US2947949A - Fast multivibrator circuit - Google Patents

Description

MlCHlYUKl NAKAMURA 2,947,949 FAST MULTIVIBRATOR CIRCUIT Aug. 2, 1960 Filed Nov. 12, 1957 5 If I cm f "I" M/:.: k H

INVENTOR. Mon run Mun/vain Mag/W624 United States Patent FAST MULTIVIBRATOR CIRCUIT Michiyuki Nakamura, Berkeley, Calif., assignor to The Regents of The University of California, Berkeley, Calif., a corporation of California Filed Nov. 12, 1957, Ser. No. 695,943

3 Claims. (Cl. 331-144) This invention relates to fast-acting multivibrators,

to external events, or monostable, or bistable.

In a multivibrator, two switching devices, e.g., electron tubes or transistors, conduct current alternately. An important characteristic of a multivibrator is the time required to complete the switching of current from one switching device to the otheri.e., the transition time between the two operating states of the multivibrator. The length of the switching or transition time has been a major limitation in the achievement of fast-acting multivibrators.

A faster-acting multivibrator may be devised for a variety of reasons; e.g., to achieve higher frequencies of operation; or to obtain pulse-type and other waveforms having shorter rise times, or shorter fall times, or both. An object of this invention is to provide improved, fast-acting multivibrators for such purposes.

A limiting factor aflecting the switching or transition time in multivibrators is the time required to charge and to discharge circuit capacitances, including parasitic or stray capacitances. These capacitances tend to delay changes in voltage at various parts of the circuit, and thus tend to lengthen the rise and fall times of the voltage waveforms. This in turn lengthens the minimum period of the operating cycle. A more specific object of this invention is to provide improved multivibrators having novel means for charging and discharging circuit capacitances, and in particular certain of the unavoidable stray capacitances, with extreme rapidity.

According to the invention, a controlled, variable-impedance load device is connected in series with one switching device of the multivibrator, and, preferably, another such load device is connected in series with the other switching device. Each of the load devices is individually controlled, as hereinafter explained, to provide a low impedance while the voltage across the associated switching device is rising, so that circuit capacitances charge-rapidly through the load device for effecting a short rise time in the voltage waveform, and to provide a high impedance while the voltage across the associated switching device is falling, so that circuit capacitances discharge rapidly through the switching device for effecting a short fall time in the voltage waveform. Multivibrators so constructed have been operated at significantly higher frequencies than any known to have been attained with prior multivibrators, with corresponding improvements in the rate of change in the voltage waveform at the transistion points.

One example of carrying the present inventive concept into practice is described in conjunction with the accompanying drawing in which there is shown a multivibrator circuit incorporating therein the novel circuitry of this invention.

In the drawing, there is shown a multivibrator circuit of the astable or Tree-running type. Although the astable circuitry is shown, it will be readily appreciated that it is only by way of example and for purposes of illustration, whereas the inventive concept can also be values of the bias resistors, etc.

. which may be astable, with or without synchronization readily applied to monostable and bistable multivibrators. The changes required to apply the concept of the present invention to bistable and monostable multivibrators requires only modifications in the parameters of the circuit, well known to those skilled in the art, such varying the In the circuit illustrated, there is a first multivibrator tube or switching device in the form of a vacuum tube 10. The cathode 12, defining one principal electrode, of the switching device or tube 10 is connected to a ground potential lead 14. The other principal electrode, anode 16, of the device or tube 10 is coupled through a first inductor 18 to one principal electrode, the cathode 20, of a first load device in the form of a vacuum tube 22. The other principal electrode, anode 27, of the first load device or tube 22 is connected to a plate-supply voltage lead 24, which may be at +300 v., and will be referred to as E The control electrode 26 of the tube 22 is connected to the anode of the switching tube 10 through a lead 28 interposed between the plate 16 and the first inductor 18. The control grid 30 of the first switching tube 10 is connected through a bias resistor 32 to a negative voltage lead 34 (which can be referred to as E and is also connected through a first grid resistor 36 and first grid capacitor 38, in parallel, to one principal electrode, the cathode 40, of a second load device in the form of vacuum tube 42, while the anode 43, or other principal electrode, of the second load tube is connected to E This completes the first section of the circuit.

