Making of Porcelain Insulators

THE MAKING OF PORCELAIN

INSULATORS.*

By F. A. C. PERRINE, D.Sc.

BY most writers on the subject, the material of porcelain is considered to be one of the two great divisions of ceramic ware, and it is distinguished from pottery by being more dense, whiter and less fusible, but particularly in being translucent. Indeed, many consider that translucency is the only means by which porcelain may be distinguished from pottery. Pottery is a ceramic ware moulded from a paste of impure hydrated silicate of alumina, containing certain amounts of free silica, lime and iron. The product is opaque and invariably porous, on account of the removal of the volatile ingredients in the baking process. Porcelain, on the other hand, consists principally of a pure silicate of alumina only slightly hydrated, called kaolin, inclosed within a matrix of hard silicate glass. In the oldest history of the art the Chinese found pure kaolin in certain clay banks, and in other banks the ingredients for a silicate glass, which, being mixed together, moulded and baked, formed what is called "natural "or "hard paste " porcelain. This was first imitated in Europe by " soft paste'' porcelain, an admixture of the kaolinic clay with an artificial glass composed of a mixture of niter, soda, gypsum and salt, the proportion of kaolin to the glass being much less than in the natural Chinese product. Between these two lie the "mixed" or "bastard" porcelains, which are uncertain in character, but which are all composed of kaolin inclosed within a bond of more or less fusible glass. To this last class belong most of the American porcelains, although some of our potteries produce a material more like pottery than porcelain in porosity by mixing the ordinary grades of pottery clay with kaolin. When first baked, all of these materials are more or less imperfect over the surface, having much roughness and little luster, which surfaces must be finally protected by a subsequent glazing, the glaze consisting of nothing more than a thin coating of glass, which may be merely and entirely an artificial glass similar to that from which bottles are made, or a natural glass made by melting over the surface a powder of pure feldspar. We see then, in the first place, that the character of the insulating material depends upon the clays used and the percentages in which they are used, since by changing the ingredients we may pass from glass, which is melted and runs at a temperature a little above the red heat, through soft paste porcelains and mixed porcelains to hard paste porcelains, which in the hardest grades cannot be melted at a temperature below 2,000° Fah. Pottery can hardly be ranged in this series, nor is it proper to range here the so-called porcelains containing pottery clays, for the reason that the materials from which these articles are made all contain substances of a volatile character, and furthermore, these clays are deficient in fusibility, so that the ware is necessarily porous on account of the volatilization of water and other contents of the materials as they are moulded before baking.

The general process of manufacture of the insulator for an electrical line follows closely the processes that have been used in the manufacture of porcelain since the earliest times, with one important change, which is the result of modern inventive methods. The clays used are intimately mixed by grinding under water into a perfectly homogeneous mass, and after the superfluous water used in grinding has been removed by settling or filtering, the clays are moulded into shape required for the finished articles by either "casting," "throwing" or "pressing." The operation of "casting" is the one by means of which we obtain thin eggshell china so much admired in our dainty teacups. For this purpose a mould of plaster of Paris is furnished, which determines the outside shape of the article to be made. This mould is filled with a so-called "slip "of clay, about the consistency of cream ; the plaster mould immediately absorbs the water and holds on its surface a layer of clay, the thickness of which is determined by the time the slip is allowed to stand. When the articles are thick enough, the remaining slip is poured out, the surface smoothed by a delicate touch or two and the teacup is ready for drying and baking. Larger and thicker articles have from time immemorial been "thrown" on the potter's wheel. For this work the slip of clay is settled and filtered, and the remaining mass of wet clay moulded and dried until the correct plasticity is obtained, when it is taken by the potter and whirled upon his wheel, while with his hands he guides the jug into the shape desired.

