13, 1989Dec 3, 1991Performance Controls, Inc.Pulse-width modulated circuit for driving a load
US5081409 *Aug 7, 1990Jan 14, 1992Performance Controls, Inc.Pulse-width modulated circuit for driving a load
US6353354Sep 26, 2000Mar 5, 2002Mts Systems CorporationPulse-width modulated bridge circuit within a second bridge circuit
US7183823Mar 15, 2005Feb 27, 2007Texas Instruments IncorporatedMethod of generating a pulsed output signal from a periodic ramp signal and a reference voltage, and a switch mode power converter
US7379856 *Aug 6, 2003May 27, 2008Verigy (Singapore) Pte. Ltd.Modeling an electronic device
DE102004016073A1 *Mar 30, 2004Oct 27, 2005Texas Instruments Deutschland GmbhVerfahren zur Erzeugung eines Impulsausgangssignals aus einem periodischen Sägezahnsignal und einer Referenzspannung, und getakteter Stromwandler
DE102004016073B4 *Mar 30, 2004Dec 23, 2010Texas Instruments Deutschland GmbhVerfahren zur Erzeugung eines Impulsausgangssignals aus einem periodischen Sägezahnsignal und einer Referenzspannung, und getakteter Stromwandler
WO1987003154A1 *Apr 24, 1986May 21, 1987Shlomo GonenOptical underwater communicator
Classifications
U.S. Classification398/191, 332/110, 327/172, 398/192, 398/195
International ClassificationH03F3/20, H03F3/217
Cooperative ClassificationH03F3/217
European ClassificationH03F3/217