High Voltage Insulation

High-Voltage InsulationBY J. L. R. HAYDEN and CHARLES P. STEINMETZ

Member, A. I. E. E. Fellow, A. I. E. E.Chief Consulting Engineer, General Electric Co., Schenectady, N. Y.

I. General reached 220 kv. Then 12 kv. was the highest satis-T HE most important chapters in electrical engineer- factory cable voltage; now we have reached 22 kv.

ing are those dealing with efficiency, heating, and a few cables of 33 kv. and higher, but cables atmagnetism and insulation. 33 kv. are still semi-experimental. The comparison

In the field of insulation, our knowledge is most shows that the advance in pushing cable voltages up tobackward. higher values, has been slower than with overhead

In regard to efficiency, with efficiencies of electrical lines, and our knowledge of liquid and solid insulation,apparatus of 90 up to over 99 per cent, nofurtherradical such as come into consideration in the cable wall andprogress appears feasible. the machine insulation, is materially less advanced

In magnetism, the losses in an alternating field have than that of air as dielectric. With regard to air, abeen reduced so far, that they have ceased to limit, by good working theory has been established, by con-their heating effect, the size of apparatus, but are merely sidering the dielectric strength of air as analogous toa factor in the efficiency. An increase of the saturation the mechanical strength of structural materials. Thedensity would decrease the size of apparatus, but is theory recognizes a definite dielectric strength of air,excluded by its inherent chemical limitations. or a disruptive breakdown gradient, of 30 kv. per cm.

Heating specifications are not made for other classes at normal air density. Puncture occurs when thisof apparatus, such as prime movers, etc., and if an dielectric strength is exceeded, just as mechanicalimportant subject in electrical engineering, it is almost disruption occurs, when anywhere in a mechanicalentirely because the insulation of the apparatus is structure the stresses exceed the elastic limit of thedestroyed by the higher temperatures. Thus the prob- material. This conception of a definite breakdownlem of the heating of electrical apparatus is essentially strength then was extended to liquid and solid dielec-one aspect of the insulation problem. trics, but with these, it failed to give a satisfactory

In our high-voltage apparatus, cables, etc., we operate explanation of the mechanism of the breakdown, morethe insulation at voltage stresses which rarely exceed particularly of the all-important feature of the timemuch the disruptive strength of air, though laboratory lag of disruption. And even with air, the theory of atests often show this insulation to have a disruptive constant breakdown gradient, as modified by the con-strength of 10 to 20 times that of air. ception of the energy distance, is satisfactory only

All phenomena of nature are very complex. There- within a certain range.fore, in calculating a phenomenon or designing an ap-

.

.

paratus, we must approximate by neglecting "second- II. Air as Dielectricary terms," and take care of these by an allowance, The present practically universally accepted theorya margin or a factor of safety. Obviously, the more of a dielectric strength of air at and near atmos-completely a phenomenon is known and understood, pheric pressure, as most completely developed bythe closer it can be calculated, in other words, the less Mr. F. W. Peek, Jr., is:is the margin or safety factor required in order to allow Air has a definite and constant "dielectric strength,"for the unknown stresses, etc. The margin or safety at which it ceases to be an insulator and becomes afactor, which experience shows as necessary, thus is an conductor, that is, breaks down electrically.indication of the exactness of our knowledge of a phe- The dielectric strength of air is proportional to thenomenon. For example when dealing with magnetic air density, and is 30. kv. per cm. at normal air densityphenomena, with efficiency, with heating, etc., we have of 0 deg. cent. and 76 cm. barometer.to allow a margin of a few per cent only. In testing The dielectric breakdown (or puncture) of air doesthe insulation of apparatus however, the A. I. E. E. not occur as soon as the voltage gradient in the di-standards specify a test voltage more than twice the electric field exceeds the dielectric strength at anydelta voltage, though the normal stress is the Y voltage. point, but the voltage gradient in the field must exceedThat is, we require a safety factor of over 3.46, a the dielectric strength over a finite distance, the so-margin above normal of over 246 per cent. called "energy distance."The insulation problem has become of increasing The energy distance depends on the convergency of

importance with the rapid advance of electrical en- the electric field at the place where the breakdowngineering into higher voltages. Not many years ago occurs, and is the less, the more convergent the field.44 kv. was the highest transmission voltage for re- The energy distance between parallel cylindersliability of operation of overhead lines. Now we have (wires) of radius R? iS 0.3 ub R; between_spheres of

Presented at the Pacific Coast Convention of the A. I. E. E., radius R it is, approximately; 0.54 Vx,6 1R, whereDeZ Monte, Cal., October 2-5, 192 3 . uS= air density (with normal air density as unity).

1029

1030 HAYDEN AND STEINMETZ: HIGH-VOLTAGE INSULATION Transactions A. I. E. E.

Thus the breakdown gradient at the surface of two For gap lengths less than the energy distance, theparallel cylinders of radius R is: stand.ard theory does not apply.

0.3 For gaps in which corona precedes the spark dis-90 = 303 (1 + charge, the law does not seem to satisfactorily agree

with experience. Between spheres, corona appearsat the surface of spheres it is only at gap lengths materially larger than those in

0.54 which it should appear according to the theory. Also9o= 30 ( 1 + 4 the disruptive voltage at which spark discharge occursV3Iib between needle points, is materially lower than cal-

For gaps of a length less than the energy distance, culated. This may partly be accounted for as follows:a higher breakdown gradient is required, the more so, (a) The space filled by corona is not a perfect con-the shorter the gap. The law is unknown. ductor, but has a finite resistance, which depends onThe space in the dielectric field in which the dielectric the corona density, thus on the voltage, frequency, etc.

strength is exceeded, becomes conducting, and, if ad- (b) It is practically an impossibility to calculatejacent to the terminals, becomes a part of the terminal. accurately the final dielectric field and its voltage gra-This changes the configuration of the dielectric field dients, as resulting from the partial short circuit by theand if thereby the dielectric strength is exceeded in corona, of regions of the initial dielectric field, even ifother parts of the dielectric field, these also break down the resistivity of the corona regions were neglected.and become conducting, until either the breakdown (c) The nature of the corona-filled space is that ofextends from terminal to terminal, the air gap punctures an unstable third class conductor, which tends toand short-circuits by a spark discharge (which tends to produce local oscillations and other transients, anddevelop into an arc), and the current rises and the thereby rapid variations of the resultant field andvoltage drops to the short-circuit values of the supply premature breakdown. Incidently, when high-fre-source. Or the breakdown in the dielectric field limits quency oscillations and other low-power transients ofitself, producing a new configuration of the dielectric unknown origin appear in a system, it is advisable tofield, containing conducting regions, in which the air look for the existence of corona somewhere in the cir-is broken down and luminous by "corona," separated cuit, as possible cause.by insulating air spaces below the breakdown gradient. The theory of the dielectric strength of air at andIn this case the discharge current is limited to the near atmospheric pressure, outlined above, is based onsmall "corona current." A further increase of voltages the distribution of dielectric stresses and thereforethen may extend the breakdown across the terminals depends on the configuration of the dielectric field. Asand change the limited discharge into an unlimited the dielectric field is essentially determined by thespark discharge. shape of the terminals, the latter thus is of fundamentalBetween needle points, corona always precedes the importance. At the same time, in the "conduction of

