US2027253A - Flasher circuits - Google Patents

Description

Jan; 7, 1936. I w. w. VIEBAHN FLASHER CIRCUITS Filed Jan. 6, 1953 WITNESSES:

INVENTOR William W Vz'ebahn ATTO EY Patented Jan. 7, 1936 UNITED STATES PATENT OFFICE FLASHER CIRCUITS of Pennsylvania Application January 6, 1933, Serial No. 650,517

11 Claims.

My invention relates to control apparatus and has particular relation to apparatus for controlling the supply of power to a system of the type that is to be periodically energized and deenergized.

It is an object of my invention to provide control apparatus of a type wherein mechanical contacts shall not be utilized for a translating systern that is to be periodically energized and deenergized.

Another object of my invention is to provide apparatus for controlling the condition of a translating system that is to be periodically deenergized and energized in which the lengths of the intervals during which the translating system is energized and deenergized shall be capable of any desired adjustment.

An incidental object of my invention is to provide an accurate time-delay element of simple structure.

According to my invention, I provide a control system of the type incorporating preferably a gas-filled electric discharge device having an anode, a cathode and a control electrode. The output circuit or" the gas-filled device feeds the input circuit of a reactor, preferably of the inductive type. The output windings of the reactor are connected through the transmitting system to be controlled and supply power thereto. A time constant circuit is coupled between the control electrode and the cathode of the electric discharge device, and its efiect is to control the excitation of the electric discharge device.

The change in the condition of the electric discharge device is produced in the practice of my invention by varying the potential impressed between the control electrode and the cathode of the electric discharge device, between a maximum value and a minimum value as an alternating potential is impressed between the anode and the cathode. The electric discharge device is of such character that if an ordinary time-constant circuit were coupled to the electric discharge device in the usual manner, the electric discharge device would be energized and deenergized during periods corresponding to the half cycles of the potential impressed between the anode and the cathode thereof, and the periodicity of the excitation of the translating system would be equal to the periodicity of the source whereby potential is applied to the electric discharge device. An important feature of my invention resides in the fact that I have provided apparatus in which the intervals during which the electric discharge device is energized and deenergized may be adjusted to any desired value.

The novel features that I consider characteristic of my invention are set forth with particularity in the appended claims. The invention itself, however, both as to its organization and its method of operation, together with additional objects and advantages thereof, will best be understood from the following description of aspecific embodiment when read in connection with the accompanying drawing, in which:

Figure l is a diagrammatic view showing a preferred embodiment of my invention;

Fig. 2 is a graph illustrating the operation of an electric discharge device of the type that is utilized in the practice of my invention; and,

Figs. 3, 4, 5 and 6 are graphs that will be utitized in explaining the changes introduced in the characteristic of the system by the various elements utilized in the practice of my invention.

The apparatus shown in Fig. 1 comprises an electric discharge device I of the gas-filled type. The anode 3 of the electric discharge device I is connected one bus line 5 of an alternating current power supply source (not shown) and the cathode 'l is connected to the other bus line 9 through the input circuit I l of a three-legged core reactor I3. The cathode is also coupled to the bus line 9 through a plurality of capacitors I 5 and H and resistors l9, 2| and 23, and a rectifier 25. The resistors l 9, 2| and 23 are connected in series with each other and in parallel with the input circuit 5! of the reactor 53, while the capacitors l5 and H and the rectifier 25 are connected in parallel with each other and each is in a separate series network with the input circuit ll. One terminal of one capacitor I5 is connected to one terminal of the outermost resistor I9; one terminal of the other capacitor I1 is connected between the resistor l9 and the central resistor 2| while the positive terminal of the rectifier 25 is connected between the resistor 2| and the resistor 23 which is connected directly to the bus line 9. It is to be noted that the rectifier 25 is conductive in the direction in which the arrow in Fig. 1 points.

The operation of my apparatus will be more clearly understood if it is noted that there are essentially three circuits involving the capacitors l5 and Il and the resistors I9, 2| and 23. These circuits may be traced as follows:

(a) From the left-hand power line 5, through the discharge path between the anode 3 and the cathode 7 of the electric discharge device, the input windings II of the reactor I3 to the other power line 9.

(b) From the power line 5, through the discharge path between the anode and cathode of the device I, the capacitor H, the resistor 2i, the resistor 23'to the other power line 9.

(c) From the power line 5, through the discharge path between the anode and cathode of the device I, the capacitor I5, the resistor I9, the resistor 2I, the resistor 23 to the other power line 9.

