German Silver, Platinoid, Silver, and Platinum are treated like iron. With regard to silver and platinum the same precautions as recommended in the case of zinc must be observed, for both these metals form fusible alloys with solder.

Gold when pure requires no flux. Standard gold, which contains copper, solders better with a little chloride of zinc.

Lead must be pared absolutely clean and then soldered quickly with a hot iron, using tallow as a flux. Since solder if over hot will adhere to lead almost anywhere, plumbers are in the habit of specially soiling those parts to which it is not intended that solder shall adhere. The "soiling" paint consists of very thin glue, called size, mixed with lampblack; on an emergency a raw potato may be cut in half, and the work to be soiled may be rubbed over with the cut surface of the potato.

Hard Carbon or gas coke may be soldered after coating with copper by an electrolytic process, as will be described.

Sec. 99. Brazing.

Soldering at a red heat by means of spelter is called brazing. Spelter is soft brass, and is generally made from zinc one part, copper one part; an alloy easily granulated at a red heat; it is purchased in the granular form.

The art of brazing is applied to metals which will withstand a red heat, and the joints so soldered have the strength of brass.

The pieces to be jointed by this method must be carefully cleaned and held in their proper relative positions by means of iron wire. It is generally necessary to soften iron wire as purchased by heating it red hot and allowing it to cool in the air; if this is not done the wire is usually too hard to be employed satisfactorily for binding.

Very thin wire—i.e. above No. 20 on the Birmingham wire gauge—does not do, for it gets burned through, and perhaps allows the work to fall apart at a critical moment.

The work being securely fastened, the next step is to cover the cleaned parts with flux in order to prevent oxidation. For this purpose "glass borax" is employed. "Glass" borax is simply ordinary borax which has been fused for the purpose of getting rid of water of crystallisation. The glass borax is reduced to powder in an iron mortar, for it is very hard, and is then made up into a cream with a little water. This cream is painted on to the parts of the work which are destined to receive the solder.

The next step is to prepare the spelter, and this is easily done by mixing it with the cream, taking care to stir thoroughly with a flattened iron wire till each particle of spelter is perfectly covered with the borax. The mixture should not be too wet to behave as a granular mass, and may then be lifted on to the work by means of the iron spatula.

Care must be taken to place the spelter on those parts only which are intended to receive it, and when this is done, the joint may be lightly powdered over with the dry borax, and will then be ready for heating.

If the object is of considerable size it is most conveniently heated on the forge; if small the blowpipe is more convenient. In the latter case, place the work on a firebrick, and arrange two other bricks on edge about it, so that it lies more or less in a corner. A few bits of coke may also be placed on and about the work to increase the temperature by their combustion, and to concentrate the flame and prevent radiation. The temperature is gradually raised to a bright red heat, when the spelter will be observed to fuse or "run," as it is technically said to do.

If the cleaning and distribution of flux has been successful, the spelter will "run" along the joint very freely, and the work should be tapped gently to make sure that the spelter has really run into the joint. The heating may be interrupted when the spelter is observed to have melted into a continuous mass. As soon as the work has fallen below a red heat it may be plunged into water, a process which has the effect of cracking off the glass-like layer of borax.

There is, however, some risk of causing the work to buckle by this violent treatment, which must of course be modified so as to suit the circumstances of the case. If the joint is in such a position that the borax cannot be filed off, a very convenient instrument for its removal by scraping is the watchmaker's graver, a square rod of hard steel ground to a bevelled point (Fig. 80).

Fig. 80.

Several precautions require to be mentioned. In the first place, spelter is merely rather soft brass, and consequently it often cannot be fused without endangering the rest of the work. A good protection is a layer of fireclay laid upon the more delicate parts, such for instance as any screwed part.

Gun-metal and tap-metal do not lend themselves to brazing so readily as iron or yellow brass, and are usually more conveniently treated by means of silver solder.

Spelter tends to run very freely when it melts, and if the brass surface in the neighbourhood of the joint is at all clean, may run where it is not wanted. Of course some control may be exercised by "soiling" with fireclay or using an oxidising flame; but the erratic behaviour of spelter in this respect is the greatest drawback to its use in apparatus construction. The secret of success in brazing lies in properly cleaning up the work to begin with, and in disposing the borax so as to prevent subsequent oxidation.

Sec. 100. Silver Soldering.

This process resembles that last described, but instead of spelter an alloy of silver, copper, and zinc is employed. The solder, as prepared by jewellers to meet special cases, varies a good deal in composition, but for the laboratory the usual proportions are:

For soft silver solder

Fine silver 2 parts Brass wire 1 part

For hard silver solder

Sterling silver 3 parts Brass wire 1 part

The latter is, perhaps, generally the more convenient.

Silver solders may, of course, be purchased at watchmakers' supply shops, and as thus obtained, are generally in thin sheet. This is snipped fine with a pair of shears preparatory to use.

As odds and ends of silver (from old anodes and silver residues) generally accumulate in the laboratory, it is often more convenient to make the solder one's self. In this case it must be remembered in making hard solder by the second receipt that standard silver contains about one-twelfth of its weight of copper—exactly 18 parts copper to 220 silver.

The silver is first melted in a plumbago crucible in a small furnace together with a little borax; if any copper is required this is then added, and finally the brass is introduced. When fusion is complete, the contents of the crucible are poured into any suitable mould.

The quickest and most convenient way of preparing the alloy for use is to convert it into filings with the assistance of a coarse file, or by milling it, if a milling machine is available.

Equal volumes of filings and powdered glass borax are made into a thin paste with water, and applied in an exactly similar manner to that described under the head of "brazing." In fact all the processes there described may be applied equally to the case under discussion, the substitution of silver for spelter being the only variation.

The silver solder is more manageable than spelter, and does not tend to run wild over the work: a property which makes it much more convenient both for delicate joints and in cases where it is desired to restrict the solder to a single point or line. Small objects are almost invariably soldered with silver solder, and are held by forceps or on charcoal in the pointed flame of an ordinary blow-pipe.

Sec. 101. On the Construction of Electrical Apparatus: Insulators.

It is not intended to deal in any way with the design of special examples of electrical apparatus, but merely to describe a rather miscellaneous set of materials and processes constantly required in its construction.

It is not known whether there is such a thing as a perfect insulator, even if we presuppose ideal circumstances. Materials as they exist must be regarded merely as of high specific resistance, that is if we allow ourselves to use such a term in connection with substances, conduction through which is neither independent of electromotive force per unit length, nor of previous history.

Even the best of these substances generally get coated with a layer of moisture when exposed to the air, and this as a rule conducts fairly well. Very pure crystalline sulphur and fused quartz suffer from this defect less than any other substances with which the writer is acquainted, but even with them the surface conductivity soon grows to such an extent as totally to mask the internal conduction.

It is proposed to give a brief account of the properties of some insulating substances and their application in electrical construction, and at the same time to indicate the appliances and methods requisite for working them.

With regard to the specific resistances which will be quoted, the numbers must not be taken to mean too much, partly for the reason already given. It is also in general doubtful whether sufficient care has been taken to distinguish the body from the surface conductivity, and consequently numerical estimates are to be regarded with suspicion. The question of "sampling" also arises, for it must be remembered that a change in composition amounting to, say, 1/10000 per cent may be accompanied by a million-fold change in specific resistance.

Sec. 102. Sulphur.

This element exists in several allotropic forms, which have very different electric properties. After melting at about 125 deg. C, and annealing at 110 deg. for several hours, the soluble crystalline modification is formed. After keeping for some days—especially if exposed to light—the crystals lose their optical properties, but remain of the same melting-point, and are perfectly soluble in carbon bisulphide. The change is accompanied by a change in colour, or rather in brightness, as the transparency changes.

The "specific resistance" of sulphur in this condition is above 1028 C.G.S.E.M. units, or 1013 megohms per cubic centimetre for an electric intensity of say 12,000 volts per centimetre. This is at ordinary temperatures. At 75 deg. C. the specific resistance falls to about 1025 under similar conditions as to voltage.

In all cases the conductivity appears to increase with the electric intensity, or at all events with an increase in voltage, the thickness of the layer of sulphur remaining the same.

The specific inductive capacity is 3.162 at ordinary temperatures, and increases very slightly with rise of temperature. [Footnote: March 1897.—It is now the opinion of the writer that though the specific inductive capacity of a given sample of a solid element is perfectly definite, yet it is very difficult to obtain two samples having exactly the same value for this constant, even in the case of a material so well defined as sulphur.]

