US 3860914 A
A digital data recorder is described having inherent circuitry characteristics that will re-establish the recorded data state previously established just prior to a power failure upon the return of the source power supply. The retentive characteristics of the recorder are supplied from an arrangement of magnetic memory cores which controllably vary the time constants associated with given portions of the circuitry of a digital counter comprising the major subunit of this invention. In one embodiment the counter is supplied with a zero inhibit to prevent reverse counting past the zero output state, thus preventing a false indication of the recorded data. The circuitry thus described, basically comprising solid state elements, provides a compact, resilient unit with no mechanical moving elements.
Claims available in
Description (OCR text may contain errors)
United States Patent Zitelli et al.
[ Jan. 14, 1975 1 DIGITAL DATA RECORDER  Inventors: William E. Zitelli; Alan F. Mandel,
both of Pittsburgh, Pa.
 Assignee: Westinghouse Electric Corporation,
 Filed: Jan. 8, 1973  Appl. No.: 321,910
7  11.8. Cl 340/172.5, 307/282, 340/174 SR  Int. Cl. G061 1/00  Field of Search..... 340/172.5, 174 SR, 174 FA; 307/282; 328/395 56 1 References Cited UNITED STATES PATENTS 3,003,141 10/1961 Myers, Jr 307/282 X 3,154,671 10/1964 Eachus 340/174 SR 3,214,606 10/1965 Wilson 307/282 3,408,505 10/1968 Coolidge, Jr. et a1 340/174 SR 3,487,381 12/1969 Michaelis 340/174 S POWER Primary ExamincrHarvcy E. Springborn Attorney, Agent, or Firm-D. C. Abeles A digital data recorder is described having inherent circuitry characteristics that will re-establish the recorded data state previously established just prior to a power failure upon the return of the source power supply. The retentive characteristics of the recorder are supplied from an arrangement of magnetic memory cores which controllably vary the time constants associated with given portions of the circuitry of a digital counter comprising the major subunit of this invention. In one embodiment the counter is supplied with a zero inhibit to prevent reverse counting past the zero output state, thus preventing a false indication of the recorded data. The circuitry thus described, basically comprising solid state elements, provides a compact, resilient unit with no mechanical moving elements.
ABSTRACT 5 Claims, 11 Drawing Figures POWER CONTROL CIRCUIT PULSE COUNTER COUNTER COUNTER COND. lst 2nd 3rd CIRCUIT w STAGE H STAGE STAGE ??BUFFER BINARY CODED DATA OUTPUT PATENTEB JAN 1 M975 UP RELAY I l l-o 0( SHEET 10F 5 PODWCER POWER CONTROL CIRCUIT PULSE COUNTER COUNTER NTER COND. IsI 2nd 3rd CIRCUIT STAGE STAGE STAGE Io 22 z e DOWN 3 RELAY BINARY CODED DATA OUTPUT %%\BUFFER FIG. I
INPUT FUNCTIONS 2| :1-1
LOGIC T- LOGIC JJ' INPUT INPUT FUNCTIONS PULSE TRIGGER RETENTIvE CIRCUIT f2? MASTER BUFFER SLAVE RESET FIG. 2
PATENTH] JAN] 41975 SHEET 2 0F 5 So E PATENTEBJAN l M975 SHEEI' u n:
42 RESETD SET FIG. 6
GROUND PATENTED 3.860.914
SHEET 5 BF 5 D C POWE R INPUT POWER CONTROL OUTPUT IOA VOLTAG E TIME FIO. IOA
DC POWER INPUT TIME IOB VOLTAGE POWER CONTROL OUTPUT TIME FIG. IOB
DIGITAL DATA RECORDER BACKGROUND OF THE INVENTION The present invention is related in general to digital data recorders and more particularly to such recorders that retain data information for later presentation upon loss of the recorder supply voltage.
In a nuclear reactor power plant, the plant power demand controls adjustment of the position of control rod banks within the reactor. An important safety feature in the operation of the plant is the ability to record the required control rod bank position and secure this information in the event of a loss of power to the recorder.
