US3356888A - Two-electrode spark gap with interposed insulator - Google Patents

Description

Dec. 5, 1967 K. J. GERMESHAUSEN ET AL TWO-ELECTRODE SPARK GAP WITH INTERPOSED INSULATOR Original Filed Dec. 27. 1960 5 Shee ts-Sheet 1 INVENTORS KENNETH J. GERMESHAUSEN JOHN TURNER 1967 K. J. GERMESHAUSEN ET AL 3,356,888

TWO-ELECTRODE SPARK GAP WITH INTERPOSED INSULATOR Original Filed Dec. 27, 1960 5 Sheets-Sheet 2 us l5 I 5 POWER SUPPLY ll? 47 liT STROBOSCOPE FREQUENCY Eggs? CONTROL Fig. 3.

I .INVENTORS KENNETH J GERMESHAUSEN JOHN L. TURNER 1967 K. .1. GERMESHAUSEN E AL 3,356,888

TWO-ELECTRODE SPARK GAP WITH INTERPOSED INSULATOR Original Filed Dec. 27, 1960 v 3 Sheets-Sheet 5 VOLTAGE GRADIENT FACTOR INCH o l I I RADIUS OF CENTER CONDUCTOR (THOUSANDTHS OF AN INCH) Fig. 5.

INVENTORS KENNETH J GERMESHAUSEN JOHN L. TURNER United States Patent 3,356,888 TWO-ELECTRODE SPARK GAP WITH INTERPOSED INSULATOR Kenneth J. Germeshausen, Weston, and John L. Turner,

Needham, Mass, assiguors to EG & G, Inc., a corporation of Massachusetts Original application Dec. 27, 1960, Ser. No. 78,587.

Divided and this application June 24, 1964, Ser.

7 Claims. (Cl. 313-325) The present invention relates to gaseous-discharge devices, and more particularly to spark gaps. This application is a division of co-pending application Ser. No. 78,- 587, filed Dec. 27, 1960, by the applicants herein.

Applicant Germshausens co-pending application, Ser. No. 598,325, filed on July 17, 1956, now US Letters Patent No. 2,977,508 issued on Mar. 28, 1961, discloses a novel gaseous-discharge device and system. The subject invention is useful in this and other gaseous-discharge devices to materially improve the performance thereof as hereinafter pointed out.

There as long been a problem of initiating regular and uniform discharges in gaseous-discharge flashtubes. It is believed that in order to initiate a discharge, there must be ions or free electrons available in the discharge region which can be affected by the electric field produced at the trigger electrode when the trigger impulse is applied. The electric field drives the ions or electrons through the gas toward one of the principal electrodes and the resulting collisions with gas molecules produce ionization thereby initiating the discharge. In the presence of ambient light, the problem of initiation is less severe because there are phontons available to produce some photoelectrons which will be affected by the electric field produced at the trigger electrode. In any particular case, there may or may not be sufi'icient photoelectrons to effect consistent breakdown, but it can be said in general that the problem becomes more serious as the intensity of the ambient light decreases, and in substantially total darkness the problem is greater. Because it is greatest in total darkness, we shall refer to this phenomenon as the dark-start problem, but this term also includes the full range of ambient light as well as complete darkness.

The dark-start problem becomes even more serious in flashtubes operating at low average power levels, such as single flashes; irregular flash frequencies as when the flashtube is triggered in response to the random occurrence of an event; or in stroboscopes or other repetitive fiashprecision applications at very low flash rates in the order of, for example, ten flashes per second or less. In such cases, flashtubes have demonstrated a tendency to skip, to fire late, or, at the low flashing rates, to fire irregularly or not at all.

It is, therefore, an object of the invention to provide a novel spark-gap having optimum discharge characteristics.

Another object of this invention is to provide a sparkgap for operation within a gaseous-discharge flashtube. In summary the present invention consists of a spark-gap of unique design in which a pair of electrodes are separated from each other by an insulative material across which a discharge may take place when the static breakdown potential is exceeded by pulsed energy.