The total residual charge, after ten minutes' charging with an intensity of 12,000 volts per centimetre, is not more than 4 parts in 10,000 of the original charge. In making this measurement the discharge occupied a fraction of a second. The electric strength for a homogeneous plate of crystalline sulphur is not less than 33,000 volts per centimetre, and probably a good deal more. If the sulphur is contaminated with up to 3 per cent of the amorphous variety, as is the case if it is cooled fairly quickly from a temperature of 170 deg. C. or over, the specific resistance falls to from 10^25 to 10^26 at ordinary temperatures; and the specific inductive capacity increases up to 3.75, according to the amount of insoluble sulphur present.

The residual charge under circumstances similar to those described above, but with an intensity of about 4000 volts per centimetre is, say, 2 per cent of the initial charge. So far as the writer is aware sulphur is the only solid non-conductor which can be easily obtained in a condition of approximate purity and in samples sufficiently exactly comparable with one another; it is the only one, therefore, that repays any detail of description.

Very pure sulphur can be bought by the ton if necessary from the United Alkali Company of Newcastle-on-Tyne. It is recovered from sulphur waste by the Chance process, which consists in converting the sulphur into hydrogen sulphide, and burning the latter with insufficient air for complete combustion. The sulphur is thrown out of combination, and forms a crystalline mass on the walls and floor of the chamber.

The sulphur which comes into the market consists of this mass broken up into convenient fragments. In order to purify it sufficiently for use as an insulator, the sulphur may be melted at a temperature of 120 deg. to 140 deg. C, and filtered through a plug of glass wool in a zinc funnel; as thus prepared it is an excellent insulator. To obtain the results mentioned in the table it is, however, necessary to conduct a further purification (chiefly from water) by distillation in a glass retort.

The sulphur thus obtained may be cast of any desired form in zinc moulds, the castings and moulds being immediately removed to an annealing oven at a temperature of from 100 deg. to 110 deg. C, where they are left for several hours. If the sulphur is kept melted for some time at 125 deg. C. the annealing is not so important.

The castings may be removed from the mould by slightly heating the latter, but many breakages result. Insulators made on this plan are much less affected by the condensation of moisture from the air than anything except fused quartz. They are, however, very weak mechanically, and apt to crack by exposure to such changes of temperature as go on from day to day. It is clear, however, that in spite of this their magnificent electrical properties fit them for many important uses.

If the sulphur be cooled rapidly from 170 deg. C. or over, a mixture of the crystalline and amorphous varieties of sulphur is obtained. This mixture is very much stronger and tougher than the purely crystalline substance, and may be worked with ordinary hardwood tools into fairly permanent plates, rods, etc. Sheets of pure thick filter paper may also be dipped into sulphur at 170 deg. C, at which temperature air and moisture are mostly expelled, and such sheets show a very considerable insulating power. The sulphur does not penetrate the paper, which therefore merely forms a nucleus.

Cakes of the crystalline or mixed varieties may be made by grinding up some purified sulphur, moistening it with redistilled carbon bisulphide, or toluene, or even benzene (C6H6), and pressing it in a suitable mould under the hydraulic press. The plates thus formed are porous, but are splendid insulators, especially if made from the crystalline variety of sulphur, and they appear to keep their shape very well, and do not crack with ordinary temperature changes.

The metals which resist the action of sulphur best are gold and aluminium; while platinum and zinc are practically unacted upon at temperatures below a red heat—in the former case,—and below the boiling-point of sulphur in the latter.

A very convenient test of the purity of sulphur is the colour assumed by it when suddenly cooled from the temperature at which it is viscous. Quite pure sulphur remains of a pale lemon yellow under this treatment, but the slightest trace of impurity, such as arises from dust containing organic matter, stains the sulphur, and renders it darker in colour.

Sec. 103. Fused Quartz.

This is on the whole the most reliable and most perfect insulator for general purposes. No exact numerical data have been obtained, but the resistivity must certainly be of the same order as that of pure sulphur at its best. The influence of the moisture of the air also reaches its minimum in the case of quartz, as was originally observed by Boys.

As yet, however, the material can only be obtained in the form of rods or threads. For most purposes rods of about one-eighth of an inch in diameter are the most convenient. These rods may be used as insulating supports, and succeed perfectly even if they interpose less than an inch of their length to electrical conduction. The sketch (Figs. 81 and 81A) shows (to a scale of about one-quarter full size) a complete outfit for elementary electrostatic experiments, such as has been in use in the writer's laboratory for five years. With these appliances all the fundamental experiments may be performed, and the apparatus is always ready at a moment's notice.

Fig. 81.

Though quartz does not condense moisture or gas to form a conducting layer of anything like the same conductivity as in the case of glass or ebonite, still it is well to heat it if the best results are to be obtained. For this purpose a small pointed blow-pipe flame may be used, and the rods may be got red-hot without the slightest danger of breaking them. They then remain perfectly good and satisfactory for several hours at least, even when exposed to damp and dusty air.

The rods are conveniently held in position by small brass ferrules, into which they are fastened by a little plaster of Paris. Sealing-wax must be avoided, on account of the inconvenience it causes when the heating of the rods is being carried out.

One useful application of fused quartz is to the insulation of galvanometer coils (Fig. 82), another to the manufacture of highly insulating keys (Fig. 83); while as an insulating suspension it has all the virtues. If it is desired to render the threads conducting they may be lightly silvered, and will be found to conduct well enough for electrometer work before the silver coating is thick enough to sensibly impair their elastic properties.

Fig. 81A.

Fig. 82 is a full-size working drawing of a particular form of mounting for galvanometer coils. The objects sought to be attained are: (1) high insulation of the coils,

(2) easy adjustment of the coils to the suspended system.

The first object is attained as follows. The ebonite ring A is bored with four radial holes, through which are slipped from the inside the fused quartz bolt-headed pins B. The coil already soaked in hard paraffin is placed concentrically in the ring A by means of a special temporary centering stand. The space between the coil and the ring is filled up with hard paraffin, and this holds the quartz pins in position. The system of ebonite ring, coil, and pins is then fastened into the gun-metal coil carrier, which is cut away entirely, except near the edges, where it carries the pin brackets C. These brackets can swivel about the lower fastening at E before the latter is tightened up.

The coil is now adjusted in the adjusting stand to be concentric with the axis of symmetry of the coil carrier, and the supporting pins are slipped into slot holes cut in the brackets, the brackets being swivelled as much as necessary to allow of this. When the pins are all inserted the brackets are screwed up by the screws at E. The pins are then cemented firmly to the brackets by a little plaster of Paris. The coil carrier can now be adjusted to the galvanometer frame by means of screws at D, which pass through wide holes in the carrier and bold the latter in position by their heads. In the sectional plan the parts of the galvanometer frame are shown shaded. The front of the frame at F F is of glass, and the back of the frame is also made of glass, though this is not shown in the section.

A represents an ebonite ring into which the wire coil is cemented by means of paraffin. B B B B are quartz pins, with heads inside the ebonite ring. C C C are slotted brackets adjustable to the pins and capable of rotation by releasing the screws E E. D D are the screws holding the coil carriage to the galvanometer framework. These screws pass through large holes in the carriage so as to allow of some adjustment.

Fig. 82.

Fig. 83.

Sec. 104. Glass.

When glass is properly chosen and perfectly dry it has insulating properties possibly equal to those possessed by quartz or crystalline sulphur. For many purposes, however, its usefulness is seriously reduced by the persistence with which it exhibits the phenomena of residual charge, and the difficulty that is experienced in keeping it dry.

The insulating power of white flint glass is much in excess of that of soft soda glass, which is a poor insulator, and of ordinary green bottle glass. The jars of Lord Kelvin's electrometers, which insulate very well, are made of white flint glass manufactured in Glasgow, but it is found that occasionally a particular jar has to be rejected on account of its refusing to insulate, and this, if I understand aright, even when it exhibits no visible defects.

A large number of varieties of glass were tested by Dr. Hopkinson at Messrs. Chance Bros. Works, in 1875 and 1876 (Phil. Trans, 1877), and in 1887 (Proc. Roy. Soc. xli. 453), chiefly with a view to the elucidation of the laws regulating the residual charge; and incidentally some extraordinarily high insulations were noted among the flint glasses. The glass which gave the smallest residual charge was an "opal" glass; and flint glasses were found to insulate 105 times as well as soda lime glasses. The plates of Wimshurst machines are made of ordinary sheet window glass, but as the insulating property of this material appears to vary, it is generally necessary to clean, dry, and test a sheet before using it. With regard to hard Bohemian glass, this is stated by Koeller (Wien Bericht) to insulate ten times as well as the ordinary Thuringian soft soda glass.

