US3870616A

US3870616A - Current controlled regulation of gas evolution in a solid polymer electrolyte electrolysis unit

Info

Publication number: US3870616A
Application number: US320165A
Authority: US
Inventors: Russell M Dempsey; Mary E Nolan; Conti Anthony B La; Robert A Torkildsen
Original assignee: General Electric Co
Current assignee: General Electric Co; Raytheon Technologies Corp
Priority date: 1973-01-02
Filing date: 1973-01-02
Publication date: 1975-03-11
Anticipated expiration: 1992-03-11

Abstract

A gas generator utilizing an electrolysis cell having a solid polymer electrolyte is described in which the gas output is controlled by controlling the current to the electrolysis unit. In a preferred embodiment a hydrogen containing compound such as water is electrolyzed to generate hydrogen and the rate of gas evolution is controlled by varying electrical current to the cell. The output gas pressure is sensed and used to regulate the current flow to the cell to control the rate of gas evolution. In one instance a form of time-ratio current control is utilized to regulate the current whenever the outlet gas pressure exceeds a preset level. Alternately, a strain gage type of pressure sensor is utilized to produce a continuous output signal proportional to the output gas pressure and the signal is fed back to control the cell current with pressure variations.

Description

United States Patent [1 1 Dempsey et a1.

1 1 CURRENT CONTROLLED REGULATION OF GAS EVOLUTION IN A SOLID POLYMER ELECTROLYTE ELECTROLYSIS UNIT [75] Inventors: Russell M. Dempsey, Hamilton;

Mary E. Nolan, Marblehead; Anthony B. La Conti, Lynnfield; Robert A. Torkildsen, Danvers, all of Mass.

[73] Assignee: General Electric Company,

Wilmington, Mass.

{22] Filed: Jan. 2, 1973 [21] Appl. No.: 320,165

[52] U.S. Cl. 204/230, 204/266 [51] Int. Cl B0lk 3/00, B0lk 3/10 [58] Field of Search 204/230, 263-66, 204/128-129 [56] References Cited UNITED STATES PATENTS 3,336,215 8/1967 Hagen 204/230 3.485.742 12/1969 Emich et al 204/230 3,489.670 1/1970 Maget 204/129 3600228 8/1971 Conti 204/129 X CONTROL NETWORK CELL CURRENT 1 Mar. 11, 1975 3,616,436 10/1971 Haas 1. 204/230 X 3,635,804 l/1972 Gritzncr et al 1 1. 204/128 3,725,236 4/1973 Johnson, Jr 204/129 X 3,755,128 8/1973 Herwig 204/230 Primary Examiner-John H. Mack Assistant ExaminerD. R. Valentine [57] ABSTRACT A gas generator utilizing an electrolysis cell having a solid polymer electrolyte is described in which the gas output is controlled by controlling the current to the electrolysis unit. In a preferred embodiment a hydrogen containing compound such as water is electrolyzed to generate hydrogen and the rate of gas evolution is controlled by varying electrical current to the cell. The output gas pressure is sensed and used to regulate the current flow to the cell to control the rate of gas evolution. In one instance a form of time-ratio current control is. utilized to regulate the current whenever the outlet gas pressure exceeds a preset level. Alternately, a strain gage type of pressure sensor is utilized to produce a continuous output signal proportional to the output gas pressure and the signal is fed back to control the cell current with pressure variations.

4 Claims, 4 Drawing Figures PATENTEU 1 I sum 3 053 CURRENT CONTROLLED REGULATION OF GAS EVOLUTION IN A SOLID POLYMER ELECTROLYTE ELECTROLYSIS UNIT Ths instant invention relates to a method and apparatus for controllably generating gases in an electrochemical electrolysis cell, and more particularly, to a gas generator in which evolution of the gas is controlled by selectively varying the current flow to the electrolysis unit.

Generating gas by electrolyzing a chemical compound into its constituent elements, one of which may be a gas, is, of course, an old and well-known technique. One recently developed form of such gas evolving electrolysis unit involves the use of a cell which utilizes an electrolyte in the form of a solid polymer ionexchange-membrane. In a cell of this sort, an ionexchange-membrane such as a sulfonated perfluorocarbon membrane has a pair of electrodes of a suitable catalyst positioned on opposite sides thereof. Through an oxidation reaction the ionic form of one of the constitutents elements (hydrogen ions, for example, when H O is electrolyzed) is produced at one electrode. The ion is transported across the ion-exchange-membrane to the other electrode where a reduction reaction takes place that the positive hydrogen ion gains electrons to.

produce hydrogen molecules. The solid polymer ionexchange-membrane electrolyte electrolysis unit is particularly advantageous because it is efficient, small in size, and does not require any corrosive electrolytes. Gas generators employing such an electrolysis unit may thus be used in many new applications which formerly required the use of stored gas and makes possible the manufacture of a small, compact assembly for producing gas.