The invention will now be described in connection with the accompanying drawings FIGURES 1A and 1B of which are standard projections of a spark-discharge device, hereinafter referred to as sparker, constructed in accordance with a preferred embodiment, FIGURE 1A being a side view partially cut away to illustrate details of ice construction, and FIGURE 1B being an end view of the right-hand end of the sparker shown in FIGURE 1A;

FIGURE 2 is a perspective view, partially cut away, of a gaseous-discharge flashtube utilizing the sparker shown in FIGURE 1;

FIGURE 3 is a schematic circuit diagram illustrating a preferred electrical system for operating the flashtube of FIGURE 2;

FIGURE 4 is a plan view, partially cut away, of the gaseous-discharge flashtube of FIGURE 2, showing the disposition of the tube elements;

FIGURE 5 is a graph depicting the effects of variations in the design of the sparker.

FIGURES 1A and 1B show a sparker, indicated generally by reference designator 50, which consists of a cylindrical block of an electrically insulative material 51 such as, for example, ceramic and having a hollow cylindrical portion 52 extending the length of the ceramic block '51 and substantially coaxial thereto. A conductive-metal probe 53 of, for example, tungsten, is inserted in the hollow portion 52, terminating at the end of the block at the right-hand side shown in FIGURE 1A and sealed to and extending out of the block 51 at the left-hand end and for purposes hereinafter indicated. A conductive metal ring 54, of, for example, Kovar, is firmly attached to the exterior surface of the block 51 at its right-hand end by brazing, crimping or the like. Attached to the metal ring 54, as by Welding or the like, is a conductive metal lead 55 Whose function will be shown below.

The sparker 50 is essentially a two-electrode spark-gap designed to produce an electric discharge between the probe 53 and the metal ring '54 when an electric impulse is fed to the probe 53, The sparker has been designed for use in a gaseous-discharge flashtube as shown in FIG- URE 2, and also to operate from the trigger impulse of the flashtube. It is, therefore, necessary that the sparker be highly eflicient and dependable. The design must be such that a discharge is assured each time the flashtube is triggered. In order to make the most efficient use of the voltage of the impulse fed to the probe 53 to effect breakdown, it is essential to develop as intense an electric field as is practicable at the surface of probe 53. Many factors are involved in designing a spark discharge device having a very intense electric field adjacent to one of the electrodes, and for this reason, a theoretical discussion of the subject will help to explain the advan tages of the subject configuration.

Starting with the case of a pair of parallel-plane conductors of infinite extension, separated from each other by a distance d and subjected to a voltage V, a uniform electric field is produced which may be expressed as =V/d Where E is the electric field at each point be tween the conductors.

It is well known that if a coaxial configuration is substituted for the parallel-plane conductors, but maintaining the voltage and the separation distance between the conductors the same, the electrical field at the surface of the center conductor is more intense than that found in the aforesaid parallel-plane system. We have discovered that the intensity of the electric field at the surface of the 7 center conductor can be further increased many fold by inserting an insulating body between the center and outer conductors and optimizing the design of the unit.

The six most important parameters which effect the intensity of the electric field at the surface of the center conductor are 1) the dielectric constant (K) of the insulating material disposed between the center and outer conductors; (2) the radius of the center conductor (r (3) the inside radius of the insulating material (r (4) the outside radius of the insulating material (r (5) the inside radius of the outer conductor (r.,); and (6) the voltage (V).

In order to simplify this discussion we shall make certain of these parameters constant and we shall use the quantities of a sparker designed for use with the fiashtube of FIGURE 2, as the constants. The insulative material should preferably be one having a high dielectric constant in the range of, say, 8.60 to 9.50 or higher, and it should, of course, have sufficient dielectric strength to withstand the electric fields produced therein. In the sparker 50, we used a high alumina ceramic (94% A1 having a dielectric constant of approximately 8.80. Our choice of ceramic cylinders was limited by the small variety in sizes that were commercially available, thereby limiting our choice of r and r Accordingly, we choose a ceramic cylinder having an outside radius (r of 0.020 inch and an inside radius (r of 0.005 inch.

We have discovered that with respect to the radius of the outer conductor (r.,) the most intense electric field is found at the surface of the center conductor when the outer conductor is sealed to the outer surface of the ceramic cylinder. If there is a spacing between the outer conductor and the ceramic, then as this spacing is increased, the intensity of the electric field at the surface of the center conductor decreases. For this reason, r., was made as close to 1' as practical. Therefore, 1' and 11, may be considered equal.