On the whole the most satisfactory laboratory practice is to employ good white flint glass. The only point requiring attention is the preparation of the glass by cleaning and drying. Of course all grease or visible dirt must be removed as described in an earlier chapter (Sec. 13), but this is only a beginning. The glass after being treated as described and got into such a state as to its surface that clean water no longer tends to dry off unequally, must be subjected to a further scrub with bibulous paper and a clear solution of oleate of soda. The glass is then to be well rinsed with distilled water and allowed to drain on a sheet of filter paper.

A very common cause of failure lies in the contamination of the glass with grease from the operator's fingers. Before setting out to clean the glass the student will do well to wash his hands with soap and water, then with dilute ammonia and finally with distilled water.

In the case of an electrometer jar which has become conducting but is not perceptibly dirty, rubbing with a little oleate of soda and a silk ribbon, followed, of course, by copious washing, does very well. If there is any tin-foil on the jar, great care must be taken not to allow the glass surface to become contaminated by the shellac varnish or gum used to stick the tin-foil in position.

Finally, the glass should be dried by radiant heat and raised to a temperature of 100 deg. C. at least, and kept at it for at least half an hour. Before drying it is of course advisable to allow the water to drain away as far as possible, and if the water is only the ordinary distilled water of the laboratory, the glass is preferably wiped with a clean bit of filter paper; any hairs which may be left upon the glass will brush off easily when the glass is dry.

In order to obtain satisfactory results the glass must be placed in dry air before it has appreciably cooled. This is easily done in the case of electrometer jars, and so long as the air remains perfectly dry through the action of sulphuric acid or phosphorus pentoxide, the jar will insulate. The slightest whiff of ordinarily damp air will, however, enormously reduce the insulating power of the glass, so that unvarnished glass surfaces must be kept for apparatus which is practically air-tight.

For outside or imperfectly protected uses the glass does better when varnished. It is a fact, however, that varnished glass is rarely if ever so good as unvarnished glass at its best. Too much care cannot be taken over the preparation of the varnish; French polish, or carelessly made shellac varnish, is likely to do more harm than good.

The best orange shellac must be dissolved in good cold alcohol by shaking the materials together in a bottle. The alcohol is made sufficiently pure by starting with rectified spirit and digesting it in a tin flask over quick-lime for several days, a reversed condenser being attached. A large excess of lime must be employed, and this leads to a considerable loss of alcohol, a misfortune which must be put up with.

After, say, thirty hours' digestion, the alcohol may be distilled off and employed to act on the shellac. In making varnish, time and trouble are saved by making a good deal at one operation—a Winchester full is a reasonable quantity. The bottle may be filled three-quarters full of the shellac flakes and then filled up with alcohol; this gives a solution of a convenient strength.

The solution, however, is by no means perfect, for the shellac contains insoluble matter, and this must be got rid off.'' One way of doing this is to filter the solution through the thick filtering paper made by Schleicher and Schuell for the purpose, but the filtering is a slow process, and hence requires to be conducted by a filter paper held in a clip (not a funnel) under a bell jar to avoid evaporation.

Another and generally more convenient way in the laboratory is to allow the muddy varnish to settle—a process requiring at least a month—and to decant the clear solution off into another bottle, where it is kept for use. The muddy residue works up with the next lot of shellac and alcohol, which may be added at once for future use.

The glass to be varnished is warmed to a temperature of, say, 50 deg. C, and the varnish put on with a lacquering brush; a thin uniform coat is required. The glass is left to dry long enough for the shellac to get nearly hard and to allow most of the alcohol to evaporate. It is then heated before a fire, or even over a Bunsen, till the shellac softens and begins to yield its fragrant characteristic smell.

If the coating is too heavy, or if the heating is commenced before the shellac is sufficiently dry, the latter will draw up into "tears," which are unsightly and difficult to dry properly. On no account must the shellac be allowed to get overheated. If the varnish is not quite hard when cold it may be assumed to be doing more harm than good.

In varnishing glass tubes for insulating purposes it must be remembered that the inside of the tube is seldom closed perfectly as against the external air, and consequently it also requires to be varnished. This is best done by pouring in a little varnish considerably more dilute than that described, and allowing it to drain away as far as possible, after seeing that it has flooded every part of the tube.

During this part of the process the upper end of the tube must be closed, or evaporation will go on so fast that moisture will be deposited from the air upon the varnished surface. Afterwards the tube may be gently warmed and a current of air allowed to pass, so as to prevent alcohol distilling from one part of the tube to another. The tube is finally heated to the softening point of shellac, and if possible closed as far as is practicable at once.

Sec. 105. Ebonite or Hard Rubber.

This exceedingly useful substance can be bought of a perfectly useless quality. Much of the ebonite formerly used to cover induction coils for instance, deteriorates so rapidly when exposed to the air that it requires to have its surface renewed every few weeks.

The very best quality of ebonite obtainable should be solely employed in constructing electric works. It is possible to purchase good ebonite from the Silvertown Rubber Company (and probably from other firms), but the price is necessarily high, about four shillings per pound or over.

At ordinary temperatures ebonite is hard and brittle and breaks with a well-marked conchoidal fracture. At the temperature of boiling water the ebonite becomes somewhat softened, so that it is readily bent into any desired shape; on cooling it resumes its original hardness.

This property of softening at the temperature of boiling water is a very valuable one. The ebonite to be bent or flattened is merely boiled for half an hour or so in water, taken out, brought to the required shape as quickly as possible, and left to cool clamped in position.

The sheet ebonite as it comes from the makers is generally far from flat. It is often necessary to flatten a sheet of ebonite, and of course this is the more easily accomplished the smaller the sheet. Consequently a bit of ebonite of about the required size is first cut from the stock sheet by a hack-saw such as is generally used for metals. This piece is then boiled and pressed between two planed iron plates previously warmed to near 100 deg. C.

With pieces of ebonite such as are used for the tops of resistance boxes, measuring, say, 20 X 8 X 11 inches, very little trouble is experienced. The sheets when cold are found to retain the flatness which has been forced upon them perfectly well. It is otherwise with large thin sheets such as are used for Holtz machines. I have succeeded fairly, but only fairly, by pressing them in a "gluing press," consisting of heavy planed iron slabs previously heated to 100 deg. C.

I do not know exactly how best to flatten very thin and large sheets. It is easy to make large tubes out of sheet ebonite by taking advantage of the softening which ebonite undergoes in boiling water. A wooden mandrel is prepared of the proper size and shape. The ebonite is softened and bent round it; this may require two or three operations, for the ebonite gets stiff very quickly after it is taken out of the water. Finally the tube is bound round the mandrel with sufficient force to bring it to the proper shape and boiled in water, mandrel and all. The bath and its contents are allowed to cool together, so that the ebonite cools uniformly.

Tubes made in this way are of course subject to the drawback of having an unwelded seam, but they do well enough to wind wire upon if very great accuracy of form is not required. If very accurate spools are needed the mandrel is better made of iron or slate and the spool is turned up afterwards. The seam may be strapped inside or at the ends by bits of ebonite acting as bridges, and the seam itself may be caulked with melted paraffin or anthracene.

Working Ebonite.

Ebonite is best worked as if it were brass, with ordinary brass-turning or planing tools. These tools should be as hard as possible, for the edges are apt to suffer severely, and blunt tools leave a very undesirable woolly surface on the ebonite. In turning or shaping ebonite sheets it is as well to begin by taking the skin off one side first, and then reversing the sheet, finishing the second side, and then returning to the first. This is on account of the fact that ebonite sometimes springs a little out of shape when the skin is removed.

Turned work in ebonite, if well done, requires no sand-papering, but may be sufficiently polished by a handful of its own shavings and a little vaseline. The advantage of using a polished ebonite surface is that such a surface deteriorates more slowly under the influence of light and air than a surface left rough from the tool. If very highly polished surfaces are required, the ebonite after tooling is worked with fine pumice dust and water, applied on felt, or where possible by means of a felt buff on the lathe, and finally polished with rouge and water, applied on felt or cloth.

