Title:
ELECTROLYSIS SYSTEM AND METHOD
United States Patent 3755128
Abstract:
An electrolysis apparatus and method including means for decomposing an electrolyte into one or more gas products, and also including means responsive to the pressure or flow condition of one of the product gases for controlling the electrical input to the electrolysis cell whereby to match the gas generation rate and the gas demand rate in the electrolysis system. Electrical power cost of the electrolysis process, a major operating expense, will be reduced if the gas generation rate responds to the demand rate.


Inventors:
HERWIG W
Application Number:
05/072250
Publication Date:
08/28/1973
Filing Date:
09/15/1970
Assignee:
Isotopes, Inc. (Westwood, NJ)
Primary Class:
Other Classes:
204/228.1, 204/228.5
International Classes:
C25B15/02; (IPC1-7): B01K3/00; C01B13/04
Field of Search:
204/228-230,129
View Patent Images:
Primary Examiner:
Mack, John H.
Assistant Examiner:
Valentine D. R.
Claims:
The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows

1. An electrolysis system for decomposing an electrolyte into at least one product gas, comprising an electrolysis cell having terminals and adapted to have an electrolyte therein, said cell being adapted to generate at least one product gas upon the connection of direct current across the terminals of said cell, the generation of said one product gas by said cell being a function of the direct current input to said cell, means in circuit relationship for supplying a variable direct current to said cell terminals whereby said one product gas is generated by said cell, means for sensing the flow rate of a product gas generated by said cell, and feedback network means coupled directly between said sensing means and said direct current supplying means for controlling the variable direct current input to said cell as a function of the condition sensed.

2. An electrolysis system as defined in claim 1 in which said condition sensed is the flow rate from said cell of said one product gas.

3. An electrolysis system as defined in claim 1 in which said means for supplying direct current to said cell comprises at least one thyratron type solid state device, and wherein said feedback network produces an output electrical signal that controls the load current conduction through said solid state device whereby to control the electrical current input to said cell.

4. An electrolysis system for decomposing an electrolyte into at least one product gas, comprising an electrolysis cell having terminals and adapted to have an electrolyte therein, said cell being adapted to generate at least one product gas upon the connection of direct current across the terminals of said cell, the generation of said one product gas by said cell being a function of the direct current input to said cell, means in circuit relationship for supplying a variable direct current to said cell terminals whereby said one product gas is generated by said cell, means for sensing a condition of a product gas generated by said cell, feedback network means coupled directly between said sensing means and said direct current supplying means for controlling the variable direct current input to said cell as a function of the condition sensed, and sensor means connected in current sensing relation between said direct current supply means and said cell and connected to said feedback network means for limiting the maximum and minimum current supplied by said direct current supply means to said cell to predetermined maximum and minimum values.

5. The method of controlling the rate of gas generation in an electrolysis system of the type including an electrolysis cell in which at least one product gas is generated by said cell upon the connection of direct current to the terminals of said cell and in which the generation of said one product gas by said cell is a function of the direct current input to said cell, which comprises the steps of sensing the flow rate from said cell of said one product gas generated by said cell, and of controlling the direct current input applied to said cell as a function of the flow rate sensed whereby to control the rate of gas generation by said cell.

Description:
The product gas demand function such as product gas pressure or flow rate is converted into an electrical control signal which is fed to a feedback network which controls the direct current output of the power supply to the cell, to thereby cause a gas generation rate equaling the demand rate. The gas demand function which is measured compensates automatically for varying operating conditions which affect gas generation rate such as ambient temperature, cell electrolyte phenomena, and long term aging characteristics of the electrolysis cell. Control signals representing various other parameters of the electrolysis system such as high and low limits for electrolysis current may also be fed into the feedback network to control the current flow from the power supply to the cell.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to an electrolysis system, and more particularly to an apparatus for and a method of matching the gas generation rate in and the gas demand rate on an electrolysis system.