The above described networks including the capacitors I5 and I1 and the resistors I9, BI and 23 are coupled to the control electrode 21 of the electric discharge device I through an additional capacitor 29, a portion of the windings 3| of a secondary 33 of a transformer 35 the primary 3'! of which is connected between the bus lines 5 and 9, and a grid-leak resistor 39.

The electric discharge device I is preferably of the type that has an energized and a deenergized condition and is capable only of abrupt transition from one condition to the other. Its manner of operation is best represented by a graph such as is shown in Fig. 2. In this view the potentials impressed between the electrodes 3, 'I and 2! of the electric discharge device I are plotted as ordinates against a time function which is plotted as abscissa. The full-line curve 2! represents the sinusoidal potential impressed between the principal electrodes 3 and E of the electric discharge device I. This potential will be termed hereinafter the principal potential. The broken-line curve 43 may be designated as the limiting control potential curve of the electric discharge device I and represents the boundary line of values of potential, impressed between the control electrode 2'! and the cathode I of the electric discharge device I, below which the device is in a deenergized condition and above which the device is in an energized condition. That is to say. when the potential impressed between the control electrode 21 and the cathode I of the electric discharge device I is, at all times during any positive half-cycle of the principal potential, of a value below the broken-line curve 43, the electric discharge device is in a deenergized condition. If this potential is raised to a value above the broken-line curve t3, the electric discharge de vice I is energized and remains energized, regardless of any subsequent variation in the control potential, until the principal potential is reduced to a value below the ionization potential of the electric discharge device. It is seen that the electric discharge device may be periodically energized and deenergized by shifting the control potential from a value below the broken-line curve to a value above the broken-line curve. The periodicity of such excitation of the electric discharge device I by reason of the lock-in characteristic of the device can of course only be smaller than the periodicity of the principal potential. The periodic .saturation of the reactor is and therefore the operation of the system is dependent on the periodic excitation of the electric discharge device I.

From a consideration of Fig. 1 it can be seen that this object is accomplished principally by the charging and discharging of the capacitor I5 that is directly connected to the cathode 5. Normally the potentials impressed between the control electrode 21 and the cathode I of the electric discharge device I are adjusted to such values that when the capacitors I5 and Il are in a substantially uncharged condition, the electric discharge device I is energized. When it is thus energized, current is supplied through the electric discharge device I to the input winding I i of the reactor I3 and the reactor is magnetically saturated. The output circuit 45 of the reactor is coupled through the translating system A! to be controlled which, as in the present case, may be a system of lamps 49 utilized in a display device such as an advertising sign for example. When the reactor I3 is magnetically saturated, the impedance of its output circuit 45 is low and the lamps 69 are energized.

However, at the same time that current is supplied to the input circuit H of the reactor I 3 a difierence of potential is established between the terminals of the rectifier 25. During the half cycles of positive principal potential, therefore, the capacitors I5 and II are charged through the resistors I9 and 21 by reason of the difierence of potential that exists between the terminals of the rectifier 25. During the intervals of negative principal potential, during which the electric discharge device I is deenergized, the magnetic field built up by the current in the input circuit II begins to collapse, and a magnetizing current is accordingly induced therein, circulating through the resistor 23 and the rectifier 25. The magnitude of the potential impressed across the rectifier by reason of this current is considerably smaller than the magnitude of the potential by reason of which the capacitors were charged and the mean efiect of the complete cycle of principal potential is to impress a potential across the capacitors the polarity of which is the same as the polarity of the potential across the rectifier produced by the half-cycle of positive principal potential. Consequently, as long as the electric discharge device I remains in an energized condition the capacitors continue to be charged and the difierence of potential between their plates continues to be increased, assuming that the charging period is not extended long enough for the capacitors to become fully charged.

It will be noted that one capacitor I5 is charged through the two resistors Iii and El in series, while the other capacitor I? is charged only through one resistor 2!. The latter capacitor I? will, therefore, be raised to a higher potential than the former capacitor I5 and the time required to raise the latter capacitor I! to this potential will be less than the corresponding time required for raising the first capacitor IE to the corresponding potential. The relationship of the potentials impressed on the capacitors I5 and I! is, of course, dependent on the magnitudes of the capacitors. In the practice of my invention the capacitors I5 and I! are preferably of equal magnitude.