Ebonite works particularly well under a spiral milling cutter, and sufficiently well under an ordinary rounded planing tool, with cutting angle the same as for brass, and hardened to the lightest straw colour.

It is not possible, on the other hand, to use the carpenter's plane with success, for the angle of the tool is too acute and causes the ebonite to chip.

In boring ebonite the drill should be withdrawn from the hole pretty often and well lubricated, for if the borings jam, as they are apt to do, the heat developed is very great and the temper of the drill gets spoiled. Ebonite will spoil a drill by heating as quickly as anything known; on the other hand, it may be drilled very fast if proper precaution is taken.

It is advisable to expose ebonite to the light as little as possible, especially if the surface is unpolished, for under the combined action of light and air the sulphur at the surface of the ebonite rapidly oxidises, and the ebonite becomes covered with a thin but highly conducting layer of sulphurous or sulphuric acid or their compounds. When this happens the ebonite may be improved by scrubbing with hot water, or washing freely with alcohol rubbed on with cotton waste in the case of apparatus that cannot be dismounted.

A complete cure, however, can only be effected by scraping off the outer layer of ebonite so as to expose a fresh surface. For this purpose a bit of sheet glass broken so as to leave a curved edge is very useful, and the ebonite is then scraped like a cricket bat. In designing apparatus for laboratory use it is as well to bear in mind that sooner or later the ebonite parts will require to be taken down and scraped up. Rods or tubes are, of course, most quickly treated on the lathe with rough glass cloth, and may be finished with fine sandpaper, then pumice dust and water, applied on felt. After cleaning the pumice off by means of water and a rag, the final touch may be given by means of vaseline, applied on cloth or on ebonite shavings.

Sec. 106. Mica.

A great variety of minerals go under this name. Speaking generally, the Russian micas coming into commerce are potash micas, and mica purchased in England may be taken to be potash mica, especially if it is in large sheets.

At ordinary temperatures "mica" of the kind found in commerce is an excellent insulator. Schultze (Wied. Ann. vol. xxxvi. p. 655) comes to the conclusion that both at high and at low temperatures mica (of all kinds?) is a better insulator than white "mirror glass," the composition of which is not stated. The experiments of the author referred to were apparently left unfinished, and altogether too much stress must not be laid on the results obtained, one of which was that mica conducts electrolytically to some extent at high temperatures.

Bouty (Journal de Physique, 1890 [9], 288) and J. Curie (These de Doctorat, Paris, 1888) agree in making the final conductivity of the mica used in Carpentier's condensers exceedingly small—at all events at ordinary temperatures. Bearing in mind that for such substances the term specific resistance has no very definite meaning, M. Bouty considers it is not less than 3.19 x 1028 E.M. units at ordinary temperatures. M. Bouty gives a note or illustration of what such numbers mean—a precaution not superfluous in cases where magnitudes are denoted logarithmically. Referring to the value quoted, viz. 3.19 x 1028, M. Bouty says, "Ce serait la resistance d'une colonne de mercure de 1mmq de section et de longueur telle que la lumiere se propageant dans le vide, mettrait plus de 3000 ans A se transmettre d'une extremite a I'autre de la colonne."

M. Bouty returns to the study of mica (muscovite) in the Journal de Physique for 1892, p. 5, and there deals with the specific inductive capacity, which for a very small period of charge he finds has the value 8—an enormous value for such a good insulator, and one that it would be desirable to verify by some totally distinct method. This remark is enforced by the fact that M. Klemencic finds the number 6 for the same constant. The temperature coefficient of this constant was too small for M. Bouty to determine. The electric intensity was of the order of 100 volts per centimetre, and the experiments seem to indicate that the specific inductive capacity would be only slightly less if referred to a period of charge indefinitely short.

I have found that the residual charge in a mica condenser, made according to Carpentier's method (to be described below), is about 1 per cent of the original charge under the following circumstances.

Voltage 300 volts on a plate 0.2 mm. thick, duration of charge ten minutes, temperature about 20 deg. C. To get this result the mica must be most carefully dried. This and other facts indicate that the so-called residual charge on ordinary condensers is, to a very large extent, due to the creeping of the charge from the armatures over the more or less conducting varnished surfaces of the mica, and its slow return on discharge.

This source of residual charge was carefully guarded against by Rowland and Nichols (Phil. Mag. 1881) in their work on quartz, and is referred to by M. Bouty, who adduces some experiments to show that his own results are not vitiated by it. On the other hand, M. Bouty shows that a small rise in temperature enormously affects the state of a mica surface, and that the surface gets changed in such a way as to become very fairly conducting at 300 deg. C. Also anybody can easily try for himself whether exposing a mica condenser plate which has been examined in presence of phosphorus pentoxide to ordinary air for five minutes will not enormously increase the residual charge, as has always been the case in the writer's experience, and if so, it is open to him to suggest some cause other than surface creeping as an explanation.

M. Bouty, using less perfectly dried mica, did not get so good a result as to smallness of residual charge as the one above quoted.

The chief use of mica for laboratory purposes depends on the ease with which it can be split, and also upon the fact that it may be considerably crumpled and bent without breaking. It therefore makes an excellent dielectric in so far as convenience of construction is concerned in the preparation of condensers, and lends itself freely to the construction of insulating washers or separators of any kind. Its success as a fair insulator at moderate temperatures has led to its use in resistance thermometers, where it appears to have given satisfaction up to, at all events, 400 deg. C.

It is worth a note that according to Werner Siemens, who had immense experience (Wied. Ann. vol. clix.), soapstone is the only reliable insulator at a red heat, but, no doubt, a good deal depends on the particular specimen investigated.

Sec. 107. Use of Mica in Condensers.

If good results are desired it is essential to select the mica very carefully. Pieces appreciably stained,—particularly if the stain is not uniformly distributed,—cracked pieces, and pieces tending to flake off in patches should be rejected. The best samples of mica that have come under the writer's observation are those sheets sold for the purpose of giving to silver photographic prints that hideous glazed surface which some years ago was so popular.

Sheets of mica about 0.1 to 0.2 mm. thick form good serviceable condenser plates, and will certainly stand a pressure of 300 volts, and most likely a good deal more. The general practice in England seems to have been to build up condensers of alternate sheets of varnished or paraffined-mica and tin-foil.

This practice is open to several objections. In the first place, the capacity of a condenser made in this way varies with the pressure binding the plates together. In the second place, the amount of mica and tin-foil required is often excessive in consequence of the imperfect contact of these substances. Again, the inevitable air film between the mica and tin-foil renders condensers so made unsuitable for use with alternating currents, owing to the heating set up through air discharges, and which is generally, though often (if not always) wrongly, attributed to dielectric hysteresis.

These imperfections are to a great extent got over by M. Carpentier's method of construction, which is, however, rather more costly both in material and labour. On the other hand, wonderful capacities are obtained with quite small amounts of mica. M. Bouty mentions a condenser of one microfarad capacity weighing 1500 grms. and contained in a square box measuring 12 centimetres on the side, and about 3 centimetres thick.

The relation between the capacity and surface of doubly-coated plates is in electro-static units:

Capacity = (sp. ind. capacity X area of one surface)/(4pi X thickness)

This may be reduced to electro-magnetic units by dividing by 9x10^20, and to microfarads by further multiplying by 10^15.

M. Carpentier begins, of course, by having his mica scrupulously clean and well selected. It is then silvered by one of the silvering processes (Sec. 65) on both sides, for which purpose the sheets may be suspended in a paraffined wood rack, so as to lie horizontally in the silvering solution, a space of about half an inch being allowed between the sheets. The silvering being finished, the sheets are dipped along two parallel edges in 75 per cent nitric acid. With regard to the third and fourth edges of the sheet, the silver is removed on one side only, using a spun glass brush; if we agree to call the two surfaces of the mica A and B respectively, and the two edges in question C and D, then the silver is removed from the A side along edge C, and from the B side along edge D. The silvered part is shown shaded in Fig. 84. By this arrangement the silver and mica plates may be built up together so as to form the same mutual arrangement of contacts as in an ordinary mica tin-foil condenser.

Fig. 84.

It need hardly be said that the sheets require very complete washing after treatment with nitric acid, followed by a varnishing of the edges as already described in the case of glass, and baking at a temperature of 140 deg. C. in air free from flame gases, till the shellac begins to emit its characteristic odour and is absolutely hard when cold.

The plates are then built up so as to connect the sheets which require to be connected, and to insulate the other set. General contact is, if necessary, secured by means of a little silver leaf looped across from plate to plate—a part of the construction which requires particular attention and clean hands, for it is by no means so easy to make an unimpeachable contact as might at first appear.