It is highly desirable in gas generators of this type to be able to control the gas flow from the generator preferably by controlling the rate at which the gas is evolved at the cell electrodes. Such control has obvious advantages in that the rate of gas evoluion may be increased when gas usage is high and correspondingly lowered when the usage is low. Furhennore, control of the gas evolution rate may be utilized to prevent buildup of excess gas pressure in the unit while leaving the generator in an operationally ready state thereby enhancing the usefulness of the device. It has been found that all of these desirable objectives may be attained by sensing an operating parameter such as the outlet gas pressure and varying the current flow to the electrolysis cell in response thereto, thereby controlling evolution of gas at the electrodes.

It is therefore a primary objective of this invention to provide a gas generating solid polymer electrolyte electrolysis unit in which the flow of gas from the unit may be automatically controlled by controlling evolution of the gas at the cell electrodes.

A further objective of the invention is to provide a solid polymer electrolyte electrolysis unit in which the flow of gas from the unit is electrically controlled to control the rate of gas evolution at the electrodes of the unit.

Still another objective of the invention is to provide a gas generator having a solid polymer electrolyte elec- The various advantages and objectives of the invention are achieved in a gas generating electrolysis cell which utilizes a solid polymer ion-exchange-membrane electrolyte to evolve the gas. The output from the cell is controlled by sensing the output gas pressure and varying the current flow to the cell to control the rate of gas evolution. The output from the gas generator may be thus controlled either continually or in an intermittent or time-ratio manner to control both the rate of flow or to control the maximum pressure buildup in the generator. In a preferred embodiment, the cell current is controlled by. a pair of silicon controlled rectifiers (SCRs) which are alternately gates by a variable pulse rate pulse generator. A comparator and error signal circuit is provided to produce a control signal proportional to a reference signal and a gas pressure signal to vary the repetition frequency of the output pulses from the pulse generator to control the firing angle of the SCRs and thus the current flowing to the electrolysis cell. Any variation in the output pressure, or alternately, the exceeding of a preset output pressure, results in a control signal from the comparator and error signal network which varies the repetition rate of the pulse generator so as to modify the current flow in the cell and thereby control the rate of gas evolution at the cell electrodes.

The novel features which are believed to be characteristic of the invention are set forth with particularity in the appended claims. The invention itself, however, both as to its organization and method of operation, as well as additional objectives and advantages thereof, will best be understood from the following description when taken in connection with the accompanying drawings in which:

FIG. I is a schematic diagram of a gas generator in which the rate of gas evolution is electrically controlled in response to the output pressure of the cell.

FIG. 2 is a schematic ofa portion of the solid polymer electrode ionexchangemembrane which is useful in understanding its mode of operation.

FIG. 3 is a circuit diagram of the pressure responsive, current control network forming part of the system of FIG. 1.

FIG. 4 is a partial illustration of a modified form of the control network of FIG. 3.

FIG. 1 shows the gas generator of the instant invention which includes means for controlling the rate of gas evolution at the electrodes of a solid polymer electrolyte electrolysis unit in response to the output gas pressure. The gas generator includes an electrolysis cell assembly shown generally at 10 of the solid polymer ion-exchange-membrane electrolyte type in which a hydrogen containing compound such as water or hydrogen chloride, for example, is dissociated electrochemically to generate the desired gas such as hydrogen, chlorine, oxygen, etc. In the following description, the invention will be discussed in connection with a hydrogen generator in which H O is dissociated to produce hydrogen as well as oxygen. It will be apparent, however, that the invention is broadly applicable to all gas generators of the solid polymer electrolysis type including those capable of dissociating other hydrogen containing compounds such as hydrogen chloride or, for that matter, any dissociable chemical compound. Cell assembly 10 includes a housing 11 which is separated into anode and

cathode chambers

12 and 13 by a solid polymer ion-exchange-membrane electrolyte 14. The solid polymer electrolyte ion-exchange member may, for example, be a thin, mils or so, sulfonated perfluorocarbon membrane of the type manufactured and sold by the Dupont Co. under their trade designation Nation and which is characterized by the fact that positive ions are transported across the membrane. Positioned on either side of the ion-exchangemembrane are electrodes Y and 16 which are energized from current control network 17. Electrodes l5 and 16 include both a platinized titanium current conducting screen pressed against a suitable catalyst which adheres to the membrane for enhancing dissociation of the water. One suitable catalyst may be a mixture of Platinum (Pt) and Iridium (Ir) in the proportions of 50 percent and 50 percent by weight for the anode and platinum for the cathode. Anode chamber 12 communicates through conduit and water inlet solenoid valve 18 with main water supply tank 19. The water is dissociated at anode electrode 15 into positive hydrogen ions plus oxygen. The positive hydrogen ions move across ion-exchange-membrane 14 and are converted into molecular hydrogen at cathode electrode 16. Associated with the cathode chamber and communicating therewith, is a hydrogen/H O accumulator chamber 20 which acts as a storage reservoir for the hydrogen evolved at the cathode electrode, as well as that water which is pumped across the ion-exchange-membrane along with the hydrogen ions, i.e., each hydrogen ion carries several molecules of water with it as it moves across membrane 14 to the cathode. Positioned in chamber 20 is a water level float 21 which actuates a suitable switch such as a reed switch, for example, when the water reaches a predetermined level. This actuates water inlet solenoid valve 18 to shut off the water supply to the anode chamber.