The capacitor I5 when it is charged simulates a battery, the negative terminal of which is connected to the capacitor 29 and the positive terminal of which is connected to the cathode. Hence, as the potential difference between the plates 01' the capacitor I5 is increased, the potential of the control electrode 27 is decreased or rendered more negative relative to the potential of the cathode I. After an interval of time predetermined by the relative values of the resistors I9 and 2! and the first capacitor I5, the potential of the control electrode 21 is so decreased that its value falls below the broken-line curve 43 of the graph shown in Fig. 2. If such a condition persists for the time occupied by more than one cycle of the principal potential the electric discharge device I is deenergized. When the electric viousiy explained. ing no longer connected to the bus-line during discharge device I is thus deenergized and the supply of current to the input circuit I I of the reactor !3 interrupted, the circulating magnetizing current in the circuit I I, 23, 25 dies out, and the impedance of the reactor I3 is correspondingly increased, thus deenergizing the illuminating units 49.

At the same time, the capacitor I5 would immediately begin to discharge through the resistors i9 and ZI and the rectifier 25, if the second capacitor I7 were not provided so as to maintain the capacitor I5 at the same or an even higher potential for at least a predetermined interval of time, as will subsequently be explained. If the second capacitor 1'! were omitted, therefore, the first capacitor l5 would be discharged sufficiently to bring the control potential above the critical value 28, thus reenergizing the electric discharge device I, almost immediately, or within a short interval of time after the potential of the control electrode 2'! attains such a value that the electric discharge I is deenergized.

Since the capacitor I5 would thus be discharged within a short interval after the electric discharge device is deenergized, and since the electric discharge device is deenergized within a short interval after the control potential falls below the limiting control potential, a slight decrease in the charge impressed on the capacitor I5 would revert the control potential to its initial value and the electric discharge device I would be reenergized as soon as the principal potential is of proper magnitude. The effect of such an arrangement would be such that the electric discharge device I would be deenergized for an interval of time corresponding approximately to only one negative half-cycle of potential impressed between its principal electrodes.

The second capacitor IT is incorporated for the purpose of maintaining the first capacitor I5 in its charged condition for at least an interval of time, which may be explained as follows. When the electric discharge device 5 is energized, it conducts currents only during the positive halicycles of principal potential, that is, when the left-hand bus-line 5 is positive and the righthand bus-line 9 is negative. Under these conditicns, the upper plates of the two capacitors i5 and I! will be substantially at the potential of the positive bus-line 5 during the positive l'alfcycles, (there being only a small potential-drop in the discharge device I), and the lower plates of the two capacitors will be connected, through the resistors I 9, 2| and 23, to the other bus-line 9.

Since the lower plates of the two capacitors are connected to each other through the resistor I9, the charging of the first capacitor I5 will lag behind the charging of the second capacitor I I. When the lower plate of the first capacitor i5 reaches a sufficiently high negative charge relative to the cathode I (which is at the same potential as the upper plate of the capacitor I5), the electric discharge device I is deenergized, as pre- The second capacitor ll, be-

the positive half-cycles of principal potential, immediately begins to discharge through the discharge-circuit 2 l, 25, but the first capacitor I5, being less highly charged, is at first at a smaller overall potential than the second capacitor I'a', so that the first capacitor I5 still continues to charge until t -e two capacitors I5 and I? are at the same voltage, whereupon the first capacitor also begins to discharge, the discharging-circuit including both resistors I9 and 2| and the rectifier 25.

When the first capacitor !5 is sufilciently discharged, the electric discharge device I will again be energized, as previously explained. But since the second capacitor I? has less resistance in its discharge-circuit than the first capacitor !5, it will have been discharging faster than the first capacitor I5, so that, at the moment of reenergization of the discharge device I, the first capacitor I5 has a larger overall potential than the second capacitor I I, and the first capacitor will continue to discharge during the first part of the charging period of the second capacitor IT, or until the two capacitors are at the same overall potentials.

In this way, it is possible to overshoot the critical grid-voltage line 43 by a considerable distance, thus avoiding flashing or energizing the electric discharge device every half-cycle, as would have been the case if the second capacitor I'I had been omitted, as previously explained. The length of time for a complete cycle depends upon the values of the resistors I8 and 2! and the capacitors I5 and IT. It is seen that by utilizing the second capacitor I I, the interval during which the elect "ic discharge device I remains in deenergized condition may be considerably increased and any desired period of time may be attained during which the illuminating units 49 are deenergized.