The condenser, having been built up, may be clamped solid and placed in its case; the capacity will not depend appreciably on the tightness of the clamp screws—a great feature of the construction. Such a condenser will not give its best results unless absolutely dry. I have kept one very conveniently in a vacuum desiccator over phosphorus pentoxide, but if of any size, the condenser deserves a box to itself, and this must be air-tight and provided with a drying reagent, so arranged that it can be removed through a manhole of some sort.

Contact to the brass-work on the lid may be made by pressing spring contacts tightly down upon the ends of the silver foils and carrying the connections through the lid. This also serves to secure the condenser in position.

Sec. 108. Micanite.

This substance, though probably comparing somewhat unfavourably with the insulators already enumerated, and being subject to the uncertainties of manufacture, has during the last few years achieved a considerable success in American electrical engineering construction. It is composed of scrap mica and shellac varnish worked under pressure to the desired shape, and may be obtained in sheets, plates, and rods, or in any of the forms for which a die happens to have been constructed.

Of course, in special cases it would be worth while to prepare a die, and then the attainable forms would be limited by moulding considerations only. The writer's experience is very limited in this matter, but Dr. Kennelly, with whom he communicated on the subject, was good enough to reply in favour of micanite for engineering work.

Sec. 109. Celluloid.

This material is composed of nitrocellulose and camphor.

It has fair insulating properties, and may be obtained in a variety of forms, but has now been generally abandoned for electrical work on account of its inflammability.

Sec. 110. Paper.

Pure white filter paper, perfectly dry, is probably a very fair insulator; the misfortune is that in practice it cannot be kept dry. Under the most favourable circumstances its specific resistance may approach 1024 E.M. units. It must therefore be considered rather as a partial conductor than as an insulator. The only case of the use of dry paper as an insulator in machine construction which has come under the writer's notice is in building up the commutators of the small motors which used to drive the Edison phonographs.

Its advantages in this connection are to be traced to the fact that a commutator so built up is durable and keeps a clean surface. Of course, the use of paper as an insulator for telephone wires is well known, but its success in this direction depends less upon its insulating properties than upon the fact that it can be arranged in such a way as to allow of the wires being partially air insulated, an arrangement tending to reduce the electrostatic capacity of the wire system.

At one time it was the custom of instrument makers to employ ordinary printed paper in the shape of leaves torn from books or the folios of old ledgers to form the dielectric of the condensers used in connection with the contact breakers of induction coils. This practice has nothing but economy to recommend it, for cases often occur in which the paper, by gradual absorption of moisture from the air, comes to insulate so badly that it practically short circuits the spark gap, and so stops the action of the coil. Three separate cases have come within the writer's experience.

Some measurements of the resistance of paper have been made by F. Uppenborn (Centralblatt fuer Electrotechnik, Vol. xi. p. 215, 1889). There is an abstract of the paper also in Wiedemann's Beiblaetter (1889, vol. xiii. P. 711). Uppenborn examined the samples of paper under normal conditions as to moisture and obtained the following results:-

Description of Paper

I

Pressure Intensity

II.

Specific Resistance corresponding to pressures as in Column I. Ohms.

III

Pressure Intensity.

IV.

Specific Resistance corresponding to Column III. Ohms.

Common cardboard 2.3 mm. thick

0.05 kilo. per 6000 sq. mm.

4.85 x 1015

20 kg. per 6000 sq. mm.

4.7 x 1014

Gray paper, 0.26 mm. thick

0.05 kilo. per 5000 sq. mm.

3.1 x 10^15

20 kg. per 5000 sq. mm.

8 x 1014

Yellow parchment paper-09 mm. thick

0.05 kilo. per 5300 sq. mm.

3.05 x 1016

20 kg. per 5300 sq. mm.

8 x 1016

Linen tracing cloth

0.05 kilo. per 6000 sq. mm.

1.35 x 1016

20 kg. per 33,000 sq. mm.

1.86 x 10^15

Sec. 111. Paraffined Paper.

Like wood and other semiconductors, paper can be vastly improved as an insulator by saturating it with melted paraffin. To get the best results a pure paper free from size must be employed—gray Swedish filter paper does well. This is dried at a temperature above 100 deg. C. for, say, half an hour, and the sheets are then floated on the top of paraffin, kept melted at 140 deg. C. or thereabout in a baking dish. As soon as the paper is placed upon the melted paraffin the latter begins to soak through, in virtue of capillary action, and drives before it the air and moisture, causing a strongly marked effervescence.

After about one minute the paper may be thrust below the paraffin to soak. When a sufficient number of papers have accumulated, and when no more gas comes off, the tray may be placed in a vacuum box (Fig. 85), and the pressure reduced by the filter pump. As the removal of the air takes time, provision must be made for keeping the bath hot.

A vacuum may be maintained for about an hour, and air then readmitted. Repeated exhaustions and readmissions of air, which appear to improve wood, do not give anything like such a good result with paper. In using a vacuum box provision must be made in the shape of a cool bottle between the air pump and the box. If this precaution be omitted, and if any paraffin splashes on to the hot surface of the box, it volatilises with decomposition and the products go to stop up the pump. Paraffin with a melting-point of 50 deg. C. or upwards does well.

The bath should be allowed to cool to about 60 deg. C. before the papers are removed, so that enough paraffin may be carried out to thoroughly coat the paper and prevent the entrance of air.

Fig. 85.

Fig. 85 is a section of a vacuum vessel which has been found very convenient. It measures about two feet in diameter at the top. It is round, because it is much easier to turn one circular surface than to plane up four surfaces, which has to be done if the box is square. Both the rim of the vessel and the approximating part of the cover require to be truly turned and smoothly finished. A very good packing is made of solid indiarubber core about half an inch thick. This is carefully spliced—cemented by means of a solution of rubber in naphtha, and the splice sewed by thick thread. The lid ought to have a rim fitting inside the vessel, for this keeps the rubber packing in place; the rim has been accidentally omitted in Fig. 85. The bolts should not be more than five inches apart, and should lie at least half an inch in diameter, and the rim and lid should be each half an inch thick.

Condensers may now be built up of sheets of this prepared paper interleaved with tin-foil in the ordinary way. If good results are required, the condenser when finished is compressed between wooden or glass end-pieces by means of suitable clamps. It can then be put in a box of melted paraffin, heated up to 140 deg. C, and exhausted by means of the water pump for several hours.

In this process the air rushes out from between the paper and foils with such vehemence that the paraffin is generally thrown entirely out of the box. To guard against this the box must be provided with a loosely fitting and temporary lid, pierced with several holes.

The real test as to when exhaustion is complete would be the cessation of any yield of air or water. Since it is not generally convenient to make the vacuum box so air-tight that there are absolutely no leaks at all, and as the paraffin itself is, I think, inclined to "crack" slightly at the temperature of 140 deg. C, this test or criterion cannot be conveniently applied.

Two exhaustions, each of about two hours' duration, have, however, in my experience succeeded very well, provided, of course, that the dielectric is prepared as suggested. At the end of the exhaustion process the clamping screws are tightened as far as possible, the condenser remaining in its bath until the paraffin is pasty.

Condensers made in this way resist the application of alternating currents perfectly, as the following tests will show. The dielectric consisted of about equal parts of hard paraffin and vaseline. A condenser of about 0.123 microfarads capacity and an insulation resistance of 2000 megohms, [Footnote: As tested by a small voltage.] having a tin-foil area of 4.23 square metres (about), and double papers each about 0.2 mm. thick, designed to run at 2000 volts with a frequency of 63 complete periods, was tested at this frequency.

The condenser was thoroughly packed all round in cotton-wool to a thickness of 6 inches, and its temperature was indicated more or less by a thermometer plunged through a hole in the lid of the containing box and of the condenser box, and resting on the upper surface of one set of tin-foil electrodes, from which the soft paraffin mixture had been purposely scraped away. The following were the results of a four hours' run at a voltage 50 per cent higher than that for which the condenser was designed—i.e. 3000 volts.

Times. Voltage Temperature Temperature Difference in Condenser. in Air.

Hrs. Min.

2 10 2750 22.8 deg. C. 23.0 deg. C. + 0.2 deg.

3 10 2700 23.0 deg. C. 23.3 deg. C. + 0.3 deg.

3 18 3200 23.1 deg. C. 23.0 deg. C. -0.1 deg.