spark discharge; between spheres at moderate distance gases in a vacuum" a theory has been developed, based(less than 3 to 4 times the radius of the spheres), on convection phenomena by carriers, the ions, and inspark discharge occurs without corona. this theory the configuration of the dielectric field andThe disruptive discharge of single gaps in air, is the shape of the terminals is practically not considered,

instantaneous, that is, no time lag is produced in the and the two conceptions thus apparently are out ofspark gap, if no corona precedes the spark discharge, line with each other, though the phenomena of theand the gap length is greater than the energy distance. Geisler tube and those of air at normal pressure shouldWhen with rising voltage corona precedes the spark differ quantitatively only, by the different value of airdischarge, a time lag exists. With gaps of less than the density 3.energy distance, and multigaps, a time lag also seems Approximately, and within a limited range, theto exist. The combination of resistance in series and voltage e required to break down a gap of length 1 atcapacity in shunt to an instantaneous spark gap gives air density 3, is constant as long as the product 13 isthe gap a time lag owing to the time required to charge constant, that is, for constant quantity of air withinthe capacity by a current limited in value by the the gap. In other words, for different air densities 3,resistance. Without a capacity in shunt, an otherwise the gap length 1 at constant voltage e varies inverselyinstantaneous sphere gap in series with a resistance proportional with the air density 3. The air densitieslags by the time of charging the capacity of the spheres in vacuum tubes usually vary from one thousandth tothemselves, through the series resistance. For high one millionth of normal air density and less, and evenvalues of resistance, this may become quite consider- the longest vacuum tubes usually therefore correspondable, especially with large spheres. at atmospheric pressure to gaps of very small length,

This standard theory of dielectric strength of air at greatly within the energy distance. In other words,and near atmospheric pressure is fairly satisfactory for the phenomena of vacuum tube conduction correspondgaps ranging between the energy distance and a larger to the phenomena occurring in very small gaps atvalue where corona begins, atmospheric pressure, where the conventional theory

October 1923 HAYDEN AND STEINMETZ: HIGH-VOLTAGE INSULATION 1031

does not apply, and the configuration of the field and Investigation shows that it is the form and intensitythe shape of the terminals become of lesser importance. of the dielectric field at and near the positive terminalInversely, the voltage phenomena of the Geisler tube, which has the greatest effect on the disruptive voltagemay also appear in the dielectric strength of large gaps of the air gap, while the field at and near the negativeat atmospheric pressure, as terminal phenomena, terminal is of secondary importance. Thus if at con-though then secondary in magnitude to the gradient stant gap length the positive terminal is graduallyvoltages of the dielectric field. changed in steps from a sharp point to a flat plate,The conventional theory considers only the voltage while the negative terminal remains unchanged in form,

gradient or intensity of the dielectric field as determin- the disruptive voltage changes correspondingly over aing the dielectric strength, but not the direction, that wide range. A corresponding change in the physicalis, assumes the same breakdown strength for either form of the negative terminal however, with the posi-polarity of the impressed voltage. This resulted from tive terminal remaining unchanged in form, varies thethe use of alternating voltages in determining the laws disruptive voltage relatively little.of the dielectric strength of air, due to high direct volt- In Fig. 1, curve I approaches the striking distance

curve between needles, as modified by the. different100 distribution of voltage gradients due to the large

AC.lMaximum , | negative terminal, and curve II similarly approaches90 / ____ the striking distance curve between 12.5 cm. spheres.

r l Z/II III X> H This suggests the explanation that the disruptive/ /' I discharge is due to carriers of current produced by the80 field at the positive terminals, the positive ions possibly.

Sphere Positive // Tests made at atmospheric pressure on very smalln70 Grounded & High | I Igaps-down to fractions of microns-show that with/ Sphere Negative decreasing gap length the voltage does not indefinitely

E / Needle GroundedI/ decrease, but reaches a finite minimum value, ofU60 Shere Negative about 320 volts, and becomes apparently constant at_ / I /' and Grounded this value. This would be the minimum voltage re-50 - _/ll quired to disrupt an air gap, no matter how small. It

means that with decreasing gap length, the voltagezI3: gradient of the air film rises to very high values, andD 40 | 1t' 4|tl1such thin air films have an extremely high dielectric

Ai | | / } l strength; gradients of 6.2 million volts per cm. haveCo3 /30 been reached.

Inversely, with increasing gap length, the voltage20L X I ; I < increases with the gap length, and proportional thereto,/,/l j j from an initial value of about 320 volts. The increase

with increasing gap length, however, is not at the rate10 ___ of the dielectric strength of air, 30 kv. per cm., as

might be expected, but is at approximately twice this0 2 rate; about 60 kv. per cm., up to voltages of abouto 2 4 6 8 10 12 14 4500. Then the increase of voltage with increasing

GAP IN CENTIMETERS gap length decreases and approaches the normal valueFIG. 1 of the breakdown strength, 30 kv. per cm.

The voltage e required to break down an air gap ofage of considerable power being unavailable until length 1 and uniform dielectric field intensity thusrecent years. It is very far from correct, however, and would consist ofin an unsymmetrical field, the disruptive strength with (1) A constant voltage el = 320 volts = 0.32 kv.the voltage in one direction, may be more than twice (2) A voltage gradient g2 of about 60 kv. per cm.that with the voltage in opposite direction. As illus- over a short length lo = 0.07 cm., giving a constanttration, Fig. 1 shows the striking distances between a voltage e2 = 60 lo kv.needle and a 12.5 cm. sphere, for various gap lengths. (3) The voltage gradient of dielectric strength,Curve I gives the results winth the needle positive, g0 = 30 kv. per cm., for the rest of the gap, I-lo~orcurve II with the needle negative, and curve III with g3 = 30 (1I-lo ); thus giving a total voltage, ap-alternating voltage. As seen, I and II differ greatly, proximately:II being more than twice the voltage of I. The alter- e = e1 + e2 + e3nating curve approximates I, as is to be expected. In = 0.32 + 30 1o + 30 lthe voltage range between I and II, such an unsym- or, if 1< 10, simplymetrical gap rectifies. e = 0.32 60 1

1032 HAYDEN AND STEINMiETZ: HIGH-VOLTAGE INSULATION Transactions A. I. E. E.

In a non-uniform field, the breakdown would occur sphere gap in oil, at moderate voltage between thefrom the positive terminal up to the distance, to which spheres, and observing the resultant motions andthe breakdown gradient g = 30 extends, and this space deformations. It is even somewhat questionableis filled with corona. whether liquid dielectrics have a definite dielectric

In Appendix I is given a suggestion of-the mechanism strength at all, or whether disruption does not alwaysof the dielectric breakdown of air by carriers issuing occur through gases produced from the dielectric byfrom the positive terminal. the electric stress. A discharge through oil seems

always accompanied by the appearance of gas bubbles,III. Liquid Dielectrics and while the gas bubbles may not be the cause, but

