US1232467A - Spark-gap. - Google Patents

Description

F. W. PEEK, In.

SPARK GAP.

APPLICATION HLED AUG-l9, 191s.

Patented July 3, 1917.

Inventor: FranK Peek. Jr, y Mp His flttorneg.

UNITED STATES PATENT OFFICE.

FRANK W. PEEK, JR., OF SCHENEC'IADY, NEW YORK, ASSIGNOR TO GENERAL ELECTRIC COMPANY, A CORPORATION OF NEW YORK.

SPARK-GAP.

Specification of Letters Patent.

Patented July 3, 191 '7.

Application filed August 19, 1915. Serial No. 46,398,

To all whom it may concern:

Be it known that I, FRANK W. PEEK, Jr, a citizen of the United States, residing at Schenectady, county of Schenectady, State of New York, have invented certain new and useful Improvements in Spark- Gaps, of which the following is a specification.

My invention relates to lightning arresters of the type having a spark gap for relieving transmission lines and other conductors of transient voltages and similar abnormal conditions.

Transient voltages, surges, and other abnormal conditions on the conductor of a transmission line or piece of electrical apparatus impose severe strains on the various dielectrics, such as air and insulation, which adjoin the conductor. To relieve these strains spark gaps are used and are so set that the dielectric of the spark gap will withstand the stress of normal voltage on the conductor but will rupture and permit a discharge to ground when subjected to some higher voltage. The voltage at normal frequency at which spark over or complete dis charge occurs is usually called the breakdown voltage of the gap. If a transient voltage rises at about the same rate as the normal voltage at normal frequency, that is, the front of the wave of transient voltage is not much steeper than the front of a wave of normal voltage, the dielectric of the spark gap will rupture before the transient voltage can rise very much above the breakdown voltage. To rupture any dielectric, whether gaseous, liquid or solid, energy must be expended in the dielectric and therefore a time element or time lag is introduced into the action of the spark gap. On account of the time lag, the application of a voltage which rises at a very rapid rate, as for example, a transient impulse, does not cause a spark gap constructed in the usual manner to spark over or completely break down at the instant the break down voltage is reached. The voltage overshoots or continues to rise above the breakdown value during the period of time required to rupture the dielectric of the spark gap. The more rapidly the transient voltage is rising the greater the height it will attain during that period of time. In a spark gap constructed in the usual manner, this period of time may be great enough to permit a traveling wave to pass the spark gap and reach the station apparatus, or a transient voltage with a steep wave front or a voltage impulse may rise to a value far above normal before the gap breaks down. In some cases the wave form of the transient voltage may be such that the spark gap does not respond to it as readily as the insulation of'some of the apparatus, and the apparatus is injured before the spark gap has time to discharge. This time lag seems to be due to the fact that in spark gaps constructed in the usual way, local break down, such as brush discharge, or corona, first occurs in the path of the arc at some lower voltage than spark over, or before spark over or complete discharge takes place.

The object of my invention is to provide a protective device which is faster in action and more elficient than spark gaps constructed in the usual manner, particularly in responding to rapidly rising voltage such as transient voltages and voltage waves with steep fronts. A further object is to provide a protective device suitable for installation out of doors and practically unaffected by rain and changes of weather. To this end, I proportion and arrange the spark gap in such a manner that the air or other dielectrio is subjected to electrostatic stress in such a manner that complete discharge or spark over occurs without any local discharge such as brush discharge or corona in the path of the arc. A spark gap constructed in accordance with my invention is more eflicient than a spark gap constructed in the usual manner, as it is very much faster, will discharge very much more quickly and for a given line voltage the electrodes can be set much closer together. The dielectric may be stressed in the desired manner in various ways, but I have found that the best results are obtained if the electrodes of the spark gap are so proportioned and positioned with relation to each other that a large part of the electrostatic stress between the proximal or nearest portions of the electrodes is so uniformly distributed that no local break Clown of the air, such as brush discharge or corona, occurs in the ath of the arc prior to spark over or comp ete breakdown of the gap when potential is applied to the elec trodes. The desired distribution of the electrostatic stress is best secured by the use of electrodes which have their coiiperating or proximal surfaces in the form of surfaces of revolution, such a cylinder or sphere. The electrostatic stress in the gap between such electrodes will be suiliciently uniform if the length of the gap is within certain definite limits, which depend upon the size and shape of the electrodes. For example, in air at atmospheric pressure at sea level, spherical electrodes should not be farther apart than about twice the radius of the sphere, else the gap will be slow. I have also found that if the electrodes are placed too close together the gap will be slow and inefficient. Spherical electrodes, for example, with air as the dielectric, should not be closer than three tenths of the square root of the radius of the sphere measured in centimeters and better results are obtained if the electrodes are separated at least one half of the square root of the radius of the sphere measured in centimeters. The exact spacing necessary to secure the best results depends somewhat upon the density of the air. Ordinarily I prefer to connect a spark gap constructed in accordance with my invention and having two electrodes properly placed with relation to each other as above, in series with some current controlling means such as a resistance, an electrolytic cellor similar device. It is also desirable in some cases to associate with a sphere gap constructed in accordance with my invention one or more gaps of different types connected in parallel with the sphere gap in order to minimize the efl'e'cts of weather and other variable conditions on the protection afforded the conductor.