The hydrogen passes from accumulator chamber 20 over conduit 22 to a dessicant chamber 23 where any moisture in the gas is removed. The gas then passes through a pressure regulator 24 to the exterior of the unit. Mounted between the pressure regulator and the outlet valve is a pressure gage 25 to indicate the gas gage pressure. A pressure sensing element 26 which is shown in FIG. 1 as being a pressure switch is coupled to outlet conduit 22 between the dessicant chamber and the pressure regulator, and senses the outlet hydrogen pressure in conduit 22. Pressure sensing element 26 is connected to cell current control network 17 to vary the current supplied to the electrolysis cell and thus, the rate of hydrogen evolution at the electrodes.

The water dissociated in anode chamber 12 to produce the hydrogen ions also produces molecular oxygen which is retained in the chamber. An outlet conduit 30' is connected to the chamber and transports the liberated oxygen and any water vapor which may be contained therein to an oxygen-water separator 31 in which the entrained water vapor is removed from the oxygen and returned to main water supply tank 19 whereas the oxygen is vented to the air or stored in a suitable container.

As will be described in detail subsequently in connection with the structure of FIG. 4, the instant invention is not limited to using pressure switch which produces control of the cell current on an on-and-off basis, thus effectively producing a time-ratio modulation of the cell current. A pressure sensing element which produces a continuously varying electrical signal responsive to the pressure variations may be used with equal facility. One type of such a continuously varying pressure sensing device which is illustrated in FIG. 4 is a pressure strain gage which produces an electrical analog output signal proportional to the pressure and which is continuously variable over a given pressure range.

The main water supply tank 19, as pointed out p eviously, supplies water to the anode electrode for dissociation. However, not all of the water which is supplied to the anode electrode is dissociated. In fact, the bulk of the water supplied to the anode electrode is transported across the ion-exchangemembrane into the cathode chamber. Part of this water returns to the anode chamber by diffusion across the ion-exchangemembrane, however the rate of protonic pumping by the hydrogen ions is much greater than the diffusion rate so that eventually a buildup of water takes place and accumulator chamber 20 is provided for this purpose. It will also be apparent that water level float 21 which controls water inlet solenoid valve 18 is provided to cut off the water supply from the main supply from the main supply tank whenever the water in the accumulator chamber rises above a predetermined level. When this occurs, solenoid valve 18 is closed shutting off the water from the main supply tank. The cell operates on the water diffusing from the cathode chamber back across the ionexchange-membrane to the anode electrode until the water level in the accumulator chamber has been reduced below the predetermined level. Float 21 then deenergizes solenoid valve 18 opening the valve so that the main water supply chamber again supplies water to the anoid. Thus, tank 19 normally supplies the water required for the generation of hydrogen but from time to time the water collected in accumulator chamber 20 is so utilized thereby minimizing the need for shutting the system down and emptying the accumulator chamber.

FIG. 2 is a partial schematic of a section of the ionexchange-member electrode of cell 10 and is useful in understanding the action taking place at the electrodes and across the ion-exchange-membrane by means of which the hydrogen is evolved. Membrane 14 which is a sulfonated perfluorocarbon has an anode electrode 15 attached to its surface. The electrode consists of a current conducting element 40 which is shown to be of a mesh or screen like construction which is pressed against a catalyst 41. Similarly, on the cathode side of the membrane, the electrode 16 consists of a current conducting mesh 42 pressed against a catalyst 43. The

current conducting electrodes

40 and 43 are connected through suitable leads, not shown, to a source of D-C voltage which estabilshes a suitable potential at the electrodes to dissociate the hydrogen containing compound and results in the evolution of the gases at the respective electrodes. Thus, as shown in FIG. 2, an oxi- .dation reaction takes place at the anode electrode 4H 4e 2H (Reduction-gain of electrons) In summary, water is dissociated at the anode to produce molecular oxygen and hydrogen ions. The hydrogen ions are transported across the ion-exchangemembrane to the cathode where a reduction reaction takes place and the hydrogen is converted into molecular hydrogen.