4 10 3200 23.3 deg. C. 23.7 deg. C. + 0.4 deg.

5 10 3100 23.6 deg. C. 23.4 deg. C. -0.2 deg.

6 10 3000 23.8 deg. C. 23.35 deg. C. -0.45 deg.

An idea of the order of the amount of waste may be formed from the following additional experiment.

A condenser similar to the one described was filled with oil of a low insulating power. It was tested calorimetrically, and also by the three voltmeter method, which, however, proved to be too insensitive. The temperature rise in the non-conducting box in air was about 0.3 deg. C. per hour, and the loss of power was found to be less than 0.1 per cent. In the present case the actual rise was only 1 deg. in four hours, and the integral give and take between the condenser and the air is practically nothing; consequently we may consider with safety that the rate of rise is certainly less than 1 degree per three hours. The voltage and frequency were about the same in both experiments, consequently the energy passed is about proportional to the capacity used in the two experiments.

From this it follows that since the specific heat of both condensers was the same (nearly), the loss in the present case is a good deal less than one-tenth per cent. The residual charge is also much less than when the condenser is simply built up of paper paraffined in an unsystematic manner, and from which the air and water have been imperfectly extracted, as by baking the condenser first, and then immersing it in paraffin or oil.

It is usual to consider that the phenomena of residual charge and heating in condensers, to which alternating voltages are applied, are closely allied. This is true, but the alliance is not one between cause and effect—at all events, with regard to the greater part of the heating. The imperfect exclusion of air and moisture, particularly the latter, certainly increases the residual charge by allowing surface creeping to occur; but it also acts directly in producing heating, both by lowering the insulation of the condenser and by allowing of air discharges between the condenser plates.

Of these causes of heating, the discharges in air or water vapour are probably the more important. Long ago a theory of residual charge was given by Maxwell, based on the consideration of a laminated dielectric, the inductivity and resistance of which varied from layer to layer. It was shown that such an arrangement, and hence generally any want of homogeneity in a direction inclined to the lines of force leading to a change of value of the product of specific resistance and specific inductive capacity, would account for residual charge.

This possible explanation has been generally accepted as the actual explanation, and many cases of residual charge attributed to want of homogeneity, which are certainly to be explained in a simpler manner. For instance, the residual charge in a silvered mica plate condenser, carefully dried, can be increased at least tenfold by an exposure of a few minutes to ordinarily damp air. The same result occurs with condensers of paraffined or sulphured paper; and these are the residual changes generally observed. The greater part must be due to creeping.

Sec. 112. Paraffin.

This substance has long enjoyed great popularity in the physical laboratory. Its specific resistance is given by Ayrton and Perry as more than 1025, but it is probably much higher in selected samples. The most serviceable kind of paraffin is the hardest obtainable, melting at a temperature of not less than 52 deg. C. It is a good plan to remelt the commercial paraffin and keep it at a temperature of, say, 120 deg. C. for an hour, stirring it carefully with a glass rod so that it does not get overheated; this helps to get rid of traces of water vapour.

Hard paraffin, when melted, has an enormous rate of expansion with temperature, so great, indeed, that care must be taken not to overfill the vessels in which it is to be heated. Castings can only be prepared by cooling the mould slowly from the bottom, keeping the rest of the mould warm, and adding-paraffin from time to time to make up for the contraction. The cooling is gradually allowed to spread up to the free surface.

The chief use of paraffin in the laboratory is in the construction of complicated connection boards, which are easily made by drilling holes in a slab of paraffin, half filling them with mercury, and using them as mercury cups.

Since paraffin is a great collector of dust, it should be screened by paper, otherwise the blocks require to be scraped at frequent intervals, which, of course, electrifies them considerably. This electrification is often difficult to remove without injuring the insulating power of the paraffin. A light touch with a clean Bunsen flame is the readiest method, and does not appear to reduce the insulation so much as might be expected. The safest way, however, is to leave the key covered by a clean cloth, which, however, must not touch the surface, for a sufficient time to allow of the charges getting away.

The paraffin often becomes electrified itself by the friction of the key contacts, so that in electrometer work it is often convenient to form the cups by lining them with a little roll of copper foil twisted up at the bottom. In this case the connecting wires should, of course, be copper. Small steel staples are convenient for fastening the collecting wires upon the paraffin; or, in the case where these wires have to be often removed and changed about, drawing-pins are very handy.

With mercury cups simply bored in paraffin great trouble will often be experienced in electrometer work, owing to a potential difference appearing between the cups—at all events when the contacts are inserted and however carefully this be done. A few drops of very pure alcohol poured in above the mercury often cures this defect. The surface of paraffin is by no means exempt from the defect of losing its insulating power when exposed to damp air, but it is not so sensitive as glass, nor does the insulating power fall so far.

Two useful appliances are figured.

Fig. 86. Fig. 87.

One, in which paraffin appears as a cement, is an insulating stand made out of a bit of glass or ebonite tube cemented into an Erlenmeyer flask, having its neck protected from dust when out of use by a rubber washer, the other a "petticoat" insulator made by cementing a flint glass bottle into a glass dish with paraffin. In course of time the paraffin will be found to have separated from the glass. When this occurs the apparatus may be melted together again by placing it on the water bath for a few minutes.

Sec. 113. Vaseline, Vaseline Oil, and Kerosene Oil.

These, when dry, insulate almost, but not quite as well as solid paraffin. H. Koeller (Wien Berichte, 98, ii. 201, 1889; Beibl. Wied. Ann. 1890, p. 186), working with very small voltages, places the final(?) specific resistance of commercial petroleum, ether, and vaseline oil at about 2 X 1027 C.G.S. This is ten times higher than the value assigned to commercial benzene (C6H6), and a hundred times higher than the value for commercial toluene.

All these numbers mean little or nothing, but the petroleum and vaseline oil were the best fluid insulators examined by Koeller. By mixing vaseline with paraffin a soft wax may be made of any desired degree of softness, and by dissolving vaseline in kerosene an insulating liquid of any degree of viscidity may be obtained.

Hard paraffin may be softened somewhat by the addition of kerosene, and an apparently homogeneous mass cast from the mixture. It will be found, however, that in course of time the kerosene oozes out, unless present in very small quantity. Koeller has found (loc. cit.) that some samples of vaseline oil conducted "vollstaendig gut," but I have not come across such samples. Vaseline oil, however, is sold at a price much above its value for insulating purposes.

Kerosene oil is best obtained dry by drawing it directly from a new tin and exposing it to air as little as possible. Of course, it may be dried by chemical means and distillation, but this is usually (or always) unnecessary.

Fig 88.

There is some danger of kerosene containing minute traces of sulphuric acid, and it and other oils may be conveniently tested for insulation in the following manner. The quartz electroscope is taken, and the insulating rod heated in the blow-pipe. The electroscope will now insulate well enough to show no appreciable collapse of the leaves in one or two hours' time. Upon the plate of the electroscope is put a platinum or copper cylinder, and this is filled with kerosene (say) up to a fixed mark.

The electroscope is placed on a surface plate, or, at all events, on a sheet of plate glass, and a "scribing block" is placed along side it and the scriber adjusted to dip into the kerosene to any required depth. This is done by twisting a bit of wire round the scribing point and allowing it to project downwards. The point itself serves to give an idea of the height to which the vessel may be filled. The liquid is adjusted till its surface is in contact with the end of the scribing point, and the wire then projects into the liquid and forms an electrode of constant area of surface. The scribing block is put to earth. A charge is given to the electroscope, and the time required for a given degree of collapse of the leaves noted.

The kerosene is then removed and its place taken by vaseline or paraffin, known to insulate well as a standard for comparison. The experiment is then repeated, and the time noted for the same degree of collapse. This test, though of course rough, is generally quite sufficient for workshop purposes, and is easily applied. Moreover, it does not require correction for electrometer leakage, as generally happens when more elaborate appliances are used.

The actual resistance of insulating oils depends so much on the electrical intensity, on the duration of that intensity, and on the previous history of the oil as to the direction of the voltage to which it has been subjected—to say nothing of the effects of traces of moisture—that quantitative experiments are of no value unless they are extremely elaborate. I shall therefore only append the following numbers due to Bouty, Ann. de Chemie et de Physique (6), vol. xxvii. p. 62, 1892, in which the effect of the conductivity on the determination of the specific inductive capacity was properly allowed for:-

Carbon

Bisulphide.