Modern invention has almost superseded the potter's wheel by the potter's press, and by this press the tough, hard paste porcelain can be made in thick objects without porosity, as was formerly only possible with glass and soft paste material. For working in the press and to obtain a material of great density, the damp clay is further dried until it is in a powder which will not adhere in masses except under considerable pressure. This powder is then filled into a matrix and a core brought down; the sharp pressure consolidates the clay into the form of the finished article, the density of the resultant baked porcelain depending not only upon the proportion of the ingredients, soft and fusible, but also upon the amount of pressure applied; since as the pressure is increased the clay may be more thoroughly dried and smaller spaces left by evaporating moisture. Moisture cannot, however, be entirely eliminated from clay until it is subjected to the heat of a pottery kiln, and that this process removes a very considerable amount of water from the mass is shown by the fact that the shrinkage in baking amounts to about as much as the shrinkage in the cooling of cast iron, namely, a shrinkage of one-eighth in the linear dimensions. The spaces remaining are filled in the baking process by the fusible material with which the kaolin is mixed.

Whether "cast," " thrown" or " pressed," the articles made from clay are of consistency not different from that of a lump of clay until subjected to the heat of a pottery kiln, when the water of hydration is driven off from the silicate of alumina and any glass contained in the mass is fused, thus rendering the finished baked porcelain article hard and solid, the hardness and toughness depending largely upon the quantity and quality of silicate of alumina, while for the solidity we must look to the presence of a certain amount of fusible glass. The proper porcelain, therefore, for insulation is that in which there is only that amount of glass present which is necessary to fill up the porosity of the dehydrated silicate of alumina, since when this proportion is attained the greatest amount of strength consistent with non-porosity is reached. Should the amount of glass be increased beyond this point, the porcelain will become brittle, although it will still be nonporous — but nonporous just as glass is nonporous — and hence without the advantage over ordinary glass supposed to be possessed by a hard paste porcelain. If the amount of glass present is only so much as will fill up the pores left by the escaping water as the silicate of alumina is dehydrated, we can readily see that such porcelain has no property by means of which wide cracks in the moulded clay can be stopped up, since at no temperature available within the pottery kiln will the mass fuse and run. We may say that the glass is drawn by the porous silicate of alumina into itself by capilliary attraction, and when spaces are present which are not capillary these spaces cannot be filled up; accordingly, we see that the solidity of the finished article in hard paste porcelain manufacture depends upon the solidity of the moulded clay, and we also see that it is erroneous to suppose that while the porcelain manufacturer can make thin articles which will be solid, it is impossible for him to obtain an equal solidity when heavy masses are dealt with. It is, indeed, more likely that the heavy articles will contain open spaces than that the latter be present in thin pieces, since whether the pressure is applied on the wheel by the hand or in the matrix by the die, it is easier to make solid a thin layer of the clay than a thick one, but at the same time that porcelain manufacturer who does not increase his pressure sufficiently to insure the solidity of thick pieces is not handling his clays properly, and the fact that certain makers are not able to produce heavy solid articles does not tend to prove that solidity is an impossibility where the clays are properly handled.

Another danger must also be avoided in the choice of a porcelain insulator of great weight, which is harder to detect than the presence of porosity, since the exact knowledge of the mixture of clays used is only possessed by the manufacturer himself. I am speaking now of the tendency toward the use of soft paste porcelain by manufacturers whose machinery is not powerful enough to compress the hard paste clay to a sufficient density for a perfectly solid heavy insulator. Of course, with what has gone before, we can at once see that, should a manufacturer find that his mixture in heavy masses does not give a thoroughly solid product, he can at once add more glass, or glass of a more fusible character, reducing the amount of kaolin, on which we depend for strength. He will then obtain great solidity at the expense of toughness and strength. This is an error in the manufacture of insulators which can only be detected with the greatest difficulty, since the highest experts on porcelain wares are often at a loss to determine whether the body of an article is a hard or a soft paste porcelain without determining the crushing strength, and of course a mechanical test is exceedingly difficult to apply to such an irregular body as a line insulator. Even the chemist is here upon ground in which he is not sure, since the differences between the clays and the glasses are chemically so slight that the result of an analysis is of very little worth. While this is undoubtedly the greatest difficulty in the matter of insulators for withstanding high potentials and carrying lines of great weight, it is the point in which the manufacturers are deficient in their apparatus, since there is hardly a factory in the country which is equipped for pressing the heavy high voltage line insulator more solidly than articles much smaller are necessarily pressed, and, in consequence, it is probable that at the present time most of the manufacturers make an insulator which is either porous in a heavy body of the insulator head or that a much softer paste is dealt with than is used for smaller articles.

