Publication: The Central Station
New York, NY, United States
Porcelain Insulators for High Voltage Lines.
BY J. A. SANPKORD, JR.
The high price of copper as well as of other material, which has been prevailing for the past year, has hastened the time when the voltage of our longer transmission lines and those carrying the greatest amounts of power should be gradually raised from 50,000 to 100,000.
Referring particularly to the insulators required, this increase of 100 per cent, in voltage represents an increase of probably 500 per cent, and an increase of a smaller degree on the cost of other items necessary in constructing a high-voltage transmission line. The one exception to this last statement is, of course, the amount of copper required.
With this increase of voltage, has come a more than proportionate increase in trouble for the porcelain manufacturer producing insulators for these lines. This difficulty, however, has been at least temporarily overcome, by the introduction of the suspension type of insulator.
It is my intention to give in a general way, an idea of what is required of a good insulator; how they are made in all factories in this country; how they are tested, and what the tendency of the best engineers of our country has been along these lines for the past few years.
REQUIREMENTS OF A GOOD HIGH TENSION INSULATOR.
1. The material must have a high dielectric strength; in other words, it must be strong to resist puncture by current. In order to fulfill this condition the material must be continuous, compact* and homogeneous, even the most minute crack or fracture being a weakness.
2. There must be sufficient resistance over the surface of the insulator so that there will be no considerable conduction or leakage of current.
3. The distance around the insulator between the wire and pin or support must be sufficient to prevent the current from arcing,
4. Its contour must be such that there will be unexposed surfaces which will not collect dirt, salt, etc., and which will always remain dry. Evidently this requirement varies with the climate. Usually a larger insulator is required for the same voltage in a cold than in a warm climate. This may explain why some insulators have worked satisfactorily in one case and entirely unsatisfactorily on the same voltage in another. Insulators for the Pacific Coast should have as few petticoats as possible, and be of the open type in order to be easily cleaned. This is especially true for places in the neighborhood of Los Angeles.
The reason for this last statement is that, owing to the prevailing salt fogs, the insulators become coated with a crusty substance which consists of dust, salt and other foreign matter, which greatly reduces the efficiency of the leakage surface. If the insulator is so designed that wind and rain can get at a large portion of it, this coating will not form so perfectly as it otherwise might. Also the insulators may be more easily cleared by hand if occasion demands.
5. The shape and arrangement of the petticoats should be such that the electrostatic capacity of the insulator will be small.
6. The internal heat losses flue to conduction and hysteresis should not be such as appreciably to heat the insulator.
7. Mechanical requirements such as strength, mounting, method of fastening wire, color, etc., are in general, dependent on conditions to be met.
TESTS REQUIRED TO SEE IF INSULATORS FULFIL THESE CONDITIONS.
1. In order to determine the dielectric strength of the insulator, the most effective method is to invert the insulator and immerse the head in salt water and fill the center piece to a depth covering the thread with the same, and then apply the voltage to these solutions. A test of five minutes should be sufficient to weed out faulty insulators. This method applies to both porcelain and glass. The factor of safety should be at least three, and higher if possible.
2. Measurement of losses over insulators is a very difficult and delicate matter and not suited for factory tests. Any appreciable loss will be evident by an excessive brush discharge and by heating of the insulator.
3. The test for arcing voltage is applied in the same manner as for dielectric strength.
4. Tests to see if the contour is effective should be made by duplicating as nearly as possible the most severe rain storms by throwing a stream of water on the insulator under test, at an angle of about 30 degrees with the horizontal, and testing for several minutes. This test need not be made on more than two or three insulators of a lot. The factor of safety which should be allowed between the voltage at which the insulator will arc over under the above conditions, and the line voltage is a very uncertain and variable quantity, but should be in all cases as great as possible.
The electrical requirements are at variance with the mechanical requirements in the following points.