In addition to the transport of hydrogen ions across the ion-exchange-membrane, each hydrogen ion in moving across the membrane transports seven molecules of water from the anode to the cathode. It is this transport of water by the hydrogen ions, i.e., protonic pumping of the water, which results in the accumulation' of water at the cathode and dictates the need for an accumulator chamber to hold the water. While water is pumped protonically from the anode to the cathode by the hydrogen ions, the accumulation of the water in the cathode chamber results in a pressure gradient which produces diffusion of water back across the ion-exchange-membrane from the cathode to the anode, as indicated by the arrow D shown in the lower portion of the ion-exchange-membrane. However, the rate of protonic pumping to the cathode side is substantially greater than the rate of diffusion to the anode side so that overall there is an accumulation of water on the cathode side of the cell. Consequently, an accumulator must be provided as well as a means for intermittently stopping the supply of water to the anode when the water accumulation at the cathode side exceeds a predetermined level.

As pointed out previously, the dissociation of water ions involves the loss of electrons on the anode side, the gain of electrons on the cathode side and the transport of positive hydrogen ions across the membrane. As a result, current flow in the external circuit and the magnitude of this current flow controls the evolution of hydrogen at the cathode with the rate of evolution being proportional to the amount of current flowing in the external circuit. By controlling the current flow to the electrochemical cell as a function of the output hydrogen pressure, the rate of evolution of the hydrogen gas may be controlled thereby providing a ready means for controlling both the flow rate of the hydrogen from the generator, as well as controlling the pressure levels in the system.

In the arrangement of FIG. I, the output gas pressure from the generator is sensed by a pressure switch 26 which is actuated to produce an electrical control signal whenever the output pressure exceeds a given level. The electrical signal from the pressure switch is then applied to cell current control network 17 to control the current to the electrolysis cell assembly to vary the current level by a time ratio mode thereby varying the rate of gas evolution at the electrodes. FIG. 2 illustrates a preferred embodiment of the circuitry for controlling the current to the electrolysis cell in response to the gasoutlet pressure. The network of FIG. 3 contains 4 major elements including a cell current source 50 which includes a pair of Silicon Controlled Rectifiers (SCRs) connected-in push-pull for supplying current to the electrolysis cell, a variable PRF trigger pulse generator 51 for controlling the firing angle of the SCRs, a comparator and error signal generating network 52 for producing a control signal which controls the repetition frequency of the triggering pulses from pulse generator 5], and a power supply circuit 53 for supplying an A-C supply voltage to the anodes of the SCRs and positive and negative unidirectional supply voltages to the remaining circuit components.

The cell current source 50 includes a pair of alternately conducting silicon controlled rectifiers 54 and 55 connected in a push-pull configuration to supply current to the electrolysis cell shown schematically at 10. Anode voltage for SCRs 54 and 55 is supplied via leads 56 from the power supply module 53. The anode voltage for SCRs 54 and .55 is an A-C voltage so that the SCR anode-cathode voltages are positive during opposite alternations of the A-C supply voltage. Consequently, SCRs 54 and 55 conduct alternately during alternate half-cycles of the supply voltage. The cathodes of SCRs 54 and 55 are connected together and to the positive terminal of electrolysis cell 10 so that conduction of each of the SCRs produces current flow through the cell, the magnitude of which is controlled by the firing angle of the SCRs. The gate electrodes 57 of each of the SCRs are transformer coupled to pulse generator 51 with the gates connected respectivelyto secondary windings 58 and 59 of a pulse transformer 60, the primary of which is connected to the output of trigger pulse generator 51. Secondary windings 58 and 59 are so wound, as indicated by the dots adjacent their junction, that the pulses from pulse generator 51 are applied in phase to the gate electrodes so that the positive pulses are applied to both gate electrodes simultaneously. However, since only one of the SCRs has a positive voltage at its anode during any given half-cycle of the supply voltage only one of the SCRs can be triggered at a time.

The output of the trigger pulse generator 51 controls the firing and thus the phase angle of conduction of the SCRs and thereby the current flow through the cell. For example, during the half-cycle when the anode of SCR 54 is positive and that of 55 is negative, the application of a trigger pulse to the gate electrode of SCR 54 causes that SCR to conduct whereas the application of the same pulse to the gate electrode of SCR 55 will have no effect since the anode-cathode path is reverse biased. The average current flowing during each SCR conducting period depends on the point in time during the positive anode voltage cycle that the SCR is triggered into conduction. That is, the conduction phase angle, and hence, the average current depends on the firing point. If the SCR is triggered early in the positive anode voltage cycle, as close to 0 as possible, the average current is high. If the SCR is triggered late in the positive half-cycle of the cycle, i.e., closer to the 180 point when anode voltage goes negative again and the SCR stops conducting, the average current is low. The value of the average current thus may be varied from a very' high value to zero current by varying the conduction phase angle between 0 and 180. The conduction phase angle in turn, depends on how early or late in the anode voltage cycle the SCR is triggered. It will also be obvious that the greater the repetition frequency of the triggering pulse, the earlier in the cycle the SCRs will fire and hence, the greater the average current. Conversely, as the pulse repetition frequency is reduced. the SCRs fire later and later in the cycle thereby reducing the average current flowing through the cell. Thus, by controlling the repetition frequency .of the pulses from pulse generator 51, the firing angle of the SCRs may be varied and the current level correspondingly controlled.