Oil and similar materials (as petrolatum in the high- the result of the discharge, there are some differencespotential cable impregnation) are the most important in the shape of the striking distance curve of an oilliquid insulating materials and are depended upon for gap from that of an air gap, which look as if the dis-the insulation of the highest voltage electrical ruptive discharge in air is the result of the action of theapparatus, the transmission power transformer. The dielectric field on the air, while the disruptive dischargefairly satisfactory agreement of the behavior of air as in oil is the result of the action of the dielectric field oninsulating material, with the theory of a definite a material produced from the oil by the dielectricdielectric strength, modified by the conception of the field, that is, the dielectric field produces dissociation"energy distance" led to the application of the same or ionization of the oil, and the disruption occurstheory to liquid dielectrics. Here, however, the agree- through the gases of ionizationment with experience is not satisfactory; and tests IV. Solid Dielectricsshow that oil does not have a practically availabledefinite dielectric strength, but successive tests made The dielectric strength of solid dielectrics is often

expressed by the disruptive voltage gradient in a30r ll -r uniform field, in kilovolts per cm., or per mil. Suchz20 _ SamplefOil -_ - - an expression however, is meaningless, unless ac-I

. companied by a statement of the thickness of thelo vV III '*I_ _ dielectric, to which it applies, as the dielectric strength

~20 - 4 -LH 4Sof solids varies with the thickness, sometimes by many20- O ___K- hundred cent. Besides this, the disruptive strength

o-1 _ _N Ai - -. . . the voltage application, the humidity, previous historyffio44444Wo Wlv+498 ~~ofthe sample, etc., and even with tests made underIlo10 20 30 40 50 60 70 90- identically the same conditions, just as wide differencesNUMBEROFBREAKDOWN occur between successive tests of solid dielectrics, as

FIG. 2 shown by Fig. 2 to occur in liquids. The "disruptivestrength of a solid dielectric" thus has only a limited

with all precautions with the same oil in the same gap meaning, and that, when the conditions are fully given.(using such a large quantity of oil as to exclude the Disruptive strength tests of solid dielectrics areeffect of deterioration) differ from each other by values usually made on sheets between metal plates as ter-many times greater than the possible errors of observa- minals, -to produce a uniform dielectric field. This istion. The upper curve on Fig. 2 shows the deviation open to the objection that at the edges of the terminalsfrom average of 100 successive tests of the disruptive the field is not uniform. However, experience showsvoltages of an oil gap, and the lower curve shows the that puncture does not always occur at the edges ofcorresponding results with an air gap. Plotting the the terminal plates, but often inside, and by throwingnumber of times each deviation from average occurs, out the tests in which the disruption takes place at thein A, as function of the value of deviation, gives a edge of the terminals, and averaging those in which the

probability curve.,disruption occurs well inside of the terminals, fairlyThe most satisfactory explanation of the mechan- representative results are secured.

ism of the breakdown of oil and other liquids seems Two phenomena come into consideration in solidto be that based on the assumption of lack of homo- dielectrics: The electric conductivity and the dielectricgeneity, namely that the breakdown of an oil gap losses in an alternating field.is the result of materials of different conductivity or The dielectric has an appreciable though usuallyspecific capacity being sucked into the gap by the very low conductivity. That is, under an impresseddielectric field, or being produced in the gap by it, and voltage, a slight current is conducted through it,so causing a distortion of the field, with local high causing, alo of wer and thereby a heating of thedensities,- at which disruption begins. This can be dielectric. The power consumed by ohmic conductionvisually shown by dropping a minute quantityoflinstheudielectriousualy is extremely small, at least atwater or other foreig nm ial i ahorizontal atmospheric temperature. The conductivity usually


increases very greatly with the temperature, often MECHANISM OF THERMAL BREAKDOWN OF THEapproximately exponentially, (for instance, increases DIELECTRIC AS THIRD CLASS CONDUCTORten-fold for every 25 deg. temperature rise, so that athigh temperature the dielectric may become a goodconductor). Often, the conductivity also increases third class conductors, that is, have in some range ofwith increasing voltage. In an alterating field, the their voltampere characteristic such a high negativeresistance loss usually is very small compared with temperature coefficient of conductivity, that in tem-the other dielectric losses. perature equilibrium the voltage decreases with in-Under an alternating voltage, losses occur in the g curent

dielectric, more or less of the nature of a dielectric Suppose a uniform sheet of solid dielectric is exposed.. . ~~~~toa constant direct voltage between two conductinghysteresis. While probably mainly proportional to t a c d v

the frequency, some of these losses may increase faster terminals. Due to the slight conductivity of the> . . . ~~~~~~~dielectric, a current then flows through it at a uniform)than the frequency, some at a lesser rate, giving in the elatter case a dielectric loss at zero frequency. though very low current density. This consumeslTtercsedielectric losses may be representedbyanelectric energy, converting it into heat, and gives aThese dielectric losses may be represented by an slgh teprtr ieo h ilcrc o,n

.. . slight temperature rise of the dielectric. Now, noenergy current and an effective conductivity of di- material can be absolutely uniform and homogeneous,electric hysteresis, and lead to a power factor of the and thus in this sheet of dielectric there will be somedielectric. If the energy current of dielectric losses is spots of a slightly higher conductivity. Howeverproportional to the frequency and to the voltage, and slight the difference, at such a spot the current densitythe dielectric power loss therefore proportional to the must be slightly greater, thus the energy consumptionfrequency and the square of the voltage, then the and the heat production slightly greater, giving locallypower factor of the dielectric circuit is constant and a slightly higher temperature rise. However littleindependent of the frequency and the voltage. An this may be, if the conductivity of the dielectric in-increase of power factor with increasing voltage shows creases very rapidly with increasing temperature, itlosses increasing faster than the square of the voltage, will lead to a slightly higher current density, a corres-such as losses due to ionization of gas spaces inclosed pondingly higher energy consumption and heat pro-in the dielectric; a decrease of power factor with duction and thus temperature rise, and so on. Twoincreasing frequency shows losses increasing less than p bilities theproportional to the frequency, or constant lossesisuchhn exist, either the heat which can beasopohmiconductionth losesuenc, orconstantlossess conducted away from such a "hot spot" due to itsas ohmic conduction losses. temperature rise, is more than the heat produced byThe observation of the power factor thus is one means the increased conductivity due to the temperature rise.

of judging the nature of the losses in the dielectric. Then the temperature of the hot spot finally limitsA dielectric of high disruptive strength in general itself, and stable thermal conditions pertain. Or the

has low conductivity and low dielectric losses, thus a low heat which can be conducted away from the hot spotpower factor in an alternating field; but the reverse into the dielectric, by the temperature rise, is less thanis not necessarily the case, that is, a dielectric may the heat produced in the dielectric by the increasedhave low conductivity or low power factor, and still conductivity due to the temperature rise. Then thebe of poor disruptive strength. conditions are unstable, that is, the temperature con-

