US1872354A - Composite conductor - Google Patents

Description

Patented Aug. 16, 1932 I 1 UNITED STATE.)

PATENT OFFICE HOWARD soccer, or wILxINsBuRe, rENNsYLvANrA, ASSIGNOR 'ro WESTINGHOUSI ELECTRIC & MANUFACTURING coMPANY, A CORPORATION or PENNSYLVANIA coMrosIrE CONDUCTOR .No Drawing.

My invention relatesto composite conductors having such a coeflicient of expansion that they may be effectively sealed into glass, and more particularly to a composite wire having an inner core composed of an composed of an outer sheath or jacket composed of copper and an inner core containing iron, nickel and cobalt, the sectional area of the copper and the inner core being maintained in such proportions that the mean coeflicient of expansion of the metal conductor will range from 5 to 7x10 per degree centigrade from zero to 250 C.

Another object of myinvention is to provide a composite metal conductor that has a sufliciently low coeificient of expansion to permit its' being sealed into hard glass having a coefl'icient of expansion ranging from i 3 to 5x10- per degree centigrade.

A further object of my invention is to provide a'method of effecting a seal between a copper-coated conductor and glass which .comprises forming a glass beading on the conductor, inserting the beaded conductor through an aperture in the glass, fusin the lass thereto, and maintaining the used oint at the lowest temperature at 'which strains are efiectively eliminated -for a period of from 1 to 24 hours before cooling to normal temperatures.

Composite wires having an inner core composed of an iron-nickel alloy and an outer sheath or jacket of copper haveheretofore been employed -for conducting electrical energyinto vitreous envelopes, such as glass. In utilizing such conductors, it has been necessary to maintain the cross-sectional area of the copper and the inner core in such proportions that the coeflicient of expansion of the conductor and the glass will be substantially the same. In order to provide a Application filed February 8, 1930. Serial No. 427,060.

conductor having such a coeflicient of expansion, the cross-sectional area of the core must be about 85% of the cross-sectional area of the composite conductor.-

Nickel steel, however, about x10" ohms per centimeter cube, and since the copper occupies only a'relatively small proportion of the total cross-sectional area, the resistivity of the composite wire is approximately 11x10 ohms per centimeter on e.

I have made the discovery that when an iron-nickel-cobalt alloy or an iron-nickelcobalt steel is employed as the core in a composite wire having an outer jacket of copper," the proportion of the cross-sectional area of the copper may be increased and a composite conductor formed which may be. effectively sealed into an envelope composed of soft glass, or the coefiicient of expansion of the composite wire may be reduced to such an extent that it may beeflectively sealed into hard glass. v

For example,'the coefficient of expansion;

of a core composed of nickel steel is about 5.5x10' per degree centigrade, while the coeflicient of expansion of an iron-nickel-cobalt alloy containing 332% nickel and-10% cobalt is approximately 3.52:10' per degree centigrade. An iron-nickel-cobalt alloy also has the additional advantage that it has a lower resistively than nickel steel. The mean coeflicient of expansion of composite wires that are employed for sealing into soft glass is approximately 8x10 per degree centigrade. When a core composed of an iron-cobaltnickel alloy having the composition specified has a resistivity of v is employed in place of nickel steel, the crosssectional area of the copper may, therefore,

be doubled and the cross-sectional area of the core correspondingly reduced without increasing the coefiicient of expansion of the composite wire. Since the resistivity of copper, however, is considerably less than the core material, the resistivity of the composite wire will be reduced to about 5x10 ohms per centimeter c'ube.

Composite wires' have heretofore'been employed only for making seals in soft glass, such as lead glass,

which has acoefiicien-t of expansion of approximately 9x10 per degree centigrade. Their use has been limited in this respect because a sufliciently low expansion could not be obtained over the required temperature range with a composite wire having a nickel-steel core and an outer jacket of copper which could be sealed into hard glass and still have a moderate resistivity. When an iron-nickel cobalt alloy, however, is employed as a core, a composite wire can be roduced which is capable of being sealed 1nto a glass or vitreous envelope having a coeflicient of expansion of 6x10 per degree centigrade. For example, a composite wire having a core composed of an alloy containing 20% to 30% nickel and 18% to 28% co alt and an outer sheath or jacket of copper occupying 10% to 30% of the total cross-sectional area, will have substantially the same coeflicient of expansion as hard lass, such as that known in the art as borosi icate thermometer glass, and will have a resistivity of about 82:10- ohms per centimeter cube.