Trigger pulse generator 51 is a simple relaxation oscillator including a programmable unijunction transistor and a R-C timing network which includes a variable resistance controlled by an error signal from comparator 52 to vary the pulse repetition frequency. Thus, trigger pulse generator 51 includes a programmable unijunction transistor 63 having a gate electrode 64, a cathode 65 and an anode 66'. Cathode 65 is connected to the A bus by resistor 67 and gate electrode 64 is connected through

voltage divider resistors

68 and 69 to the A- and A+ busses of the power supply to establish the firing voltage for unijunction 63.

An R-C timing network controls thevoltage level at anode 66 and hence, the rate at which the unijunction transistor is driven into conduction. The timing network includes a capacitor 71 connected between the anode and the negative A- supply bus. A fixed resistor 72 is connected in series with the emitter-collector path of transistor 73 which functions as a variable resistance between capacitor 71 and the positive A+ supply bus. Storage capacitor 71 charges through resistor 72 and the emitter collector path of transistor 73 toward the positive voltage at the regulated A+ bus. When the voltage at anode 66 becomes sufficiently positive to forward bias the anodegate junction, unijunction transistor 63 conducts, rapidly discharging capacitor 71. This rapid discharge produces current flow through cathode resistor 67 and a short positive pulse is generated each time the unijunction transistor conducts. This pulse is coupled to the base of NPN transistor amplifier 74. Transistor 74 has its collector connected through a collector resistor 75 and a diode 76 to the A+ bus and its emitter directly to the A bus. The output from the transistor 74 is coupled to primary winding 61 of pulse transformer 60 and thence, through secondary windings 58 and 59 to the gating electrodes of SCRs 54 and 55. A capacitor is connected between the junction of resistor 75 and diode 76 and charges to the positive A+ supply. The voltage on capacitor 77 maintains the voltage across transistor 74 even when the A-lsupply voltage which is a clipped. full rectified sine wave goes to zero twice during each cycle of the rectified A-C supply. Diode 76 is provided to prevent discharge of capacitor 77 when the voltage on the A+ bus goes to zero since it is poled to block current flow when the A+ bus is a zero volt and capacitor 77 is charged to a positive voltage.

When capacitor 71 discharges the voltage drops and the anode-gate junction is again reverse biased terminating conduction so that capacitor 71 again begins to charge towards the voltage at the A+ bus. The rate at which capacitor 71 charges, i.e., the R-C time constant of the R-C network, and hence, the repetition rate of the output pulses from this relaxation oscillator is controlled by the resistance of the emitter-collector path of the PNP transistors 73. Transistor 73 thus functions as a variable resistor which controls the time constant of the network and hence, the repetition rate of the out- .put pulses from the oscillator. The resistance of transispulse rate. The signal from comparator 52 changes the voltage at thebase electrode to control the resistance of the emitter and collector path of transistor 73. Thus, the output or error signal from comparator 52 controls the pulse rate of the pulses from trigger pulse generator 51 and hence, the firing angle of the SCRs and the current level supplied to electrolysis cell 10.

Comparator and error signal generator 52 includes a summing amplifier 81 which integrates the various input signals to the amplifier and includes a pair of

input terminals

82 and 83. Terminal 82 is connected to a current setting potentiometer 84 consisting of

resistors

85 and 86 shunted by zener diode 87 which maintains the voltage across the potentiometer constant. A movable slider 88 is positioned along the lower resistor 86 and couples a reference signal which establishes the reference current level to the summing amplifier. In the absence of any other input signals, the output from the summing amplifier as established by the setting of slider 88 controls the resistance of the emitter collector path of transistor 73 to establish a pulse rate from pulse generator 51 which results in a given current level from the SCRs. Input terminal 83 of summing amplifier 81 receives a feedback signal from the cell which is proportional to the actual current flowing through the cell. To this end, a current shunt resistor 90 is connected between one terminal of cell 10 and ground. The voltage drop across resistor 90 which is proportional to the current flowing in the cell is applied over lead 91 to input terminal 83 and is thus compared withv the reference signal from potentiometer slider 88. The feedback signal from cell 10 corrects for current variations due to variations in line voltage, temperature, cell conditions, etc. That is, if due to line voltage variations or conditions in the cell itself, the actual current flowing through the cell differs from the predetermined current level established by the position of slider 88 on the current setting potentiometer 84, a signal is fed back of magnitude and polarity to produce an output from the summing amplifier which varies the pulse repetition frequency rate sufficiently to bring the current to the desired level.