1. In order to in increase the dielectric strength, reduce the capacity and brush discharge, the thickness of the head of the insulator has to be increased. This increases the distance between the pin and wire and more strain is thus brought upon the insulating material. If of porcelain the liability to crack is increased.
2. If the wire is at the side, the surface distance is decreased, and hence the petticoats must be extended, also the diameter of the head has to be increased to get the required dielectric strength at the side. Practically this condition exists on all insulators owing to the. use of the tie wire which passes around the neck of the insulator.
3. Up to within a short time no logical or safe arrangement of circuits had been found whereby the lines could be placed otherwise than on the tops of the insulators. In this position the surface of the insulator is exposed to the elements at least as far as the edge of the piece next to the wire, and the effect is to aggravate the cause for leakage for a certain distance, where it must be checked. The suspension type of insulator overcomes this difficulty but it is too expensive for voltages for which the ordinary type of insulator can be economically produced.
4. The requirement for a larger insulator means one which is more breakable; if of glass, one apparently beyond the present knowledge of how to mould, or how to anneal.
THEORY CONNECTED WITH THE DESIGN OF HIGH TENSION INSULATORS.
The design of insulation for high pressure should involve a consideration of all the effects of electrical tension on the dielectric in the vicinity of the conductors. In the case of a line insulator, air is always a dielectric in combination with glass, porcelain, wood, or other materials. Wherever there is a difference of electrical potential there exists in the surrounding media a certain state of strain called an electrostatic field. This state of strain is the result of electrical stress applied to the insulating material. Dielectrics possess a sort of atomic elasticity, and electrical tensions produce a displacement in the molecular structure which, if carried beyond a certain limit, result in disruptive breakdown of the material.
Before a difference of potential can exist, current must flow into the dielectric, thus producing a state of strain equal to the electrical stress applied. If the material be not strained beyond its limits of molecular elasticity, current will flow from the material whenever the tension is removed or reduced, and a path provided. All dielectrics possess the quality of receiving strain before rupture, but not to the same degree.
Due to this fact, we have the common so-called brush discharge, caused by the fact that the air breaks down and acts as a conductor before the material to which it is adjacent will do so.
Consider that alternating electrical tension be applied to a solid insulating disc: if the pressure be low, only charging current will flow, but if sufficiently increased, the air about the electrodes will be ruptured, and brush discharge will occur. The air thus affected becomes a conductor, and thus enlarges the electrodes. As the tension is further increased this zone of conducting air spreads out over the surface of the disc, increasing the capacity and hence the charging current. At this stage streamers form on the disc. These offer a path of low resistance to the outer edge of the disc, and this process would continue, were it not for the cooling influences met with during the spreading out process. If the pressure be increased far enough the streamers may unite around the edge, forming a short circuit. This is not due to surface leakage, but to electrostatic capacity and local failure of the air as a dielectric.
Surface insulation has little to do with the performance of line insulators, and the surface leakage may be neglected altogether excepting cases where the surface is covered with water or other foreign material.
During rains the top surface of an insulator is at the same potential as the conductor, and the ground is at least as near as the cross arm, and, with steel towers, arms and pins, is up in the insulator. The dielectric in this case, then, consists of the porcelain or glass, and the air, between the pin and the conductor.
If the pressure applied sets up an electro-static field strong enough, the air will rupture, and brush discharge occur, which may spread out enough to make the insulator arc over. This may be prevented in two ways: by increasing the size of the surfaces, and by increasing the thickness of the dielectric, thus reducing the strength of the electrostatic field to such a point that brush discharge will not occur. These discharges are wasteful of energy and together with charging current cause the burning of pins, etc. The general dimensions of the insulator should be such that the distance from conductor to cross arm or pin will prevent the air rupturing.
