This invention relates to electrical wires, and especially to wires that employ electrical insulation based on aromatic polymers.

Electrical wire and cable that use aromatic polymer insulation have been used for many years in numerous applications. For example wires that employ polyimide wraps or tapes usually bonded with fluoro- polymer adhesive layers have been used extensively as aircraft wire, for both civil and military applica¬ tions. Other examples of aromatic insulation that have been used for equipment wire or "hook-up" wire, air frame wire and in wire harnesses include aromatic polyether ketones, polyether ether ketones, modified polyphenylene oxide, and polyimide amides. Highly aro¬ matic polymers ^• have been used successfully in many applications because they have a range of desirable properties especially high strength and toughness, abrasion resistance, temperature resistance, dielectric strength and are often inherently highly flame- retarded.

The combination of these properties has enabled wire and cable fabricated from these polymers to be

used in small lightweight constructions. Such constructions have been used increasingly in both mili¬ tary and civil applications due to the high density and complexity of modern electrical systems.

However, these highly aromatic polymers suffer from a major problem: they are particularly suscep¬ tible to tracking. Tracking is a phenomenon associated with the formation of permanent and progressive con¬ ducting paths on the surface of the material by the combined effects of an electrical field and external surface pollution. Once commenced, the carbonaceous conducting deposits often extend quickly in dendritic fashion to give a characteristic "tree" pattern until failure occurs across the surface. Electrical tracking can occur when a damaged energised bundle of wires become wet e.g. from electrolytes or condensation. This tracking may lead to flashover and arcing that -causes additional wires in the bundle to become damaged. A catastrophic cascade failure can result from a fault to a single wire if adjacent wires that are at a different electrical potential are also susceptible to tracking or if the bundle is in contact with a grounded structure. Tracking can occur at low voltages e.g. 100V a.c. or less but becomes less likely as the voltage is reduced.

A related phenomenon, to which these polymers are also highly susceptible, is that of breakdown due to arcing. In this case a potential difference between two conductors, or between a conductor in which the insulation has been mechanically damaged, and ground, can result in the formation of an arc between the con-

ductors or between the conductor and ground. The high temperature of the arc causes the polymer to degrade extremely rapidly and form an electrically conductive carbonaceous deposit which can extend rapidly, as with wet tracking, and lead to catastrophic failure in which many or all of the wires in a bundle are destroyed. Arcing can occur at very low voltages, for example 24V d.σ. or lower, and since, unlike tracking, no electro¬ lyte or moisture is involved, it is a particularly hazardous phenomenon. Arcs may also be struck by drawing apart two conductors between which a current is passing as described for example by J.M. Somerville "The Electric Arc", Methuen 1959.

Another phenomenon that can be associated with tracking and arcing is erosion. In this case insu¬ lating material is removed by a vaporization process originated by an electrical discharge without the for¬ mation of electrically conductive deposits so that failure of the insulation will not occur until complete puncture of the insulation occurs .

According to the present invention, there is pro¬ vided an electrical wire which comprises an elongate electrical conductor, an inorganic arc-control layer surrounding the conductor, an insulating layer which surrounds the arc-control layer and comprises an aroma¬ tic polymer and a secondary tracking control layer surrounding the insulating layer, the secondary tracking control layer having a comparative tracking index of at least 300V. The "comparative tracking index" (C.T.I.) of a polymer is defined below.

By forming a wire in accordance with this inven¬ tion it is possible to utilise the very good thermal, mechanical and electrical properties of aromatic poly¬ mers without the wire exhibiting a high propensity to track which is associated with highly aromatic poly¬ mers .

The aromatic polymer will normally have a car¬ bonaceous char residue of at least 25%, preferably at least 30%, some polymers having a char residue of at least 40% and even at least 50%. This does not mean to say that a high char value is desired for its own sake, but simply that good mechanical and physical properties of these aromatic polymers including temperature stabi¬ lity and fire, retardancy, are usually associated with high char residues.

The char residue of the polymer components in the -electrical wire according to the invention can be measured by the method known as thermogravimetric ana¬ lysis, or TGA, in which a sample of the polymer is heated in nitrogen or other inert atmosphere at a defined rate, e.g. 10 ^β C per minute to a defined tem¬ perature and the residual weight, which is composed of char, is recorded. The char residue is simply the quantity of this residual char expressed as a percen¬ tage of the initial polymer after having taken into account any non polymeric volatile or non-volatile com¬ ponents. The char residue values quoted herein are defined as having been measured at 850°C and at a heating rate of 10°C per minute.