Turpentine.

Benzene (C6H6) at 20 deg. C.

Benzene at

60 deg. C.

Specific inductive capacity

2.715

2.314

2.21

2.22

Specific resistance in ohms per cubic centimetre

1.5 x 1013,

1.75 x 1012

1.56 x 1011

7.9 x 1011

[Footnote: Professor J. J. Thomson, and Newall (Phil. Proc. 1886) consider that carbon bisulphide showed traces of a "residual charge" effect; hence, until this point is cleared up, we must regard Bouty's value as corresponding only to a very short, but not indefinitely short, period of charge. On this point the paper must be consulted.

March 1897—The writer has investigated this point by an independent method, but found no traces of "residual charge."]

Information as to the specific inductive capacity of a large number of oils may be found in a paper by Hopkinson, Phil. Proc. 1887, and in a paper by Quincke in Wiedemann's Annalen, 1883.

Sec. 114. Imperfect Conductors.

Under this heading may be grouped such things as wood, slate, marble, etc—in fact, materials generally used for switchboard insulation. An examination of the insulating power of these substances has recently been made by B. O. Peirce (Electrical Review, 11th January 1895) with quite sufficient accuracy, having in view the impossibility of being certain beforehand as to the character of any particular sample. The tests were made by means of holes drilled in slabs of the material to be examined. These holes were three-eighths of an inch in diameter, and from five-eighths to three-quarters of an inch deep, and one inch apart, centre to centre. A voltage of about 15 volts was employed. The following general results were arrived at:-

(1) Heating in a paraffin bath always increases the resistance of wood, though only slightly if the wood be hard and dense.

(2) Frequent exhaustion and readmission of air above the surface of the paraffin always has a good effect in increasing the resistance of wood.

(3) When wood is once dry, impregnating it with paraffin tends to keep it dry.

(4) Red vulcanised fibre, like wood, absorbs paraffin, but it cannot be entirely waterproofed in this way.

(5) The resistance of wood with stream lines along the grain is twenty to fifty per cent lower than when the stream lines cross the grain.

(6) The "contact" resistance between slabs of wood pressed together is always very high.

The following table will sufficiently illustrate the results obtained. The stone was dried in the sun for three weeks in the summer (United States), and the wood is described as having been well seasoned:-

CURRENT WITH THE GRAIN

Lowest Resistance Highest Resistance Lowest Specific Highest Specific between two Cups between two Cups Resistance in Resistance in in Megohms. in Megohms. Megohms. Megohms.



Ash.

550 920 380 700

Cherry

1100 4000 2800 6000

Mahogany

430 730 310 610

Oak

220 420 1050 2200

Pine.

330 630 360 1470

Hard pine.

10 48 17 1050

Black walnut

1100 3000 320 2100

Red fibre

2 4 3 60

Slate

184 280

Soapstone.

330 500

White marble

2000 8800

Sec. 115. As to working the materials very little need be said.

Fibre is worked like wood, but has the disadvantage of rapidly taking the edge off the tools. In turning it, therefore, brass-turning tools, though leaving not quite such a perfect finish as wood-turning tools, last much longer, and really do well enough. Fibre will not bear heating much above 100 deg.C—at all events in paraffin. At 140 deg. C. it becomes perfectly brittle. Its chief merit lies in its great strength. So far as insulation is concerned, Mr. Peirce's experiments show that it is far below most kinds of wood.

Slate. This is a vastly more useful substance than it is generally credited with being. It is very easily worked at a slow speed, either on the shaping machine or on the lathe, with tools adjusted for cutting brass, and it keeps its figure, which is more than can be said for most materials. It forms a splendid base for instruments, especially when ground with a little emery by iron or glass grinders, fined with its own dust, and French polished in the ordinary way. Spools for coils of considerable radial dimension may be most conveniently made of slate instead of wood or brass, both because it keeps its shape, and because it insulates sufficiently well to stop eddy currents—at all events, sufficiently for ordinary practice. An appreciable advantage is that slate may be purchased at a reasonable rate in large slabs of any desired thickness. It is generally cut in the laboratory by means of an old cross-cut saw, but it does not do much damage to a hard hack saw such as is used for iron or brass.

Marble. According to Holtzapffell, marble may be easily turned by means of simple pointed tools of good steel tempered to a straw colour. The cutting point is ground on both edges like a wood-turning tool, and held up to the work "at an angle of twenty or thirty degrees" (?with the horizontal). The marble is cut wet to save the tool. As soon as the point gets, by grinding, to be about one-eighth of an inch broad it must either be drawn down or reground; a flat tool will not turn marble at all.

A convenient saw for marble is easily made on the principle of the frame saw. A bit of hoop iron forms a convenient blade, and is sharpened by being hammered into notches along one edge, using the sharp end of a hammer head. The saw is liberally supplied with sand and water—or emery and water, where economy of time is an object. The sawing of marble is thus really a grinding process, but it goes on rapidly. Marble is ground very easily with sand and water; it is fined with emery and polished with putty powder, which, I understand, is best used with water on cloth or felt. As grinding processes have already been fully described, there is no need to go into them here. I have no personal knowledge of polishing marble.

Sec. 116. Conductors.

The properties of conductors, more particularly of metals, have been so frequently examined, that the literature of the subject is appallingly heavy. In what follows I have endeavoured to keep clear of what might properly appear in a treatise on electricity on the one hand, and in a wiring table on the other. The most important work on the subject of the experimental resistance properties of metals has been done by Matthieson, Phil. Trans. 1860 and 1862, and British Association Reports (1864); Callender, Phil. Trans. vol. clxxiii; Callender and Griffiths, Phil. Trans. vol. clxxxii; The Committee of the British Association on Electrical Standards from 1862 to Present Time; Dewar and Fleming, Phil. Mag. vol. xxxvi. (1893);

Klemencic, Wiener Sitzungsberichte (Denkschrift), 1888, vol. xcvii. p. 838; Feussner and St. Lindeck, Zeitsch. fuer Inst. 'Kunde, ix. 1889, p. 233, and B. A. Reports, 1892, p. 139. Of these, Matthieson, and Dewar and Fleming treat of resistance generally, the latter particularly at low temperatures.

[Footnote: The following is a list of Dr. Matthieson's chief papers on the subject of the electrical resistance of metals and alloys: Phil. Mag. xvi. 1858, pp. 219-223; Phil. Trans. 1858, pp. 383-388 Phil. Trans. 1860, pp. 161-176; Phil. Trans. 1862, pp. 1-27 Phil. Mag. xxi. (1861), pp. 107-115; Phil. Mag. xxiii. (1862), pp. 171-179; Electrician, iv. 1863, pp. 285-296; British Association Reports, 1863, p. 351.]

Matthieson, and Matthieson and Hockin, Klemencic, Feussner, and St. Lindeck deal with the choice of metals for resistance standards. Callender's, and Callender and Griffiths' work is devoted to the study of platinum for thermometric purposes.

The bibliography referring to special points will be given later. The simplest way of exhibiting the relative resistances of metals is by means of a diagram published by Dewar and Fleming (loc. cit.), which is reproduced on a suitable scale on the opposite page. For very accurate work, in which corrections for the volumes occupied by the metals at different temperatures are of importance, the reader is referred to the discussion in the original paper, which will be found most pleasant reading. From this diagram both the specific resistance and the temperature coefficient may be deduced with sufficient accuracy for workshop purposes. In interpreting the diagram the following notes will be of assistance. The diagram is drawn to a scale of so-called "platinum temperatures"—that is to say, let R0, R100, Rt be the resistances of pure platinum at 0 deg., 100 deg., and t deg. C. respectively, then the platinum temperature pt is defined as

pt = 100 X (Rt-R0)/(R100-R0)

This amounts to making the temperature scale such that the temperature at any point is proportional to the resistance of platinum at that point. Consequently on a resistance temperature diagram the straight line showing the relation between platinum resistance and platinum temperature will "run out" at the platinum absolute zero, which coincides more or less with the thermodynamic absolute zero, and also with the "perfect gas" absolute zero. Platinum temperatures may be taken for workshop purposes over ordinary ranges as almost coinciding with air thermometer temperatures. The metals used by Professors Dewar and Fleming were, with some exceptions, not absolutely pure, but in general represent the best that can be got by the most refined process of practical metallurgy. We may note further that the specific resistance is only correct for a temperature of about 15 deg. C, since no correction for the expansion or contraction of material has been applied.