When the insulators are made up of several parts cemented together, the dielectric material is no longer homogeneous, and the distribution of the electro-static strain may be materially altered. The cement between the sections is in series with the dielectric material of the insulator, and is exposed to the same electro-static field of force. The strata of cement in some cases redistributes the electrostatic charge. Under other conditions the pressure is conducted directly to the cement through the ruptured air. In this case the semi-conducting cement becomes charged with the full terminal pressure, and excessive tension may thus be applied to a section of the insulator not designed to stand it. This frequently results in sectional-breakdowns, and the insulator fails in detail.
The resistance to disruptive breakdown or puncture of the solid dielectric is also of importance. Good porcelain or glass has, however, such great strength in this respect that if the insulator be so designed that the electro-static field cannot start a brush discharge, this other defect will be taken care of.
Of the various substances available, glass and porcelain have been used almost exclusively for high-tension insulators. Glass has excellent dielectric qualities, and can readily be obtained in desirable shapes, at reasonable cost. Its greatest defect is its mechanical weakness, which is due almost entirely to internal strains developed during manufacture. Consistent design of the surfaces so as to obviate, as far as possible, shrinkage strains, and careful annealing have improved the conditions of many glass insulators, so as to render them reliable for service, but they still do not possess the mechanical strength of the best porcelain. Glass, however, is a more reliable dielectric material, and from an electrical standpoint gives better and more uniform results. The best porcelain has great mechanical strength, and good dielectric qualities. It is, however, difficult of manufacture in considerable thickness, and is very apt to develop flaws and surface cracks.
It has been the policy of the writer not to recommend glass insulators for voltages higher than 15,000. Glass insulators for this voltage and lower are in many instances giving good satisfaction, but the sizes required for voltages higher than this are liable to develop internal strains and break without warning. It is the opinion of the most conservative engineers in most cases, that the small additional cost of porcelain insulators over glass for the same voltage is money well expended.
Other substances have been proposed for high-voltage insulators, but nothing has been found as yet, that does not age or deteriorate with time, with the exception of porcelain, which, as far as this point is concerned, is practically indestructible.
THE MANUFACTURE OF PORCELAIN HIGH-TENSION INSULATORS.
Before proceeding to describe the process of manufacture of high-tension insulators, I will attempt to define the word, porcelain, and tell briefly of what a porcelain body consists, and how the different substances affect the ware, with a view to give the reader some idea as to what care must be taken to produce desired results.
True porcelain consists of a finely-ground mixture of kaolin, quartz, and feldspar, being fixed until it is absolutely vitrified. In this state it is not greatly different from glass. It is, therefore, not advisable to fire an electrical porcelain any higher than is absolutely necessary to give it the required electrical strength.
The efficiency of an electrical porcelain is dependent upon the proportion of its constituent parts. The materials used are, as above stated, certain kinds of kaolin, which are very plastic, quartz and feldspar. Clays are always of secondary origin, that is to say, they are the result of decay or decomposition of rocks, most frequently those containing feldspar. Whenever the parent rock is feldspar, kaolin is the result and may be considered a basic hydrated aluminus silicate. If this residual clay or kaolin is transported by water action and deposited in a sedimentary layer, together with impurities gained in transportation, we have ordinary clay; if, there are but few impurities, especially if the transported kaolin contains little or no iron compounds, and possesses good tensile strength, i.e., prasticity, when wet, it is termed a ball-clay.
The body will be more resistant the higher the per cent, of kaolin and feldspar; also the finer the grain of kaolin used, the higher will be the maturing point of the porcelain produced. The feldspar is a basic substance, and unites with silica to form silicates, thus producing an intimate mixture of silicates. Therefore, the contents of kaolin must be in a definite ratio to the quartz and feldspar in regard to the insulating properties.
Too much quartz makes the body too little plastic and difficult to work, and the ware shows impressions and some warping due to the tendency of the clay substance to contract. Too high a content of feldspar makes the body soft and liquid and increases the tendency of crack. A body low in feldspar does not produce sufficient vitrification, and hence decreases its insulating efficiency. A porcelain body may shrink 15 to 18 per cent, and it is very important that this figure be accurately known. No crazing or shivering of the glaze is allowable. If the glaze is too infusible the ware will be dull; if too fusible, it is likely to be covered with blisters.