The capacitor 29 that is connected between the first capacitor and the control electrode 21 is charged from a potentiometer 5!, that is energized from the secondary 33 of the transformer across which it is connected through a rectifier 53, charging current flows in such direction that the capacitor-terminal nearest the controlelectrode 2'! is positive. The rectifier 53 is preferably a disk of the copper oxide type or of similar structure. 29, when once charged, will be maintained at a constant potential. The function of the capacitor 29 will be understood from a consideration of Figs. 3 and 4.

Fig. 3 shows the control-electrode potentials when the capacitor 29 is absent from the system and the first capacitor I 5 is directly connected to the control electrode 21, and Fig. 4 illustrates to the condition when the capacitor 29 is utilized. The broken-line curve 43 shown in Fig. 3 represents the limiting control potential with respect to the cathode-potential 55, and is the same as the curve 43 shown in Fig. 2. The horizontal lines 57 and 55 represent the maximum and minimum possible values of the control potential if the first capacitor I5 were successively charged to its fullest extent and completely discharged again. It is seen that, when the capacitor I5 first begins to charge, the control potential quickly falls below the limiting control potential (represented by the curve 43) and the electric di charge device is deenergized at the end of the first half-cycle, and remains deenergized for an interval corresponding to capacitor IE to return to a potential less (in amount) than the critical voltage 43. During the half-cycle that the discharge device I is energized, the capacitor I5 is charged to a negative potential considerably higher (in amount) than the small critical value 43, and must return to this critical value before the discharge device I will again be energized.

In the tubes utilized in the practice of my invention the potential dillerence between the lowest point of curve 43 anc iine is approximately 90 2 volts. As can be seen from Fig. 3, therefore,

t is to be noted that this capacitor i the interval during which the electric discharge device I is energized is comparatively small while the interval during which the discharge device is deenergized is comparatively large. It is often desirable, on the other hand, to more evenly distribute the periods during which the electric discharge device I, and therefore the illuminating units 49, are energized and deenergized and for this reason the third capacitor 29, is introduced.

The capacitor 29 is ordinarily charged to a potential that is opposite in polarity to the potential impressed on the first and second capacitors I5 and I? and the eifect of so charging the capacitor 29 is to decrease the effective limiting control potential and to drop the limiting control potential curve 43 to a position represented by the lower broken-line curve 59 as shown in Fig. 4. It is to be noted that, as the charge on the capacitor I5 is now varied from its minimum value to its maximum value, the control potential passes through the limiting control potential values considerably later than it did in the system represented by Fig. 3, and the interval during which the illuminating units are energized is considerably increased. By properly adjusting the movable tap SI of the potentiometer 5I, the relative proportions of the intervals during which the illuminating units 49 are energized and deenergized may be fixed at any desired values.

The function of the third capacitor 29 may be perhaps more clearly understood by noting that the effect of charging the third capacitor 29 as shown in the drawing is to apply a positive charge to the control electrode 21. In view of the fact that such a positive charge is applied to the control electrode 21, the negative charge that must be applied to counteract the positive charge and to cause the electric discharge device I to become deenergized must be correspondingly greater than it would be if the capacitor were merely a conductor. Consequently, the ultimate result of utilizing the capacitor is to shift the limiting control potential curve 43, to a lower position as already described.

In Figs. 5 and 6, the function of the alternating potential that is superposed by the windings 3| on the varying potential impressed through the capacitors I5 and I1 is illustrated.

Fig. 5 shows the potentials obtained when the windings 3I are not present, and Fig. 6 corresponds to the condition represented in Fig. .1.

If the windings 3I were not present in the system, under conditions as represented in Fig. 5,

an alternating potential is not superposed on the varying direct-current potential applied by the capacitors I5 and H and the ultimate result may be graphically represented simply by the movement of a horizontal line 63 from the minimum value of control potential to the maximum value of control potential. It is to be noted, however, that the limiting control potential curve 43 is concave, substantially symmetrical about the maximum ordinate 62 of the curve 4! representing the corresponding half-cycle of principal potential. As the control potential is varied and the line 63 representing its variation moves upwardly from its lowest value, it first intersects the limiting control potential curve 43 at its cen- -tral point and as it continues to move upwardly 5 intersects the time axis.