It is doubtful whether a true dielectric strength of tinuous to rise and the conductivity to increase in-solid insulation exists, that is, a definite voltage definitely, until the energy concentration at the hotgradient at which the dielectric is disrupted directly spot becomes destructive, and the dielectric is destroyedby the intensity of the dielectric field; or if it exists, by "puncture."it is so far beyond the disruptive strength observable, The successive steps of this phenomenon have beenas to be of no practical consideration. That is, observed by limiting the current density by suitablepuncture of a solid dielectric probably in practise always terminals and can be calculated from tests made on theoccurs as the result of a more or less rapid progressive conductivity of the dielectric at various temperaturesdeterioration, far below the true dielectric strength, and voltages.and the latter may be of moment only under direct As the energy consumption by the conduction currentlightning condition, for instance, in the puncture of a varies with the square of the voltage, there is thus asheet of insulation by a powerful Leyden Jar discharge definite voltage-for a given set of conditions-at(giving the characteristic appearance of an internal which instability is reached and puncture results, andexplosion), this voltage is the "breakdown voltage," its volt-The mechanism of the breakdown of a dielectric age gradient the "dielectric strength"' of the

under electric stress, may probably be any of a number material under the conditions of test. This voltageof possibilities, thermal, mechanical, chemical, physical, depends upon the initial temperature, and considerablyetc. decreases with increasing temperature; at higher


temperature, a lesser temperature rise is necessary to thermal instability leading to puncture at "hot spots"reach unstable conditions, and at the higher initial thus results.temperature the phenomenon starts with a higherconductivity, that is, greater power consumption and MECHANISM OF DISRUPTION DUE TO MECHANICALheat production. The heat conductivity of the di- INSTABILITYelectric and of the terminals and other surrounding Suppose a dielectric encloses a particle of a higherobjects, and their ability to quickly absorbing heat, conductivity or higher specific capacity. An electricthat is, their specific heat, are essential factors in field then exerts a mechanical force on this particle,determining the value of the puncture voltage. An which tends to elongate it in the direction of theessential factor also is the relation between temperature electric field, and to compress it at right angles thereto,and conductivity of the dielectric, etc., so that the that is, tend to form it into a filament short-circuiting"dielectric strength" thereby determined, is very far the dielectric field. With liquid dielectrics, this canfrom a constant for the material. be observed visually by dropping a little water into a

Suppose the "hot spot" is filamentary in shape. horizontal sphere gap in a transparent oil. EachThat is, the diameter of the spot of slightly higher droplet, as it is sucked into the field between the spheres,conductivity (or temperature) at which the breakdown lengthens into a filamentary conductor which bridgesstarts, is small compared with its length in the direction between the terminals, and then is destroyed with aof the lines of force, so that heat conduction from it is flash by the heat of the current conducted by it. Aessentially into the dielectric. Then a change of thick- similar phenomenon occurs with moisture in a solidness of the dielectric will not affect the heat conduc- dielectric, except that the puncture usually is per-tion, and in this case it follows that the disruptive manent; occluded moisture moving under electro-voltage is proportional to the thickness of the dielectric, static forces through the pores of the solid dielectric,its dielectric strength independent of the thickness. may form a conducting bridge between the terminals

Suppose however the "hot spot" is plate-shaped, and by the heat of the current in the moisture filamentthat is, its dimension parallel to the lines of force start destruction of the dielectric. Or the moisture-(from terminal to terminal) is small compared with thread may partially bridge between terminals andits diameter parallel to the terminals, so that the heat locally short-circuit the field, which will cause excessiveconducted from it flows into the terminals. Then an dielectric stresses in the part of the gap in series withincrease of thickness of the dielectric gives an increase the moisture filament. In these over strained portionsof heat to be eonducted through a longer distance of the field, destruction starts by thermal or chemicalthrough the same area, thus instability is reached at instability, etc.a lower voltage gradient. In this case it is found that In impregnated insulation, 'where the impregnatingthe puncture voltage is proportional to the square root material is liquid or viscous, motions of the impreg-of the thickness of the dielectric, and the dielectric nating material through the impregnated material maystrength inversely proportional to the square root of result from the mechanical forces caused by differencesthe thickness. in specific capacity or conductivity of the materials,

In general then, in dielectric breakdown by tem- leading to a redistribution of the impregnating material.perature instability, depending on the path of the heat Conductivity and specific capacity here act in the sameconduction, the puncture voltage varies between the direction, in general. The differences in specificcapacityfirst and the 0.5th power of the thickness of the di- of differentt dielectrics are however relatively small;electric, and the dielectric strength or breakdown rarely more than 1 in 6 in materials which are suitablegradient varies between independence of the thickness, for apparatus insulation; while the same constituentand the-0.5th power of the thickness of the dielectric. materials may differ in conductivity over an enormous

This phenomenon of dielectric breakdown by thermal range. The final effect of conductivity as comparedinstability is the result of the increase of energy loss to specific capacity is likely therefore, to be muchin the dielectric, with increasing temperature, and greater. On the one hand, the mechanical forces due tooccurs wherever the losses greatly increase with the differences in specific capacity appear instantaneously,temperature. While its mechanism has been illus- and thus are present also in alternating fields. On thetrated above on ohmic conduction in a direct-voltage other hand the differences due to difference in con-field, the same phenomenon occurs in the same manner ductivity appear gradual, being accompanied by thein an alternating field, and as the result of the specific formation of local internal electrostatic charges, andlosses of the nature of a dielectric bysteresis, as far as may require seconds or even minutes for their com-these losses increase with the temrperature. The in- pletion. These local changes would not be present orcrease-and usually very rapid increase-of the power only partly present in alternating fields, and thefactor with the increase of temperature of most di- mechanical actions of alternating fields thus differ fromelectrics shows that the losses in the dielectric in an those of continuous fields. This can conveniently bealternating field, increase with the temperature, and observed visually on sphere gaps in oil.


MECHANISM OF DISRUPTION BY CHEMICAL action, and such ionization due to enclosed air probablyDETERIORATION is a frequent cause of disruption. It is guarded against

Suppose the dielectric contains a particle of higher by excluding the air by impregnation with a materialof higher dielectric strength and higher specific capacity,specific.caact or hihrcnutvt. Leus1 such as oil. It means however, that the process of

assume, at first, this particle to be spherical in shape,and of relatively infinite specific capacity or con- impregnation to be effective must be perfect.

ductivity, that is, short-circuiting -thefield(forinstan It is interesting to note that some of the mechanismsductivity, ta is rt-irtingth feld for istanc of breakdown, as thermal instability and mechanicalaitionoi of 1 ielectricof0megohms resistivity)woulh .c forces, are reversible, that is, at the withdrawal of thediTion inhad.itr of 1000 megohms fresinstiv voltage, the original condition may gradually return,of the sphericaltycof theelin esodl tricfre side leaving the dielectric undamaged, while the chemicalrofndthe sheicapricl isathr ieast th .atin .ths deterioration by the electrostatic cutting edges or byrounding dielectric, and-at least temporarily, until ionization is irreversible, that is, what damage has beena redistribution of internal charges has occurred-the done is permanent and remains at the withdrawal ofdensity of the lines of dielectric force and thus the the voltage, and at the next applica-tion of voltage,voltage gradient in the dielectric outside the poles of the deterioration progresses further.the spherical particle would be three times normal. In non-homogeneous dielectrics, such as laminatedIt would be less than three times normal, if the specificcapacity orconductivity ofthe sphesinsulation, due to the differences in the ratios of thedapieleceri respective specific capacities and the specific conduc-less than given above from the surrounding delectrc.tivities of the component dielectrics, gradual changesBut it would be greater if the particle is not a sphere, may occur in the distribution of the voltage gradientsbut more irregular in shape, and would assume much through the dielectric, with the formation of interalhigher values at the edges and at points of the charges, which continue for some time, thus resultingparticle. Thus in the dielectric adjacent to edges orpoints of an enclosed particle of higher specific capacityor conductivity, very high-voltage gradients occur, and V. Time Lag of Insulationmay be far beyond the dielectric strength of the ma- T t imporan op onsulationterial, and lead to local breakdown of the material by dieltricpsrtant tielg Thelforemaswhat may be called "electrostatic cutting edges. that trenis hlii in the lage gae uprto 'whichWith organic insulation, the effect usually would be an insulaio a hold bcthe voltage ,andu whencarbonization~ ~ ~~by hihlcltmeaue,aceia an insulation can hold back the voltage, and whenco by hthis voltage gradient, the "dielectric strength" orchange which in general increases the conductivity. " sThe mechanical electrostatic forces brought about "'disruptive strength" of the insulation, is reached,hereby are in the direction of the field, so that the shape becomesia cuctor.of the product of chemical deterioration tends towardbecomesthe form of a conducting needle, with excessive voltage Experience shows the following condition: Atgradients in front of it, gradually piercing the dielectric least with very many insulations, the voltage at which