2. Description of the Prior Art

Electrolysis systems are known in the prior art in which a direct current is applied across a pair of electrodes in contact with an electrolyte to cause decomposition of the electrolyte into one or more product gases. One such system, involving an aqueous electrolyte and hydrogen and oxygen product gases, is shown in U.S. Pat. No. 3,410,770, granted to Lester W. Buechler on Nov. 12, 1968. The system of the Buechler patent just mentioned consists essentially of six parts: (1) an electrolysis module (a stack of cells); (2) a recirculating electrolyte loop; (3) water addition equipment; (4) oxygen manifolding; (5) hydrogen manifolding; and (6) a direct current power supply. The system of the aforementioned Buechler patent as well as other known electrolysis systems encounter operating problems when the operating requirements are increased from those of a continuous steady state operation to a variable demand with minimum operator attendance. In prior art electrolysis systems, operator attendance is required for adjustment of the electrolysis current due to: (1) changes in gas demand; (2) changes in ambient conditions affecting the rate of gas production; (3) changes in the system parameters affecting the rate of gas production; and (4) cell aging characteristics affecting rate of gas production.

SUMMARY OF THE INVENTION

Accordingly, it is an object of the present invention to provide an electrolysis system and apparatus which includes means for automatically matching the gas generation rate and the gas demand rate, whereby to operate the system at optimum efficiency and to effect economy in the operation of the system.

It is another object of the invention to provide an electrolysis system, apparatus and method in accordance with which the electrolysis current is applied to the electrolysis cell or module as a function of the gas demand rate whereby to cause the gas generation rate of the system to equal the gas demand rate.

It is a further object of the invention to provide an electrolysis system and apparatus which is substantially automatic in its operation and which does not require operator attendance for adjusting the gas generation rate.

It is a further object of the invention to provide an aqueous electrolysis system, apparatus and method in accordance with which the electrolysis current is automatically adjusted to compensate for changes in gas demand, for changes in ambient conditions affecting the rate of gas production, for changes in system parameters such as electrolyte temperature affecting the rate of gas production, and to compensate for cell aging characteristics affecting the rate of gas production.

In achievement of these objectives there is provided in accordance with this invention an electrolysis apparatus and method including means for decomposing an electrolyte into one or more gas products, and also including means responsive to the pressure or flow condition of one of the product gases for controlling the electrical input to the electrolysis cell whereby to match the gas generation rate and the gas demand rate in the electrolysis system. Electrical power cost of the electrolysis process, a major operating expense, will be reduced if the gas generation rate responds to the demand rate.

The product gas demand function such as product gas pressure or flow rate is converted into an electrical control signal which is fed to a feedback network which controls the direct current output of the power supply to the cell, to thereby cause gas generation rate equaling the demand rate. The gas demand function which is measured compensates automatically for varying operating conditions which affect gas generation rate such as ambient temperature, cell electrolyte phenomena, and long term aging characteristics of the electrolysis cell. Control signals representing various other parameters of the electrolysis system such as high and low limits for electrolysis current may also be fed into the feedback network to control the current flow from the power supply to the cell.

Further objects and advantages of the invention will become apparent from the following description taken in conjunction with accompanying drawings in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic view of an electrolysis apparatus and system embodying the automatic control features of the invention;

FIG. 2 is a schematic diagram of control circuitry which may be incorporated in the system and apparatus of the present invention to control the gas generation rate as a function of the gas demand rate and also as a function of other factors;

FIG. 3 is a schematic diagram of another control circuit which may be used to control the gas generation rate as a function of the gas demand rate also as a function of other factors; and

FIG. 4 is a vector diagram showing the various electrical relationships in the circuit of FIG. 3.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring now to FIG. 1, there is shown an electrolysis system and apparatus which except for the automatic control system to be hereinafter described, is substantially similar to the electrolysis system shown in the aforementioned U.S. Pat. No. 3,410,770, granted to Lester W. Buechler, on Nov. 12, 1968, the teachings of which patent are hereby incorporated by reference into this application.