Thus it will be noted that as the control potential passes through the limiting potential the electric discharge device first becomes energized at a point corresponding to the point of intersection of the line 63 and the limiting potential curve 43, and the electric discharge device continues to be energized during the remainder of that half-cycle. The magnitude of the current transmitted by the electric discharge device I is of course dependent on the point of intersection of the line 63 and the curve 43. As the intersection point is moved towards the left the current output of the device is increased. The negative half-cycle of potential is next applied to the electric discharge device and the device remains deenergized. During the interval following the first intersection of the straight line 63 and the curve 43, the straight line 63 continues to move upwardly, and when the positive half-cycle of potential is again applied to the electric discharge device, the potential applied to the control electrode 2'! corresponds to a condition in which the straight line 53 intersects the brokenline curve 43 at two points 65 and 6'! on both sides of the axis 62 of symmetry. It is seen that by reason of the fact that the first intersection 65 is now towards the left of the axis 62 of symmetry, the output of the electric discharge device is increased over what it was during the preceding half-cycle.

The output of the electric discharge device I thus increases gradually when the control potential is increasing, and decreases gradually as the control potential decreases. The result-of this condition is that the desired cut-01f efiects are not produced, and the illuminating units 49 do not become energized and deenergized with the desired briskness, but rather fade in and fade out. 7

To obtain a highly desirable sharpness of operation, the alternating potential of the winding 3| is impressed on the varying control potential, and the resulting condition is represented graphically by curves such as are shown in Fig. 6. In this view the lower full-line curve 69 represents the varying control potential which is composed of the alternating potential provided by the secondary section 3i of the transformer 35 and the varying potential provided through the capacitors I5 and 29. As the negative charge on the capacitor I5 is decreased, the curve 69 is shifted from a position below the limiting control potential curve 43 to a position above the limiting control potential curve. t is seen that, as the per position and passes through the limiting control position curve 43, it intersects this curve at two points II and 13 adjacent to the points at which the principal potential curve 45 intersects the time axis. The corresponding physical condition is represented by a system in which the electric discharge device I when once energized delivers substantially its maximum current out put at once, and continues to deliver the maximum current output until the control potential is decreased to a value below the limiting control potential.

From the relationships shown in Fig. 6, it is evident that the winding 3!, as rep-resented by the resultant control-potential curve 69, is impressing a negative half-wave of potential on the control electrode 2? at the same time that the alternating-current source, as represented by the principal-potential curve 4!, is impressing a positive half-cycle on the anode 3.

A specific system that I have found to be of considerable utility in the practice of my invention incorporates an electric discharge device I which is sold commercially under the name of Grid-Glow tube KIT-621. This device has an oxide-coated filament l, and is capable of passing an average of approximately .64- ampere. The system is ordinarily energized. from 110 volt, 60- cycle source which feeds the bus lines 5 and 9. The capacitors I5 and i1 connected in series with the output of the electric discharge ievice are of the order of 2 microfarads, while the resistors l9 and 21 associated with the capacitors l5 and ii may vary from megohm to 10 megohms. The resistor 23 is of the order of 100 ohms. The capacitor 29 that is maintained in a constant charged condition is ordinarly of the order of microiarad and is maintained at a potential which may be varied from approximately zero to 50 volts, depending on the particular relationship that is desired between the energized periods and the deenergized periods of the illuminating units #9. The alternating potential superposed on the varying potential by the Winding Si is of the order of 5 volts, and the transformer 35 whereby the potential is applied is wound to yield this potential. The grid resistor 39 is of the order of 5 megohms.

My invention has been shown herein as applied to a specific system. It is apparent that a number of modifications may be made without changing the essential features of my invention. For example, the asymmetric discharge device i, that is shown as being utilized in the practice of my invention and that is actually utilized, in the preferred practice of my invention, may be replaced by another asymmetric device. It may be replaced, for example, by a device incorporating two cathodes and a control electrode which is adapted to pass current during both half cycles of the potential applied to the principal elec trodes.

Moreover, while I prefer to utilize a gas-filled electric discharge device in the practice of my invention, such a device is not absolutely essential and may be replaced by a high vacuum electric discharge device, or by a mercury pool device, or by other electric discharge devices of similar character.

Finally it should be noted that while my invention has been shown and described herein as applied to a load comprising illuminating units, it may be applied to a great number of other types of leads. I have, for example, found it to be highly useful in the timing of the operation of spot-welding equipment.

Although I have shown and described certain specific embodiments of my invention, I am fully aware that many modifications thereof are possible. My invention, therefore, is not to be restricted except insofar as is necessitated by the prior art and by the spirit of the appended claims.

I claim as my invention:

1. In combination an electric discharge device having an anode, a cathode and a control electrode, a capacitor connected in seri s with said anode and cathode and between said cathode and said control electrode, the connection in series with said anode and cathode including a resistor connected in series with said capacitor, whereby said capacitor is charged through said resistor by the current transmitted between said anode and cathode, and another capacitor connected in series with said anode and cathode but shunting said first-named capacitor and said resistor.