until fia.untrocrbtenh ei disruption or breakdown occurs, depends on the timeuntil final puncture occurs between the terminals. .... .. fl ~~~orduration of voltage application. The lower theThus in a laminated insulation consisting of very many

layers, a foreign particle in one of the layers though voltage, the longer the time it has to be applied. Thereis a minimum voltage, which continuously applied,originally forming only an insignificant part of the still just disrupts the insulation, and inversely, thetotal thickness of the dielectric, may gradually but hcumulatively, in the course of time, pierce and destroy hithas toevaped.the insulation by its electrostatic cutting edges, the ithastobeapplied

voltage gradients within the dielectric, . This time, during which a given voltage has to beaverage being..applied to cause disruption, is called the "time lag" ofstill very low compared with the tested "dielectricstrength" of the material. The destruction will be his vlt atio o iparticularvoltage ofthe faster, the higher the local voltage gradient. This brief apiation the voltag n,wic anentlyseems to be the main reason why such excessive margins api b dmrrvarins impulse ratio" of the time of voltage application.alr Irequirditfnh i Time lag bears to dielectric strength, in electricalengineering, a relation similar to the one between

IONIZATION elasticity and mechanical strength in mechanical en-The specific capacity of the common solid dielectrics gineering; if it were not for elasticity, there would

is from 2 to 8 times that of air. Thus if air pockets be no mechanical engineering. A pebble dropped onare contained in the dielectric, the electric stress in the an armor plate would shatter it,-since theoretically,air is much higher than in the solid, and as the break at the point of impact, without elasticity infinite me-down gradient of air is low, it breaks down with the chanical forces would be produced.formation of "corona," giving heat and chemical So without the phenomenon of time lag, there would


be no electrical engineering, as there would be no thus are designed for a puncture voltage much higherpossibility of insulation, since every insulation, even a than the flashover voltage, and 60 cycles tests as welllow-voltage lighting circuit, is theoretically constantly as high-frequency tests show such insulator strings toexposed to transients of infinite voltage (though flash over and not to puncture. Nevertheless lightningnegligible energy), that is, far beyond its possible punctures them not infrequently. The time lag ofdielectric strength, by the inductive and capacity flashover is greater than that of puncture, due to theeffects of any change of circuit conditions, and thus relatively high capacity of the insulator string. There-saved only by the time lag of its insulation, fore, under lightning- conditions, that is, very rapid

Just as the relations between mechanical strength application of voltage of considerable energy, theand elasticity give the wide variety of structural voltage reaches puncture values in less than the timematerials, on which the mechanical engineer depends, lag of flash over, while in low-frequency tests flashoverand which we denote by brittle, tough, ductile, elastic, limits the voltage, and in high-frequency tests the raterigid, flexible, yielding, etc., etc., so the relation of of voltage application is reduced, that is, the wavetime lag to dielectric strength gives us insulating front flattened by the capacity of the insulator string,materials of widely different properties and correspond- unless there is very great power back of the voltage,ingly widely different uses-but our knowledge in this so as to maintain it against the short-circuiting effectfield is unfortunately still very limited. of the capacity of the testing appliance (so-calledTo illustrate the importance of the time lag: Ex- "lightning generator").

perience as well as calculation shows that in 2300-voltprimary distribution circuits during thunder storms, TIME LAG OF SOLID AND LIQUID INSULATIONpotentials of short duration of the magnitude of When the mechanism of the dielectric breakdown ofhundred thousand volts are not infrequently produced the insulation consists of a cumulative thermal, me-by the setting free, by the lightning flash, of the chanical or chemical effect, as discussed above, itbound charge of the atmospheric electrostatic field. inevitably involves a time lag, and usually a con-The lighting transformers distributed over these cir- siderable one.cuits are not, and economically cannot be insulated to This time lag may be as short as a fraction of astand this voltage continuously. Hence, they must second, such as occurs in the electrostatic field of an oildepend on the time lag of their oil insulation to stand gap sucking in a moisture droplet, stretching it into athe lightning voltage until it is dissipated or discharged filamentary conductor, bridging between the electrodesby the lightning arrester. Inversely, the time lag of and flashing over. In the formation of hot spots inthe lightning arrester must be so short, and its dis- solid insulation, it may take minutes and hours, untilcharge rate so high, as to discharge the lightning voltage the process leads to the final disruption, and may extendin a time less than the time lag of the transformer, to days and years in the chemical action of ionization,bringing the voltage down to values safe for the etc., so that here the time lag of the insulation break-transformer. down gradually merges into the aging or deteriorationFrom the pnenomenon of time lag it results that the of the insulation.

rate of voltage application has a discriminating effect. For instance, under overload a cable may get over-Suppose two insulations are used in parallel, the one of heated and some hot spots form in the insulation andlower dielectric strength but higher time lag than the finally, after some hours, lead to a puncture. Ifother. (As for instance, the oil insulation ofa transformer however the load is taken off before the final breakdown,and the surface air insulation of its entrance bushings). the cable cools and the hot spots disappear and leaveAt very rapid voltage application, the voltage may the insulation undamaged, and we say that the cablerise beyond the dielectric strength of the stronger has been saved by the time lag of insulation and noinsulation of shorter tirre lag, in less than the time lag harm done by the temporary overload, because in thisof the weaker insulation, and the former thus punctures, case of approaching thermal breakdown the process iswhile with a slower voltage application the weaker reversible.insulation of greater time lag would puncture. Thus If however, under high-voltage test, ionization occursunder lightning conditions, the transformer bushings in the cable and by chemical action begins to destroymay flash over, short-circuit and blow the fuses without the insulation, we also may say that the cable is savedany damage to the transformer, while under high by the time lag of chemical breakdown, if the over-potential test the oil insulation may puncture far voltage is taken off before failure has occurred. How-below the voltage at which the bushings flash over, ever, as this process is notreversible, some damage hasTo illustrate the importance of this discriminating been done, and at the next overvoltage further damage

effect of rate of voltage rise: The insulation of high- is done and adds itself, until final disruption occurs.voltage transmission lines depends on sectional or In some respect, we may thus consider the gradualstring insulators. With these, it is very desirable that and slow aging and deterioration of the insulationin case of a failure the insulator disks should flash over during the years of use, which limits its final life, as arather than puncture. The transmission insulators progressive breakdown, and the entire life of the


insulation as the time lag of breakdown; however, the final condition at disruption. Then the phenom-this rather extends the meaning of the term. enon would be exponential, that is, start at maximum

rate and gradually decrease, thus theoretically takeELECTRICAL TIME LAG infinite time. It is customary to consider as the dura-