While an alloy having specific proportions of iron, nickel and cobalt has been mentioned, it will be understood that the proportions of these ingredients may be varied widely, depending upon the cross-sectional area of the copper and the composition of the glass into which the conductor is to be sealed. A small proportion of manganese is also preferably added to such alloys to increase their forgeability and the usual impeurities associated with the ingredients may present, although the amount of carbon should preferably be maintained below .5%. Alloys containing 45% to 66% iron, 20% to 34% nickel, 4% to 28% cobalt and 0 to 1% manganese have been found satisfactory, the proportion of the ingredients in the alloy depending upon the expansivity desired and the cross-sectional area of the copper. An alloy containing 25% to 34% nickel, 4% to 17 cobalt, 0 to 1% manganese and the remainder iron and minor impurities has been found suitable for sealing into soft glass when the cross-sectional area of the copper is maintained between 10% and 40%. By reducing the cross-sectional area of the co per to approximately 5% to 20%, the coe cient of expansion of'the composite conductor can be reduced to 3 to 5x10 per degree centigrade and can, therefore, be effectively sealed through an envelope composed of hard glass, such as that which, is commercially known as Nonex.

The values given in the foregoin exam les do not represent the lowest resistivities ut only those which may be obtained by conventional methods. It is possible to produce a core having a lower resistivity by certain refinements,.such as by reducing the manganese content or by increasing the ratio of cobalt to nickel. By such methods, it is possible to obtain a copper-jacketed core which is capable of being successfully sealed into hard glass having a coefficient'of expansion of a proximately 4x10" per degree centigra e which will have a fairly high conductivity.

My improved composite wire may be effectively sealed into the glass envelope in any desired manner. I prefer, however, to first form'a glass beading on the conductor which may have the same composition as the glass envelope or a composition having a coeflicient of expansion intermediate the coefiicient of expansion of the glass and that of the composite wire. The beaded glass may then be inserted through an aperture in the glass envelope, and the contacting glass of the envelope and the beading are fused and pressed thoroughly in contact with each other.

In forming an effective seal between glass and a metal which is wetted by glass, such as copper it is only necessary to have the mean coeilicient of expansion of the metal the same as that of the glass from atmospheric temperatures to that at which strains in either the glass or metal are ra idly released. The softening temperature 0 copper is below that of glass. Consequently, with a copperjacketed conductor, it is only necessary that the mean coeflicient of expansion of the com-. posite wire shall be equal to that of glass up to the softeningtemperature of copper. In certain constructions, and especially where good conductivity is desired, I have found that if, after sealing the conductor into the glass, the temperature is maintained at the lowest point at which strains are efiectively eliminated for a period of one to twenty-four hours, the expansivity of the com osite wire ma be substantially reduced. or copperja eted wires having an iron-nickel-cobalt core, I have found that the strains may be eliminated by maintaining the seal at a temperature of from 200 C. to 300 C. for one to twenty-four hours. When such a process is followed, the cross-sectional area of the copper-jacket may be increased and the conductivity correspondingly improved. The

above process also enables the seal to be cooled to normal" temperature at a comparatively fast rate.

While I have described my invention in considerable detail and have given numerous examples, it will be understood that my in vention is not limited-to the specific details set forth in the foregoing examples which should be construed as illustrative and not by way of limitation and in view of the numerous modifications which may be efl'ected therein without departing from the spirit of my invention, it is desired that only such limitations shall be imposed as are indicated in the appended claims.

I claim as my invention:

1. A composite wire comprising a core composed of an iron base alloy containing 25% to 34% nickel, 4% to 17% cobalt and a small ioo I percentage but not more than 1% of manganese, and an outer acket formed of copper; the cross sectional area of the copper being maintained at 10% to 40% of the cross sectional areaof the wire. 'Y v 2. A composite Wire-comprising a core composed of an iron base alloy containing to 34% nickel, 4%- to 17% cobalt and a small percentage but not more than 1% of manganese; and an outer jacket of copper, the

cross sectional area of the copper being maintained between 5% to.20% of the cross sectional area of the wire, the coefficient of expansion of the composite conductor varying from 3 to 5x10 per degree centigrade.

In testimony whereof, I have hereunto subscribed my name this 6th day of February, 1930.

HOWARD SCOTT.