2. In combination an electric discharge device having an anode, a cathode and a control electrode, said device being of the type having an energized and a deenergized condition and being capable only of abrupt transition from one condition to the other, means for impressing a periodic potential between said anode and cathode, a capacitor connected in series with said anode and cathode and between said cathode and said control electrode, a resistor connected in series with said anode and cathode and means, including means for impressing an alternating potential of uniform amplitude, coupled between said control electrode and said cathode, for energizing said discharge device precisely at a predetermined instant in each cycle of said periodic potential by applying potential between said control electrode and said cathode which is greater than 20 a predetermined critical value.

3. In combination with a voltage source, a first capacitor and means to control the rate of flow of a charging current to said capacitor from said source, a second capacitor connected to said first capacitor through an impedance and means for charging it at such a rate that its voltage rise is more rapid than that of said first capacitor, and means for interrupting the flow of charging current from said source to both said capacitors when said first capacitor is charged to a predetermined voltage.

4. In combination with a voltage source, a first capacitor and means to control the rate of flow of a charging current to said capacitor from said source, a second capacitor connected to said first capacitor through an impedance and means for charging it at such a rate that its voltage rise is more rapid than that of said first capacitor, means for interrupting the flow of charging current from said source to both said capacitors when said first capacitor is charged to a predetermined voltage, and means for gradually discharging said second capacitor.

5. In combination with a voltage source, a first capacitor and means to control the rate of flow of a charging current to said capacitor from said source, a second capacitor connected to said first capacitor through an impedance and means for charging it at such a rate that its voltage rise 1 is more rapid than that of said first capacitor, means for interrupting the flow of charging current from said source to both said capacitors when said first capacitor is charged to a predetermined voltage, means for gradually discharging said second capacitor, and means for restarting the flow of charging current to said capacitors when said first capacitor is charged to a predetermined voltage.

6. In combination with an electrical discharge device having an anode, a cathode and a control electrode, a capacitor having one of its terminals connected to said cathode .and its other terminal connected to a load circuit supplied with current by said device, a connection from the last-named terminal of said capacitor to said control electrode, and means for discharging said capacitor when said discharge tube is substantially nonconductive at a rate slower than the charging rate of said capacitor when said discharge tube is conductive.

7. In combination with an electrical discharge device having a pair of principal electrodes and a control electrode, means for impressing a pcriodic voltage between said principal electrodes,

a timing circuit comprising a capacitor connected in circuit between said control electrode and one of said principal electrodes, a rectifier and an impedance connected to charge said capacitor by means of current flowing through the principal electrodes of said discharge device, and a second capacitor connected in shunt around said impedance and said first capacitor.

8. In combination with a source of periodic voltage, an electrical discharge device having a pair of principal electrodes and a control electrode, a timing circuit comprising a capacitor connected in circuit between said control electrode and one of said principal electrodes, a rectifier and an impedance connected to charge said capacitor by means of current flowing through the principal electrodes of said discharge device, a second capacitor connected in shunt around said impedance and said first capacitor, and a constant potential source connected in series in the circuit between said first capacitor and said control electrode.

9. In combination with a source of periodic voltage, an electrical discharge device having a pair of principal electrodes anda control electrode, a timing circuit comprising a capacitor connected in circuit between said control electrode and one of said principal electrodes, a rectifier and an impedance connected to charge said capacitor by means of current flowing through the principal electrodes or said discharge device, a second capacitor connected in shunt around said impedance and said first capacitor, and a source of periodic voltage of the same frequency as said firstnamed source connected serially in the circuits with said first capacitor and said control electrode.

10. In combination with .a voltagesource, a first capacitor and means to control the rate of flow of a charging current to said capacitor from said source, a second capacitor connected to said first capacitor and means for charging it at such a rate that its voltage rise is more rapid than that of said first capacitor, and means for interrupting the flow of charging current from said source to both said capacitors when said first capacitor is charged to a predetermined voltage.

11. In combination with a voltage source, a. first capacitor and means to control the rate of flow of a charging current to said capacitor from said source, a second capacitor connected to said a rate that its voltage rise is more rapid than that of said first capacitor, means for interrupting the fiow of charging current from said source to both said capacitors when said first capacitor is .first capacitor and means for charging it at such 20 charged to a predetermined voltage, means for 25

Images