The preferred aromatic polymers which are used in this invention are well known to those skilled in the

art, and reference may be made for example to U.S. Patents Nos . 3,025,605, 3,306,874, 3,257,357, 3,354,129, 3,441,538, 3,442,538, 3,446,654, 3,658,938, 3,677,921, 3,838,097, 3,847,867, 3,953,400, 3,956,240, 4,107,147, 4,108,837, 4,111,908, 4,175,175, 4,293,670, 4,320,224, and 3,446,654, British Patents Nos. 971,227, 1,369,210 and 1,599,106 and European Patent Applications Nos. 170,065, 124,276 and 178,185. Such polymers include polyketones, polyether ketones, polyether ether ketones, polyether sulphones, polyether ketone/sulphone copolymers, polyether imides and polyphenylene oxides. Blends of different polymers can be used. The polymers may be wholly aromatic or they may include one or more aliphatic moieties which may comprise pendant alkyl groups or may comprise alkylene groups in the polymer backbone. Preferably the or each aliphatic moiety has not more than 4, and more pre¬ ferably not more than 3 carbon atoms. In the case of pendant alkyl groups, each group is most preferably a methyl group, while in the case of alkylene groups each group preferably has not more than 3 carbon atoms, and especially only one carbon atom, in the chain backbone, for example a methylene or isopropylidine group.

Preferred aromatic polymers are polymers with a melting or softening point of at least 250°C, par¬ ticularly at least 300°C and which may be crystalline or amorphous. Softening points of amorphous polymers may conveniently be measured by thermomechaniσal analy¬ sis (TMA), in which case the softening point refers to the temperature at which the probe has reached 60% penetration. In one class of such polymers the polymer comprises, and preferably consists essentially of, units of the formula

-Ar-Q-

the units being the same or different,

wherein Ar represents an unsubstituted or substituted divalent aromatic radical and Q represents -0-, -S-, -SO2-, -CO-, -NH-CO- or -COO-, or Ar represents a poly¬ valent radical and Q represents

-CO-

-N

CO-

each bond of the Q radical preferably being bonded directly to an aromatic carbon atom.

One preferred class of polymer comprises the polyphenylene oxides of the repeating unit

Ri

in which the groups R_, which may be the same or dif¬ ferent, each represents a hydrogen or halogen atom or a hydrocarbon atom having no tertiary alpha carbon atom.

In another class ^* of aromatic polymers the aromatic polymer is a crystalline polyarylene ether comprising recurring units of the formula

-0-E-0-E - -

where E is the residue of a dihydric phenol and E' is the residue of an aromatic compound having an electron withdrawing group in at least one of the positions ortho and para to the valence bonds, the E and E ¹ radi¬ cals being linked to the -0- radicals through aromatic carbon atoms. In one preferred sub-class, E is a radi¬ cal of the formula

wherein R2 is a divalent radical; x is 0 or 1; Y is a radical selected from halogen atoms, alkyl radicals containing 1 to 4 carbon atoms and alkoxy radicals con¬ taining 1 to 4 carbon atoms; y is O, 1, 2, 3 or 4; Y' is a radical selected from halogen atoms, alkyl radi¬ cals containing 1 to 4 carbon atoms and alkoxy radicals containing 1 to 4 carbon atoms; z is 0, 1, 2, 3 or 4, and E* is a radical of the formula

wherein R3 is a sulphone, carbonyl, vinyl, sulphoxide, azo, saturated fluorocarbon, organic phosphine oxide or ethylidene radical. In this class preferred poly- sulphones are those in which y and z are 0, x is 1, R3 is a sulphone radical and R2 is a radical of the for¬ mula

*4

-C- i R ₄

wherein each of R4 is independently selected from the group consisting of hydrogen; alkyl radicals containing 1 to 4 carbon atoms; halogen-substituted alkyl radical containing 1 to 4 carbon atoms; aryl, alkaryl and aralkyl radicals containing 6 to 10 carbon atoms; and halogen-substituted aryl alkaryl and aralkyl radicals containing 6 to 10 carbon atoms.

In another class of aromatic polymers, the polymer is a polyether imide or polysulphone imide which comprises recurring units of the formula

where Q is -0- or -SO2-, Z is a trivalent aromatic radical, R5 is a divalent aromatic radical and Rg is a divalent organic radical. Preferably the aromatic polymer has the general repeat unit

in which D represents a group of the formula:

R' represents an arylene group.

Another class of polymers is the polyetherketones that have repeating groups comprising aromatic ether and aromatic ketone groups together with an imide, amide, ester, benzoxazole or benzothiazole group. •Examples of such polymers are those having repeating units of the formula:

where R7 represents an imide, amide or ester group.