The rod position demand data is independent of the actual rod position data obtained from sensors positioned along the control rod housing and therefore it is specifically desirable to record the rod position demand data so that it can be compared with the actual rod position to insure that the rods are following their respective control signals. Additionally, in the event of a plant power supply failure, the apparatus assigned to record data will be required to maintain the information for presentation upon the resumption of power to enable the plant operator to determine the rod position demand at the time of failure.
At present, there are two designs of apparatus available for supplying the desired data recorder characteristics; both electromechanical. The first utilizes pulsed signals generated from relay contact closures to simulate discrete rod movement demands. The pulsed signals are communicated to a solenoid, which when energized, mechanically forces a stepping relay in the form of a wheel to be rotated one step. The wheel is numbered from zero to nine and is mechanically connected to other similarly numbered wheels geared in a manner to record and display a decimal output count. The decimal number is then representative of the position demand of the rod control bank. Since the stable output states of the counter are independent of the source power supply, the data can be maintained irrespective of a loss of plant power.
In the second arrangement pulsed signals generated from relay contact closures representative of discrete rod movement demands are communicated to a pulse conditioning circuit. This circuit modifies the pulses to drive an up-down counter formed basically from diode transistor logic gates. The outputs of the counter are then connected to relay drivers which energize latching reed relays. The digital output displayed by the latching reed relays is then proportional to the rod position demand and is retained in the state of the relays which can only be altered by further pulsing of the counter. In the manner, any loss of power to the system will not result in loss of the recorded information.
While both apparatus designs will theoretically function as desired they are subject to most of the deficiencies associated with electromechanical equipment in that they are bulky and subject to increased wear due to the adverse operating environment encountered within a reactor containment. The electromechanical characteristics exhibited by the prior art recorder designs thus render the output information suspect unless the equipment is continually checked.
SUMMARY OF THE INVENTION Briefly, this invention provides a compact, resilient,
electronic, digital data recorder having the inherent circuitry characteristics of re-establishing the recorded data state previously established just prior to a power failure upon the return of the source power supply. The retentive features of the recorder are supplied from a novel arrangement of magnetic memory cores which controllably vary the time constants associated with part of the recorder circuitry within the feedback loops of an incorporated digital counter. In one embodiment, the counter is supplied with a zero state inhibit to prevent counter stepping in reverse count past the zero data output state, which would otherwise exhibit itself as a false indication of the recorded data. The entire recorder unit, excluding memory cores, is amenable to solid state construction and is operable in extremely adverse environments, such as are found in nuclear reactor applications, to provide a reliable and durable output function.
BRIEF DESCRIPTION OF THE DRAWINGS For a better understanding of the invention, reference may be had to the preferred embodiment, exemplary of the invention, shown in the accompanying drawings, in which:
FIG. I is a block diagram of one embodiment of the data recorder of this invention;
FIG. 2 is a block diagram of a one bit stage of the counter identified in FIG. 1;
FIG. 3 is a schematic circuitry diagram of theone bit stage previously identified in FIG. 2;
FIG. 4 is a schematic circuitry diagram of the first two stages of the counter identified in FIG. 1;
FIG. 5 is a schematic circuitry diagram of the third stage of the counter previously identified in FIG. 1;
FIG. 6 is a simplified schematic circuitry diagram illustrative of the retentive characteristics of the circuits previously described in FIGS. 4, S and 6;
FIG. 7 is a graphical illustration of the hysteresis loop of the magnetic memory core utilized in the circuits of FIGS. 4, 5 and 6;
FIG. 8 is a schematic circuitry diagram of the pulse conditioning circuit previously identified in FIG. 1;
FIG. 9 is a schematic circuitry diagram of the power control circuit previously identified in FIG. I; and
FIGS. 10A and 10B show graphical illustrations of the input and output signals exhibited by the circuit of FIG. 9 in response to changes in the source supply voltage.
DESCRIPTION OF THE PREFERRED EMBODIMENT In order to overcome the disadvantages of the prior art this invention provides a compact, resilient, electronic, digital data recorder having the inherent circuitry characteristics of re-establishing the recorded data state previously established just prior to a power failure upon the return of the source power supply. The recorder, excluding a novel arrangement of saturable magnetic cores within an incorporated digital counter, is amenable to solid state construction and the entire unit is operable in extremely adverse environments, such as are found in nuclear reactor applications, to provide a reliable and durable output function.
