Cross Reference to Related Applications
The subject matter of the subject invention is related to the subject matter of the application entitled "Switching Circuit Having Hysteresis", Ser. No. 967,825 and "Comparison Circuit Having Bidirectional Hysteresis", Ser. No. 967,826, by Marc T. Stein which are filed concurrently herewith and assigned to Motorola, Inc.
Background of the Invention
This invention relates to pulse generator circuits and more particularly to multivibrator circuits for producing constant width output pulses in response to applied alternating input signals.
In a control system it is quite often desired to generate a constant width pulse in response to an applied alternating input signal. For instance, it might be desired to provide a precision one shot function where, in response to either the positive or negative going transition of the applied input alternating signal, a constant width output pulse is generated. The output pulse can then be integrated to provide an analog voltage. This analog voltage may be used to control or adjust some parameters of the system in which the multivibrator circuit is utilized. Moreover, it might be desired to generate a doubling one shot function wherein fixed output pulses are generated at both the negative and the positive going transitions of the applied alternating input signal.
It is important in such pulse generating systems that the output pulses be generated for the particular transition of the applied alternating input signal. A problem can occur when the frequency of the input alternating input signal is increased such that the period becomes quite small. In this mode it is quite possible that in prior art circuits the generation of the pulses will occur very rapidly, becoming crowded, such that the pulses tend to crowd together to become one and are no longer distinct from one another. If this were to happen, and if the pulses were integrated in the system, then an analog voltage having an incorrect magnitude could be generated. Thus it is important that crowding of the pulses be prevented during high frequency modes of operation of the control circuit.
Summary of the Invention
Accordingly, an object of the present invention is to provide an improved pulse generator.
Another object of the present invention is to provide an improved pulse generator for producing a precision one shot function.
Still another object of the present invention is to provide a pulse generator system for producing a doubling one shot function.
A further object of the present invention is to provide a pulse generator system which prohibits pulse crowding when the system is in a high frequency mode.
A still further object of the present invention is to provide an improved pulse generator system for producing a precision one shot or doubling one shot function which is suitable to be manufactured in monolithic integrated circuit form.
In accordance with the foregoing and other objects there is provided a pulse generator circuit for producing output pulses of fixed width in response to applied alternating input signals. In one embodiment of the present invention the output pulses are generated with respect to the positive going transition of the applied input signal. In this embodiment, the pulse generator circuit comprises a triggering circuit adapted to receive the alternating input signal to generate triggering signals on the positive going transition of the alternating input signal, and a control circuit which is responsive to the applied triggering signals and to logic control signals supplied thereto for producing control signals at an output thereof. Additionally, the pulse generator circuit includes an output switching device which is responsive to the control signals from the control circuit for producing the output pulses and logic circuitry which produces the logic control signals. A high frequency control circuit is also provided which is enabled when the frequency of the applied input alternating signal becomes greater than a predetermined value in order to ensure that output pulses are generated at each positive going transition of the applied alternating input signal.
In another embodiment of the invention there is included a pulse generator system for producing output pulses having fixed widths upon both the negative going and the positive going transition of the applied alternating input signal. In this embodiment, the above described pulse generator circuit is modified such that the triggering circuit provides triggering signals on both the negative going and positive going transition of the applied alternating input signal. In this manner, the control circuit produces said control signals upon both negative and positive going transitions of the applied input signal which are then utilized to drive the output switching circuit to generate the output pulses at the appropriate times.
Brief Description of the Drawings
FIG. 1 illustrates in partial schematic and block diagram form the multivibrator or one shot pulse generator of the first embodiment of the invention;
FIG. 2 is a waveform diagram useful for understanding the function of the embodiment of FIG. 1;
FIG. 3 is a detailed schematic diagram of the preferred embodiment of FIG. 1;
FIG. 4 is a partial schematic and block diagram form of a doubling one shot pulse generator of a second embodiment of the invention; and
FIG. 5 is a waveform diagram useful for understanding the function of the embodiment of FIG. 4.
Detailed Description of the Preferred Embodiments
Turning to FIGS. 1 and 3, there is shown multivibrator or one shot pulse generator 10 of one embodiment of the present invention. One shot 10 is triggered by the positive going edge of triggering input signal F.sub.IN which is applied to input 12. As will be discussed in more detail, the application of f.sub.IN causes a constant width output pulse to be produced at output terminal C.sub.0. It is understood that in agreement with the objectives of the present invention that the output pulse C.sub.0 can be integrated and utilized to provide an analog voltage having a magnitude that is proportional to the frequency of f.sub.IN.