Theoretically, for such a coaxial structure of infinite lengths and perfect concentricity, the electric field E at the surface of the center conductor is given by the following formula:

(1) E=VG where Therefore by using the value G, the voltage gradient factor which may be defined as that factor which when multiplied by the voltage V applied between the inner and outer conductors, results in the electric field E at the surface of the center conductor, the effect of varying the radius of the center conductors r upon the voltage gradient factor, G, can be shown independent of the voltage applied. By Formula 1 the intensity of the electric field may be obtained. FIGURE 5 is such a graph.

The curve of FIGURE 5 which was plotted with r r r, and K held constant demonstrates the variation in the voltage gradient factor G and therefore, in the electric field E also, as the center conductor varies from an infinitesimally thin wire to one approaching the size of the hollow portion of the ceramic insulator r It can be seen that the maximum G is obtained when the radius 1', is a minimum. Obviously, when the radius r reaches zero, there is no center conductor and G is also zero. This curve tends to show that the most intense electric field is obtained by using a center conductor with the smallest radius. This is theoretically true, but practical disadvantages sometimes make it more advantageous to utilize the increase in G obtained by large radius center conductors as shown by the curve at the right-hand side of FIGURE 5. The disadvantages involved in using extremely fine wire for the center conductor include the fragility of such a wire causing erosion which, in turn, results in a limited service life and the difficulty in handling, assembling, and accurate and stable positioning.

These disadvantages may be less important in other applications of sparkers but for use with fiashtubes of the type hereinafter described, practical considerations require that the up-swing of the curve at the larger radii be used. It should be pointed out, however, that the increase in G with the increase of r does not continue to the point where r, is a maximum or r equals r It has been found experimentally that when r and r are equal, that is, when the center probe is sealed to the ceramic insulator, the sparker requires a very high voltage to cause breakdown (low effective G). The reason for this is easily found in the theoretical analysis of the electric field within such a sealed, idealized structure and the electric field throughout the volume of the sparker is identical to that in the case of no ceramic at all. It should be further pointed out that there may be another restriction in the approach of r to r to effect maximizing of G on the right-hand side of the curve of FIGURE 5. As the gap between r, and r approaches approximately one mean-free-path of ions and electrons in the gas under the conditions of use, an incipient discharge may be starved for atoms or molecules to ionize, and so the idealized minimum of breakdown voltage may not be achieved for this reason.

The curve of FIGURE 5 indicates that the most intense electric fields on the surface of the center conductor is obtained when the radius of center conductor is somewhat less than the inside radius of the insulator. Any point, however, on this curve is a substantial improvement over the corresponding values for coaxial systems without a ceramic insulator, and parallel-plane systems. This is graphically illustrated in FIGURE 5 by the point A which shows a center conductor 0.0035 inch has voltage gradient factor of approximately 556 per inch while the corresponding points for a coaxial system without an insulator is approximately per inch (point B) and for the parallelplane system is approximately 61 per inch (point C). The electric field indicated by point A is more than three times that of point B and more than nine times that of point C. The radius 0.0035 inch was chosen for this example because it is the value which we used in the sparker shown in FIGURE 1A. A center conductor with a greater radius could have been chosen to further increase the electric field, but it was convenient for production purposes to use the value chosen in order to use a wire size identical to that of the trigger probes, 21, 23, 25, 27 and 29 (see FIG- URE 2); and also for ease of assembly, still giving adequately low breakdown voltage, commensurate with tube triggering requirements and trigger probe breakdown voltages.

The curve of FIGURE 5 is theoretical for a coaxial unit of infinite length and perfect .concentricities and, therefore, any physical embodiment of these principles will, of course, result in modifications in the curve. Even with these modifications, the curve indicates the relationship between the named factors and is highly useful for this purpose.

Although the probe 53 is shown terminating at the end of and disposed in the center of, the ceramic block 51, such termination and disposition are not essential and modifications thereof may be resorted to without departing from the spirit and scope of this invention.

The ceramic block 51 and the probe 53 are held in fixed relationship by sealing the probe 53 to the left end of the ceramic block as shown in FIGURE 1A, and by ring 54 at the right end which is attached to the ceramic 51 and to the cathode support 17 (see FIGURE 2).