The aqueous electrolysis system and apparatus of FIG.1 comprises a cell 10 having a pair of gas permeable electrodes, namely, cathode 11 and anode 12, in direct contact with the opposite surfaces of the electrolyte matrix 13 which is saturated with an aqueous electrolyte. Electrodes 11 and 12 are connected by leads 15 and 16 to a source of direct current generally indicated at 14 to be described more fully hereinafter. The quantity of gas produced is directly proportional to the number of electrolysis cells and to the current flow through each cell. Electrolyte matrix 13 is a porous diaphragm of asbestos fibers or other material which is resistant to attack by caustic alkali solution or other electrolyte. The thickness of the electrodes 11, 12 and the matrix 13, as shown in FIG. 1, have been exaggerated for purposes of clarity in description. The electrodes 11 and 12 and the electrolyte matrix 13 are supported in housing 17 which forms a gas chamber 18 with the cathode 11 and an electrolyte chamber 19 with the anode 12. The aqueous electrolyte 20 is contained in the electrolyte chamber 19.

The electrolyte matrix 13, saturated with the aqueous electrolyte, serves a dual function. The matrix 13 maintains intimate contact between the interface of the electrolyte and the gas permeable electrodes 11 and 12 and also serves as a gas impervious barrier to prevent mixing of the gas products.

Electrolyte is supplied to the cell by the circulation of electrolyte solution from separator 22 to electrolyte chamber 19 by means of pump 23 through conduits 24, 25 and 26. The gas generated at the electrode 12 adjacent the electrolyte chamber 19 is removed from the electrolyte chamber entrained in the circulating electrolyte solution.

During the operation of the cell, the aqueous electrolyte 20, impelled by the driving force of the gas bubbles present in it, is forced from the electrolyte chamber 19 through conduit means 24 and passes to electrolyte separator 22 where the gas produced at the electrode 12 and entrained in the aqueous electrolyte 20 is extracted and separated from the circulating electrolyte solution. Design details for the separator 22 are not shown as separators of this type are well known in the art. Water consumed in electrolysis is replaced by means of water storage means 30 from which water is fed in controlled amounts to separator 22 via conduit 31 to maintain the electrolyte at optimum concentration. Valve 32 regulates the addition of water to the electrolyte solution in separator 22 at a rate determined to replace the water consumed by electrolysis.

The oxygen gas produced at anode electrode 12 is withdrawn overhead from the separator 22 via conduit 33, through differential pressure regulator or control valve 58, and through connected conduit 59. Oxygen not required for maintaining system pressure is discharged through conduit or "plumbing" 59. The degassed electrolyte passes out of the separator 22 via conduit means 25 to pump 23 wherein it is directed via conduit means 26 back to the electrolyte chamber 19.

The hydrogen gas produced at cathode electrode 11 is evolved into gas chamber 18 and the chamber 18 is maintained under pressure by regulating the passage of gas out of the chamber. The gas that is removed from gas chamber 18 is passed by conduit 50 through valve 60 and thence externally of the system through conduit 61. Demand for hydrogen gas is generated by valve 60 or by a system of valves and control devices in response to the variable demand requirements of a gas demanding load external of the system. Hydrogen demand is discharged by conduit or "plumbing" 61. Valve 60 and conduit 61 are both located external to the electrolysis and control system but interface with the electrolysis and control system at 62.

When a direct current is initially applied to the apparatus at cathode 11 and anode 12, as will be described hereinafter, gas generated at electrode 11 is collected in the chamber 18 until the pressure of the gas in the chamber 18 exceeds the pressure of the gas generated at electrode 12. This occurs in a relatively short time interval and if any electrolyte is present in chamber 18 it will be forced through the capillary matrix to chamber 19. The differential pressure control valve 58 in oxygen conduit 33 maintains a preset pressure differential between the hydrogen gas pressure in gas chamber 18 and the pressure of the oxygen gas in conduit 33 leading from separator 22. The hydrogen gas pressure in conduit 50 is monitored to valve 58 by conduit 72 connected between conduit 50 and differential pressure control valve 58.