Yet another class of aromatic polymer is the polyarylates. Polyarylates that may be used include those that are derived from dihydric phenols and at least one aromatic dicarboxylic acid. Examples of such polymers include those derived from a dihydric phenol of the general formula

in which the groups Y, which may be the same or dif¬ ferent, each represent a hydrogen atom, a Ci to C4 alkyl group, or a chlorine or bromine atom; b is 0 ir an integer from 1 to 4; R8 represents a divalent saturated or unsaturated hydrocarbon group, e.g. an alkylene, alkylidine, cycloalkylene or cycloalkylidine group, an oxygen or sulphur atom or a carbonyl or sulphonyl group; and c is 0 or 1.

Preferred aromatic polymers consist essentially of repeating units having one of the following formulae

( 5 )

wherein each of x, m and n is 0 or 1, with n being 0 when x is 1, p is an integer from 1 to 4, with m being 1 and x being 0 when p is greater than 1, e.g..

(8)

or

in which units derived wholly from isophthalic acid or terephthalic acid or a mixture of both are present.

Other polymers having aromatic moieties e.g. poly 1,12-dodecamethylene pyromellitimide or 1,13-tridecamethylene pyromellitimide, as described in U.S. patent No. 3,551,200, may be used.

Blends of any two or more of the above polymers may be employed as may copolymers based on any two or more of these polymers.

The preferred aromatic polymers will usually have a molar C:H ratio of at least 1.0, preferably at least 1.2, more preferably at least 1.3 and especially at least 1.4. The- toughest polymers such as the polyaryl ether ketones, which are associated with high char residues, will have C:H ratios greater than 1.5.

The insulating layer may, in some cases, have an inner surface that has a C.T.I, of at least 250V. Preferably the C.T.I, of the inner surface of the prin¬ cipal insulating layer is at least 300V and most espe¬ cially at least 500V. Such values for the C.T.I, will normally be associated with carbonaceous char levels of

not more than 10% by weight preferably not more than 5% by weight more preferably not more than 2% and espe¬ cially 0%. The C.T.I, of the inner surface of the insulating layer may be achieved in certain cases by selection of the appropriate aromatic polymer . Alternatively it may be appropriate to blend the aroma¬ tic polymer with an aliphatic polymer in order to reduce the char level or increase the C.T.I, of the layer . We have found that the incorporation of an aliphatic polymer in the aromatic polymer can lead to a disproportionate reduction in the char residue of the blend, as described in our co-pending British patent application entitled "Electrical Wire and Cable" filed on even date herewith (Agent's ref: RK336), the disclo¬ sure of which is incorporated herein by reference. The preferred aliphatic polymers that may be blended with the aromatic polymer are the same as those referred to below for use as the secondary tracking control layer.

Alternatively or in addition, the insulating layer may be in the form of two or more layers, the outermost layer comprising the aromatic polymer, either alone or as a blend with one or more other polymers, and the inner layer comprising a polymer having a low char residue, e.g. not more than 10% by weight. The inner layer may, and preferably does, comprise one or more of the aliphatic polymers mentioned below.

The total thickness of the insulating layer will usually be in the range of from 50 to 250 micrometres, and especially from 75 to 200 micrometres. When it is in the form of more than one layer, the inner layer will normally have a thickness of from 75 to 125 micro-

metres, and the outer layer will normally have a thickness of up to 150 micrometres, preferably up to 100 micrometres, and especially up to 75 micrometres.

The purpose of the arc-control layer is to control the growth of the arc roots which, according to the invention, is achieved by controlling the arc diameter through the provision of a refractory, preferably adherent, layer on the electrical conductor, which will aid quick conductor melting and severence before more extensive wire or bundle damage can occur. Thus, for example, a layer of alumina, silica, silicon nitride, aluminium nitride, magnesium oxide, and titanium dioxide may be applied by any appropriate means, for example a vapour deposition method. Examples of such methods includes sputtering, chemical vapour deposi¬ tion, flame spraying, plasma ashing, reactive ion plating, electron beam evaporation or by other tech¬ niques. Examples of refractory coatings and methods of forming them on a conductor are described in our patent application No. GB-2,144,260A (US 818,854), the disclosure of which is incorporated herein by reference. Alternatively a weathered mica coating may be employed, for example as described in our copending European applications filed on even date herewith and claiming priority from British applications Nos . 8716303 and 8716309, the disclosures of which are also incorporated herein by reference.

The thickness of the arc-control layer will depend on, amongst other things, the material from which it is formed. In the case of refractory layers, the thickness will normally be in the range of from 0.1 to

10 micrometres preferably 5 to 10 micrometres, and the layer will preferably adhere to the electrical conduc¬ tor for example by provision of a metallic or refrac¬ tory interlayer. For other materials, thicknesses up to 100 micrometres may be used.

