C. F. FRANK.

PROTECTIVEDEVICE.

APPLICATION FILED MAYIS, 1914.

Patented Nov. 2, 1915.

2 SHEETS SHEET I.

a Irv/enter Crosbs F. Frank Witnesses 5 His Attorn'e c. F. FRANK.

"PROTECTIVE DEVICE.

I APPLICATION FILED MAY I5. 1914. 1,159,205. Patented Nov. 2, 1915.

2 SHEETSSHEET 2- I .20 J0 6'0 .90 M0 Wit fis ses'. I Inventor 2 I Crosb FTFranl HisAttorngg.

UNITED STATES PATENT orrrcn CROSBY FIELD FRANK, OF SCHENECTADY, NEW YORK, ASSIGNOR T0 GENERAL ELECTRIC COMPANY, A CQRPOBATION OF NEW YORK.

PROTECTIVE DEVICE.

- Specification of Letters Patent.

Patented Nov. f8, 11ers.

Application filed m 15, 1914. Serial No. 838,675.

T 0 all whom it may concern Be it known that I, CROSBY FIELD Fnsxn, a citizen of the United States, residing at Schenectady, county of Schenectady, State of New York, have invented certain new and useful Improvements in Protective Devices, of WlllCl'FtllG following is a specification.

The present invention relates to the protection of electrical distribution systems and conductors from the abnormal conditions due to lightning, static disturbances, resonance, switching, etc.

The object of my invention is to provide apparatus which will relieve the line of these abnormal conditions, particularly abnormally high potentials, by providing a path to earth of comparatively low resistance and which will at the same time prevent any heavy flow oi the dynamic current from the line over the same path.

In carrying my invention into reflect ll make use, as a part of the path to earth, of a body or finely divided granular material possessing certain predetermined electrical characteristics. Carborundum is especially suitable for this purpose though other materials such as boron or its compounds, ilmenite, etc, having the desired characteristics may be employed.

The most marked characteristics of the material used is its rapid decrease in resistance when subjected to voltages above av certain critical value. If the potential across a mass of loose carborundum granules be increased gradually, at a certain point the current will suddenly begin to increase very rapidly, in this respect much resembling the volt ampere curve oi an aluminum cell. As soon as the voltage decreases to its normal "alue the resistance increases rapidly to its initial value.

My invention will be best understood by reference to. the accompanying drawings forming a part of this specification and in which:

Figure 1 shows a cross-section of one form of my apparatus; Fig. 2 is a partial plan view thereof; Fig. 3 shows a modification;

Fig. -l is a. diagrammatic representation oi the relation between the size of the granules employed and the equivalent needle gap of the arrester under static conditions and corresponding to dynamic breakdown; Fig. 5 illustrates diagrammatically a system for connecting the arrester to the line which may be desirable in some cases, especially when used on a high voltage system.

In the particular form of my invention shown in Fig. 1, a container 1 of porcelain or other suit-able insulating material holds a mass 2 of granules of carborundum or other suitable material. At the bottom of this mass 1 provide an electrode 3 having a rounded edge and having a smaller area than the cross-sectional area of the container. At the top of the mass is placed a second electrode l which is also smaller than the cross-sectional area of the container. A coiled spring 5 between the cover 6 and the electrode insures a uniform contact pressure between the electrode and the mass of granline but will break down under any appreciable increase in the potential applied to this case I have shown a gap of the it. horn type, the horns 7, 7 being suitably mounted on projections 8, Son the cover of the arrester. A. terminal 9 is provided for connecting the line wire to one of the horns and the other horn is connected by means of the bolt 10 and the flexible connector ll to the electrode 4. The cover may be held in place by bolts 12 which .may be concealed by suitable water-proofing compound 13 to P6111 der the arrester thoroughly waterproof and suitable for outdoor use.

lln the form of my device shown in Fig. 3, a spark gap is provided of the compression chamber type. llhis comprises electrodes l l made of thin sheets or disks of metal suitably dished or cencaved as shown. These electrodes are Spaced, so as to have a gap *hetvveen them, by a ring 15 of insulating macreases its tendency to extinguish the are which forms. The cover in this case is shown as held in place by clamping rings 16 and the pressure of the upper electrode 17 upon the mass of granules may be adjusted to a certain extent by means of the clamping bolts 18.

In determining'the characteristics of the material tobe used. I have discovered that with pure carborundum granules the breakdown voltage of a mass of the granules of a given length is directly dependent upon the size of the granules used. This variation in the breakdown voltage with the size of granules is illustrated in Fig. 4 in which the ordinates represent the value of the equivalent needle gap and the abscissee represent the size of the granules used as determined 1 by passing them through a series of screens having from 10 to 100 meshes per inch.