My invention will best be understood in connection with the accompanying drawing in which, merely for purposes of illustration, I have shown diagrammatically some of the various forms in which my invention may be embodied and in which Figure 1 is a diagram of a sphere gap constructed in accordance with my invention; Fig. 2 of a modified form of gap embodying my invention; Fig. 3 a diagram of a composite spark gap embodying my invention and proportioned to respond to practicall any abnormal condition on the line, and ig. 4: is a diagram matic illustration of another modification, showing partly in longitudinal section another form of high speed gap which responds very quickly to abnormal potentials.

In the particular form of device shown in Fig. 1, a conductor 1 such as a conductor of a transmission line has a discharge path through an electrolytic cell 2 which for purposes of illustration is shown connected to ground. In series with the cell and between it and the conductor 1 is a sphere spark gap constructed in accordance with my invention and adapted for use in the air at sea level. This sphere gap comprises two spherical electrodes 3 separated by an air gap 4, of which the length does not exceed the diameter of eitherof the spheres 3, but in air at sea level is greater than one third or one half the square root of the radius of either f them measured in centimeters. In practice I prefer to make the spheres 3 the same size and make the gap t about equal to their radius. In some cases I may shunt the gap by a condenser 5 while in other cases the condenser is omitted. In a sphere gap constructed in this manner, the electrostatic stress between proximal surfaces of the electrodes appears to be sufficiently uniform so that no local discharge such as brush discharge or corona, occurs in the path of the are between said surfaces prior to spark over and complete discharge. As a result thereis no appreciable time lag between application of break down potential and complete break down of the gap. If the electrodes 3 are spaced either farther apart or closer together than the limits just mentioned, the complete break down of the gap is preceded by local discharge in the path of the arc and the objectionable time lag of the common forms of spark gap is present.

A sphere gap is not particularly efficient in extinguishing the are which follows the discharge through the gap, and therefore, I prefer to use the electrolytic cell 2 in series with it. As soon as the discharge through the gap 4 and the cell 2 to ground has reduced the potentia'lof the line to normal the electrolytic cell 2 prevents further flow of current.

The shape of the electrodes of the gap may be modified in various ways but the best results are obtained with electrodes which have their proximal surfaces in the form of surfaces of revolution. The particular form which I prefer is the sphere. Whatever the absolute size of the sphere, the minimum time lag in air at sea level is obtained when the spacing between their proximal surfaces is between twice the radiusof the sphere and one half of the square root of that radius measured in centimeters. In particular, the best results are obtained when the length of the spark gap in air is about equal to the radius of the smaller sphere. A horn gap which is set to have a much lower break down "oltage at normal frequency than the sphere gap may be connected in parallel with the sphere gap yet transient voltages with steep wave fronts will discharge through ,the sphere gap instead of the horn gap in spite of the higher break down Voltage of the sphere gap at normal frequency.

A sufficiently uniform distribution of enough of the electrostatic flux to eliminate a great deal of the time lag can also be obtained with electrodes having substantially plane proximal surfaces. I prefer to make such electrodes in the form shown in Fig. 2, in which the electrodes 5 are in the form of disks having their edges curved to a radius sufficiently great to prevent formation of corona at the normal potential of the line. As in Fig. 1, the gap is preferably connected in series with an electrolytic cell 2 to prevent flow of current at normal potential.

The break down voltage of a sphere gap at normal frequency is lowered, if the spheres become wet, and consequently the electrodes should preferably be kept dry by a suitable housing. If the gap must be exposed to the weather, the spacing of the electrodes must be greater than Where it is protected. I may provide a gap of a different kind in parallel with the sphere gap to secure maximum protection under all conditions of weather and to secure other desirable results For example, as shown in Fig. 3, the main sphere gap 4: is connected in shunt with a horn gap comprising two elongated electrodes 6 which approach each other closely at points 7 intermediate their ends and diverge in both directions from these points to form a horn gap of the usual type. In some cases spheres 8 or other electrodes having their proximal surfaces in the shape of surfaces of revolution are mounted adjacent the points 7 of the horns and positioned with relation to each other to form a sphere gap embodying my invention. Where I use more than one gap, I may provide resistance between the electrodes of the different gaps. I may for example, insert resistance 9 between each electrode 3 of the main sphere gap and the corresponding electrode on the auxiliary gap. The horn gap is preferably mounted above the sphere gap in position to receive the are rising from the sphere gap. If an arc persists in the sphere gap it is transferred to the horn gap and broken by it. I may also associate with the sphere gap a needle gap having pointed electrodes 10 in such relation to the horn gap that an arc in the needle gap is transferred to the horn gap. I may also in some cases provide a condenser in shunt to the sphere gap.