The flashtube in FIGURE 2 is shown having a glass, fused quartz of similar light-transparent envelope 1 with a planar top and cylindrical side walls. This particular configuration has the advantages of maximum light output through the planar top and minimum space required for mounting the flashtube. The gaseous medium, such as exnon and the like, may be sealed within the envelope 1, as, for example, by closing off the gas-filled inlet tube 3, in the base of the envelope 1. For the purposes of variablefrequency stroboscopes and the like, it is preferable that the gas be maintained at a high pressure of the order of, say, one-third to three atmospheres, more or less. An anode electrodeS and a cathode electrode 7, preferably both of the same construction, are supported spaced from one another within the envelope 1 by conductive supports 15 and 17 that extend outside the base of the envelope through the bottom wall thereof. The cathode 7 and the anode are preferably both of the sintered cold-cathode type disclosed in applicant Germeshausens prior United States Letters Patent No. 2,492,142 issued Dec. 27, 1949, and they are illustrated in FIGURE 2 as substantially similar rectangular-surface pills disposed substantially parallel to one another. Such sintered electrodes are capable of withstanding the gaseous bombardment inherent in the operation of closely spaced electrodes at substantial voltages in a high-pressure gas.

Disposed within the space between the substantially parallel opposing surfaces of the anode 5 and the cathode 7 are a plurality of probe-type trigger or control electrodes 21, 23, 25, 27 and 29, of, for example, tungsten. While five such trigger electrodes are illustrated, more or less trigger electrodes may be employed consistent with the separation between the anode 5 and the cathode 7 and the hereinafter described required discharge-conducting or guiding function of the plurality of trigger electrodes. Trigger electrodes 21, 23, 25, 27 and 29 are supported by and electrically attached to, by welding or the like, conductivesupport pins 121, 123, 125, 127 and 129 respectively. These pins enter the fiashtube through the base thereof at points spaced from the side walls. The trigger electrodes are attached to the pins at approximately right angles, with the free ends of the trigger electrodes terminating in the space between the anode 5 and the cathode 7.

The free ends of trigger electrodes 23, 25 and 27 lie along a straight line L (shown dotted) between the center points at the top of the facing surfaces of the anode 5 and the cathode 7. The free ends of trigger electrodes 21 and 29 lie along a line (not shown) extending from the top corners of the facing surfaces of the anode 5 and the cathode 7 on the same side of the envelope that these trigger electrodes 21 and 29 are attached to their respective supports 121 and 129. This configuration is shown more clearly in FIGURE 4. The free ends of trigger electrodes 21 and 29 are so terminated to insure that the arc breakdown between these trigger electrodes and the principal electrodes adjacent to each, takes place only at the free ends of these trigger electrodes and not at different places along the trigger electrodes which might be the case if these electrodes were extended to line L.

Sparker 50 is disposed within the gas of the fiashtube but remote from the discharge path between the cathode 7 and the anode 5. Sparker 50 is positioned so that the light produced when it is energized will impinge, either immediately or by reflection, on one of the principal electrodes to assist in initiating a discharge in the fiashtube.

In accordance with the present invention, a voltage is applied between the anode 5 and the cathode 7 that is of if itself insuflicient to produce a discharge therebetween through the gas. A trigger impulse is applied to break down the gas in the neighborhood of either the anode 5 or the cathode 7 between the same and the adjacent trigger electrode 29 and 21, respectively. In this embodiment, breakdown is first initiated at the cathode 7. Simultaneously, the trigger pulse is also fed to probe 53 of sparker 50 through conductive-pin support 60 causing a discharge to take place between probe 53 and metal ring 54 which is shown connected to the cathode support 17 by conductive wire 55. The electrode arrangement and geometry of the sparker 50, as has been pointed out, are such that the required breakdown potential between probe 53 and metal ring 54 is less than that between cathode 7 and adjacent trigger electrode 21 thereby insuring that a discharge will take place between probe 53 and ring 54 each time the fiashtube is triggered. The light from sparker 50 impinges upon the cathode 7 at the same time that the trigger impulse is applied to adjacent electrode 21. As the light from the sparker 50 impinges upon the cathode 7, electrons are released from the cathode 7 which are subjected to the electric field produced by the trigger impulse at trigger electrode 21. The electrons, under the force of the electric field, collide with the gas molecules thereby producing ionization of the gas and the initiation disclosed in the aforesaid co-pending application of applicant Germeshausen.