As the pressure of the gas produced at the electrode 12 is raised or lowered, the valve 58 opens or closes in response to this pressure to provide a decrease or increase in the oxygen gas pressure in oxygen conduit 33 so as to maintain an appropriate differential pressure across the cell 10 and avoid leakage of electrolyte solution through the matrix 13 to chamber 18. In this manner, the electrolyte in the capillary matrix 13 adjacent electrode 11 is constantly replenished with electrolyte solution. Since the gas produced at electrode 11 is not entrained in the electrolyte, the need for a separator unit to disentrain gas produced at electrode 11 from the electrolyte is eliminated.

The hydrogen manifold pressure in conduit 50 is the independent or reference pressure for the differential pressure valve 58 and the oxygen manifold 33 pressure is the dependent or following pressure. Oxygen not required for maintaining manifold 33 pressure is discharged through conduit or "plumbing" 59.

It will also be understood that in the pressure differential control system, the valve 58 could be placed in conduit 50 instead of in conduit 33 as shown, in which case the oxygen pressure in conduit or manifold 33 would be used as the reference pressure.

Also, while in the system illustrated in FIG. 1, the demand function transducer device 51 senses a gas flow or pressure condition in hydrogen manifold 50, the transducer device 51 could instead sense a gas flow or pressure condition in oxygen manifold 33.

If the differential pressure valve 58 is located in hydrogen manifold 50 and the pressure in oxygen manifold 33 is used as the reference pressure for the differential pressure system, then the demand function transducer device 51 should also be located in oxygen manifold 33.

As is obvious to one skilled in the art, a multicellular apparatus comprising a plurality of the unit cells may be connected in series and clamped into a compressed face-to-face relationship along a common axis to form a module or stack of cells. In commercial applications such assemblies are preferred for efficient, quantitative production of hydrogen and oxygen.

In the illustrated embodiment of FIG. 1, the polarity connections of the direct current input power to the cell electrodes 11 and 12 are such that electrode 12 is the anode at which oxygen is evolved. The reason for this, as explained in the aforementioned U.S. Pat. No. 3,410,770 to Lester W. Buechler, is that the volume of oxygen produced at the anode is one half that of the hydrogen produced at the cathode and consequently a smaller separator 22 is required to disentrain the O2 from the circulating electrolyte. However, it will be understood that instead of the preferred arrangement just referred to, the polarity connections of the power supply to the electrolysis cell could be the reverse of those just described, so that electrode 11 is the anode at which oxygen is evolved and electrode 12 is the cathode at which hydrogen is evolved.

It will be noted in the schematic diagram of FIG. 1 that the power supply generally indicated at 14 has its electrical output connected by conductors or cables 15 and 16 to the electrodes 11 and 12 of the cell 10. An electrical transducer device 51 which may sense either gas pressure or gas flow conditions in the hydrogen manifold 50, depending upon the particular type of transducer device 51 which is used, is electrically connected by conductor means schematically indicated at 75, to the input of the feedback network 53.

As will be explained in more detail hereinafter, the feedback network 53 in response to a control signal or signals applied thereto controls the power supply 14 via conductor means 73 to provide a variable electrolysis current to the electrolysis cell 10 via cables 15 and 16. The electrolysis current is therefore a function of the gas pressure or of the gas flow condition sensed by transducer 51, depending upon whether transducer device 51 is of a type which senses gas pressure or gas flow. In this manner a given pressure or flow rate of the hydrogen can be maintained.

If the transducer device 51 is of a type which senses pressure, a transducer device which may be used is manufactured by Robinson-Halpern of 5 Union Hill Road, West Conshohocken, Pennsylvania 19428, under the designation "Variable Set Point Pressure Transducer Model 107A". This pressure transducer senses pressure variations or departures from a reference set point pressure and provides an output voltage or error signal which is proportional to the deviation of the pressure from the set point. Thus, for example, if transducer 51 detects a pressure drop in hydrogen manifold 50 which is indicative of increased demand for hydrogen, it will transmit a proportional signal to feedback network 53 which will cause feedback network 53 to cause an increase in the direct current flow to cell 10 from power supply 14 sufficient to increase the gas generation rate of cell 10 until the pressure in hydrogen manifold 50 is returned to the set point.

