US2920191A - Trigger circuit - Google Patents

Description

Jan. 5, 1960 M. ROSENBERG TRIGGER CIRCUIT Filed Oct. 28. 1952 .l f i} i E'y. Z.

llllj HIP . 1N VE N TOR.

I M/fazzlwedarg PHASE I I i I By M 4 ti if! 77/14;

11 TTORNEY United States Patent TRIGGER CIRCUIT Milton Rosenberg, Trenton, N.J., assignor to Radio Corporation of America, a corporation of Delaware Application October 28, 1952, Serial No. 317,192

4 Claims. (Cl. 250-27) This invention relates to a signal responsive device of the type known as a trigger circuit in which the usual electron discharge tubes are replaced by binary transformer elements.

In modern information handling apparatus, such as digital computers, considerable utility is made of electronic switching and trigger circuits which are used to perform various control and storage functions. A basic form of such circuits is the Eccles-Iordan or flip-flop circuit which responds to input signals to produce two different outputs. The two outputs are produced by two different states of the circuit, which may be characterized by one of two tubes being either conductive or nonconductive. Two tubes are generally cross-coupled so that a change in a tube to one state produces a change to the opposite state in the other tube. Thus, the circuit has an essential binary character and mode of operation.

Due to the limited reliable life of an electron discharge tube, there are attendant problems of tube replacement in electron tube circuits. To reduce the number of tubes required, a magnetic flip-flop has been developed which performs essentially the same function as electronic trigger circuits but which dispenses with the electron tube as the basic circuit element. 'One such circuit is described in the Carter et al. Patent No. 2,591,406. See also the article Static Magnetic Storage and Delay Line by An Wang and Way Dong Woo in the Journal of Applied Physics, volume 21, page 49, 1950. The basic circuit element of the magnetic flip-flop shown in the cited references is a magnetic core having a rectangular hysteresis loop whereby the coremay be put in either of two stable conditions of saturation, namely saturation in a positive or a negative polarity. The magnetic core has an input Winding wound thereon whereby the polarity of the core may be reversed, and a transfer winding wound thereon whereby reversal of polarity may be detected and a current pulse transferred. A circuit using a pair of such cores also has a binary character or mode of operation. By means of a simple and economical device of this type, electron tube circuits may be replaced in various applications.

Accordingly, it is an object of this invention to provide a new and improved signal responsive device having a binary mode of operation in which a magnetic core is used as a basic element.

Another object of this invention is to provide a simple and economical bistable device utilizing a pair of magnetic cores.

Still another object of this invention is 'to provide a simple and reliable trigger circuit which utilizes a magnetic core as a basic circuit element.

These and other objects of this invention are achieved by means of a pair of magnetic cores having a substantially rectangular hysteresis loop. Each core is wound with an input winding and a transfer winding. The input windings are connected in series with an electronic switch for simultaneously energizing them in response to an Patented Jan. 5, 19fi0 input signal. The transfer windings are coupled together through an electrical delay line.

Initially the cores are magnetized to opposite polarities of saturation. Upon the application of an input signal pulse, both input windings are energized to apply magnetizing force pulses of the same direction on their respective cores. Since the cores originally had opposite polarities, only one of the cores (hereafter known as the first core) is reversed in polarity by the applied magnetizing forces. A current is induced in the transfer winding of the first core by the change in magnetic flux, and it is applied to the transfer winding of the second core after a delay due to the delay line coupling. By that time the input pulse has terminated, and the polarity of the flux in the second core is reversed by the current in the second core transfer winding. The polarities of saturation of both cores are thus reversed by the application of a signal pulse. A cycle is completed by the application of another pulse to restore the cores to their original condition. The process is the same, except that the second core is reversed initially. Output signals may be produced by detecting changes in magnetic polarity in either one of the cores.

The organization and method of operation of the invention may be best understood from the following description and the accompanying drawing in which:

Figure 1 is a graphical diagram of the hysteresis loop of a magnetic core used in this invention;

Figure 2 is a schematic diagram of circuitry embodying this invention; and

Figure 3 is a time-waveshape diagram.