The preferred embodiment of this invention will therefore be described within a reactor environment, to best illustrate its advantageous characteristics.
Most nuclear reactors utilize three types of neutron absorber rods respectively known as shut down rods, part length rods and control rods. With each type of rod there is associated a number of rod banks, and within each bank there is contained a number of rod mechanisms. An example of such a nuclear reactor rod system is more fully described in US. Pat. No. 3,654,607 to Andre Wavre et al entitled Signal Sequencing System," issued Apr. 4, 1972 and assigned to the assignee of the present invention. The shut down rods are operable to be either driven full out" when the reactor is at full power or full in" when the power is shut down. The control rods, on the other hand, are moved either up or down incrementally in banks, with all the rods within a given bank being moved to the same level simultaneously; the level depending upon the power demand of the plant. The part length rods act as verniers and are never entirely out of or in the reactive region, but are driven either up or down within a small range of steps. The full length control rods are moved up and down within pressurized housings by magnetic jack drive mechanism more fully described in US. Pat. No. 3,158,766, issued to E. Frisch and assigned to the assignee of the present invention. It takes approximately 232 pulsed steps to cover the total distance of control rod travel, which amounts to approximately a total distance of 144 inches. Other mechanisms generally exist for driving the part length rods to obtain their vernier control.
The pressure vessel containing the control rods makes it practically impossible for an operator to directly determine the actual position of the rods. A control rod drive shaft position sensor, normally located along the control rod housing is commonly provided for this purpose to sense the position of the control rod drive shafts which are affixedly coupled to the control rods. The position information thus obtained is then translated to actual control rod position. An example of such a monitoring device can be found in the application to .l. A. Neuner, F. T. Thompson and L. Vercellotti entitled Digital Multiplex Position Indication and Transmission System, Ser. No. 320,792, filed Jan. 3, 1973, (WE. 44,067). i
In order to assure the safe operation of the reactor plant it is desirable to compare the discrete signals employed to command control rod position with the signals received from the control rod monitoring equipment to assure that the command has been executed. Furthermore, it is desirable to know the last command issued by the reactor control equipment prior to a power failure in order to assure the safe and continued operation of the reactor.
For these purposes, the present invention provides a resilient, electronic, data recorder having the inherent circuitry characteristics of re-establishing the recorded data state previously exhibited just prior to a power failure upon the return of the source power supply. The entire recorder unit is designed to be operable in the extremely adverse environment of a reactor containment to provide reliable and durable output information which can be used for the comparisons required in nuclear applications. The digital data recorder thus contemplated is generally illustrated in FIG. 1 as basically comprising a pulse conditioning circuit 10, an updown retentive counter 12 and a power control circuit 14.
The pulse conditioning circuit 10 will receive either an up or down control rod drive pulse from the relay contact closure 16 or 18 respectively, simulating discrete rod position movement demands. The signals are processed by the pulse conditioning circuit 10 to a form compatible with the up-down retentive counter 12. Zero detect circuitry, forming a portion of the pulse conditioning unit 10, monitors the counters output, shown in binary coded form by reference character 20, and is responsive to the zero counter state to inhibit further reverse counting which could otherwise result in returning the counter to maximum count.
The up-down retentive counter 12 is basically constructed from high threshold integrated logic circuits in a toggle flip-flop configuration. A block diagram of a one bit circuitry section of the counter is shown in FIG. 2. The separate counter stages are synchronized by the up-down rod drive pulses emanating respectively from the relay outputs 16 and 18 shown as the input functions 21 in FIG. 2. The up and down pulses 21 are supplied to a series of logic circuits which process the information conveyed to a form compatible with the digital counter output desired. The output from the logic unit is then applied to a trigger circuit 25 which drives a multiple configuration of toggle flip-flops. The toggle flip-flops basically comprise two stages of RS flip-flops. The first stage 27, commonly known as the master, reacts to the falling edge of the input pulse, while the second stage 28 which forms the output circuit, commonly known as the slave, is responsive to the rising edge of the pulse to provide one bit of the digital output through an isolation buffer 30. A retentive circuit which provides the desired memory characteristics of this invention is formed as an integral part of the master section of each toggle flip-flop and employs the principle of either setting or resetting, upon power turn off or turn on, the RS flip-flops to their prepower failure state. The set-reset decision is dependent upon the magnetic flux density existing within an inductive core located within a feedback loop of the master section of the flip-flop.