Also connected to input terminal 82 is an electrical signal responsive to the outlet gas pressure for modifying the current flow through the cell. To this end, single pole double throw switch element 91 which is controlled by the pressure sensor is connected between the B supply bus and input terminal 82. Switch 91 has a movable armature 92 which is normally maintained in the open position so that switch 91 has no effect on summing amplifier 81 and the flow of current through cell 10. In the particular arrangement illustrated in FIG. 1, when the outlet gas pressure reaches a predetermined level, pressure switch 26 is actuated thereby moving armature 92 of switch 91 to connect the B- voltage bus to input terminal 82 through a current limiting resistor. As a result, the output from summing amplifier 81 and from comparator 52 goes heavily positive. With the signal from summing amplifier 81 going more positive, PNP transistor 73 becomes less conductive and the resistance of the emitter-collector path rises thereby increasing the time constant of the R-C timing network associated with unijunctions transistor 63. The repetition rate of the pulses from generator 51 is correspondingly reduced to a very low value and, in fact, the pulse rate drops below the SCR supply voltage frequency. Consequently, SCRs 54 and 55 are not triggered at all during each half-cycle of the supply voltage and the current supplied to cell 10 goes to zero. When the current goes to zero, gas evolution at the cell electrodes ceases and the outlet pressure begins to drop. When the outlet pressure again drops below the critical value, pressure switch 26 is deactivated moving armature 92 of the switch 91 into the open position and removing the B- voltage from input 82. The output from the comparator circuit becomes less positive which decreases the resistance of the emitter collector path of transistor 73. The time constant of the timing network is reduced and the repetition rate of the pulse generator is increased sufficiently so that SCRs 54 and 55 are again triggered and conduct current. This sequence, with the SCRs being disabled for the entire cycle of the supply voltage, continues until the outlet gas pressure of the generator reaches and stabilizes at a voltage below the critical pressure level. Thus, in effect, a time ratio control of the cell current is produced by enabling and disabling the SCRs until the average current level is reduced sufficiently to reduce the rate of gas evolution to maintain the output pressure below the critical level.

The power supply module 53, which supplies both the A- Csupply voltage for SCRs 54 and 55 as well as the positive and negative unidirectional supply voltages for the remaining circuitry, includes an iron core transformer 95 having a primary winding 96 connected to an A-C supply and a center tapped secondary winding 97. The center tap of secondary winding 97 is grounded so that two secondaries produce alternating output voltages which are 180 out of phase. The anodes of SCRs 54 and 55 are connected via the leads 56 to intermediate taps on opposite sides of the grounded center so that positive anode voltages are supplied to SCRs 54 and 55 on alternate half-cycles of the supply voltage. Positive and negative unidirectional supply voltages for the B+, B, A+ and A- busses are provided by full wave rectifying circuits connected to winding 97. To this end, a pair of

diodes

98 and 99 are respectively connected to the opposite ends of winding 97 and are so poled as to conduct during positive alternations of the supply voltage to produce a positive rectified voltage on lead 103. Diodes 100 and 101, on the other hand, are so poled as to conduct during the negative alternations. Consequently, these rectifier pairs supply positive and negative voltages for the B+ and B- supply busses over leads 103 and 104. Connected between the supply busses and ground are the

filter capacitors

105 and 106 which conduct any alternating or ripple cur rent components in the rectified voltage to ground. A pair of

zener diodes

107 and 108 are connected respec' tively to the 3+ and B supply bus to regulate the rectifled D-C voltage which is then filtered by

capacitors

105 and 106. The zener diodes function, in a manner well known to those skilled in the art, to regulate the voltage and maintain it at a given level. Thus, as the rectified positive voltage on the B+ bus exceed the level established by the particular zener, the zeners begin to conduct clipping the rectified voltage. Simi larly, zener 108 is so poled as to conduct if the negative voltage on the B- bus exceeds a predetermined level to clip the negative rectified voltage. The A+ and A busses of the trigger pulse generator are connected to the 5+ and B- busses respectively by

resistors

70 and 109 and the voltage at the A-- and A+ busses is maintained at a different and lower level by means of zener diodes 110 and 111. The voltages at the A+ and A busses have the wave form illustrated by Curve 112 since there are no filter capacitors provided at these busses. The A+ and A- voltages therefore, go to zero twice during each supply voltage cycle in order, as will be explained in detail later, to reset the pulse generator each time the supply voltage to the SCRs goes to zero.