Conversely, if transducer 51 detects a pressure increase in hydrogen manifold 50 which is indicative of decreased demand for hydrogen, it will transmit a proportional signal to feedback network 53 which will cause feedback network 53 to cause a reduction in the direct current output of power supply 14 to the electrodes 11 and 12 of cell 10 to thereby decrease the gas generation rate of cell 10 until the pressure in hydrogen manifold 50 is returned to the set point.

The thermal and mass flow considerations require that the maximum and minimum electrolysis currents be limited to preselected values. A sensor 55 connected in current sensing relation to cable 16 (or 15) leading from power supply 14 to one of the electrodes of cell 10 is used to quantitatively sense the electrolysis current and to feed a signal via conductor means 77 into the feedback network 53 to limit the maximum and minimum current supplied by power supply to electrolysis cell 10 to maximum and minimum values.

The feedback network and associated power supply may be of the type shown and described, for example, in U.S. Pat. No. 3,333,178 granted to Roland L. Van Allen and Charles E. Hardies on July 25, 1967, and the teachings of U.S. Pat. No. 3,333,178 to Van Allen and Hardies are hereby incorporated by reference into the present application. Thus, for example, the feedback network 53 and power supply 14 may assume the form shown in FIG. 4 of the aforementioned Van Allen et al. patent which is substantially embodied in FIG. 2 of the present application. There is shown in FIG. 2 of the present application a feedback network and thyristor power supply utilizing solid state devices generally indicated at 100 and 100' of the type having thyratron characteristics. A number of solid state devices of this character are now available to the industry. General Electric Company offers devices identified as "Silicon Controlled Rectifiers" (SCRs) and Westinghouse Electric Corporation produces devices of this type identified by the name "Trinistors."

The solid state devices of the type indicated at 100 and 100' in FIG. 2 include three terminals, two which (102, 108) may be considered as main current carrying terminals and the third (120) as a gate or control terminal. These devices are so characterized that with respect to the main current carrying terminals the device at all times presents a high impedance to current flow in one direction, that is a high reverse impedance, and in this respect, the device exhibits the characteristics of a rectifier, while the device presents a high impedance to the flow of current between the main terminals in the opposite direction that is, a high forward impedance, until the device is energized or fired upon application of a control signal to the gate or control terminal. When the device is fired, the forward impedance with respect to the main current carrying terminals abruptly drops to an extremely low value, and the flow of current through the device between the main current carrying terminals is independent of and does not require application of a control or energizing signal to the gate terminal and will continue so long as a potential is maintained across the main current carrying electrodes. When the potential across the main current carrying electrodes is extinguished, the device returns to its normal or unenergized condition, presenting a high forward impedance, and will not pass current although a potential is applied across the main current carrying electrodes, providing a breakdown potential is not reached, until a control signal is applied to the gate terminal to again energize the device.

It is apparent from the foregoing that this class of solid state devices possess certain operational characteristics of a thyratron and may therefore be described as "solid state thyratrons" or "solid state devices possessing thyratron characteristics." The latter terms are used throughout this description and in the appended claims to define solid state devices of the class described hereinbefore. Furthermore, as a descriptive aid, the main current-carrying terminals of the solid state devices of the class described hereinbefore are referred to herein and in the appended claims as "anode terminal" and herein and in the appended claims as "anode terminal" and "cathode terminal" although the terms "anode" and "cathode" are not generally employed in connection with solid state devices, and probably would not be used to designate the components of these solid state devices to which the main current carrying terminals are connected.