In conjunction with the waveforms illustrated in FIG. 2, the operation of one shot 10 can be described with the aid of FIG. 1. With one shot 10 in its monostable state, transistor 14 is conducting and transistor 16 is nonconducting. Hence, capacitor 18 is fully discharged by transistor 14 as is shown by waveform 62. Transistors 14 and 16 are maintained in their aforementioned states by the Q output (waveform 60) and its complementary output Q from D-type flip-flop 20 being applied to the respective base electrodes thereof. Thus, the output C.sub.0 (waveform 66) of the one shot is nearly equal to the supply voltage V+ which is supplied via resistor 22 to the collector of transistor 16.
Triggering occurs on the positive going edge of input signal f.sub.IN (waveform 52) when the magnitude thereof becomes greater than the voltage supplied at the positive terminal of comparator 24. Comparator 24, comprising operational amplifier 26, feedback resistor 28 and resistors 30 and 32 functions as a comparator having hysteresis as described in the aforementioned Stein application, which is incorporated herein by reference, entitled "Comparison Circuit Having Bidirectional Hysteresis". When comparator 24 is triggered, its output goes negative (waveform 54). This negative transition of C.sub.1 cause the Q output of D-type flip-flop 34 to go high (waveform 56). The Q output of flip-flop 34 is applied to an input of NAND gate 36. Assuming the other input to NAND gate 36 is also high, the output of comparator 38 being high when capacitor 18 is discharged, the output C.sub.2 of the gate is forced low (waveform 58). This negative transition of C.sub.2 is applied to the C input terminal of flip-flop 20 causing the Q and Q outputs thereof to change states. This change of state causes transistor 14 to become nonconducting and transistor 16 conductive. The output C.sub.0 will then be forced low and capacitor 18 starts charging through resistor 40; ramping toward the power supply voltage V+. The forcing of C.sub.2 low (output of gate 36) also resets flip-flop 34 through inverter 42.
When the magnitude of the voltage developed across capacitor 18 reaches the value 1/2 V+, comparator 44 is triggered and its output goes low. In this condition, the output R (waveform 64) of NAND gate 46 is forced high which in turn resets flip-flop 20. Resetting of flip-flop 20 renders transistor 14 conductive again to discharge capacitor 18 and the aforedescribed cycle is initiated once again. Hence, in response to the triggering signal f.sub.IN an output pulse C.sub.0 is produced having a fixed width which is determined by the RC time constant of capacitor 18 and resistor 40. If one shot 10 is a monolithic integrated circuit with resistor 40 and capacitor 18 provided off "chip" the width of pulse C.sub.0 can be made substantially independent to temperature variations.
D-type flip-flop 48 is utilized only in a high frequency mode, i.e., when the frequency of f.sub.IN increases to a point where capacitor 18 cannot be charged completely to 1/2 V+ between positive going edges of f.sub.IN. Because output pulses C.sub.0 may be integrated to provide an analog voltage proportional to the frequency of f.sub.IN it is important that each positive going edge of f.sub.IN is recognized or errors could be created in the integration scheme. Flip-flop 48 resets flip-flop 20 to discharge capacitor 18 if a new trigger edge of f.sub.IN is received from comparator 24 while capacitor 18 is still being charged: ramping toward V+. The operation of flip-flop 48 during the high frequency mode is explained hereinafter.
If f.sub.IN goes positive while capacitor 18 is ramping, the negative transition of C.sub.1 causes the Q output of flip-flop 48 to go low because the D input thereof is forced high by the Q output of flip-flop 20 being high during this period. This causes NAND gate 46 to change output states thereby applying a high output signal to the reset terminal of flip-flops 20 and 48. The resetting of flip-flop 20 causes the Q output therefrom to go high turning transistor 14 on to discharge capacitor 18. Resetting of flip-flop 20 also places flip-flop 48 into its nominal state (Q output high). A race condition between resetting of flip-flops 20 and 48 is prevented by the nominal gate delay through NAND gate 46. Thus, even in a high frequency mode, each positive edge of f.sub.IN is recognized.
The same negative edge of C.sub.1 which triggered flip-flop 48 also triggers flip-flop 34 such that the Q output therefrom goes high as previously described. However, as capacitor 18 is ramping at this time the output from comparator 38 is caused to be low so that there is no change in output state of NAND gate 36. However, the causing of capacitor 18 to be discharged through the triggering of flip-flop 48 causes the output of comparator 38 to go high which in conjunction with the Q output of flip-flop 34 being high then causes NAND gate 36 to change state. When the output of C.sub.2 goes low, capacitor 18 begins charging again as discussed above.