Thus, in the arrangement shown schematically in FIG. 1 and the current control circuitry shown in detail in FIG. 3, the current level in the cell is set by means of a potentiometer from zero to a maximum current with the signal from the potentiometer being summed with a feedback signal from the cell and an external signal responsive to outlet gas pressure. All these signals are integrated in a summing amplifier to produce an increasingor decreasing current output in response to an error input, thus resulting in a very closely controlled output current. Current for the cell is derived from a transformer secondary and controlled by two phase controlled SCRs which produce a variable full wave rectified current output. Control of this current and triggering of the SCRs is accomplished by a unijunction oscillator which is transformer coupled to the SCRs. The phase variation of the firing of the rectifiers is accomplished by varying the pulse repetition frequency according to the error signal form the summing amplifier which error signal is in'turn, controlled in response to a pressure sensitive element from the gas outlet section of the generator and in response to the actual current flow in the generator. Thus, a time ratio servo control system is provided which closely controls the current flow through the cell to control the rate of gas evolution at the cell and in turn, the outlet pressure and gas flow from the generator.

It will also be apparent that since the supply voltages on the A+, A busses go to zero twice during each cycle, as shown by Curve 112, the voltage at gate electrode 64 of unijunction 63 also goes to zero. With the bias voltage at zero, the rectifying junction at emitter 66 is no longer reverse biased and the resistance at that junction is very low. This permits capacitor 71 to discharge very rapidly through this junction and its voltage goes to zero. In this fashion, the unijunction relaxation oscillator is reset each time the supply voltage goes to zero. Since A-C supply voltage also goes to zero at this time, it can be seen that the oscillator is reset each time current conduction is switched from one SCR to the other.

In the control circuitry of FIG. 3, an arrangement is illustrated in which the rate of gas evolution at the cell is varied by controlling the current through the cell in response to a pressure sensing element which actuates the control circuitry whenever the gas pressure exceeds a predetermined level and then cycles the system until the outlet pressure stabilizes at a lower pressure level. The invention, however, is not limited to a time-ratio or an on-an-off system which is actuated only if the pressure exceeds a predetermined value. The system may also function so that the outlet pressure is continually sensed and compared with the reference value es tablished by the current setting potentiometer to control the current through the cell continuously in response to pressure variations. FIG. 4 illustrates such an arrangement in which a pressure transducer is utilized to sense the outlet gas pressure to produce a control signal which varies continually with pressure. To this end, a pressure transducer having a plurality of strain gages connected as the four active armsof a Wheatstone Bridge may be utilized as'the pressure sensing element. In FIG. 4 a pressure transducer [which may for example, be a pressure transducer of the kind sold by the CBC/Transducer Div. of the Bell & Howell Co. located at Munrovia, Calif. under its designation-Series 4-236] is connected to the outlet conduit to sense the outlet pressure. The outlet pressure varies the resistance of four-strain gages, 113 connected as the arms of a Wheatstone Bridge as a function of pressure. A source of D-C potential is connected across one diagonal of the bridge and a voltage proportional to gas pressure is produced across the other diagonal of the bridge. The output voltage from the Wheatstone Bridge is connected by a pair of cables to an amplifier 114 where it is amplified and applied as one of the inputs to an input terminal 115 of a summing amplifier 181. The other input to the amplifier is a reference voltage from a current setting potentiometer similar to that described in connection with FIG. 3. That is, the current setting potentiometer includes a pair of series connected

resistors

116 and 117 which are connected be tween the B+ terminal and ground potential. A zener diode 118 is connected across the two resistors to maintain the voltage across the potentiometer constant. A movable slider 119 is connected along the lower resistor 115 and establishes the reference voltage from the current setting potentiometer. This reference voltage is applied to input 118 of the summing amplifier. The output of summing amplifier 81 thus produces an error signal which controls the repetition rate of the trigger pulse generator which controls the firing angle of the current supplying SCRs connected to the electrolysis cell.

In the arrangement illustrated in FIG. 4, the current level established initially by the position of the slider 119 establishes the current level in the cell and the rate of gas evolution at the electrodes. The evolution of gas is established by reference signal from the current setting potentiometer will under normal conditions establish an output pressure level which is sensed by the strain gage transducer and fed back to the input of the summing amplifier. As the gas pressure varies above or below the value desired, due to usage or other variables the feedback signal from the strain gage pressure transducer varies thereby varying the output from the summing amplifier and controlling the pulse repetition rate of the pulse generator which controls the firing angle or phase angle of the current supplying SCRs. By controlling the firing angle of the SCRs, the amount of current through the cell is controlled as is the rate of evolution of the gas and hence, the outlet pressure from the generator. The strain gage pressure transducer thus provides a continuous signal which varies with outlet pressure to maintain the current through the cell at a level such as to maintain the rate of evolution of the gas at a level adequate to maintain the desired gas pressure as conditions such as use, temperature, etc., vary to cause changes in fluctuations in gas pressure.