A simplified circuitry schematic of a one bit section of the counter circuitry is illustrated in FIG. 3 with the counters three stages 22, 24 and 26 shown in more detail in FIGS. 4 and 5. FIG. 6 provides a simplified schematic of the actual retentive circuit of the master section of one portion of the counter with the corresponding hysteresis loop of the inductor being illustrated in FIG. 7. The figures in combination with the following description will provide a complete understanding of the operation of the circuits illustrated.
RETENTIVE DIGITAL COUNTER The circuitry associated with a one bit section of the counter is illustrated in FIG. 3 and forms the basic building block of the counter, which comprises a plural arrangement of toggle flip-flops. As is known in the art, a toggle flip-flop is constructed from an arrangement of two RS flip-flops; the first forming the master and the second forming the slave stage. The master stage is generally indicated in FIG. 3, by the NAND gates 42 and 44 and the slave stage is indicated by reference characters 34 and 38. The memory characteristics of the circuit are essentially supplied by the inductive core L and its corresponding time constant associated with the feedback loop of the master stage. In addition, a time constant is associated with the reference supply voltage input Vt provided at one input to NAND gate 44 and is dependent upon the value of the capacitor C, the resistor Re and the internal pull up resistor associated with the gate. NAND gates 32 and 36 form the triggering portion of the circuit previously identified in FIG. 2 by reference character 25 and function to assure that the master stage is responsive to the falling edge and the slave stage is responsive to the rising edge of the input pulse. NAND gate 40 appearing at the output serves to provide output isolation and acts as the buffer 30 previously noted in FIG. 2, to prevent current drain on the circuit.
FIG. 4 illustrates a four hit counter output stage and is employed to form the first and second stages 22 and 24 of the counter 12. It is readily apparent that the cir cuit illustrated in FIG. 3 supplies the basic building block for the individual counter output bit circuits 46, 48, 50 and 52, with the digital output bits 20 being identified by A, B, C and D, respectively. NAND gates 54 form the logic circuitry previously identified by reference characters 23 in FIG. 2, and function to condition the inputs so that circuitry of the counter can provide the digital outputs in the desired form. The diodes 56 OR the NAND gate outputs to the appropriate counter bit circuit inputs, while the NAND gates identified by reference character 58 monitor the output bits of the counter to identify the zero count state. The latter information is then conveyed to the pulse conditioning circuit at output terminal 60 for combination with the other counter monitoring outputs provided by the other two counter stages.
FIG. 5 illustrates the last stage of the counter previously identified by reference character 26 in FIG. 2, which provides a two bit output. Circuits circuits 62 and 64 function in the same manner as the circuits previously described by reference characters 46, 48, 50 with 52, and the appropriate counter output bits taken at terminals 66 and 68.
The operation of the retentive memory cell which forms a portion of the feedback loop of the master section of the toggle flip-flop can best be understood by reference to the simplified schematic provided in FIG. 6 and the corresponding graphical illustration shown in FIG. 7. The basic principle of operation is dependent upon a comparison to two time delays T and T corresponding to the time constants 1' and 7 associated with the toggle flip-flops. 'r, is a fixed time constant dependent upon the characteristic of the capacitor C, the resistor R and the internal pull up resistance associated with the reset input to NAND gate 42 in FIG. 6. 1', is a time constant which varies depending upon the last output state 0 of the flip-flop. As will be appreciated by those skilled in the art 1', is governed by the state of the magnetic core generally associated with the inductor L.
The saturable magnetic core L is desirably constructed from a magnetic material having magnetic characteristics which provide a substantially rectangular hysteresis loop, such as the one illustrated in FIG. 7. Therefore, the output 0 will assume either of two stable states, high or low. The following two examples which illustrate the two output states Q can assume, will completely describe the operation of the memory core in the event of a power failure.