Turning now to FIG. 3, a more detailed schematic of one shot 10 is shown. It is to be understood that one shot 10 can be fabricated using monolithic integrated circuit techniques. Furthermore, the logic functions of one shot 10 are fabricated using known integrated injection logic (I.sup.2 L) techniques and the linear circuits are fabricated using linear integrated circuit techniques. It is to be understood that the I.sup.2 L circuits utilize open collector transistors with a constant current source at each input thereof.
Comparator 24, which is described in the aforementioned Stein application entitled "Comparison Circuit Having Bidirectional Hysteresis" is briefly described. NPN transistors 70 and 72 are matched to one another during fabrication of one shot 10 as well as are PNP multi-collectors 74 and 76. Collector electrodes 78 and 80 are coupled to a current mirroring circuit comprising transistors 82 and diode 84, which is well known in the art, to provide a single ended output from amplifier 26 that is taken at the collector electrode of transistor 82.
As is understood, the trip point of amplifier 26 will occur when the magnitude of f.sub.IN exceeds V.sub.88, the voltage at node 88. Switching current is provided by current source 86. If the magnitude of f.sub.IN is less than the voltage V.sub.88, transistor 70 is either nonconductive. In this state, transistor 94 is nonconductive whereas transistor 96 is conductive. The magnitude of V.sub.88 is determined by the resistor divider comprising resistors 90, 92, and 98 which are connected between V+ and the collector-to-emitter voltage of transistor 96. Transistor 96, being in a saturated condition, causes the output from inverter 100, waveform C.sub.1, to be in a high state as shown in FIG. 2. Now, when f.sub.IN becomes greater than the magnitude of V.sub.88, amplifier 26 is tripped which causes C.sub.1 to go negative. In this state, transistor 70 is conductive and transistor 72 is nonconductive. Excess current from transistor 74 provides drive to the base of transistor 94 rendering it conductive. With transistor 94 conducting, transistor 96 is rendered nonconductive forcing the output C.sub.1 low. Moreover with transistor 94 conducting, the magnitude of V.sub.88 is caused to be decreased which introduces hysteresis into the circuit. Therefore, as described, comparator 24 is triggered on the positive edge of f.sub.IN to trigger flip-flop 34 and 48.
Amplifier 38 is shown as including current source 102 which provides current to differentially connected PNP transistors 104 and 106. These transistors are coupled in a quasi-Darlington amplifier configuration to transistors 108 and 110 respectively. The collector electrodes of transistors 104 and 106 are coupled to a single ended output circuit comprising transistor 112 and diode 114 as well as transistor 116 in a similar manner as discussed with regards to amplifier 26. The output of amplifier 38 which is coupled to NAND gate 36 (shown as inverter 119) is provided at the output of inverter 118.
In operation, with capacitor 18 discharged, transistors 104 and 108 conduct more heavily than transistors 106 and 110 because the latter transistors are biased by V.sub.REF which is supplied to the base of transistor 110 is of greater magnitude than the bias potential supplied at the base of transistor 108. In this state, excess current is provided at the collector of transistor 104 to render transistor 116 conductive. Hence, the potential appearing at the collector electrode of transistor 116 will be near ground potential. Inverter 118 which is coupled to the collector of transistor 116 is then caused to produce a logic "1" at the output of comparator 38 which is in agreement with the aforedescribed operation. The output of amplifier 38 remains high until such time as the potential supplied at the base of transistor 108 becomes greater than the magnitude of V.sub.REF. When this condition occurs transistor 106 conducts more heavily than transistor 104. Transistor 112 then becomes saturated such that the base drive current is "starved" from transistor 116 rendering it nonconductive and thereby causing the output, via inverter 118, of comparator 38 to go to a logic "0". Therefore as described immediately above, the output from comparator 38 is maintained and a logic "1" when capacitor 18 is discharged through transistor 14 whereby NAND gate 36 is triggered by the change of logic states at the output of flip-flop 34 caused by the triggering signal f.sub.IN.
Comparator 44 operates in substantially the same manner as comparator 38 to drive one input of NAND gate 46 which is shown as a two input inverter gate. For example, with capacitor 18 discharged, NPN transistor 120 is more conductive than NPN transistor 122. Multi-collector PNP transistor 124 is therefore more conductive than multi-collector PNP transistor 126. Because current source 127 sinks a constant current from transistors 124 and 126, the excess current from transistor 124 to diode 128 causes transistor 130 to be in a saturated state. This in turn prevents base current drive from being sourced to transistor 132 which renders it nonconductive. Transistor 132 being nonconductive establishes a logic "1" to the input of NAND gate 46. Subsequent to triggering of one shot 10, when the magnitude of the potential appearing across capacitor 18 becomes equal to 1/2 V+, the output of comparator 44 switches causing resetting of flip-flop 20. Switching of the comparator occurs because transistors 122 becomes more conductive than transistor 120. In this condition, transistor 126 conducts more heavily than transistor 124 and transistor 130 is no longer saturated. Thus, the excess current is supplied to the base of transistor 132 rendering it conductive to force its collector to go low generating a logic "0".