As stated above, the secondary tracking control layer has a C.T.I, of at least 300V. Preferably it has a C.T.I, of at least 400V, more preferably at least 500V and especially at least 600V. This normally will mean that the material from which the secondary tracking control layer is formed will have a car¬ bonaceous char residue of not more than 10%, more pre¬ ferably not more than 5%, most preferably not more than 2% and especially 0% by weight. The secondary tracking control layer preferably comprises an aliphatic polymer or blend of aliphatic polymers, although it is possible for the aliphatic polymer to include one or more aromatic moieties in addition to its aliphatic moieties, and indeed a number of preferred polymers do so. However the polymer should have sufficient alipha¬ tic nature that its molar C:H ratio is not more than 1. Examples of aliphatic polymers and polymers containing aliphatic moieties (which will be referred to herein simply as aliphatic polymers for the sake of brevity) include olefin homopolymers and copolymers of olefins with other olefins and with other monomers e.g. vinyl esters, alkyl acrylates and alkyl alkacrylates, e.g. low, medium and high density polyethylene, linear low density polyethylene and ethylene alpha-olefin copoly¬ mers, ethylene/propylene rubber, butyl rubber, ethylene vinyl acetate, ethylene ethyl acrylate and ethylene acrylic acid copolymers, and linear or radial styrene

diene di- or tri-block copolymers e.g. styrene/buta- diene, styrene/isoprene copolymers, styrene/butadiene/- styrene and styrene/isoprene/styrene, and hydrogenated versions of these block copolymers especially styrene ethylene/butylene/styrene block copolymers . A par¬ ticularly preferred class of low charring polymers is the polyamides . Preferred polyamides include the nylons e.g. nylon 46, nylon 6, nylon 7, nylon 66, nylon 610, nylon 611, nylon 612, nylon 11, nylon 12, nylon 1212, and aliphatic/aromatic polyamides, polyamides based on the condensation of terephthalic acid with trimethylhexamethylene diamine (preferably containing a mixture of 2,2,4- and 2,4,4-trimethylhexamethylene diamine isomers) , polyamides formed from the conden¬ sation of one or more bisaminomethylnorbornane isomers with one or more aliphatic, cycloaliphatic or aromatic dicarboxylic acids e.g. terephthalic acid and optionally including one or more amino acid or lactam -e.g. £-caprolactam comonomers, polyamides based on units derived from laurinlactam, isophthalic acid and bis-(4-amino-3-methylcyclohexyl) methane, polyamides based on the condensation of 2,2-bis-(p-aminocyclo- hexyl) propane with adipic and azeleic acids, and polyamides based on the condensation of trans cyclo- hexane-l,4-dicarboxylic acid with the trimethylhexa¬ methylene diamine isomers mentioned above. Other aliphatic polymers that may be used include polyesters e.g. polyalkylene terephthalate and especially poly- tetramethylene terephthalate, and cycloaliphatic diol/terephthalic acid copolymers e.g. copolymers of terephthalate and is _. ophthalate units with 1,4-cyclo- hexanedimethyloxy units, polysulphides, polyethers e.g. polybutylene ether copolymers, and especially polyether

esters such as those having polytetramethylene ether and polytetramethylene terephthalate blocks; aliphatic ionomers e.g. those based on metal salts of ethylene (meth)acrylic acid copolymers or sulphonated olefins such as sulphonated EPDM, and the like. Preferred aliphatic polymers include the polyamides mentioned above, polyethylene, polybutylene terephthalate, iono¬ mers based on metal salts of methacrylated polyethy¬ lene, acrylic elastomers e.g. those based on methyl, ethyl or n-butyl acrylate or alkoxy-substituted ethyl or n-butyl acrylate polymers containing a cure site monomer and optionally ethylene comonomer, and block copolymers having long chain ester units of the general formula:

0 0 II II

-OGO-C-R-C-

and short-chain ester units of the formula

0 0 II II

-ODO-C-R-C-

in which G is a divalent radical remaining after the removal of terminal hydroxyl groups from a polyalkylene oxide) glycol , preferably a poly (C2 to C4 alkylene oxide) having a molecular weight of about 600 to 6000; R is a divalent radical remaining after removal of carboxyl groups from at least one dicarboxylic acid having a molecular weight of less than about 300; and D is a divalent radical remaining after removal of hydroxyl groups from at least one diol having a molecular weight less than 250.

Preferred copolyesters are the polyether ester polymers derived from terephthalic acid, polytetramethylene ether glycol and 1 ,4-butane diol. These are random block copolymers having crystalline hard blocks with the repeating unit:

-(CH ₂ and amorphous , elastomeric polytetramethylene ether terephthalate soft blocks of repeating unit

having a molecular weight of about 600 to 3000, i.e. n = 6 to 40.