A combination of gaps as shown in Fig. 3 is desirable where the gaps are to be mounted out of doors, as wetting the electrode surfaces of a spark gap lowers its spark over voltage to a greater or less extent depending upon the electrodes. When the electrodes of the spark gaps shown in Fig. 3 are wetted the normal spark over voltage of the horn gap is lowered to a lem extent than that of the sphere gap, while the change in normal break down or spark over voltage of the needle gap is very slight. To prevent break down of the gaps at normal line potential when the electrodes are wet by rain, for example, the spacing of the sphere gap must be made greater than would be necessary were the spheres to remain dry: the horn gap spacing need be changed to a less extent; and the needle gap spacing need be changed scarcely at all. In such a combination of gaps connected in multiple, and exposed to the weather, the same normal are over voltage for all the gaps when the electrodes are wet can be obtained only by setting the gaps so that when the electrodes are dry, the normal spark over voltage of the sphere gap is higher than that of the horn gap and also higher than that of the needle gap. Setting the gaps as above described makes a combination especially suitable for use out of doors, and with the advantage that a selective action for different kinds of voltage waves is secured. The needle gap and horn gap have lower spark over voltages than the sphere gap for voltage waves of normal or sloping wave fronts and will respond to these voltage waves, while higher voltage waves with steep wave fronts will discharge through the sphere gap, in preference to the other gaps. By using a plurality of gaps of different characteristics and set to have different normal spark over voltages, a marked selective action can be obtained.

This selective action for voltage waves of diflerent heights and wave fronts may also be obtained by nurling all or part of the surfaces of the electrodes of the sphere gap, as for example, by smoothing the proximal portions between which discharge occurs and roughing the remainder of the surfaces, by adding lugs or projections to the electrodes or by otherwise modifying them.

In Fig. 4, I have shown diagrammatically another form of high speed gap comprising electrodes 11 which have between them a member or body 12 of porcelain or other insulating material in such a relation to the electrodes that the path of discharge between the electrodes is along the surface of the'insulating body 12. I have found that the insulating body arranged as above described causes the electrostatic stress between the electrodes ll to be distributed in such a manner that a voltage impulse, a voltage wave with a steep front, will cause a discharge between the electrodes 11 with much less time lag than in the spark gap of the ordinary type. As shown in Fig. 4 a horn gap 13 may be connected in parallel with the high speed gap and preferably in position to take any are which forms in the high speed gap.

What I claim as new and desire to secure by Letters Patent of the United States, is

1. A protective device comprising two cooperating electrodes having their proximal surfaces spherical and shaped apart to form a spark gap which resists normal potential and has a length not exceeding twice the radius of either spherical surface but greater than about three tenths of the square root of said radius in centimeters.

2. A protective device comprising two spheres spaced apart a distance not exceeding twice-the radius of said spheres but greater than substantially one half the square root of said radius in centimeters.

3. A protective device comprising two cooperating spherical electrodes separated to forms, spark gap, two elongated electrodes mounted immediately above said spherical electrodes and diverging to form a horn-gap connected'in parallel with said spark gap and in the path-of arc gases rising from said spherical electrodes, and a resistance connected between one of said spherical electrodes and one of said elongated electrodes.

4:. The combination of two elongated electrodes mounted vertically and separated at a point intermediate their ends by a spark gap, said electrodes diverging upwardly and downwardly from said gap, and auxiliary electrodes mounted below said gap and having their proximal surfaces sphericaland-separated by a gap not exceeding twice the radius of curvature of said surfaces and greater than about one half the square root of the radius in centimeters of said surfaces.

5. The-combination of a sphere gap having spherical electrodes separated by a gap not exceeding twice the radius of each sphere and greater than about one half the square root of said radius in centimeters, a needle-gap in parallel with said sphere gap and comprising pointed electrodes immediately above said-sphere gap, and a horn gap in parallel with said sphere gap and mounted above saidneedle gap, said gaps being so set that when all are dry the are over voltage of said horn gap is less than that of said sphere gap and greater than that ofsaid needle gap.

6. A protective device for electric conductors comprising two cooperating elec- 8. A- protective device comprising a sphere gap having spherical electrodes spaced apart-a distance not exceeding the diameter of either of said electrodes but greater than one half the square root of the radius in centimeters of'either electrode, an electrolytic cell-in series with said gap and a condenser connected in shunt to said sphere gap.

9. 'A high speed spark gap comprising two electrodes having between them a bridge of solid insulation'with its ends in contact with said electrodes and so shaped that the discharge path of minimum length is along the surface of said insulation.

10. A -high speed spark gap comprising two electrodes and a cylindrical bridge of porcelain in the axis of said gap with its ends in contact with'saidelectrodes to pro- I vide a discharge-path of minimum length along the surface of said bridge.

Inwitness whereof, I have hereunto set my-hand'this 14th day of August, 1915.

FRANK w. PEEK, JR.

Copies-of. this-patent may be obtained for-fivecents each; by addressing'the "Commissioner-of Patents, Washington, iDJC.

Images