Air apparently has no time lag, at least, no appreci- tion of such an exponential phenomenon (for instanceable time lag, and the dielectric breakdown of the air the duration of an exponential transient) the timegap between spheres at a distance greater than the which would be required if the phenomenon con-energy distance and less than the corona distance, tinued to its end at constant initial intensity. As-with a negligible impedance between spheres and the suming this for the time lag, then the duration T wouldsource of voltage supply, is as nearly instantaneous as be the time lag at impulse ratio U = e.can be measured. Time lags (or impulse ratios) with, In the adjustable time gap consisting of an in-air gaps therefore are due either to the configuration of stantaneous sphere gap in air, shunted by a capacity Cthe gap, or due to the conditions of the supply circuit. and in series with a resistance r, the duration T, asA sphere gap with a considerable resistance in series defined above, then is:

has an appreciable time lag, the greater, the higher the T = r C,resistance, due to the time required to charge the thus has a physical meaning as the rate of condensercapacity of the spheres over the resistance. charge.By shunting an instantaneous sphere gap by a small The capacity of two spheres of ratio R, at distance 1

from each other, is, if:160 - __ _ _ __

14012

8r 0 _ 1 1120 g ; l R + 9 t R2 +R100 i

{~~~~~~~.13 101 0b

20 I_ la_f_.0_m eoecns

0

60

40 ________ and a sphere gap of 12.5 cm. spheres, at 5 cm. distance,in series with 1000 ohms resistance, would have a time

20 lag of 0.05 microseconds.Two 25-cm. spheres in air, at 0.7 cm. distance, have

o_ a time lag of about 0.5 microsecond. (Within the-4 -3 -2 -1 0 l 2 3 4

LOG OF TIME energy distance).104 i03 0.01 0.1 1 10 100 1000 10,000 Two 1-cm. spheres in air at 3 cm. distance, have aTIME IN SECONDS time lag of about 1 microsecond. (Corona)

FIG. 3A needle gap of 5 cm. in air has a time lag of the

magnitude of 1 to 2 microseconds.capacity C, with a non-inductive resistance r in series, A sphere gap of 2 mm. in oil, between 2.5 cm. spheres,any desired time lag can be produced by the proper has a time lag of about 20 microseconds.choice of C and r, and such a combination thus forms Whether, and how far and in what manner the timean adjustable time gap, convenient for the testing of lags depend on configuration, size, voltage, etc. istime lags. still largely unknown, and the entire field, though of

This raises the question of measuring the time lag. high importance, is almost entirely unexplored.The usual way of expressing a time lag is by the impulseratio. This however is not a constant, but a function Appendixof the duration of the voltage application, as shown by THE DIELECTRIC BREAKDOWN OF AIR AS Athe curve Fig. 3, which gives the impulse ratio of acable, for duration of the voltage application from con- CONDUCTIVITY PHENOMENONtinuous down to a microsecond. -1. Assume that the dielectric breakdown of air, andAn approximate expression of the electrical time lag the conductivity produced by it, are due to conducting

by a constant may sometimes be given as follows: particles or carriers, that is, particles which carryAssuming that proportionality exists,-a usual as- electric charges and thereby conduct the current.sumption with such phenomena-that is, that the We may assume that in free air, there are always somechange of the condition in the dielectric is proportional such carriers present. In an electrostatic field, theyto the difference between the existing condition and are set in motion with an acceleration a proportional


to the dielectric field intensity g, and therefore in the or in general, in a non-uniform field:time t acquire the velocity v, in a uniform field:

v = g t (1) c (g-go) dl (10)0

In a non-uniform field, the velocity would be given is the total production of conducting particles in theby: distance 1, or the ionic density or conductivity produced

v=Jwg d t (2) by the gradient g in the distance 1.If we assume that the conducting particles or car-. riers are produced from the gas molecules by collision,

Let us assume that there is a critical vvelocity vLet us asm ththro it the ionic density or conductivity c must reach a finiteat which these carriers produce additional carriers by maximum value, at which "complete ionization'' ofcollision with the atoms or molecules of the air or gas. the gas has occurred, and the conductivity thus reachedThe time t1 required to reach collision velocity vo its maximum or "short circuit" value co.

then is, Dielectric breakdown thus would be characterizedvO by

=i V(3) bg cO = (g - go) lo = constant = e1 (11)

and the distance, within which this is reached: or, in a non-uniform fieldt1 i Vo V02 1(221vO o2g (4) cO = (g -go) d = constant = el (12)

0

thus: This quantity el is of the dimension of an e. m. f.vO2 and may be called the "corduction voltage."

11 g = 2 - const. (5) To cause dielectric disruption of an air gap, theremust thus be available a dielectric field giving

or, in a non-uniform field: (a) A terminal drop eo = 320 volts.v2 (b) An excess gradient g, above the critical gradient

g d l = 2 const. (6) go (the so-called "dielectric strength of air"), extending2over such a distance lo as to integrate to a definite

This is of the dimension voltage. That is, a definite voltage el, required to produce complete ionization orvoltage, maximum conductivity of the air.

vo2 (c) The normal or critical gradient go, required toeo = 2 (7) maintain the conductivity, and extending as far as the

breakdown extends.is required to raise the free conducting carriers of the 3. Applying this to the air gap between sphericalair to the collision velocity vo, at which they produce electrodes:additional carriers from the atoms or molecules. With spherical electrodes of radius r, assuming first,The voltage eo may be considered as of the nature of that the other electrode is of such shape or so far away

a terminal drop, which must be present before the for- as not to distort the field, if we denotemation of conducting carriers can take place. gr the gradient at the surface of the sphere of radius r.

It then would be the minimum voltage, approached ro the radius, at which the gradient has decreasedfor minute air gaps: to go

eo = 320 volts. we have, at radius x,2. As the conductivity of the air rapidly disappears r 2

with the withdrawal of the dielectric field, these con- gx = 0 ) go (13)ducting carriers must be very short lived. x

Let then go be the voltage gradient, at which the rate andof production of conducting particles equals their rate ro ro rO \2of decay, and the number of conducting particles, that co = ) (g - go) dxz = go )qi( - 1L dxis, the conductivity of the air, remains constant (what- r r Xever may be its value, whether hightor low). /8= __ ____+r2o)=g(r-)

then is the rate of increase of conducting particles, or it follows, solving for r0the "ionizing force,'' and

c = GlIr = r 1+, c=(g-go)lI (9) go r )(14)


(m ) is, the gap length 1 is less than the length lo, within whichr t 1 + k the density of conducting particles increases from the

low density existing in the free air, to the maximumand the voltage gradient at the surface of the electrode density, where maximum conduction is produced, andsphere is the dielectric gradient thereby has dropped to go, so

2 that no further increase of ionization occurs.Ur = go ( ) The conductivity throughout the gap then varies\rI from the low initial value to the maximum value of

f ) 2 gradient go, and its average is half the maximum value,= 1 + (15) and the average gradient throughout the gap thus isI /r twice the minimum value go.

or approximately That is, in a short gap, the average gradient must be2 go, and the voltage consumed by the gradient (which2 m might be called "stream voltage," in distinction from

gr= g + (16) the voltage eo of (7) or the "terminal drop") thus is:where e, = 2 go I (22)