Other preferred aliphatic polymers include those based on polyether and polyamide blocks, especially the so called a "polyether-ester amide block copolymers" of repeating unit:

-C-A-C-O-B-0- ll II 0 0

wherein A represents a polyamide sequence of average molecular weight in the range of from 300 to 15,000, preferably from 800 to 5000; and B represents a linear or branched polyoxyalkylene sequence of average molecu¬ lar weight in the range of from 200 to 6000, preferably from 400 to 3000.

Preferably the polyamide sequence is formed from alpha,omega-aminocarboxylic acids, lactams or diamine/- dicarboxylic acid combinations that include C4 to C 4 alkylene carbon chains, and the polyoxyalkylene sequence is based on ethylene glycol, propylene glycol and/or tetramethylene glycol, and the polyoxyalkylene sequence constitutes from 5 to 85%, especially from 10 to 50% of the total block copolymer by weight. These polymers and their preparation are described in UK Patent Specifications Nos. 1,473,972, 1,532,930, 1,555,644, 2,005,283A and 2,011,450A.

The aliphatic polymer preferably has a C:H ratio of not more than 0.9, more preferably not more than 0.75, most preferably not more than 0.65 and especially not more than 0.55.

The aliphatic polymer may be unsubstituted or substituted. One class of aliphatic polymer that is particularly useful is the fluorinated polymers, pre¬ ferably those containing at least 10%, more preferably at least 25% fluorine by weight. The fluorinated polymer may be a single fluorine containing polymer or a mixture of polymers one or more of which contains fluorine. The fluorinated polymers are usually homo-or copolymers of one or more fluorinated, often per- fluorinated, olefinically unsaturated monomers or copo¬ lymers of such a comonomer with a non-fluorinated olefin. The fluorinated polymer preferably has a melting point of at least 150°C, often at least 250°C and often up to 350°C, and a viscosity (before any crosslinking) of less than 10^ Pa.s at a temperature of not more than 60°C above its melting point. Preferred

fluorinated polymers are homo- or copolymers of tetra- fluoroethylene, vinylidine fluoride or hexafluoropropy- lene, and especially ethylene/tetrafluoroethylene copolymers e.g. containing 35 to 60% ethylene, 35 to 60% tetrafluoroethylene by mole and up to 10% by mole of other comonomers, polyvinylidine fluoride, copoly¬ mers of vinylidine fluoride with hexafluoropropylene, tetrafluoroethylene and/or hexafluoroisobutylene, poly- hexafluoropropylene, and copolymers of hexafluoropropy¬ lene and tetrafluoroethylene. Alternatively C1-C4 perfluoroalkoxy substituted perfluoroethylene homo- polymers and copolymers with the above fluorinated polymers may be used.

The secondary tracking control layer may have a thickness of up to 100 micrometres, e.g. from 50 to 100 micrometres in the case of a polymeric layer. Thinner secondary tracking control layers may be provided by other means .

The polymeric insulation of the wire is preferably cross-linked. In general, however, the aromatic poly¬ mers will exhibit a lower degree of crosslinking than the aliphatic polymers, and in many cases the aliphatic polymers may be highly crosslinked while the aromatic polymers remain substantially uncrosslinked.

The polymeric composition may be cross-linked, for example, by exposure to high energy radiation.

Radiation cross-linking may be effected by expo¬ sure to high energy irradiation such as an electron beam or gamma-rays . Radiation dosages in the range 20

to 800 kGy, preferably 20 to 500 kGy, e.g. 20 to 200 kGy and particularly 40 to 120 kGy are in general appropriate depending on the characteristics of the polymer in question. For the purposes of promoting cross-linking during irradiation, preferably from 0.2 to 15 weight per cent of a prorad such as a polyfunc- tional vinyl or allyl compound, for example triallyl cyanurate, triallyl isocyanurate (TAIC), methylene bis acrylamide, metaphenylene diamine bis maleimide or other crosslinking agents, for example as described in U.S. patents Nos. 4,121,001 and 4,176,027, are incor¬ porated into the composition prior to irradiaton.

The polymers used for the various layers may include additional additives, for example reinforcing or non-reinforcing fillers, stabilisers such as ultra¬ violet stabilisers, antioxidants, acid acceptors and anti-hydrolysis stabilisers, pigments, processing aids such as plasticizers, halogenated or non-halogenated flame retardants, hydrated metal oxides such as alumina trihydrate and magnesium hydroxide, or decabromodi- phenyl ether or combinations thereof, fungicides and the like.

The wires and cables according to the invention may be formed by conventional techniques . For example the polymers may be blended together if necessary in a mixer, together with any additional components, pelle- tised, and then extruded onto a wire conductor. Other non-preferred wires may be formed by a tape-wrapping method. For example _, a polyamide, e.g. trimethylhexa- methylene terephthalamide may be dissolved in a solu¬ tion of a polyamic acid and, after the solvent has been

evaporated, the aromatic polymer may be imidized by heat and cut into tapes which can then be wrapped around the conductor and underlying layers. However, it is preferred for the polymers to be melt-shapeable, so that the wire insulation can be formed by extrusion.