Having given the above rough account of how the different substances affect the body of an insulator, I will now describe briefly the process of manufacture according to the jollying process.
The different substances having been weighed up in the proper proportion, are now dumped into a large revolving steel cylinder similar to a boiler. The proper amount of water is run in and the entrance closed. This cylinder is set in motion, revolving for about two and a half hours. The clay is thus ground up very fine by the large balls that the cylinder contains—hence the name "ball mill."
This machine gets its name from the fact that, in the manufacture of porcelain not required for electrical purposes, large steel balls are used for grinding the clay. In grinding the clays, for high-tension insulators, however, large smooth stones of a very hard quality, taken from the glaciers of the Arctic regions are used instead of the steel balls in order to reduce the chances for the body to collect small metallic particles.
The liquid is then drawn off through a fine screen into a pit containing a revolving mixer. From the pit it is pumped by means of a force pump into a press consisting of several iron disks, between which are fitted canvas bags. The arrangement of these is such that when all are pressed tightly together there is a disk-like cavity between each two iron plates, and a hole through the center of the whole set. As these cavities fill up, the water squeezes through the canvas and runs off, leaving the clay behind. When the final pressure of 95 pounds per square inch is reached, scarcely any water is left in the clay.
The press is now opened up and the disks of clay removed and piled up in the aging cellar, where they are allowed to stay about two weeks. This aging is an action of the bacteria in the clay, and has to be watched very closely in order not to allow it to proceed too far.
When the clay is taken from this cellar it is put through the pug mill to be thoroughly worked to a uniform plasticity, just enough water being added to bring it to the right softness for use in the moulds. From the pug mill the clay is carried directly to the jolly wheels where the moulder's assistant fills the moulds, and the moulder places them in the wheels and forms the inside of the piece with his hands, and the shoe, the mould forming the outside. The mould containing the piece is then placed in the dryer racks, where it remains until the clay has dried and shrunk enough to allow a helper to remove the piece from the mould. The threading of the center piece is done before the mould goes into the dryer.
After the piece is taken from the mould it is put on a revolving form, and the surfaces which were next to the mould are finished with a scraper and wet sponge. The ware is then placed on the rack again and is dried for about four days, the temperature being kept very even, as this is the most critical point of the process, excepting the burning. The ware at the end of this time can be safely handled, and is ready for the glaze, which is mixed in a ball mill similar to that used for the body, only smaller. The ware is dipped in a glaze contained in a tub, and then allowed to stand, preferably at least a day before being set in the kiln.
When ready to set a kiln, the ware is placed in fire clay vessels called saggars, in order to protect it from direct contact with the fire. These saggars are piled up in the kiln, care being taken to allow the proper amount of air spaces in all parts of the kiln. Between each two fire bags (usually ten or twelve in number) are placed saggars containing pyrometric cones, by which the temperature is gauged, they being observed through holes in the side of the kiln. A set is also placed at the bottom of the kiln near the stack and observed through a place that may be opened in the bottom of the door. When the fires are started they are run very slowly until practically, all the moisture is driven from the ware. They then begin to force the fires and as the cones begin to melt and bend over, the temperature is noted. Just as the last cone (No. 12, corresponding "to approximately 1370 C.) begins to bend, the fires are forced fearfully until it is melted down, when the firing ceases and gradually the fires die out. When the kiln has cooled enough, so that no red can be seen in it, the door may be pulled down and rapid cooling allowed to take place. The process of firing takes place from 42 to 45 hours, and the whole cycle for one kiln about 80 hours. When sufficiently cooled, the ware is removed from the saggars and is then ready for testing, which has been previously described.
GENERAL PRACTICE IN LIN