The magnetic cores used in this invention have a substantially rectangular hysteresis loop as shown in Figure 1. As a result of this type of hysteresis loop, the application of a positive magnetizing force greater than a critical value, the coercive force H drives the core in the direction of +B and produces a large positive residual magnetism +B in the core. The application of a magnetizing force in the negative direction greater than the coercive force -H drives the core in the direction -B and produces a large negative residual magnetism -B,. As can be seen from the hysteresis curve, the saturation flux density B of the core is substantially the same as the residual flux density B Therefore, if a magnetizing force in a positive direction is applied to a core which has a positive magnetic polarity, essentially no change in the flux density of the core takes place. If the mag netizing force in a reversing direction is less than the coercive force H the flux density is not changed beyond the knee of the curve and, therefore, the residual magnetism is substantially unchanged. Thus, the magnetic core has two states of substantial stability and it remains in either one of those states in the absence of a large magnetizing force in the opposite direction. A detailed description of magnetic cores of this type may be found in the above cited article. Additional desirable characteristics of the material used in the magnetic cores are that it have a fairly low coercive force, and that it go from one state of saturation to another in as short a time as possible.

Referring now to Figure 2, a first and a second magnetic core 10, 12, having the characteristics described, are shown. The cores are shown as toroidal shape, by way of illustration only. Each core has an input winding 14, 16 and a transfer winding 18, 20 wound thereon. The input windings 14, 16 of both cores 10, 12 are connected in series and energized from a suitable source 22 of direct current. The energization of the input windings 14, 16 is controlled by means of a switching device shown as an input electron tube 24 having an anode 26, a cathode 28 and a control grid 30. The input Windings 14, 16 are connected in the anode circuit of the input tube 24. An input terminal 32 is coupled to the control grid 30 through a condenser 34. A negative potential biasing the input tube to cut off is applied to the grid through a biasing resistor 36. The transfer windings 18, 20 are coupled in a circuit by means of an electrical delay line 38 which is shown as an inductsince-capacitance lumped constant type of line. Output signals may be taken from the device by any appropriate means for detecting changes in the permeability of one of the cores. As shown, an output winding 40 on one of the cores is used. Alternative methods of producing output signals are described below.

The magnetic polarity of the core, in the sense that it is used here, is relative, in that it depends upon the sense of the input windings on the core as well as the direction of the energizing current applied to the windings. The magnetic polarity of a core may be assumed to be positive when a positive energizing current does not change the residual magnetism in the core, and negative when a positive current reverses the cords magnetism. Initially, the cores are magnetized to saturation in opposite polarities. The first core is assumed to be in a negative state and the second core 12 in a positive state as indicated by arrows in the drawing.

if a positive signal pulse is applied to the input terminal 32, the negative bias of the control tube 24 is overcome, and the tube conducts. The resulting current pulse in the input windings 14, 16 produces magnetizing force pulses which have the same relative direction. The direction of the magnetizing forces as shown tends to drive the cores to a positive polarity, in the convention assumed. Thus, the second core 12, which is already substantially saturated in the positive direction, is unaffected. However, the magnetism of the first core 10, which originally is negatively saturated, is reversed to a positive polarity. As a result, there is a current pulse induced in the transfer winding 18 of the first core 10 which is transmitted through the delay line 38 to the transfer winding 20 of the second core 12. The duration of the delay produced by the delay line 38 is made to be longer than the duration of the input pulse necessary to turn over either one of the two cores. Therefore, when the transferred current pulse has arrived at the second core transfer winding 20, the input magnetizing force pulse is terminated. The transferred current pulse changes the flux density in the second core 12 and reverses its polarity to negative saturation. Both cores again have relatively opposite magnetic polarities which are the reverse of the original ones. Both cores remain in that state until the application of another input signal pulse. The application of another input pulse reverses the polarity of the second core from negative to positive saturation, and the first core remains unchanged in a positive polarity until the application of a delayed transfer pulse. The transferred pulse then reverses the core polarity to a negative state. The cores again have relatively opposite magnetic polarities. It is seen that this invention provides a device having two stable states which reverses polarity upon application of each input pulse, and which is capable of fast action.

Each time that the second core 12 changes from a positive residual magnetism to a negative state, a voltage pulse is induced in the output coil. This voltage pulse has a polarity depending upon the sense of the winding of the coil; it is a negative pulse in the arrangement of Figure 2. When the second core changes from a negative to a positive polarity, a voltage pulse of positive polarity is induced in the output winding. Thus, two distinct outputs are produced corresponding to the two different conditions of stability or equilibrium of the device.

The device of this invention may be used as a basic component in the circuits of computers and other information handling systems. For example, a binary counter may be formed by cascade coupling a plurality of the bistable devices, with the output of one being connected through a rectifier (not shown) to form the input of the next. The rectifier permits pulses of only one polarity to pass, and an output pulse of that one polarity is produced for every two input pulses received. In this manner, the output of one bistable device may serve to drive the input windings of the next device in place of the electron tube drive.