Curve A represents the measure of the abil discharges for a longer period without weld-' ing together. On this account I have found it best in practice to adopt a compromise between the two qualities and use only granules which will pass through the 40 mesh screen and will not pass through the 80 mesh. By using granules which come between these. limits I secure a breakdown effect approximately the Same as that of the mesh granules and it will be noted that with granules of that size the breakdown voltage is practically the same for dynamic as for static. Inasmuch as the voltage of the disturbances which it is desired to remove from the line are always appreciably higher than the line voltage, this size of granules will be entirely suitable, whereas if larger.

v granules were used the static breakdown voltage would be much greater than the dynamic and-the arrester would fail in some cases to afford the desired relief to the line.

In some cases especially when high voltage lines are to be protected it will be desirable to employ several units in series rather than to. use a single unit with a long column of granular material. An arrangement of this kind is illustrated in Fig. 5. The top unit 19 which is connected to theline in this case may be of the form shown in Figs. 1 or 3 while in the other units 20 the spark gap will preferably be omitted. order to secure a uniform distribution of potential over the different units they may be shunted by high resistances 21 as indicated. It may also be. desirable to place condensers in shunt with each unit.

I am aware that lightning arresters have previously been constructed in which, the

breakdown characteristic of carborundum at a critical voltage has been employed. As usually constructed, however, the carborundum granules have been mixed with a binder and molded into blocks. \Yith this form of arrester it has been found diflicult in the process of manufacture to accurately predetermine the equivalent spark gaps of the discharge blocks as well as the opposition offered bythem to the passage of dynamic current after discharges. The heat dissipating properties of'this form of arrester have also been so poor that they have soon failed when subjected to a continued discharge.

\Vith an .arrester constructed according to my method however, the equivalent spark gap is known to depend directly upon the distance between the electrodes and the size of the granules used and hence may be accurately predetermined. An arrester of this nature also has much better heat dissipatwith ing qualities than one constructed molded carborundum blocks.

In constructing my arrester I have found that if the electrodes are made of the same size as the cross-sectional area of the container, there is a tendency for the discharge to follow paths along the wall of the container and that for some reason the granules weld together along these paths and the resistance of the arrester is greatly de-' creased. To overcome this difficulty, I have made the electrodes of less area than the interiorof the container so that the discharge must pass through a larger number of granules in order to find a path along the wall. By this construction the discharge is uniformly distributed throughout the entire mass of granules. I have also found that carborundum granules as obtained com- 'merclally are not suitable for my purpose pgobably because of the presence of impurities such as, graphite or metallic particles.

Apparently these impurities cause the granules to weld together more easily. By takmg the carborundum masses it comes from the furnace, crushing it dry and carefully sifting it however, I secure material which seems to be entirely free from impurities.

This is apparently due to the fact that the impurities are softer than the carborlmdum and hence are ground much finer and also that. inasmuch as no wateris used in the process-the impurities do not stick to the carborundum granules. Microscopic examination shows that granules prepared accordmg to my invention are clear and bright and have sharp edges while those prepared of and a mass of granules of carborundum v between said electrodes, said granules being of such a size that they will pass through 'a screen having forty meshes to the inch but will not pass through a screen havin eighty meshes to the inch. I

'2. In a protective device in combination a container, electrodes of less area than the container at opposite ends thereof and a breakdown voltage between said electrodes, the granules of said mass being of such a size that they" will pass through a screen having forty meshes to the inch but will not pass through a screen having eighty meshes to the inch.

3. In a protective device in combination a container, electrodes of less area than the" container at opposite ends thereof, a mass of granular material having a critical breakdown voltage between said electrodes,- the granules of said mass being of such a size that they will pass through a screen having forty meshes to the inch' but will not pass through a screen having eighty meshes to the inch, and a spring for holding said electrodes firmly in contact with said mass of granular material.

4. Ina protective device in combination a container, electrodes ofless area than the container at opposite ends thereof, a mass of granules of carborundum between said electrodes, said granules being free from im purities and of such a size that they will pass through a screen having forty meshes to an inch but will not pass through a screen having eighty meshes, to an inch.

' In witness whereof, I have hereunto set my hand this 13th day of May, 1914.

, CROSBY FIELD FRANK.

Witnesses:

BENJAMIN B. HULL, MARGARET E. WoorLEY.