The second section of the circuit is similar to the first section. In the second section of the circuit, a second multivibrator tube or switching device, taking the form of a vacuum tube 44, has one of its principal electrodes, cathode 46, connected to the ground potential lead 14. The other principal electrode, anode 48, is connected in series with a principal electrode, cathode 40, of second load device or vacuum tube 42 through a second inductor 5t} interposed between the devices or tubes 42 and 44. The control electrode 52 of the second load tube 42 is connected through lead 54 to the plate 48- of the second switching tube 44 at a'poiut between theplate and the inductor 50.

The control electrode 55 of the second switching tube 44 is connected through a second bias resistor 56 to the negative voltage (-E and the control electrode 55 is also connected through second grid resistor 58 and second grid capacitor 60, in parallel, to the cathode 20 of the first load tube 22. From the foregoing description of connections, it is evident that the control electrode of each switching tube is coupled to the anode of the other switching tube, the impedances of inductors 18 and 50 being small compared to the impedances of load tubes 22 and 42', and that the control electrode of each load tube is coupled to the anode .of

' the switching tube in the same circuit branch.

The stray capacitances of the switching tubes 10 and 44 are shown in broken lines of the drawing. Among the stray capacitances indicated are a first input capacitance 62 from the control electrode 30 of the first switching tube 10 to the ground lead 14, a second input capacitance 64 from the second switching tube 44 to the ground lead 14, a first output capacitance 66 from the anode 16 of the first switching tube 10 to the ground.

trode of the first switching tube 10, which had been non-conducting, suddenly shifting the tube 10 from its non-conducting to a conducting state. The sudden flow of current to anode 16 creates a voltage drop across the first inductor 18. This supplies a negative voltage between the control electrode 26 and the cathode 20 of the first load tube 22, which tends to cut this tube oil so that it appears as athigh impedance. The impedance from anode 27 to anode 16 may then be expressed as [(p.+l)X +r The value of the plate resistance r for the load tube 22 is very high when cut-E is approached and the reactance X of the first inductor 18 is multiplied by the amplification factor a of the load tube. With a high impedance in the anode circuit, most of the electron current of the first switching tube is available for dissipating the positive charge on the first output capacity 66 and for negatively charging the second input capacity 64. The impedance path from the anode 16 of the tube 10 to the second input capacity 64 is quite low since the second grid capacitor 60 is essentially a short circuit at the high frequency considered. The reactance of the first inductor 18 is relatively low in such impedance path since the reactance of the first inductor is not multiplied by the amplification factor a of the load tube 22. The electron flow through the first inductor 18 to the second input capacity 64 creates a voltage drop which tends to hold the first load tube 22 non-conductive, and thus as a high impedance until the capacity 64 is charged. By substantially eliminat ing the retarding effect of these stray capacities, the circuit may be changed from the non-conducting to the conducting state in a very short time, and multivibrator operating frequencies up to 60 megacycles have been obtained.

' Asthe first switching tube 10 is changing from its non-conducting to its conducting state, load tube 22. presents a high impedance, as explained above, and a rapidly decreasing or negative-going potential is applied to the control electrode 55 of the second switching tube 44. Thus, tube 44 is changed from its conducting to a non-conducting state. A voltage is induced in the second inductor 50 by the decaying magnetic field of the inductor as the current through tube 44 and inductor 50 decreases driving the control electrode of the second load tube 42 positive with respect to the cathode 40.

The decaying magnetic field tends to maintain an electron current flow in the second inductor 50. Since the second switching tube 44 has been rendered non-conducting, the electrons are taken from the second stray output capacity 68. The decaying magnetic field in the second inductor 50 also causes the second load tube 42 to have a low impedance, the low limit being the reciprocal of the transconductance of the tube. The first input capacity 62 is then eifectively connected to the direct current supply (E through a low impedance. Thus, the circuit appears to have the characteristics of a cathode follower, and at the same time, have the attendant advantages of a low impedance output.