The electric circuits for operating the fiashtube of the present invention may assume the form illustrated in FIG- URE 3 in which a flash capacitor or capacitors 6 is or are charged through a limiting impedance 4 "from a powersupply energy source 2. The upper and lower terminals of the capacitor 6 are shown connected by conductors and 117, respectively, to the pins 15 and 17 connected with the anode 5 and the cathode 7. The voltage thus developed between the anode and cathode is, as'before explained, insuflicient in and of itself to effect a discharge 1 therebetween. A trigger circuit 8 may comprise a thyratron or other switching circuit adapted to discharge a capacitor 10 through the primary winding 46 of a trigger transformer 45 to produce a trigger pulse. The trigger circuit 8 may be controlled by a stroboscope frequency control 12, as of the type disclosed in United States Letters Patent No. 2,331,217 issued to applicant Germeshausen on Oct. 12, 1943. The trigger pulse so produced causes sparker 50 to fire and initiates or triggers the successive electrode breakdown before-discussed in order to permit the energy stored in the capacitor 6 to discharge between the anode 5 and the cathode 7, thereby to produce a high intensity brief flash or repetition of flashes. Trigger circuits of this character are also disclosed in United States Letters Patent No. 2,478,901, issued on Aug. 16, 1949 to Harold E. Edgerton.

Although We have described our invention with a certain degree of particularity in connection with the preferred embodiment, the invention has a much broader scope and all changes and modifications are .deemed to fall within the spirit and scope of this invention as defined in the appended claims.

We claim:

1. A spark gap comprising, in combination:

a first electrode;

a second electrode spaced from said first electrode;

an electrically-insulative member sealed to said first electrode and spaced from said second electrode a distance greater than the mean-free-path of the ions and electrons in the gas therebetween; and

means for connecting a source of electrical energy across said first and second electrodes.

2. A spark gap comprising, in combination:

an electrically-insulative member having an opening therethrough;

a first electrode disposed within said opening and spaced from said member a distance greater than the meanfree-path of the ions and electrons in the gas therebetween;

a second electrode sealed to the external surface of said member adjacent one end of said opening and spaced from the first electrode; and

means for connecting said first and second electrodes across a source of electrical impulses of suflicient potential to cause a discharge to take place therebetween.

3. A spark gap comprising, in combination:

an electrically-insulative member having an opening therethrough;

a first electrode disposed within said opening, spaced from said member a distance greater than the meanfree-path of the ions and electrons in the gas therebetween and terminating adjacent one end of said member;

a second electrode sealed to the external surface of said member, adjacent said end of the member and spaced from said first electrode; and

means for connecting said first and second electrodes across a source of electrical impulses of sufficient potential to cause a. discharge to take place therebetween.

4. A spark gap as claimed in claim 3 and in which the said second electrode passes completely around said member and said first electrode.

5. A spark gap as claimed in claim 4 and in which the electrically-insulative member, the opening in said member and the first electrode are all substantially concentric cylinders, and the said member has a high dielectric length having a cylindrically shaped opening passing therethrough from a first end to a second end thereof;

a first cylindrically shaped electrode disposed within said opening, sealed to the first end of said member and terminating adjacent the second end of said member;

a second cylindrically shaped electrode of a length less than said predetermined length, sealed to the external surface of said member and terminating adjacent the second end thereof, said first electrode being spaced 5 from said member for a length at least equal to the I length of said second electrode, said spacing being greater than the mean-free-path of electrons and ions in the gas therebetween, said second end of said member and the adjacent terminating first and second electrodes all lying substantially along a common plane; and

means for connecting said first and second electrodes across a source of electric impulses of sufiicient potential to cause a discharge to take place therebetween.

References Cited UNITED STATES PATENTS DAVID IQ GALVIN, Primary Examiner.

Claims (1)

1. A SPARK GAS COMPRISING, IN COMBBINATION: A FIRST ELECTRODE; A SECOND ELECTRODE SPACED FROM SAID FIRST ELECTRODE; AN ELECTRICALLY-INSULATIVE MEMBER SEALED TO SAID FIRST ELECTRODE AND SPACED FROM SAID SECOND ELECTRODE A DISTANCE GREATER THAN THE MEAN-FREE-PATH OF THE IONS AND ELECTRONS IN THE GAS THEREBETWEEN; AND MEANS FOR CONNECTING A SOURCE OF ELECTRICAL ENERGY ACROSS SAID FIRST AND SECOND ELECTRODES.

Images