Should a manufacturer attempt to mix and press a hard paste clay which shall give a thoroughly solid line insulator of the greatest size, he would also encounter the difficulty that clay when pressed is not a liquid body, and in consequence pressure applied in one direction is likely to produce cleavage lines in the insulator, and in consequence we may not at the present time say that insulators of the form now used can be made solid from dry hard-pressed clay without danger of the production of such cleavage lines, and in consequence we must use a softer paste until the proper form of insulator shall be developed, or perfect the moulding of the hardest paste used in the driest possible state and sufficiently hard pressed for complete solidity after baking. These cleavage lines are especially apt to occur where irregularities in pressure are sustained by the mass of clay in moulding. That such irregularities must almost inevitably occur will be evident to one who examines carefully any insulator with long thin petticoats; and furthermore, this same examination will point to a danger of the petticoat splitting away from the head of the insulator in baking, when, as has been said, the shrinkage of one-eighth occurs. The effect of this shrinkage is well known to iron moulders, and no competent patternmaker would attempt to join long thin plates to a heavy body without in some way taking care of their tendency to split away from each other when shrinkage occurs. In consequence, the most successful designers of articles for manufacture in pressed porcelain are those who have had thorough training In moulding complicated cast iron shapes. Whether the petticoats are broken away from the mass 0f the insulator in pressing or in shrinking when they are baked, we find that most of the failures of such insulators when pierced by high voltages occur along cracks which are found in the porcelain at about the point where the petticoat is joined to the heavy head. Coming from the kiln, articles made of hard-pressed porcelain, however smoothly moulded and however solid in the interior, are found to have a surface which is dull and somewhat rough, and where a smooth, shiny surface is desired, glaze must be applied. Glaze is a glass applied to the exterior surface of the baked porcelain and fused on by subsequent baking. The glaze is perfect when it flows evenly over the surface of the article, and when it does not crack on coming from the kiln, or subsequently when exposed to varying temperatures. Cracking of the glaze is technically termed " crazing," and crazing is due entirely to a difference in the coefficient of expansion between that of the glaze and that of the underlying porcelain biscuit. It would seem at once to be the most reasonable proceeding to glaze our porcelain biscuit with the material of its body; but if this body is that of a hard paste porcelain, it is not sufficiently fusible for use as a glaze, and in consequence the surface must be covered with something more fusible. Accordingly, the selection of a glaze which shall have the same coefficient of expansion as that of the porcelain body is a matter of the greatest difficulty. This difficulty increases as the thickness of the glaze is increased, on account of the fact that while a thin glaze may have sufficient elasticity to give with movements of the underlying porcelain, a heavier body of glaze will not yield so readily.

That this should be true is apparent at once by the examination of a glass thread, which can be bent and even woven, while the glass rod from which it is made seems to be almost absolutely rigid. Tiles have been made with a glass face as much as one-sixteenth of an inch in thickness, but this was accomplished only after a long series of experiments, with repeated failures, and only on the surface of a rough tile made of porous fire brick. The natural hard paste porcelain of the Chinese is supposed to be glazed with pure feldspar, but a more readily fusible glass is required by almost all the European and American porcelains, while for each porcelain body a different glass is required; and not only is there probably but one glass which will glaze a particular porcelain, but also that porcelain manufacturer is to be congratulated for the care and success of his research after he has found a glaze which under no circumstances of changing temperature will craze. Josiah Wedgwood spent many years of his life in ascertaining this fact and in determining the proper glazes tor his different wares, while in this country thousands of dollars have been consumed by the different porcelain manufacturers in finding satisfactory glazes. In many cases such glazes have not yet been found, and in consequence but very few of our porcelain manufacturers can make a satisfactory line insulator without a crazed surface. That we should require a surface which does not craze is due to the fact that, although we do not depend upon the glaze itself as an insulation, we do depend upon it for preventing the retention of moisture upon the surface of the insulator. The nonhydroscopic character of porcelain as compared with glass therefore depends upon the character of the glazing surface and also upon its mechanical perfection. There is here a danger in the use of some porcelains, that while when made the smooth glaze is nonhydroscopic, the glass necessary to use for glazing will be soluble, and when weathered will present a rough surface, even though no crazing should ever occur. Indeed, it is probable that all porcelain glazes will ultimately weather, since the hardest glass we know of is that made of pure feldspar, which, even a slight study of geology tells us, is destroyed by atmospheric influences, and this is an additional reason for turning our attention from the glaze to the porcelain body when we speak of insulation resistances; and even were this weathering of the glaze removed to a remote possibility, we do know that any glaze will be worn away from the insulator by the contact of the line wire which the insulator is destined to support.