Referring now more specifically to FIG. 2, the combined feedback network and power supply shown in FIG. 2 comprises a pair of solid state devices having thyratron characteristics 100 and 100', each including an anode terminal 102, 102' which are respectively connected to opposite terminals 104, 104' of a center tapped transformer secondary generally indicated at 106. The circuit of FIG. 2 is analogous to a full-wave center tap rectifier and provides a controlled flow of load output current to the input electrical terminals of the electrolysis cell during each half cycle of the applied voltage across transformer secondary 106, the current conduction period during each half cycle depending upon the cumulative effect of the control signals applied to the feedback network 53, as will be explained more fully hereinafter. The solid state devices 100, 100' also each respectively include a cathode terminal 108, 108', respectively, which are connected to each other and also to the output or load terminal 116 which is connected to cable 16 leading to electrode 12 of the electrolysis cell. The other cable 15 leading to the electrolysis cell is connected to the center tap 118 of the transformer secondary 106.

The solid state devices 100 and 100' each include a gate or control terminal 120, 120', respectively. The apparatus further includes a pair of saturable magnetic cores 122 and 122', respectively, preferably constructed of a material presenting a substantially rectangular hysteresis characteristic. Two control windings 124, 126 are provided for the saturable magnetic cores 122 and 122', each of the two windings 124, 126 being common to both of the saturable magnetic cores 122 and 122'. Windings 124, 126 are adapted to be energized with direct current control voltages to control the degree of saturation of saturable magnetic cores 122, 122'. The control voltage across control winding 124 may be derived, for example, from the output voltage signal of transducer device 51 which monitors either the gas pressure or the gas flow in the hydrogen manifold 50, depending on the type of transducer device 51 which is used. The control voltage across control winding 126 for example, may be a signal derived from the current sensing device 55, FIG. 1, and diagrammatically shown as being connected by conductor means 77 to the feedback network to limit the direct current supplied by the power supply 14 to electrolysis cell 10 to maximum and minimum values. The minimum input current to cell 10 is set at a value which provides a minimum gas product rate sufficient to provide dilution of gas from leaks or gas migration from cavity 20 across the capillary matrix 13 to cavity 18. In this way, a maximum gas purity in cavity 18 is maintained.

Upon conditions of low hydrogen gas demand, the gas generation rate may exceed the demand rate. This condition results in an increased hydrogen pressure which is relieved by regulator valve 79 connected to hydrogen manifold line 50, regulator valve 79 permitting the escape of hydrogen gas through relief conduit 81.

Connected in circuit with each of the solid state thyratron devices 100 and 100' is a gate winding 130, 130'. One end of each gate winding 130, 130' is connected to the terminal 104 or 104' of the secondary winding 106 and to anode terminal 102, 102' in series with a solid state rectifier 132, 132'. The opposite end of each gate winding 130, 130' is connected through a resistor 134, 134', to the gate or control element terminal 120, 120' of the solid state thyratron device 100, 100'. Each of the respective gate windings 130, 130' is mounted on or in magnetic association with one of the respective saturable cores 122, 122', respectively.

The control windings 124, 126 and the gate windings 130, 130' each are shown with a dot which indicates winding sense or polarity relationship.

In the operation of the apparatus of FIG. 2, assume that the alternating current power supply at transformer secondary winding terminals 104 and 104' has passed through zero potential and is beginning a half cycle which is positive relative to the anode terminal 102 of solid state device 100 and negative with respect to the anode terminal 102' of solid state device 100'. Under this condition, the solid state device 100 presents a high forward impedance since the control element 120 of solid state device 100 is nonenergized, and the solid state device 100' presents a high reverse impedance. Rectifier 132' blocks current flow through gate winding 130' of saturable core 122'. Gate winding 130 in the circuit of solid state device 100 presents a high impedance, since core 122 is not saturated at this instant and no power is applied across the load output terminals 115 and 116. Current, however, will flow through the gate circuit of coil 130, i.e., through coil 130, resistor 134, gate terminal 120, through a portion of solid state device 100, cathode terminal 108, and across load terminals 115, 116, back to center tap 118 of transformer secondary 106, to thereby carry the magnetic material of core 130 toward one level of saturation, the parameters of the gate circuit being selected to insure adequate current flow to effect this performance.