The wires may be used individually as equipment or "hook-up" wires, or airframe wires, or in bundles and harnesses, both jacketted and unjacketted, and may be used in multiconductor cables . The wires, harnesses or cables may be unscreened or they may be provided with a screen to protect them from electromagnetic inter¬ ference, as well known in the art. In addition flat cables may be formed using the insulation materials according to the invention, either employing flat con¬ ductors or round conductors .

The invention will now be described by way of example with reference to the accompanying drawings in which:

Figure 1 is an isometric view of part of an electrical wire according to the invention;

Figure 2 is a schematic view of the test arrange¬ ment for wet tracking; and

Figure 3 is a schematic view of the test arrange¬ ment for dry arcing.

Referring initially to figure 1 of the accom¬ panying drawings , an electrical wire comprises a con¬ ductor 11 which may be solid or stranded as shown and

is optionally tinned. A 10 micrometre thick silica or alumina layer 12 is formed on the conductor by a vacuum deposition process, e.g. by sputtering, 5 micrometre intermediate layer formed from for example aluminium optionally being provided in order to improve adhesion of the layer 12 to the conductor 11. A 100 to 200 micrometre thick aromatic polymer insulation 13 is then extruded onto the coated conductor followed by a 100 micrometre thick tracking control layer 14 formed e.g. from an ethylene-tetrafluoroethylene copolymer. After both layers have been extruded, the insulation is irra¬ diated by high energy electrons, to a dose of about 120 kGy to crosslink the tracking control layer 14 and optionally also the aromatic polymer layer (depending on the aromatic layer).

The following Examples illustrate the invention. In the Examples the following test procedures were used:

WET TRACKING TEST

This test is designed to simulate the condition occuring when a ^' damaged wire bundle comes into contact with an electrolyte. Under actual conditions, the electrolyte may be moisture containing dust particles or other ionic contaminant. Damage to the bundle may occur through a number of reasons e.g. abrasion, hydro¬ lysis of the insulation, aging, etc. Current flow through the electrolyte results in heating and evapora¬ tion of the solution. This causes one or more dry bands to appear across which the test voltage is

dropped, resulting in small, often intense, scin¬ tillations which damage the insulation.

Figure 2 shows the sample set-up. A wire bundle 1 is prepared from seven 14cm lengths 2 of 22AWG tinned- copper or nickel-plated copper conductor coated with a layer of the wire insulation under test. The bundle 1 is arranged with six wires around one central wire and is held together using tie wraps 3 so that the wires are not twisted. Two adjacent wires are notched cir- cumferentially to expose 0.5mm bare conductor on each wire. The notches 4 are arranged such that they are 5mm apart with the tie wraps 5mm either side of them. One end of each wire is stripped to enable connections to be made to the power supply via insulated crocodile clips. The sample is held at an angle of 30 degrees to the horizontal using a simple clamp made of an electri¬ cally insulating resin so that the damaged wires are uppermost and the stripped ends are at the upper end of the bundle. A piece of filter paper 5 20 x 10mm wide is wrapped around the bundle approximately 2mm above the upper notch; this is best held in place with the upper tie wrap.

A peristaltic pump conveys the electrolyte from the reservoir to the sample via a dropping pipette 6, and a power supply is provided to energise the bundle. The electrolyte used is 2% sodium chloride and 0.02% sodium dodecyl sulphate surfactant in distilled or deionised water. The pump is set to deliver this solu¬ tion at a rate of approximately 60mg per minute through the pipette 6 which is positioned 10mm vertically above the filter paper 5.

The power is supplied by a 3-phase 400Hz 115/200V generator of at least 5kVA capacity or a single phase 50Hz 115V transformer of at least 3kVA capacity. A device for recording time to failure is provided which records the time when either a wire goes open circuit, or when a circuit breaker comes out. Leakage currents can be followed with the use of current clamps surrounding the wires and connected to a suitable oscilloscope.

In the case of the three phase supply, adjacent wires of the bundle are connected to alternate phases of the power supply via 7.5A aircraft type circuit breakers with the central wire connected directly to neutral. In the case of the single phase supply, alternate wires are connected to neutral, with the remaining wires including the central conductor to live. A few drops of electrolyte are allowed to fall onto the filter paper to ensure saturation prior to starting the test. The power is switched on and the timer started as soon as a leakage current has been detected. The test is allowed to continue until:

a) one or more circuit breakers come out; b) a wire becomes open circuit; or c) 8 hours have elapsed.