Assume Q is high just prior to a power failure, rendering the output of gate 42 high and the output of gate 44 low. The current through the core will then have been flowing out of the dotted side of the magnetic core. Hence, the flux density within the magnetic core will correspond to the ordinate B,,, indicated in FIG. 7. When the power fails, this flux density will rise to B,, which is approximately 0.9 B,,.. When power is re-established, gate 44 will transition high due to the characteristics of the capacitor C, which initially acts as a short circuit. The output of gate 42 will also assume its high state inasmuch as the inductor L initially appears as an open circuit. Current will therefore begin to flow into the dotted side of the inductor L due to the higher potential appearing at the output of gate 44. The inductance of the core will then tend to resist this current flow since the flux density is initially at B,-. Thus, with the values of R and C appropriately chosen, 1-, will be much less than 1' and the voltage across C will increase much faster than the voltage across R Therefore, the output of gate 44 returns to its low state before forcing Q to its low state, thus latching Q in its high state.
On the other hand, assuming Q was low just prior to power failure, the outputs of gates 42 and 44 will be low and high respectively. Thus, current will flow into the dotted side of the magnetic core and its flux density will assume the value indicated by the ordinate +B After a power failure, the flux density of the core will drop to +8, which is approximately 0.9 B,,,. When the power is restored, the output of gates 42 and 44 will both intially transition to a high state as a result of the capacitance and inductance respectively exhibited by C and L. However, this time the inductance of the core will provide little resistance to the current flowing into the dotted portion of the coil L inasmuch as the flux density will retain a value approximately equal to the ordinate +B,. Therefore, 1' will be much less than 1' and Q will return to its low state before the output of gate 44 goes to its low state, thus latching the outputs of gates 42 and 44 in their prepower failure state.
Accordingly, the retentative characteristics provided by the magnetic core L, within the feedback loop of the master section of the toggle flip-flop, retains the output established for each bit of the counter just prior to a source power failure and re-establishes this prior output state upon the resumption of the source power sup- PULSE CONDITIONING CIRCUIT The pulse conditioning circuit previously identified by reference character 10 in FIG. 1 is shown in schematic circuitry form in FIG. 8. The up and down pulses, simulating the rod position demand data, are respectively received at terminals and 72. The resistor capacitive filter provided by resistors and capacitors 78 establish a sufficient time delay to avoid multiple counts which might otherwise occur due to contact bounce of the relays l6 and 18. When a sufficient time has elapsed for the capacitor 78, associated with the relay energized, to charge to a voltage value sufficent to raise the corresponding circuit node 82 to a higher value than node 84, then unijunction transistor 74 will conduct, sharply dropping the voltage level appearing at node 84 to a negligible value. This negligible voltage level will continue to appear until the respective input relay contact energized is opened, at which time unijunction transistor 74 will effectively open circuit, sharply raising the voltage level appearing at terminal 84 to its former value. Thus the output pulse appearing at terminal 84 will be characterized by sharp falling and rising edges and will be further modified by NAND gates 86 and 88 for distribution to the counter circuitry at terminal 90 and 92 through the interface logic gates 54.
The zero inhibit, as previously described, is responsive to the zero counter state to prevent further down rod command pulses from entering the counter. NAND gates 58 illustrated in FIG. 4 were described as the mechanism for monitoring the individual toggle flipflop circuits to identify when the zero state had been assumed by the circuits respective outputs. When the zero output state is identified the respective outputs 60 from the first two stages are then applied to the input terminal 60 of the pulse conditioning circuit illustrated in FIG. 8. In a similar manner, NANDgates 94 are used to monitor the last stage of the counter 12 to provide a corresponding output indicative of the zero state. The respective monitoring outputs from the three stages of the counter are then applied to the input of NAND gate 96 which inhibits NAND gate 88 upon the occurrence of the zero state, effectively disconnecting the output from further down stepping the counter.
Thus, the pulse conditioning circuit modifies the input data pulses to a form compatible with the counter and effects the zero inhibit to prevent counter stepping, in reverse count, past the zero data output state.