It will be appreciated that a very simple and effective arrangement for controlling the outlet pressure and the flow rates of a gas generator have been described in which all these desirable end results are achieved by sensing the output pressure and controlling the current flow to the electrolysis cell as a function of the pressure. Control of the current in turn, controls the rate of evolution of the gas at the cell electrodes thereby regulating boththe pressure and the rate of gas flow from the cell in a simple and effective manner. I

Although a particular embodiment of this invention has been shown, it will, of course, be understood that the invention is not limited thereto since many modifications both as to the arrangement and the circuitry utilized therein may be made. It is contemplated by the appended claims to cover any such modifications as may fall within the true spirit and scope of this invention.

What is claimed as new and desired to be secured by Letters Patent of the United States is:

1. In a gas generator having a controllable gas output the combination comprising:

a. an electrolysis cell having:

1. a solid polymer, ion-exchange electrolyte membrane,

2. catalytic electrodes positioned adjacent to opposite surfaces of said membrane to dissociate a chemical compound to evolve gas,

b. a source of electrical power coupled to the electrodes of said cell,

0. means to furnish a chemical compound to one of said electrodes for dissociation, the ionic form of one of the elements of the dissociated compound being transported across said ion-exchange membrane to evolve the gas,

(1. means for sensing the output gas pressure and producing a control signal in response thereto;

e. means responsive to the control signal responsive to the sensed gas pressure for controlling the current flow from said source of electrical power on a time ratio basis to vary the current level to the cell electrodes thereby to control the rate of gas evolution including:

1. a current source including a pair of alternately conducting solid state switching devices for supplying the cell current,

a voltage responsive variable repetition frequency, pulse generator coupled to said switching devices having a voltage responsive timing network for controlling the repetition pulse frequency, said pulse generator producing triggering pulses for said switches to control the average current level to said cell,

3. means coupling the signal from said gas pressure sensing means to said pulse generator timing network to vary the pulse repetition rate of the pulses from said generator in response to said signal to vary the cell current as a function of the outlet gas pressure.

2. In a gas generator according to claim 1 wherein said pulse generator comprises a relaxation oscillator including a solid state switching element, a timing network coupled to said switching including a storage element and an electrically controlled variable resistance element for coupling the signal from said pressure responsive means to said variable resistance element to vary the resistance of said resistance element and thereby the time constant of said timing network to vary the output pulse repetition frequency from said relaxation oscillator in response to the gas pressure.

3. The gas generator according to claim 2 wherein said solid state switching device is a unijunction transistor and said timing network includes a resistance capacitance network.

4. In the gas generator according to claim 1 wherein said current source includes a pair of Silicon Controlled Rectifiers connected in push-pull.

i i I *1 k

Claims

1. In a gas generator having a controllable gas output the combination comprising: a. an electrolysis cell having: 1. a solid polymer, ion-exchange electrolyte membrane, 2. catalytic electrodes positioned adjacent to opposite surfaces of said membrane to dissociate a chemical compound to evolve gas, b. a source of electrical power coupled to the electrodes of said cell, c. means to furnish a chemical compound to one of said electrodes for dissociation, the ionic form of one of the elements of the dissociated compound being transported across said ion-exchange-membrane to evolve the gas, d. means for sensing the output gas pressure and producing a control signal in response thereto; e. means responsive to the control signal responsive to the sensed gas pressure for controlling the current flow from said source of electrical power on a time ratio basis to vary the current level to the cell electrodes thereby to control the rate of gas evolution including: 1. a current source including a pair of alternately conducting solid state switching devices for supplying the cell current, 2. a voltage responsive variable repetition frequency, pulse generator coupled to said switching devices having a voltage responsive timing network for controlling the repetition pulse frequency, said pulse generator producing triggering pulses for said switches to control the average current level to said cell, 3. means coupling the signal from said gas pressure sensing means to said pulse generator timing network to vary the pulse repetition rate of the pulses from said generator in response to said signal to vary the cell current as a function of the outlet gas pressure.

1. In a gas generator having a controllable gas output the combination comprising: a. an electrolysis cell having:

1. a solid polymer, ion-exchange electrolyte membrane,

2. a voltage responsive variable repetition frequency, pulse generator coupled to said switching devices having a voltage responsive timing network for controlling the repetition pulse frequency, said pulse generator producing triggering pulses for said switches to control the average current level to said cell,

2. catalytic electrodes positioned adjacent to opposite surfaces of said membrane to dissociate a chemical compound to evolve gas, b. a source of electrical power coupled to the electrodes of said cell, c. means to furnish a chemical compound to one of said electrodes for dissociation, the ionic form of one of the elements of the dissociated compound being transported across said ion-exchange-membrane to evolve the gas, d. means for sensing the output gas pressure and producing a control signal in response thereto; e. means responsive to the control signal responsive to the sensed gas pressure for controlling the current flow from said source of electrical power on a time ratio basis to vary the current level to the cell electrodes thereby to control the rate of gas evolution including: