Title:
Electrical servo system
United States Patent 2437313


Abstract:
13 Claims This Invention relates to electrical servo sys ters, and more particularly to improvements ii the art of stabilizing electrical servo systems o the type employing A.-C. control signals. An electrical servo system, as the term is usel herein, is defined as comprising an output shaft...



Inventors:
Bedford, Alda V.
Application Number:
US57062444A
Publication Date:
03/09/1948
Filing Date:
12/30/1944
Assignee:
RCA CORP
Primary Class:
Other Classes:
318/675, 327/290, 333/19, 333/166, 342/79
International Classes:
G05D3/14
View Patent Images:
US Patent References:
2298177Electric filter circuit1942-10-06
2233415Position control system1941-03-04
2186268Recording oscillographic apparatus1940-01-09
1554698Surge preventer1925-09-22



Description:

13 Claims This Invention relates to electrical servo sys ters, and more particularly to improvements ii the art of stabilizing electrical servo systems o the type employing A.-C. control signals.

An electrical servo system, as the term is usel herein, is defined as comprising an output shaft an electric motor coupled to said shaft, mean providing a "displacement signal" related in som characteristic, such as its amplitude, to the dif ference between the actual position of the outpu shaft and the position to which it is to be driven and means for energizing the motor in respons to the displacement signal, in such manner tha the motor tends to drive the shaft to reduce thi effect of the displacement signal, and hence th( motor energization, to zero. It is well known t those skilled in the art that such systems tend t( be Inaccurate and sluggish if the displacemeni signal produces too little effect on the motor, anc tend to overrun the correct position and "hunt' If the displacement signal Is made sufficiently ef. fective to overcome the effects of friction. It Is common practice to combat these difficulties by adding to the displacement signal further signals which are in effect time derivatives of the disPlacement signal. If the displacement signal ls a D.-C. voltage of variable magnitude, the proper auxiliary sigrials are also D.-C. voltages, having magnitudes which vary as the first and highei order time derivatives of the displacement signal. If the displacement signal Is an A.-C, voltage of variable, amplitude, the auxiliary signals must also be A.-C. voltages, of the same frequency as the displacement signal, varying in amplitude in accordance with the time derivatives of the amplitude of the displacement signal. It is important to note that the auxiliary voltage waves are not time derivatives of the displacement voltage wave.

The principal object of the present invention is to provide improved servo systems of the type wherein a variable amplitude A.-C. voltage Is employed as the displacement signal, including means for producing auxiliary "derivatives" signals.

Another object is to provide improved methods of and means for deriving from a variable amplitude A.-C. voltage, further A.-C. voltages varying in their amplitudes In accordance with the time derivatives of the amplitude of said first voltage.

A further object is to provide an improved servo system adapted for automatically positioning directive radio antennas in response to radio signals picked up thereby.

. (CL 318-28) 2 These and other objects will become apparent n to those skilled in the art upon consideration of f the following description, with reference to the accompanying drawings, of which d 5 Figure 1 is a schematic diagram of a radio loS cator system embodying the instant invention, s Figure 2 is a graph illustrating a variation of e the position of the output shaft of the system of - Figure 1 with respect to the correct position of t 10 said shaft, S Figure 3 is a graph illustrating variations with e time of the amplitude of a radio signal received t in the operation of the system of Figure 1, e Figure 4 is a graph illustrating the output of e 15 the radio receiver of the system of Figure 1 under o the conditions represented by Figures 1-3, S Figure 5 is a graph illustrating the voltage of t Figure 4 after being delayed.

S Figure 6 is a graph of the difference between 20 the voltages of Figures 4 and 5, Figure 7 is a schematic diagram of a modificaS tion of one of the subcombinations of the system of Figure 1, S Figure 8 is a further modification of one of the 25 subcombinations of Fig. 1, Figure 9 is a graph illustrating the voltage of Figure 4 after being delayed by a different amount than that corresponding to Figure 5, and Figure 10 is a graph illustrating the sum of the 30 voltages of Figures 4 and 9.

Referring to Figure 1, only those elements of a radio locator system which are necessary to an explanation of the present invention are shown.

A radio transmitter I is connected through a two35 way switch device 3 to a pair of radiators 5 and 7.

The radiators 5 and 7 are directive, and are positioned upon a supporting member 9 in such manner that their directive patterns overlap with the maximum directivity of the radiator 5 lying in a 40 line to the right of the equisignal axis, and that of the radiator 7 lying to the left. The supporting member 9 is rotatable by means of a shaft I I, which is coupled to a motor 13.

The switch 3 is coupled to a synchronous motor 45 15, arranged to be energized from an A.-C. source, not shown. The motor 15 may be arranged to drive the switch 3 at a constant speed of 3600 R. P. M., for example, connecting the transmitter I alternately to the radiators 5 and 7 at a rate 50 of: 60 cycles per second.

The motor 13 is illustrated as a commutator type A.-C. motor, although it is to be understood *that any known type of reversible A.-C. motor may be used. The field winding of the motor 13 55 is connected through a phase shifter 17 to the A.-C. supply. The armature of the motor 13 is connected to the output circuit of an amplifier 19.

A radio receiver 21, provided with an antenna 23, is tuned to the frequency of operation of the transmitter I. The output circuit of the receiver 21 Is coupled to the control grid of an electron discharge tube 25. The tube 25 is provided with two load resistors 27 and 29, connected in the anode and cathode circuits respectively. A grid leak 31 is also provided, with its lower end connected to a D.-C. source of biasing potential. The tube 25, with its associated resistors, constitutes a well known type of phase invertor. It will be apparent that any other known type of phase invertor may be substituted. The phase invertor is designated generally by the reference numeral 33 in Figure 1.

The anode of the tube 25 is coupled through a blocking capacitor 35 directly to the control grid of a tube 37, and through a resistor 39 to the control grid of a tube 41. The cathode of the tube is coupled through a blocking capacitor 43, a delay network 45, and a resistor 47 to the control grid of the tube 41. The output end of the delay network 45 is also coupled through an amplifier 42 to the input circuit of a phase invertor 33', similar to the phase invertor 33.

The phase invertor 33' is coupled to the control grid of a tube 49 through resistors and a delay network, in exactly the same manner as the phase invertor 33 is coupled to the tube 41. The elements in the connections from the phase invertor 33' which are similar to those associated with the phase invertor 33 are denoted by corresponding reference numerals, primed. The tubes 37, 41 and 49 are provided with a common load resistor 51, which is coupled through a& capacitor 53 to the input circuit of the amplifier 19.

The time delay networks 45 and 45' illustrated in Figure 1 are of the same general construction as low pass filters, and in fact are low pass filter circuits. They are designed to pass at least the fundamental frequency of the output signal of the receiver 21, and are terminated in such manner and provided with the proper number of sections to introduce a delay of substantially one cycle, i. e. if the signal is 60 cycles per second, the networks 45 and 45' each cause a delay of Yoo second. If desired, the networks can be designed to pass higher frequencies as well, subject only to the condition that the required Vo0 second delay is provided. The networks are illustrated as comprising series inductors and shunt capacitors It will -be understood by those skilled in the art that series resistors may be used instead of series inductors, and other types of networks may b( substituted for those shown in Figure 1.

The operation of the system of Figure 1 is aw follows: The transmitter provides radio frequency output which is applied alternately to thi radiators 5 and 7 through the switch 3. Althougl the transmitter I may be modulated, it is as. sumed for the sake of simplicity of descriptioi that it merely provides a continuous wave. Th4 operation of the servo system is substantially thi same whether or not the transmitter is modu lated. Signal is radiated alternately by the radi ators 5 and 7. If a reflecting target lies on a lin midway between the directive axes of the radi ators, the strength of the reflected signal is th same regardless of which radiator is energized However, if the target is to the left of the equi signal line, the reflected signal is stronger whil the radiator 7 is energized, and weaker when th radiator 5 is energized.

Figure 2 shows a typical variation of the deviation of the equisignal line from the line of sight to the target, such as would be caused by motion of the target toward the left. Referring to Figure 3, the amplitude of the reflected signal varies accordingly, the pulses L2, Ia etc. representing energy transmitted from the antenna 7 becoming larger, and the right pulses R2, R3 etc. becoming smaller. It should be understood that each of the pulses L and R of Figure 3 represents only the amplitude of the reflected wave. The pulse frequency is 60 cycles per second, i. e. 60 L pulses and 60 R pulses occur each second, but the frequency of the signal itself may be several hundred megacycles per second.

The reflected signals are picked up by the antenna 23, and amplified and detected by the receiver 21, providing an output voltage represented by the graph of Figure 4. This is a 60 cycle wave, increasing in amplitude with increase in the deviation or displacement of the line of sight from the equisignal axis. If the deviation were to the right, rather than the left, a wave similar to that of Figure 4 would be produced, but of opposite phase.

The receiver output is applied to the control grid of the tube 25, causing corresponding variations of the anode current thereof, and hence similar variations in the voltage drops across the resistors 27 and 29. Upon increase of the anode current of the tube 25, the voltage at the anode becomes less positive, referred to ground potential, and that at the cathode becomes more positive. The blocking capacitors 35 and 43 pass only the A.-C. components of these voltages; thus the voltage at the control grid of the tube 37 is similar in form to that-applied to the input of the delay network 45, but 1800 out of phase with it. The voltage input to the delay network is in phase with the receiver output. The output of the delay network 45 is similar to the input, but delayed one cycle. This voltage is represented by theigraph of Figure 5.

The current through the resistor 39 is proportional to the A.-C. component of the anode voltage of the tube 25. The current through the resistor 47 is proportional to the output voltage of the network 45. Both of these currents flow through the resistor 40. The voltage drop across the resistor 40 is thus substantially proportional to the sum of the A.-C. anode voltage of the tube 25 and the output voltage of the delay network.

The voltage across the resistor 40 is represented t by the graph of Figure 6. Since the voltage apr55 plied through the resistor 39 is identical with that e represented by Figure 4, but reversed in phase, the wave of Figure 6 is actually the difference between s those of Figure 4 and Figure 5. Therefore, the magnitude of each pulse of the wave of Figure e 60 6 is proportional to the difference between sucI cessive pulses of the wave of Figure 4. The dif- ference between successive pulses is proportional n to the time rate of change of amplitude. Thus, e the amplitude of the wave of Figure 6 is propore 65 tional to the rate of change of amplitude of the - wave of Figure 4. When the wave of Figure 4 is - increasing in amplitude, the derivative wave of e Figure 6 is in phase with it. When the wave of Figure 4 is decreasing in amplitude, that of Fige 70 ure 6 is out of phase-with it.

1. The phase invertor 33' and the delay network 45' operate upon the differential signal appeare ing across the resistor 40 to provide at the control e grid of the tube 49 a wave corresponding in amplitude to the rate of change of amplitude of the differential signal. This voltage, appearing across the resistor 40', is proportional to the second derivative of the displacement. It will be apparent that further derivative signals may be produced by adding further networks similar to those illustrated.

The displacement voltage is amplified by the tube 37. The first and second derivative voltages are amplified by the tubes 41 and 49 respectively.

The three voltages are combined in the common load resistor 51, which is made of a resistance considerably less than the impedances of the tubes 37, 41 and 49 so that the anode currents of the three tubes will be independent of one another. The relative magnitudes of the displacement component and the derivative components may be adjusted by varying the values of the resistors 39, 39', 40 and 40'.

The composite voltage across the resistor 51 is amplified by the amplifier 19 and applied to the motor 13. The motor 13 is energized thereby to rotate the shaft 1 and direct the antennas 5 and 1 toward the target. Initially, while the displacement is increasing, the derivative component aids the displacement signal, providing increased motor torque to overcome friction. The second derivative component also aids the displacement signal while the rate of change of displacement is increasing, to overcome inertia during acceleration of the motor 13, and bucks the displacement signal while the rate of change of displacement is decreasing. As the displacement signal decreases, the first derivative component bucks it, tending to deenergize the motor more rapidly so that the system will coast to a stop without overshooting. The second derivative signal also bucks the displacement signal while the rate of change of displacement is decreasing, tending to overcome the momentum of the moving parts. Thus, by properly proportioning the resistors to control the amplification of the derivative signals in accordance with the friction and mass of the motor 13 and its mechanical load, the system may be made to operate smoothly and accurately, without lag or hunting.

In the operation of the system of Figure 1, the voltage appearing at the amplifier 19 comprises three components: an undelayed displacement Ssignal, a signal similar to the displacement signal but delayed by one cycle, and a signal similar to the displacement signal lut delayed by two cycles. Considering the operation of the system from this viewpoint, rather than that of successive derivatives, it becomes apparent that some of the elements of the system of Figure 1 may be omitted without altering the mode of operation of the overall system. Referring to Figure 7, a single phase invertor tube 33" is connected like the phase invertor 33 of Figure 1 to anode and cathode load resistors 27" and 29" respectively.

The cathode load resistor 29" is coupled through a blocking capacitor 33" to a delay network 45" which, like the delay networks of Figure 1, is designed to provide a delay of one cycle. The output of the network 45" is connected to a second identical delay network 45"'. The anode of the tube 33" is coupled through a blocking ca-, pacitor 35" and a resistor 70 to the input circuit of the power amplifier 19. The output of the network 45" is similarly coupled through a resistor 71 to the amplifier 19 and the output of the network ?5"' is likewise coupled to the amplifier 19 through a resistor 72.

IiY/the operation of the circuit of Figure 7, the displacement signal is transmitted without delay, through the capacitor 35" and the resistor 70 to the amplifier 19. It is transmitted in reverse phase through the network 45", which introduces a delay of one cycle, and then through the resistor 71 to the amplifier 19. The third component, delayed by two cycles, travels through both of the networks 45" and 45"' and the resistor 72 to the amplifier 19. The relative values of the resistors 70, 71 and 72 may be adjusted to provide the required proportionality between the three components. Thus the composite voltage applied to the amplifier 19 is identical with that applied to the amplifier 19 in the system of Figure 1, although the derivative voltages are not produced separately at any point in the circuit.

Refer to Figure 8. The circuit including the phase invertor 33 and delay network 45 of the system of Figure 1 may be replaced by a delay network 61, bridged by a resistor 63. The network 61 is similar in construction to the network 45, but designed to provide a delay of only one-half cycle. The delayed signal is applied to a resistor 65. The original signal is also applied to the resistor 65 through the resistors 63 and 67. The delayed signal is represented by the graph of Figure 9. This is added in the resistor 65 to the original signal, represented by the graph of Figure 4. Owing to the half cycle delay of the network 61, each pulse of the resultant voltage is proportional in magnitude to the difference between successive left and right pulses of the original signal. The first pulse has a magnitude L2-RI, which, in the illustrated case, is merely L2. The second pulse has a magnitude R2-L2, etc. Thus, the amplitude of the A.-C. component of the wave of Figure 10 is at all times proportional to the rate of change of amplitude of the wave.of Figure 4. The low frequency component of the wave of Figure 10 is of no effect, since it is removed by the blocking capacitors in the power amplifier.

Although the invention has been described in connection with an electrical servo system associated with a radio locator, it will be understood that it is equally applicable to any A.-C. servo Ssystem, and may be employed as well for any other purpose which requires differentiation of the envelope of an A.-C. wave. The various graphs in the drawing illustrate rectangular waves. However, voltages of sinusoidal or other wave forms may be used, without altering the design or operation as described.

I claim as my invention: 1. In an. electrical servo system including an output shaft, a motor/coupled to said shaft, means for producing an A.-C. displacement signal of frequency / cycles per second and amplitude substantially proportional to the difference between the actual angular position of said shaft and the position to which said shaft is to be driven, and means for applying said displacement signal to said motor, anti-hunt means Including a time delay network designed to provide 65 a delay of length 2f seconds, wherein n is an integer, means for applying said displacement signal to said delay 70 network to produce a delayed displacement signal, and means for applying said delayed displacement signal to said motor, in addition to said original displacement signal.

2. In an electrical servo system including motor means adapted to be fnergized by an alter2f seconds, wherein n is an integer, and / is the fundamental frequency of said displacement signal, means for applying said displacement signal to 10 said network to produce a delayed A.-C. displacement signal, means for combining said delayed displacement signal with said original displacement signal to produce a resultant alternating current, and means for applying said resultant 15 current to said motor means.

3. In an electrical servo system including motor means adapted to be energized by an alternating current, means for producing an A.-C. displacement signal, anti-hunt means compris- 20 ing a time delay network arranged to provide a delay of 2f 25 seconds, wherein f is the frequency of said displacement signal, means for applying said displacement signal to said network to produce a delayed A.-C. displacement signal, means for combining said delayed signal with said original 30 displacement signal in phase opposition thereto to produce a resultant alternating current, and means for applying said resultant current to said motor means.

4. In an electrical servo system including mo- 35 tor means adapted to be energized by an alternating current, means for producing and utilizing in known manner an A.-C. displacement signal, anti-hunt means comprising a time delay network arranged to provide a delay of 40 1 f seconds, wherein I is the fundamental frequency 45 of said displacement signal, means for applying said displacement signal to said network to produce a delayed A.-C. displacement signal, means for combining said delayed signal with said original displacement signal in phase opposition 50 thereto to produce a resultant alternating current, and means for applying said resultant current to said motor means.

5. In an electrical servo system including means for producing an A.-C. displacement signal, a motor, and an amplifier connected to energize said motor, phase invertor means, means for applying said displacement signal to said phase invertor means to produce two outputs, both similar to said displacement signal but 180° out of phase with each other, means for applying 60 one of said outputs substantially without delay to said amplifier, and means for applying the other of said outputs to said amplifier with a delay of 65 S f seconds, wherein / is the fundamental frequency of said displacement signal.

6. In an electrical servo system including means for producing an A.-C. displacement signal, a motor, and an amplifier connected to energize said motor, means for applying said displacement signal substantially without delay to said amplifier, and further means for applying seconds, wherein f is the fundamental frequency of said displacement signal, and n is an integer.

13. In an electrical servo system including means for producing an A.-C. displacement signal, a motor, an amplifier connected to energize said displacement signal to said amplifier with a delay of 1 2f seconds, wherein / is the fundamental frequency of said displacement signal.

7. In an electrical servo system including means for producing an A.-C. displacement signal, a motor, an amplifier connected to energize said motor and means for applying said signal to said amplifier, comprising a phase invertor including two output circuits, means for applying said displacement signal to said phase invertor, a voltage combining network connected to one of said output circuits, a time delay network connected between the other of said output circuits and said combining network, and means for applying the ,output of said combining network to said amplifier.

8. In an electrical servo system including means for producing an A.-C. displacement signal, a motor, an amplifier connected to energize said motor and means for applying said signal to said amplifier, comprising a phase invertor including two output circuits, means for applying said displacement signal to said phase invertor, a voltage combining network connected to one of said output circuits, a time delay network connected between the other of said output circuits and said combining network, means for applying the output of said combining network to said amplifier, a second phase invertor including two further output circuits, a second voltage combining network connected to one of said further output circuits, a second time delay network connected between the other of said further output circuits and said second combining network, and means for applying the output of said second combining network to said amplifier.

9. The invention as set forth in claim 8 wherein said delay network is designed to provide a delay of 1 f seconds, wherein / is the fundamental frequency of said displacement signal.

10. The invention as set forth in claim 7 wherein said delay networks are each designed to provide a delay of 1 f seconds, wherein f is the fundamental frequency of said displacement signal.

11. The invention as set forth in claim 13 wherein said delay network is designed to provide a delay of 1 f seconds, wherein f is the fundamental frequency of said displacement signal.

12. The invention as set forth in claim 13 wherein said delay networks are each designed to provide a delay of said motor and means for applying said signal to said amplifier, comprising a voltage combining circuit in the Input circuit of said amplifier, means for applying said signal directly to said combining circuit, means including a delay network for a deriving a difference voltage whose amplitude is proportional to the difference between the amplitude of each cycle of said displacement signal and the next succeeding cycle, means for applying said difference voltage to said combining circuit, and means for applying the output of said combining circuit to said amplifler.

ALDA V. BEDFORD.

Number 2,298,177 2,233,415 1,554,698 2,186,268 Name Date Scott --------------. Oct. 6, 1942 Hull -----.--------. Mar. 4, 1941 Alexanderson --.... Sept. 22, 1925 Rakala -----------. Jan. 9, ledP