Core 122 will be driven to saturation at some point during the half cycle of the alternating current supply source which is positive relative to the anode terminal 102 of solid state device 100, as determined by the combined effect of the vectorial sum of the control signals applied to the windings 124 and 126.

At the instant core 122 saturates, the impedance of gate winding 130 abruptly drops with a concomitant increase in current through gate winding 130. The abrupt current variation through gate winding 130 is applied as a control signal through resistor 134 to gate terminal 120 of solid state device 100 to energize or fire the device 100 within a microsecond or less following saturation of core 122. When fired, the forward impedance of solid state device 100 abruptly drops very close to zero and the power supply is connected across load terminals 115, 116 through the remaining portion of the half cycle. The solid state device 100 when fired presents a substantially complete short circuit across gate winding 130 and thereby terminates the control signal from gate winding 130 to gate terminal 120 of the solid state device 100. Thus, the arrangement for producing and applying a control signal to the gate terminal of the solid state device provides an automatic clipping or limiting operation which prevents application of a control signal of a magnitude which exceeds the design limitations of the solid state device.

Upon the next half cycle of applied alternating current voltage, when the input power of transformer secondary winding 106 is positive with respect to the anode terminal 102' of solid state device 100' and negative with respect to the anode terminal 102 of solid state device 100, the respective solid state devices 100 and 100' and the circuits associated therewith reverse their respective relationships from those described for the first half cycle of applied voltage. That is, the solid state device 100' now presents a high forward impedance and the solid state device 100 presents a high reverse impedance. The solid state device 100' becomes conductive or fires at a predetermined angular condition in the second half cycle of applied voltage in the same manner as described in connection with the solid state device 100 during the first half cycle of applied voltage. When fired, the forward impedance of the solid state device 100' abruptly drops very close to zero and the power supply is connected across the load terminals 115, 116, through the remaining portion of the second half cycle. During the second half cycle, current flow in gate winding 130' induces voltage in gate winding 130 through the control circuit, and this induced voltage together with the combined effect of the control signals applied to control windings 124, 126 acts to effect resetting of core 122 to the initial saturation level for response to the next half cycle. A corresponding action takes place during alternate half cycles to reset core 122'.

Other control systems which may be used instead of the phase angle shifting system of the U.S. Pat. No. 3,333,178 to Van Allen et al, hereinbefore described, are set forth in Sprague Technical Paper No. 63-9, published by Sprague Electric Company, North Adams, Massachusetts, entitled "The Silicon Controlled Rectifier in Proportional Power Control," the teachings of which technical paper are hereby incorporated by reference into the present patent application. One phase angle control circuit substantially as disclosed in the aforementioned technical paper is shown in FIG. 3 of the present application and the voltage vector diagram for the circuit of FIG. 3 is shown in FIG. 4 of the present application.

Referring now to FIG. 3, there is shown what is known as the "Silicontrol gate drive" which is a special patented form of wide angle phase-shifting circuit controlled by a saturable reactor. This special form offers a full 180° range of linear phase shift using a small (4 millwatt) direct current control signal and provides a steeply rising gate pulse which triggers the firing of the thyratron type solid state device. This Silicontrol gate drive is manufactured and sold by Sprague Electric Company of North Adams, Massachusetts. Referring now to FIG. 3, there is shown an input transformer generally indicated at 150 including a primary winding 152 to which is applied, for example, 115 volt, 60 cycles per second alternating current electric power. Transformer 150 includes a secondary winding 154. Across the terminals B and X of the transformer secondary winding 154 is connected a fixed phaseshift network RC1 which establishes a "base line" voltage ER across resistor R as shown in the vector diagram of FIG. 4. Connected across resistor R is a series resonant circuit consisting of inductor L and capacitor C. The voltages across L and C are represented in the vector diagram of FIG. 4 by vectors A'P' and P'B', separated by small angle φ (the loss angle of the inductor). In the vector diagram of FIG. 4, the primed letters such as A'P', etc., correspond to the same unprimed letters of the circuit diagram of FIG. 3.