For optimum output signal to noise ratio, the output pulses may be strobed, that is to say, sampled at an appropriate time, in the manner shown in Figure 2. One end of the output winding 40 is connected to the cathode 42 of a gating tube 44. The control grid 46 of the gating tube 44 is negatively biased through a biasing resistor 48, and the cathode 42 is positively biased through the output winding 40. The gating tube control grid 46 is driven by a delay line 50 which is coupled to a cathode resistor 52 in the circuit of the input tube 24. The gating tube biasing resistor 48 is preferably matched to the gating delay line 50. An operating potential is applied to the anode 54 of the gating tube 44 through a load resistor 56. The output is taken at a terminal 58 connected to the gating tube anode 54.

The different polarity voltage pulses in the output winding 40 on the second core 12 are induced at different time periods relative to the time the input pulse is applied to the input tube 24. When the polarity of the second core 12 is changed from negative to positive by an energizing current pulse in the input winding 16, a positive pulse is induced in the output winding 40 at the same time that the input pulse is received. This may be seen in the time-waveshape diagram of Figure 3 during time period t When a second input pulse is received during time period 1 the second core 12 already has a positive residual magnetism, and there is only a small change in magnetism in the core to saturation and back to the residual level. This small change induces a small pulse in the output winding at time t After the input pulse terminates, there is a delayed current pulse, resulting from the turnover of the core 10, applied to the second core transfer winding 20, which reverses the polarity of the second core 12 from positive to negative to induce a negative pulse in the output winding at time period 1 The signal to noise ratio at time periods t and t, is a maximum since the input pulse is then terminated. Therefore, the signal in the output winding 40 may be best strobed after termination of the input pulse to detect the state of the trigger circuit. The gating tube 44 is used for this purpose. The negative bias potential at the grid 46 of the gating tube, and the positive bias potential at the cathode 42 may each be made equal to cut-off potential. Thus, the gating tube is biased to twice cutofi. A positive gating pulse is applied to the control grid 46 from the input tube cathode resistor 52 through the delay line 50. The duration of the delay produced by the gating delay line 50 is made to be longer than the duration of the input pulse, so that the tube is gated on during time periods t and 1 During time period 1 there is no significant signal induced in the output winding 40 and the gating tube cathode, therefore, remains substantially at cut-off potential. The gating tube does not conduct and the voltage at the anode of the tube is unchanged. At time period t however, the coincidence of the negative pulse in the output winding 40 and the positive gating pulse at the control grid 46, overcome the respective bias potentials at cathode and grid to drive the tube conductive. The anode of the gating tube goes negative producing a negative output signal pulse to indicate the negative magnetic polarity in the second core 12. Thus, an output signal pulse and the absence of such a signal respectively indicate the two stable states of the trigger circuit. Of course the output from the gate tube 44 can be used to drive a subsequent circuit.

It is, therefore, evident from the above description that this invention provides a fast acting bistable device which has a binary mode of operation. It is simple in construction and economical in the components described and may be used to replace electron tube circuits in various applications.

What is claimed is:

l. A signal responsive device comprising two magnetic cores adapted to be substantially saturated in opposite polarities, an input winding and a transfer winding on each of said cores, means responsive to a signal for simultaneously energizing said input windings to apply magnetizing forces of the same direction to said cores, an electrical delay line connected between said transfer windings, and output means coupled to one of said cores for detecting changes of polarity therein, said energizing means including switching means having a control element, an input terminal coupled to said control element for receiving signals, said input windings being connected in series with said switching means, and said output means including an output winding on said one core.

2. A signal responsive device as recited in claim 1 wherein said output means includes means coupled to said output winding for sampling signals induced in said ,output winding when the signal to noise ratio is a maximum.

3. A signal responsive device as recited in claim 1 wherein said energizing means includes means for applying a unidirectional voltage to said input windings, said 6 switching means includes a grid-controlled electron mbe having an anode and cathode, said input windings being connected in series with each other and with said voltage applying means and the anode-cathode current path of said switching tube and said delay line is of the lumped constant type.

4. A signal responsive device as recited in claim 3 wherein said output means includes a gating tube having an anode, a cathode and a control grid, one end of said output winding being connected to said gating tube, a delay line connected between said switching tube and said gating tube control grid, means for applying an operating potential to said gating tube anode, and an output terminal connected to said gating tube.

References Cited in the file of this patent UNITED STATES PATENTS OTHER REFERENCES Article MagneticTriggers, by An Wang, pages 626 to 629, Proceedings of the IRE, June 1950.

Images