CO *this gives as the total voltage consumed by the dielec-m = (17) tric breakdown of a very small gap of length 1,go

e = e0j+-e1go= goo, where 3 = air density and goo equals

critical gradient at normal air density 3 = 1, at 0 = eo + 2 go l (23)deg. cent. and 76 cm. barometer. b. Assume the gap length 1 to be greater than theIt thus is ionizing distance lo, that is, the distance in which equa-

tion (1) produces maximum conductivity. Dielectric{=r1 + MO breakdown of the gap then requires the voltage

ro= r I+ r0r}(18) to 1 rdcsmxmmcnutvt.Deetia/6r eo = (g - go) lo

(1MO9 as the voltage producing the conducting stream, or the9 = goo +d 3+ r- P (19) "terminal drop" (though not limited to the terminaltx/6r )

~~~~~~alone).2m2m ei = 2 go lo

=oo t1 (20) as the voltage consumed by the average, gradient 2 gOin the distance lo, in which the conductivity is pro-

where duced by (1)/O co 21 e = 90 (1- lo)m g oo (21) as the voltage consumed in the rest of the gap (l - lo)

and co it (olouiby the gradient go required to maintain conductivity.and co iS the (total) conduction or ionizing voltage, The total voltage thus is,go0 the critical gradient at unit air density, and 3 the e = eo + e1 + e2air density. Equations (14), (16), (18) and (20) are = eo + 2 go l'o + go (1 - lo)the equations of energy distance and of surface gradient, = eo + 9O to + g0 l (24)given by Peek, and are here derived from the concep-tion of the dielectric breakdown as a conductivity = Eo + go Iphenomenon. where

4. With small gaps between plates or spheres of a Eo = eo + go lo (25)diameter large compared with the gap length, the.diamter argecompred ith he gp legth,the s the total excess voltage over that consumed by thedielectric field and the gradient are uniform throughout s tthe gap, and the maximum gradient, or gradient at the critical gradient go in the gap l.electrode surface, equal to the mean gradient through- c. Assume now that the gap is larger and the fieldiS not uniform.OUt the gap. As Uo is the minimum gradient of ionic conductivity,

g =e/l that is, the gradient required to maintain constantand the conduction voltage required to produce maxi- conductivity, the conductivity of the gap space canmum conductivity is by (11) given as extend only up to the distance from the electrodes,

el= (g- go) I (11) within which the gradient g is greater than Uo.The dielectric breakdown thus extends up to the(a) Assume the gap length is so small, that the distance of the gradient Uo, and within this distance 1',

ionic density increases throughout the entire gap, that the voltage consumed must exceed the value of break-


down gradient go times distance l', by the total ionizing 6. The polarity of the charge of the conductingvoltage Eo of (25), that is, must be particles or carriers, which give the dielectric break-

e' + Eo + go 1' down of air gaps at atmospheric pressure, can be deter-or, since mined by the action of unsymmetrical gaps at direct

voltage. Consider an unsymmetrical air gap such ase' = g d l that between a needle point and a plate, or between a

J small and a large sphere. In such a gap, with increasingit is applied voltage, the critical gradient go is first reachedand thereby the production of conducting particles

Eo =f(g-90go) dl eo + go 10 326)that is: 2e--oo(6 50 30j

~200The voltage c in paragraph 3, which gives the >150constant m in Peek's equation (14) etc., is not merely 100 -I 20__-the excess voltage eo required to produce the conducting 10-'305 1.020005stream, but is the sum of this voltage plus the excess CENTIMETERS I -1voltage Yo lo required by the excess gradient in thatpart of the field, in which the conductivity has not yet . I oreached full value. ( D.C

A. C. MaximumF0 = ~~~~~ ~~~~~(27) ,.Peek A. C. j

5. Consider now numerical values. 10-3 xlO 20 30 40From test values of small gaps between large spheres NIG 4

and between a large sphere and a plate, we find,e = 320 + 59 1= eo + 2 go l started at that terminal, which has the greater curva-

This gives ture and thus the higher gradients, the point or smalleo = 320 volts sphere. If then the polarity of this terminal is that

2 go = 59 is in good agreement with go = 30, as the of the conducting carriers, these move outwards awaytemperature correction has not been applied. from the terminal, towards the opposite terminal, and

For medium gaps, (case 4b,) the tests give, thereby short circuit the gap and cause disruption ase = 2370 + 30 i soon as the average gradient is sufficiently high to

Thus maintain the conducting stream across the gap. IfEo = 2370= co

This gives by (17)15,000-

inMgood m4 2)7 .VO079 0.28 0igodagreement with Peek's value: mn 0.27 or 10 10,000g = g 1 1 + - r CENU0.56>

= go{i+ Jr ~~~~~~~~~~55000

while Peek gives- -( 0.54 0 00 0.1 0.2 0.3 0.4 0.5

$7 90{l+ - (CENTIMETERSVr J FIG. 5

The limit lo between equation (23) and (24) ise = eO + 2 g0 l = 0.320 6010o however, the larger terminal-plate or large sphere-e =m F0 go I = 2.370 3010o is of the polarity of the carriers, a higher voltage across

68 X 10-3 ~~~~~~thegap IS required before the critical gradient iS ex-a68X1 cm. ceeded and the swarm of conducting carriers startedwhich is in close agreement with the experimental data at this larger terminal and disruptive discharge occurs.given in the tables and figures, of lo =m 69. At the smaller terminal, in this case, the formation of

In Figs. 4, 5 and 6 are given numerous test values conducting carriers has started already at lower volt-taken under various conditions, as indicated, with the age, but these carriers do not move away from thetheoretical curves shown by the drawn lines, terminal and across the gap, and thus do not cause


break down, but move towards the terminal near which unlimited current, and when it begins, thus short-cir-they started, and form a local corona in the space at cuits the gap and drops the voltage. Electronic con-and near this terminal. duction however is limited in current, and when itThus with unsymmetrical gaps, the direction of the occurs, the voltages can be maintained and still further

polarity of the applied voltage should make a difference raised. At such high vacuua, where the voltage ofin the disruptive voltage, and this voltage be higher ionic conduction has risen beyond that of electronicwith the voltage in one direction, than with the voltage conduction, and the conduction become electronic, thein the opposite direction. As the average gradient voltage can be raised beyond the value of ionic conduc-in the space in which the conducting particles are tion through the traces of residual gases, and thenformed, was shown to be about twice the breakdown ionic conduction again begins and "short-circuits" the

electron tube, thus limiting electronic conduction.Therefore a practically perfect vacuum is needed forpure electronic conduction in the modern high-powerelectron tube.

loo ttSOn the other side, with pressures higher than atmos-pheric, the disruptive voltage increases with increasingpressure, and approximately proportional thereto, asshown by experiment. It may be expected then, thatat some high pressures the voltage of ionic conductionincreases beyond that of electronic conduction, and thelatter limits the increase of disruptive strength of gases

C0 2.5 7.55 10 at high pressures. While this field has been littleCENTIMETERS investigated, experiments indicate that the dielectricFIG. 6 gradients of air and gases with increasing pressures

reach a limiting value somewhere at 1000 kv. per cm.gradient of air, the disruptive voltage of such anunsymmetrical gap should be about twice as high with Discussionthe voltage of the carriers on the large electrode, than R. W. Sorensen: My first point is in connection with7thewith the voltage of the carriers in the opposite direction. statement that we use our insulations under stresses which rarelyFig. 1 shows this, and also shows that the disruptive exceed the breakdown voltage of air, though tests show a strengthvoltage is lower with the small terminal (the needle of 10 to 20 times that of air for many of the insulations used.point in this case) positive. This plea for a more strenuous use of insulating materials is

It follows from this, that the conducting particles orcarriers, which carry the disruptive discharge in air at 1iatmospheric pressure, are positively charged, that is, 0 I - - I -they may be the positive ions, but cannot be the nega- 0 80 - A 1tive electrons. < Y

7. Dielectric conduction thus may be either ionic, 60 - - - / |or electronic. The relation possibly is the following:At atmospheric pressure the voltage required for

a40