The condition of the final bundle and the time to failure is noted in all cases. Where failure has occurred due to breakers coming out, the power is then reapplied and each breaker is reclosed in turn until there is no further activity. The condition of the bundle is again noted.

Where a wire has gone open circuit, the power is reapplied and the test continued for a total of 8 hours. Wire insulations exhibiting "failures" due to (b) or (c) may be considered to be good non wet- tracking insulation if no more than two of the origi¬ nally undamaged wires become open circuit.

Dry Arc Test

This test is designed to simulate what happens when a fault in a wire bundle causes arcing under dry conditions . A graphite rod is used to initiate the arc which causes thermal degradation of the insulation. Continuation of the fault current can only occur through the wire bundle under test due to shorting across adjacent phases through a conductive char, or direct conductor-conductor contact such as might occur if the insulation is totally removed by the duration of the arc.

Figure 3 shows the sample set-up. A wire bundle 21 is prepared from seven 10cm lengths 22 of 22AWG tinned-copper or nickel-plated copper conductor coated with a layer of the wire insulation under test. The bundle 22 is arranged with six wires around one central wire and held together with tie wraps spaced about 5cm apart. One of the outer wires is notched circumferen- tially between the tie wraps to expose 0.5mm bare con¬ ductor and one end of each wire is stripped to enable connections to be made via insulating crocodile clips .

A rod 23 is provided which is made of a spectrographically pure graphite, diameter 4.6mm, with

an impurity level not more than 20ppm. It is prepared before each test by sharpening one end using a conven¬ tional pencil sharpener of European design to give an angle of 10 degrees off vertical with a tip diameter of 0.4±0.1mm. A lOOg weight 24 is clamped onto the top of the rod 23 to maintain contact during the arc ini¬ tiation and also acts as a device to limit the depth of penetration of the rod by restricting its downward tra¬ vel. The rod passes through a PTFE bush which allows it to slide freely up and down.

The arrangement of levers enables precise posi¬ tioning of the rod 23 on the wire bundle 21 which is held securely in place by means of a simple clamp 25 made of an electrically insulating resin and mounted on a block 26 made of the same material.

The power source can be either:

a) a 3-phase 400Hz 115/200V generator of at least 5kVA capacity b) a single phase 50Hz 115V transformer, at least 3kVA capacity c) 24V d.c. supplied by two 12V accumulators.

The fault current is detected by means of current clamps surrounding the connecting leads and the voltage at failure is measured using a 10:1 voltage probe. The transducer signals are fed into a multi-channel digital storage oscilloscope where they can be displayed and manipulated to obtain power curves (voltage x current) and energy (integration of power curve).

The wire bundle 21 is positioned in the clamp 25 so that the notched wire is uppermost. Adjacent wires of the bundle are connected to different phases of the supply through 7.5A aircraft type circuit breakers, and the central wire is connected directly to neutral. In the case of single phase or d.c. supplies, alternate wires are connected to neutral or the negative ter¬ minal, with the remaining wires, including the central wire, connected through circuit breakers to live or the postive terminal. The carbon rod is also connected to neutral or the negative terminal and positioned so that the point is in contact with the exposed conductor. The gap between the lOOg weight and the PTFE bush is adjusted to the diameter of the insulated wire under test, using a suitable spacer to limit the penetration of the rod into the sample. A voltage probe is con¬ nected across the damaged wire and the rod, and current clamps positioned on each of the three phases, or on the wires connected to the live side of the supply. A protective screen is placed in front of the test set-up and the power switched on. A material is deemed to pass this test if:

a) no circuit breakers come out and the activity is relatively non-eventful, or b) there is no further activity on resetting the breakers after a non-eventful test.

In addition, non-tracking materials will have relatively few spikes in the current trace with a correspondingly low total energy consumed. Tracking materials, on the other hand, show many spikes usually on all three phases, which are accompanied by violent crepitation and large energy consumption.

CTI Test

This method is a modification of IEC 112 which measures the low voltage track resistance (up to 600V) as Comparative Tracking Index (CTI) of materials in the presence of an aqueous contaminant.

The samples of each polymeric layer are prepared by either compression or injection moulding plaques with minimum thickness of 3mm and with sufficient area to ensure that during the test no liquid flows over the edge of the sample. Before testing, the surface of the sample is cleaned with methanol to remove any surface contamination.

The test apparatus is as described in IEC 112. It consists of two platinum electrodes, each with one end chisel-shaped to an angle of 30 degrees. The electro¬ des are symmetrically arranged such that the opposing chisel faces are vertical and 4.0±0.1mm apart when placed on the surface of the specimen.