Turning now to FIGS. 4 and 5 there is shown pulse generator 140 which functions in substantially the same manner as one shot 10 but produces two fixed width output pulses for each complete cycle of the input signal C.sub.r. In FIG. 4 the components which correspond to like components of FIG. 1 are numbered with the same reference numerals. Briefly, pulse generator 140 responds to the positive and negative going edges or transitions of C.sub.r to produce output pulses at output terminal C.sub.0 of the generator.
In operation, assuming transistor 14 is conducting (the Q output of flip-flop 20 being high or at a logic "1") capacitor 18 will be discharged as illustrated by waveform 174. This condition produces a logic "1" at the output of comparator 38 which is applied to one input of NAND gate 36. The output at terminal C.sub.0 will also be in a low state (waveform 176) as transistors 16 is also conducting. Operational amplifiers 142 and 144 comprise comparator 146. The construction of comparator 146 is the same as the circuit disclosed in the application entitled "Switching Circuit Having Hysteresis" referenced above and is incorporated by reference herein. Functionally, the outputs of amplifiers 142 and 144 (waveforms 162 and 164 respectively) are the complementary of each other and trip at a value of the input signal C.sub.r that is the magnitude V.sub.0 apart from each other to set and reset flip-flop 34.
As illustrated, a constant reference voltage developed across potentiometer 148 is applied to both the noninverting input of amplifier 142 and the inverting input of amplifier 144. Hence, with the magnitude of C.sub.r greater than the magnitude of V.sub.148, the output R from amplifier 142 is low and output S from amplifier 144 is high. On the negative transition of C.sub.r, amplifier 142 is caused to be tripped when the magnitude of C.sub.r is substantially equal to the magnitude of V.sub.148 and the output therefrom then goes to a high state. Amplifier 144 is tripped by a further decrease in the magnitude of C.sub.r, which in the preferred embodiments is 36 millivolts, to cause the output S to go to a low state. With S low, the Q and Q outputs of flip-flop 34 (waveforms 166 and 168) are clocked high and low respectively.
D flip-flops 150 and 152 which act in conjunction to function as a frequency doubler, receive the Q and Q outputs respectively from flip-flop 34. The negative transition of Q output applied to flip-flop 152 causes the Q output therefrom to go negative. Similarly, when C.sub.r goes positive and its magnitude becomes greater than V.sub.148 the outputs from amplifiers 142 and 144 change states which cause the output Q of flip-flop 150 to be triggered and to go negative. Hence, in response to each transition of input signal C.sub.r, the Q outputs of flip-flops 150 and 152 are alternatively clocked to a low state. When the Q output of flip-flop 152 is clocked low a high or logic "1" is applied to one input of NAND gate 36 via inverter 154. Since both inputs are high NAND gate 136 is then caused to change states. The negative transition of the output of NAND gate 36 both resets flip-flop 152 via inverter 42 and clocks the Q outputs of flip-flop 20 to a low state (waveform 172). Resetting of flip-flop 152 also resets the output of NAND gate 36 to a high state. This action renders The Q outputs of flip-flop 20 being clocked low cause transistors 14 and 16 to be rendered nonconductive. Hence, the output at terminal C.sub.0 is forced to a high state, waveform 176.
Transistors 14 and 16 will remain off until capacitor 18, charging through resistor 40, is ramped to a value equal to 1/2 V+ which trips comparator 44 causing its output to go low. The low state at the output of comparator 44 forces the output of NAND gate 46 to go high which resets flip-flop 20 causing the Q outputs therefrom to again go to a high state. This renders transistors 14 and 16 conductive to discharge capacitor 18 (waveform 174) and to cause the output at terminal C.sub.0 to go low (waveform 176) respectively.
Similarly, when flip-flop 150 is triggered by the positive going transition or C.sub.r, the above cycle is repeated. Thus, a fixed width output pulse is generated at the output of transistor 16 upon each negative and positive transition of input alternating signals C.sub.r. Again, these pulses may be used in a control system to provide an analog control voltage.
As described for one shot 10 of FIG. 1, flip-flop 48 is utilized to ensure that each transition of input signal C.sub.r is recognized even when the circuit is in a high frequency mode of operation wherein the RC time constant of resistor 40 and capacitor 18 prevents capacitor 18 from becoming fully charged to the value 1/2 V+ during a full cycle of operation.