When the inductance of L is altered slightly, the relative lengths of vectors A'P', P'B' are effectively varied and if the angle φ is made to stay constant, by careful design, the point P', will in effect, move around the dotted circle of the vector diagram of FIG. 4. A pulse forming network generally indicated in block diagram form at 156 in FIG. 3 is connected across the center tap terminal O of the transformer secondary 154 and also across the terminal P which is the junction point between the inductance L and the capacitance C in the series resonant network across the resistance R.

By careful design, the circular locus is made to pass through the input potential point X' on the vector diagram, and since angle X' A' B' is a right angle, the vector X'B' is a diameter of the circle and the input center tap O' is at the center of the circle.

The vector O'P' then represents an output that can be taken from the network, varying in phase angle by approximately 300° while remaining constant in amplitude.

Two silicon controlled rectifiers SCR No. 1 and SCR No. 2 (not shown) may be connected to provide rectified direct current to the electrodes 11, 12 of cell 10, the magnitude of the current flow to cell 10 being controlled during both half cycles of alternating current across transformer 150 by the circuit of FIG. 3.

The inductor L takes the form of a specially designed saturable reactor 155 whose inductance is varied by saturating its core to a greater or less degree by passing varying amounts of direct current through the control windings 1-2, 3-4. The output voltage across the terminals O, P, are formed into pulse spikes by the pulse forming network 156 which drives the coupling transformer generally indicated at 158. Due to the blocking action of the two output diodes 162A, 162B, the pulse spikes are delivered on alternate half cycles from terminals G1 and G2, which are each respectively connected to the gate or control terminal of one of a pair solid state thyratron devices (not shown) which in this case are the pair of silicon controlled rectifiers SCR No. 1 and SCR No. 2, the terminals K1, K2 of the circuit of FIG. 3 being connected to the cathode terminals of the respective solid state devices.

The control windings 1-2, 3-4, which control the degree of saturation of the saturable core reactor 155 and hence control the value of the inductance L in the phase shift circuit, derive their signal voltages from the sources as set forth in the example of FIGS. 1 and 2. For example, the voltage signal across control winding 1-2 of FIG. 3 may be the signal from the transducer device 51 indicating the gas pressure or flow condition in the hydrogen manifold 50. The voltage signal across control winding 3-4 may be derived from the current transducer device 55 in current sensing relation to conductor or cable 16 between the thyristor power supply 14 and the electrode 12, to indicate the direct current flow to cell 10.

It will be understood that the feedback and power supply systems shown in FIGS. 2 and 3 are by way of example only and that other suitable types of feedback and power supply systems may be substituted therefor.

It can be seen from the foregoing that there is provided in accordance with the invention an apparatus for and method of matching the gas generation rate with the gas demand rate in an electrolysis system by measuring a demand function such as gas pressure or gas flow, and adjusting the input direct current to the electrodes of the electrolysis cell in accordance with the sensed demand function whereby to match the gas generation and demand rates. This matching of gas generation rate with gas demand rate results in important savings in the cost of electrical current input to the electrolysis cell.

It will also be noted that a very important advantage of the apparatus and method is that in sensing the gas pressure or gas flow and controlling the input current to the electrolysis cell as a function of the sensed gas pressure or gas flow as taught by the present invention, not only is the gas output of the cell varied in accordance with the varying gas demand rate of the external gas demanding load, but also automatic and inherent compensation is simultaneously made without further adjustment for such varying operating conditions as ambient temperature, module electrolyte phenomena, and long term aging characteristics of the electrolysis cell.

While there have been shown and described particular embodiments of the invention, it will be obvious to those skilled in the art that various changes and modifications may be made therein without departing from the invention and, therefore, it is aimed to cover all such changes and modifications as fall within the true spirit and scope of the invention.