The following notes on alloys suitable for resistance coils will probably be found sufficient.

Sec. 117. Platinoid.

This substance, discovered by Martino and described by Bottomley (Phil. Proc. Roy. Soc. 1885), is an alloy of nickel, zinc, copper, and 1 per cent to 2 per cent of tungsten, but I have not been able to obtain an analysis of its exact composition. It appears to be difficult to get the tungsten to alloy, and it has to be added to part of the copper as phosphide of tungsten, in considerably greater quantity than is finally required. The nickel is added to part of the copper and the phosphide of tungsten, then the zinc, and then the rest of the copper. The alloy requires to be remelted several times, and a good deal of tungsten is lost by oxidation.

The alloy is of a fine white colour, and is very little affected by air—in fact, it is to some extent untarnishable. The specific resistance will be seen to be about one and a half times greater than that of German silver, and the temperature coefficient is about 0.021 per cent per degree C. (i.e. about nineteen times less than copper, and half that of German silver). To all intents and purposes it may be regarded as German silver with 1 per cent to 2 per cent of tungsten. It does not appear to have been particularly examined for secular changes of resistance.

118. German Silver. This material has been exhaustively examined of late years by Klemencic and by Feussner and St. Lindeck. Everybody agrees that German silver, as ordinarily used for resistances, and composed of copper four parts, zinc two parts, nickel one part, is very ill-fitted for the purpose of making resistance standards. This is due (1) to its experiencing a considerable increase in resistance on winding. Feussner and St. Lindeck found an increase of 1 per cent when German silver was wound on a core of ten wire diameters.

(2) To the fact that the change goes on, though with gradually decreasing rate, for months or years;

(3) to the fact that the resistance is permanently changed (increased) by heating to 40 deg. C. or over. By "artificially ageing" coils of German silver by heating to 150 deg. C, say for five or six hours, its permanency is greatly improved, and it becomes fit for ordinary resistance coils where changes of, say, 1/5000 do not matter.

It is a remarkable property of all nickel alloys containing zinc that their specific resistance is permanently increased by heating, whereas alloys which do not contain zinc suffer a change in the opposite direction. The manufacturers of German silver appear to take very little care as to the uniformity of the product put on the market; some so-called German silver is distinctly yellow, while other samples are bright and white.

It is noted by Price (Measurements of Electrical Resistance, p. 24) that German silver wire is apt to exhibit great differences of resistance within quite short lengths. This has been my own experience as well, and is a great drawback to the use of German silver in the laboratory, for it makes it useless to measure off definite lengths of wire with a view to obtaining an approximate resistance. In England German silver coils are generally soaked in melted hard paraffin. In Germany, at all events at the Charlottenburg Institute, according to St. Lindeck—coils are shellac-varnished and baked. In any case it appears to be essential to thoroughly protect the metal against atmospheric influence.

Sec. 119. Platinum Silver.

In the opinion of Matthieson and of Klemencic the 10 per cent silver, 90 per cent platinum alloy is the one most suitable for resistance standards. At all events, it has stood the test of time, for, with the following exceptions, all the British Association coils constructed of it from 1867 to the present day have continued to agree well together. The exceptions were three one-ohm coils, which permanently increased between 1888 and 1890, probably through some straining when immersed in ice. One coil changed by 0.0006 in 1 between the years 1867 and 1891. According to Klemencic, absolute permanency is not to be expected even from this alloy.

Its recommendation as a standard depends on its chemical inertness, its small temperature coefficient (0.00027 per degree), and its small thermo-voltage against copper, as the following table (taken from Klemencic) will show:-

Thermo-voltages in Micro-volts per degree against Copper over the Range 0 deg. to 17 deg. C.

Platinum iridium 7.14 micro-volts per degree C.

Platinum silver 6.62 micro-volts per degree C.

Nickelin 28.5 micro-volts per degree C.

German silver 10.43 micro-volts per degree C.

Manganin (St. Lindeck) 1.5 micro-volts per degree C.

Mechanically, the platinum silver is weak, and is greatly affected as to its resistance by mechanical strains—in fact, Klemencic considers it the worst substance he examined from this point of view—a conclusion rather borne out by Mr. Glazebrook's experience with the British Association standards already referred to (B. A. Reports, 1891 and 1892).

Taking everything into account, it will probably be well to construct standards either with oil insulation only, or to bake the coils in shellac before testing, instead of soaking in paraffin. Fig. 89 illustrates a form of an oil immersed standard which is in use in my laboratory, and through which a considerable current may be passed. The oil is stirred by means of a screw propeller.

Fig. 89.

Fig. 89 represents a standard resistance for making Clerk cell comparisons by the silver voltameter method. The framework on which the coils are wound consists of a base and top of slate. The pillars are of flint glass tube surrounding brass bolts, and cemented to the latter by raw shellac. Grooves are cut in the glass sleeves to hold the wires well apart. These grooves were cut by means of a file working with kerosene lubrication. A screw stirrer is provided, and the whole apparatus is immersed in kerosene in the glass box of a storage cell. The apparatus is aged to begin with by heating to a temperature a good deal higher than any temperature it is expected to reach in actual work. After this the rigidity of the frame is intended to prevent any further straining of the wire. The apparatus as figured is not intended to be cooled to 0 deg. C, so that it is put in as large a box as possible to gain the advantage of having a large volume of liquid. The top and bottom slates measure seven inches by seven inches, and the distance between them is seven inches. The inner coil is wound on first, and the loop which constitutes the end of the winding is brought up to a suitable position for adjustment. The insulation of the heavy copper connectors is by means of ebonite.

Sec. 120. Platinum Iridium.

Platinum 90 per cent, iridium 10 per cent. This material was prepared in some quantity at the cost of the French Government, and distributed for test about 1886. Klemencic got some of it as representing Austria, and found it behaved very like the platinum silver alloy just discussed. The temperature coefficient is, however, higher than for platinum silver (0.00126 as against 0.00027). The mechanical properties of the alloy are, however, much better than those of the silver alloy; and in view of the experience with B. A. standards above quoted, it remains an open question whether, on the whole, it would not be the better material for standards, in spite of its higher price. Improvements in absolute measurements of resistance, however, may render primary standards superfluous.

Sec. 121. Manganin.

Discovered by Weston—at all events as to its application to resistance coils. A glance at the diagram will exhibit its unique properties, on account of which it has been adopted by the Physikalisch Technischen Reichsanstalt for resistance standards. The composition of the alloy is copper 84 per cent, manganese 12 per cent, nickel 4 per cent, and it is described as of a steel-gray colour. Unfortunately it is apt to oxidise in the air, or rather the manganese it contains does so, so that it wants a very perfect protection against the atmosphere.

Like German silver, manganin changes in resistance on winding, and coils made of it require to be artificially aged by heating to 150 deg. for five hours before final adjustment. The annealing cannot be carried out in air, owing to the tendency to oxidation. The method adopted by St. Lindeck (at all events up to 1892) is to treat the coil with thick alcoholic shellac varnish till the insulation is thoroughly saturated, and then to bake the coil as described. The baking not only anneals the wire, but reduces the shellac to a hard and highly insulating mass.

Whether stresses of sufficient magnitude to produce serious mechanical effects can be set up by unequal expansion of wire and shellac during heating and cooling is not yet known, but so far as tested (and it must be presumed that the Reichsanstalt tests are thorough) no difficulty seems to have been met with. In course of time, however, probably the best shellac coating will crack, and then adieu to the permanency of the coil! This might, of course, be obviated by keeping the coil in kerosene, which has no action on shellac, but which decomposes somewhat itself.

The method of treatment above described suffices to render coils of manganin constant for at least a year (in 1892 the tests had only been made for this time) within a few thousands per cent. Manganin can be obtained in sheets, and from this material standards of 10-2, 10-3, and 10-4 ohms are made by soldering strips between stout copper bars, and these are adjusted by gradually increasing their resistance by boring small holes through them. The solder employed is said to be "silver."

Mr. Griffiths (Phil. Trans. vol. clxxxiv. [1893], A, p. 390) has had some experience with manganin carrying comparatively heavy currents, under which circumstances its resistance when immersed in water was found to rise in spite of the varnish which coated it. Other experiments in which the manganin wire was immersed in paraffin oil did not exhibit this effect, though stronger currents were passed.

On the whole, manganin appears to be the best material for coil boxes and "secondary" resistance standards. Whether it is fit to rank with the platinum alloys as regards permanency must be treated as an open question.