Having now examined carefully the necessary details in the manufacture of porcelain for insulators, we are prepared to give some guide for the writing of proper specifications and the pointing out of tests which should be made upon a porcelain insulator in order to obtain the best product that our manufacturers can at the present time produce. It is useless for the engineer to attempt to limit the porcelain manufacturer to certain pastes or glazes, since the clays used by different makers, though bearing the same names in different parts of the world, are not at all the same in ultimate character — one or two per cent. difference in the amount of water contained by the kaolin changing the whole character of the mixture necessary in the production of hard porcelain. And, as we have said, a different glaze is required for each mixture of clay; therefore these details must be left entirely to the manufacturer himself, since he is better able to make the choice necessary for a successful product than any engineer, no matter how carefully the latter has studied the use of insulators and the manufacture of porcelain. But, at the same time, the user should endeavor to ascertain which manufacturer uses the strongest paste and produces the densest insulator. Strong pastes can be chosen from soft pastes by mechanical tests alone, and the ability of the insulator to withstand shocks as well as pressure should be carefully tested.

In applying electrical tests it seems foolish to attempt only tests upon the finished glazed articles, since if the glaze is perfect it will completely mask the most serious extent of porosity, and even twice the potential of the line may not pierce a defective insulator, which will fail during the first rainstorm after the glaze has been worn away by the wire. The proper time, therefore, to apply any electrical test to the porcelain insulator is while the insulator is in the biscuit and before it has been glazed. It is always perfectly easy to test for porosity and to stamp the accepted insulators under the glaze by a mark which cannot be counterfeited after glazing has been performed. Even a galvanometer test at moderate potentials in this stage will furnish more accurate knowledge of the state of the porcelain than a high potential test after the glazing has been applied. Especially is inspection at this time important, as the glaze will mask cracks as well as porosity, which cracks would be apparent to the eye of the inspector before glazing, although completely covered by skillful work in the glazing process. The final inspection of the finished insulator should turn upon the perfection of the glazed surface, although it is foolish to attempt to obtain an insulator glazed in every part, since the insulator must be carefully supported in the glazing furnace, and where it is supported the surface will remain rough. It is indeed possible to specify the position upon the insulator of this rough surface, but one cannot expect to obtain a piece of porcelain covered with glaze in every part.

Finally, it should be pointed out that the relative positions of the wire and the petticoats demand some study, since if the petticoats are likely to be cracked away from the body of the insulator at their point of attachment with the head, it seems wise to remove the wire from this danger point as far as possible, although, of course, actual contact of the wire with a crack is not necessary for piercing through that crack, since the high poten tials now used will leap along the surface of an insulator from three to five inches, and, in consequence, if the path from wire to pin through a crack is less than the sparking distance of the potential used, leaks will occur through a crack even though neither pin nor wire be in contact with it.

It certainly seems, therefore, that, although we may improve our present insulators by a careful study of their material and form, it is at present unnecessary to attempt any redesigning of the insulators in use until we have thoroughly tested the capabilities of our porcelain manufacturers to produce insulators of the present form which shall be sufficiently strong to carry the wire and sufficiently solid to prevent the escape of electricity through the body of the porcelain without reference to the character or existence of glazing.

*An abstract from an article in Electrical Engineering.