The power supply consists of a 0.5kVA transformer capable of supplying an a.c. voltage in the range 100-600V at 50Hz. A rheostat is incorporated into the circuit so that the short circuit current may be adjusted to give l.O±O.l amp. An over-current relay is provided which shuts off the HV supply when a current of at least 0.5 amps flows for 2 seconds, the criteria for failure.

A device for dropping electrolyte solution between the electrodes is provided. This consists of a

peristaltic pump which draws liquid from a reservoir and pumps it out of a needle situated at height of 30-40mm above and between the electrodes. The dropping rate is set to 1 drop every 30±5 seconds with a drop volume of 20±3 mm*-*. Before the test, the needle is cleaned and purged with several drops of electrolyte to ensure the correct concentration of reagent is used.

The electrolyte solution used in these tests is 0.1±0.002% ammonium chloride and 0.01% sodium dodecyl sulphate surfactant in deionised water and has a resistivity of 405±5 ohm.cm at 23°C.

By using a surfactant, reproducible results have been obtained on homogeneous samples. However, the type of surfactant used can greatly affect the CTI values obtained. Sodium dodecyl sulphate was chosen as this gave a working range of below 150 (for known badly tracking materials) to above 600 for non-tracking materials .

Test Procedure

The specimen -is put into position and the electro¬ des lowered on to the surface. A suitable voltage is chosen and the short circuit current adjusted accor¬ dingly. The electrolyte is then allowed to drop bet¬ ween the electrodes until either

a) tracking occurs b) the sample erodes through to the supporting base c) at 600V, the sample withstands 50 drops.

For those materials which track, the test is then repeated at at least three other voltages such that two voltages give results above 50 drops, and two give results below 50 drops. A curve of drops to failure versus applied voltage is plotted from which is read the voltage at which 50 drops would cause failure. This voltage is the comparative tracking index (CTI).

For those materials which withstand 50 drops at 600V, CTI is then quoted as >600 and the erosion of the sample is measured by preweighing a plaque of the material, running the test at 400V for 101 drops of electrolyte, and then reweighing the sample after removing any loose surface debris. The erosion is quoted as mg of material lost.

For those samples which penetrate through to the supporting base during the test, erosion is quoted as mg lost at 400V after 101 drops, or after complete penetration, stating the number of drops required for this to occur e.g. 75mg (80 drops).

The following wire constructions were prepared by use of a 20mm Baughan extruder for the polymeric layers and by sputtering for the inorganic layers. In the cases where a blend has been used, it has been prepared using a Baker Perkins twin screw extruder. The conduc¬ tor was 22 AWG nickel plated copper unless otherwise stated. The results for the wet and dry tracking tests are shown in the table.

Example 1

A 10 jam silica arc control layer on top of a 5 pi aluminium interlayer was coated with 160 pm of poly- aryletheretherketone as the primary insulation layer and 75 μm crosslinked ethylenetetrafluoroethylene as the secondary tracking control layer.

Example 2

A 6-7 pm alumina arc control layer on top of a 7 urn aluminium interlayer was coated with 125 pm of poly- aryletheretherketone as the primary insulation layer and 125 urn of a blend of polytetramethylene terephtha¬ late and a poly(ether-ester) block copolymer comprising approximately 57% by ^" weight polybutylene terephthalate hard blocks and approximately 43% by weight poly(buty- lene glycol polyether terephthalate) soft blocks in the ratio of 70:30 as the secondary tracking control layer. (22 AWG copper conductor).

Example 3

A 6 pm silica _, arc control layer on top of a 5.7 pm aluminium interlayer was coated with 125 pm of poly- aryletheretherketone as the primary insulation layer and 125 pm of a blend of polytetramethylene terephtha¬ late and a poly(ether-ester) block copolymer comprising approximately 57% by weight polybutylene terephthalate hard blocks and approximately 43% by weight poly(butylene glycol polyether terephthalate) soft blocks in the ratio of 70:30 as the secondary tracking control layer.

Example 4 ( Control )

A 6-7 pm silica arc control layer on top of a 7 pm aluminium interlayer was coated with 125 um of ^' poly- aryletheretherketone as an insulating layer.

Example 5 (Control)

100 pm polyaryletheretherketone as the inner insu¬ lating layer with 100 pm of a blend of polytetramethylene terephthalate and a poly(ether-ester)block copolymer comprising approxima¬ tely 57% by weight poly(tetramethylene terephthalate) hard blocks and approximately 43% by weight poly(tetramethylene glycol polyether terephthalate) soft blocks in the ratio of 70:30 as the outer insu¬ lating layer.

Example 6 (Control)

A 125 pm polyaryletheretherketone layer as the sole insulation.

TABLE

6 (Control ) ΪEEK Fail Fail

The CTI of each of the secondary tracking control layers of Examples 1 to 3 and 5 was greater than 600 V .