ionic conduction, that is, disruptive discharge, is much z 20 - -lower than the voltage required for electronic conduc- - - - 1 1 -tion, and the conduction, that is, the disruptive dis- c0 2 4 6 8 10 12 14charge through air, is ionic, that is, by positive carriers. GAP LENGTH IN CENTIMETERSWith decreasing air pressure, the voltage required for FIG. 1-ARCING POTENTIAL VS. GAP LENGTHionic co-nduction decreases, approximately proportional Between Needle and Sphere, 12.5 cm. dia.Curve I-Direct Current, Sphere Negativeto the air pressure, and the conduction thus remains II- ts II It Positiveionic. At some low pressures a minimum value of III-Alternating Current, Maximum Values.voltage (or rather voltage gradient) is reached. Then,with further decrease of air pressure, the voltage of interesting in the light of a request in another paper given thisioicodcto agi inrass du to th ,eces morning, in which the author recommends a decrease in theonlc onclctlonagal lnceases eueto tn aecease test voltage a.pplied to transformers with one terminal grounded.

Of carriers, and finally passes beyond the voltage of In Fig. 1 of the paper the curves show the interesting factelectronic conduction. The phenomenon then changes that a given potential will cause spark over between a point andits character, ionic conduction ceases and electronic a sphere for much greater spacing when the sphere is negaUtive,conduction begins. At high vacuua, the conduction than will be the case writh the needle points negative and thethu is elcroi. sphere positive. The results shown in Fig. 3 have been dupli-tnus ISelectronlc. . a~~~~~atedby Messrs. Otis and Mendenhall, two students abt Cali-There iS a difference between ionic and electronic fornia, Institute of Technology a.s shown by the curves in Fig. 1conduction: Ionic conduction permits practically accompanying this discussion. In making these tests the high-


voltage alternating current was rectified by means of a two- much lower than the voltage required for electronic conduction."segment commutator driven by a two-pole synchronous motor. This is explained further on in the paper. If you have a perfectThe a-c. curve is very much like that obtained by Hayden vacuum, as stated, how would you have the pure electronic

and Steinmetz, but the d-c. curves differ in shape because of the conduction? I presume tubes in which there are heated fila-pulsating current obtained with the commutator, whereas ments have been used.Hayden and Steinmetz had very steady current delivered by a C. P. Steinmetz: The paper deals with a subject which hasfour kenetron rectifier. An explanation of the dip in these assumed, in the last year, a still greater importance than it hadcurves would be very interesting. before, that is, the problem of high-voltage insulation and

In John S. Townsend's "Ionization of Gases by Collision" is mechanical breakdowns.found this statement "When the point is negative, the strong This is such a vast field and so much work has and is beingfield is near the negative electrode, so that the potential re- done, and can be done, that the paper must necessarily be onlyquired to produce a discharge is less than when the point is a general part of the preliminary announcement of the resultspositive." Also in "Conduction of Electricity through Gases,"1906 edition, by J. J. Thompson on page 498 we find this state- bandnoediatiso futureocaon.ment ".this minimum potential depends upon thesharpness of the point, the pressure and nature of the gas, and There is, however, one feature which begins to get clear,the sign of the electrification of the point, being less if the point namely, that our conception of insulation and of breakdown ofis negatively than if it is positively electrified." Hence we have insulation again begins to change and to be subjected to multi-from these authorities statements which, at first reading, appear plication. To members in the early days, insulation was merelyquite contrary to those given in this paper. a boundairy bar. We knew, by experience, that a tenth of an

There is, however, some confusion as to the definition of the inch of insulation of a conductor would protect it against 2300term "spark discharge," in the texts referred to the term does volts. Then, when it came to higher voltages we realized thatnot seem to apply to an arc current and the voltage required there is something occurring within the boundary bar, of im-to cause it, but means the point at which a leak discharge only portance, and it is not merely the material, but there is a dielec-and not a complete. are over occurs between -the electrodes. tric field with potential radius and other things within theIn fact in one treatise on the subject of ionization this definition boundary bars which require consideration and study, and whichappears: "Sparking potential may be defined as the potential we are studying.which is required to maintain a very small current in the gas." Now, it seems that our views are just beginning again to get

Dr. Millikan has explained the Hayden-Steinmetz results on a multiplication with respect to at least the failure of insulationthe basis of the increased difficulty experienced in extending the and the mechanism of breakdowns. It seems to be clear thationization envelope with negative points, as compared to that the mechanism of breakdowns, under the failure of insulationphenomena for positive points, hence the required higher poten- on high voltage is a phenomena of instability. In other words,tials for a breakdown over a given distance. it is not that insulation fails, that dielectric breaks down, when

In discussing the mechanism of "thermal breakdown of electric stresses are beyond limits and value, but it is said thatdielectrics" the paper follows the work of Mr. Wagner as pub- under conditions very much lower than those gradients inlished in the JOURNAIT for December 1922, but Wagner does not lightning conditions of instability occur which gradually bringsdeduce the same law. Also some tests we have made in our about the multiplication and changes leading ultimately to alaboratory do not conform the deduction that puncture voltage breakdown of insulation. It is, therefore, a condition of instabil-is proportional to the square root of the thickness of dielectric. ity of constants of material which instability brings about,To emphasize "time lag" is indeed worth while as our experi- largely upon lesser changes, which leads either to destruction

ence shows it an important factor in making an analysis of volt- or breakdown. Therefore, the mere reduction or stress on theage stresses on insulations, and indicates that an intensive study insulation is not a factor which saves breakdowns, but that theshould be made of the laws which govern it. It may be of new problem of insulation seems to assume the shape of arranginginterest to note here that in testing thousands of porcelain of designs in the dielectric field of insulation, so as to get theinsulators I have found that very few which stand a potential condition of stability and not instability. That is the futuretest at minimum arc over voltage for fifteen seconds, fail when which seems to impress itself upon us. The more we study thethe potential is applied for a longer time. In testing apparatus problems of insulation we find that it is not existing stresses thatinsulated with organic insulating materials very often potentials cause this, but largely it is the result of stability, or instability.apparently harmless when applied for short periods will cause Now, that is not only true in solids, but probably in the air.breakdown when applied for longer periods of time. In this respect we could point out the same idea that the dis-At the bottom of the last page: "It follows from this, that charge makes its own gradient. Now, you have the same con-

the conducting particles or carriers, which carry the disruptive ception there, that it is the discharge which is taking place in thedischarge in air at atmospheric pressure, are positively charged, dielectric field about conditions which are unstable. We allthat is, they may be the positive ions, but cannot be the negative know that if we had 2,000,000 volts spread over, I don't knowelectrons." how many thousand inches, there could be a gradient that

If an atom is ionized and you have your positive ions turned would be so low there would be no puncture and for that dis-loose what becomes of the negative ions, where do they go? charge under such conditions. By the discharge making its ownThen follows: "At atmospheric pressure the voltage re- gradient there would be produced a dielectric condition of

quired for ionic conduction, that is, disruptive discharge, is instability which would finally lead to self-destruction.

Documents

High Voltage Insulation