POWER CONTROL CIRCUIT The power control circuit illustrated in FIG 1 is schematically shown in FIG. 9 with a graphical illustration of the corresponding inputs and outputs provided in FIG. 10. The circuit is designed to force the supply voltage furnished to the retentive counter to be switched on and off much faster than the shortest time constant of the counter in order to assure the retentiveness of the information maintained therein. The input time constant of the circuit, dependent upon the component values of resistor 102 and capacitors 98 and 100, controls the rise time of the input power supply voltage applied between terminals 118 and ground. When the input supply voltage rises above the threshold voltage level governed by the diode drops appearing across transistors 104 and 108 and Zener 106, transistors 104 and 108 conduct, essentially grounding the collector of transistor 108, which, in turn, saturates transistor 110 and abruptly reverse biasesrtransistor 112. As transistor 110 abruptly transitions to its ON state, current is fed back to the base of transistor 104 forcing a sharp rising edge on the voltage displayed at output terminal 114. This output voltage is then applied to the supply voltage inputs of the recorder circuitry. FIG. A graphically illustrates the power on inputs and outputs of the power control circuit with a threshold voltage indicated by V The Zener and transistors are chosen to supply a threshold voltage sufficient to support the working operation of each of the elements of the data recorder so that when the power supply reaches the threshold voltage level, upon turn on, an almost instantaneous operating voltage appears at the supply inputs to the data recorder circuitry.
In a similar fashion, when the input supply voltage drops to a level sufficient to reverse bias transistor 104, the transistor transitions to its OFF state, turning off transistor 108. The collector of transistor 108 will then effectively rise to a value sufficient to turn off transistor 110 and turn on transistor 112. This essentially grounds removed from the base of transistor 104 which further reverse biases transistor 104 and provides a sharp falling edge as indicated by the graphs supplied in FIG. 108. Thus, a sharp falling edged pulse is provided at output terminal 114, almost instantaneously turning off power to the individual element units of the data rccorder. The transition point of the output 114 indicated by V in FIG. 10B is slightly different from the threshold level indicated in the graph shown in FIG. 10A due to the inherent hysteresis of the circuit. However, both threshold levels are chosen well within the working voltages of the individual element units of the recorder and a slight discrepancy in threshold level will not affect the operation of the circuit.
It is desirable to have the power control circuit supply the input supply voltage to each of the subunits of the recorder in order to assure minimum turn on and turn off times so as to prevent alteration of the data output state retained by the counter. The power control circuit is provided for this purpose.
Accordingly, the individual element units described are amenable to solid state construction and are arranged in the configuration illustrated in FIG. 1 to exhibit the desired characteristics, in extremely adverse environments such as are found in nuclear reactor applications, to provide a reliable and durable output function.
We claim as our invention:
1. A digital data recorder having an electronic counter responsive to a data input to provide a representative digital output exhibited by a plurality of bistable output bits, the counter including a plurality of toggle flip-flop output circuits corresponding to the plurality of output bits, each ofsaid toggle flip-flops including a master and a slave stage with the corresponding output bit being provided by the slave stage, the master stage including a variable reactance within a feedback loop in the master stage, the value of the variable reactance being dependent upon the stable state of the output bit to control and vary a time constant of the feedback loop in a manner to re-establish the output bit last exhibited by the slave stage prior to a failure of supply voltage power to the counter after the supply power has been resumed.
2. The digital data recorder of claim 1 wherein the master stage of said toggle flip-flop has an input from the supply voltage powering the counter and exhibits a separate fixed, predetermined, time constant independent of said variable reactance which controls the rise time of the supply voltage input and is substantially greater than the value of the time constant of the feedback loop controlled by the value of the variable reactance corresponding to one of the bistable states of the output bit of the slave stage and is substantially less than the value of the time constant of the feedback loop controlled by the value of the variable reactance corresponding to the other of the bistable states of the output bit of the slave stage.
3. The digital data recorder of claim 1 wherein said variable reactance comprises an inductor wound on a saturable magnetic core exhibiting rectangular hysteresis with the saturable state of the core dependent upon 5. The digital data recorder of claim 1 wherein said counter advances count in response to the data input in either a forward or reverse counting direction corresponding to the data input receiver, including means for inhibiting said counter from advancing count in reverse count past the counter bit output state representative of a zero data count.