It will now be assumed that switching tube 44 is cut off and that first switching tube 10 is conducting. A switching signal, derived either internally or externally, is transmitted to the grid of the first switching tube It) and tube 10 tends to cut oil. Consequently, the current flowing through the first inductor 18 decreases, whereby the grid-to-cathode voltage of the first load tube 22 to become more positive. This action tends to boot strap the cathode potential of tube 22 upward toward E This action is then transmitted to the grid of the second switching tube 44, making tube 44 conductive. The current in the second inductor 50 tends to increase, producing a voltage which tends to cut off the second load tube 42, which in turn permits the cathode voltageof tube 42 to fall. As the cathode voltage of tube 42 falls, the first switching tube 10 is further driven toward cut ofi. It will be readily appreciated, that the entire action above described is regenerative, and the action is completed when tube 44 is cut off and tube 10- is conducting.

The action of the second inductor 50 and load tube 42 may be thought of as substantially aiding the switching speed as tube 44 becomes more conductive because not all of the electrons conducted by switching tube 44 flow through load tube 42 since load tube 42 is tending to cut oil. These electrons may now concentrate on charging the stray capacitances 68 and 62' associated with the plate circuit of tube 44 and the grid circuit of tube 10 respectively.

The voltage induced across the first inductor 18 as tube 10 becomes less conductive makes the load tube 22 a low impedance for discharging the stray capacitances associated with the grid of switching tube 44. It will be readily appreciated that the inductor 18 has an inductance and a resistance, L and R respectively, the sum of which is a complex impedance that may be represented by Z Likewise, the second inductor 50 has its inherent inductance and resistance, L and R respectively, the sum of which is a complex impedance represented as Z Therefore, the output impedance (Z at the cathode 20 of tube 22 is expressed where Z =total Of RK1 and LE1. r =plate resistance of tube 10. r =plate resistance of tube 22. =arnplificati0n factor of tube 22.

The voltage gain (without load) at the cathode of tube. 22 over the signal applied to the grid of tube 10 is where ,u equals amplification factor of tube 10.

Therefore, it will be evident that the output impedance (Z at the cathode of switching tube 22 is the ratio of the open-circuit voltage gain Equation 2 to the short-circuit gain, as set out in Equation 1.

Synchronization signals, from any source, not shown, may be introduced at the control electrode or the anode of the switching tubes following procedures well known to those versed in the art. Where triggering signals are to be used, the parameter values of the described circuit may be changed to provide either one or two stable states for the multivibrator following well-known techniques in the art.

It is further within the scope of this invention to utilize transistors in place of the electron vacuum tubes of the described circuit. This change is feasible by using those modifications well known to those skilled in the art.

What is claimed is:

1'. In a multivibrator, the combination of four electron tubes each having an anode, a cathode and a control electrode, said four tubes being here designated for convenience the first switching tube, the first load tube, the second switching tube, and the second load tube, an inductor connected between the cathode of the first load tube and the anode of the first switching tube, forming a first circuit branch including two tubes in series with an inductor therebetween, another inductor connected between the cathode of the second load tube and the anode of the second switching tube, forming a second circuit branch including two tubes in series with an inductor therebet-ween, connections for applying voltage between the anode of the load tube and the cathode of the switchingtube of each circuit branch, means coupling the control electrode of the switching tube in each branch to the anode of the other switching tube via the inductor cathode junction in the other branch for varying the current in each branch in inverse relation to the current in the other branch, and means coupling the control electrode of the load tube in each branchto. the anode of the switching tube in the same branch so that each load tube has a high impedance when the current in its circuit branch is increasing and a low impedance when the same current is decreasing.

2. The combination set forth in claim 1, wherein the coupling means between the two switching tubes comprises a resistor and a capacitor connected in parallel between the cathode of the first load tube and the control electrode of the second switching tube, another resistor and capacitor connected in parallel between the cathode of the second load tube and the control electrode of the first switching tube, means providing a negative potential relative to' the cathodes of the switching tubes, and two resistors connected between the last-mentioned means and the respective control electrodes of the two switching 15 tubes.

5 connected to the anode of the second switching tube.

References Cited in the file of this patent UNITED STATES PATENTS 10 2,545,924 Iohnstone Mar. 20, 1951 2,550,116 Grosdolf Apr. 24, 1951 2,651,722 Bergfors Sept. 8, 1953 2,745,955 Havens May 15, 1956 2,754,376 Grosdoif July 10, 1956 2,778,935 Ropiequet Jan. 22, 1957

Images