US 1904162 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
April 18, 1933.
H. MILLIKEN ELECTRICAL cABLE Filed Aug. 15, 1930 `2 sheets-sheet lll 00| April 18, 1933.` H. MLLEKEN ELECTRIGAL CABUE a sheets-snee@ Filed. Aug, l5, 1950 Patented Apr. l8, 1933 UNITED STATES rA'rl-:Nr OFFICE Humains mLLIxnN, or noUNr ROYAL, Kommt, QUEBEC, enana vnnnernrca'r. canna i Application mcd .August 13,A 1930,. Serial No. 475,094.
The object of this invention is to provide a cable in which the skin eect of large power cables is reduced to a minimum'. l
, Skin effect is that well known phenomenon 5 of alternating currents in which the internal induction in a conductor causes the current density to be a minimum at the center of the" lcross section of cable and a maximum at the lo periphery., If the conductor is a cable, that the skin effect consists in each of the outer strands carryin more current than each of the inner stran s. The greater the distance of a strand from the center, the more current it carries. Considering all of the strands of a cable as so many equal conductors connected in multiple, it can be shown by elementary mathematics, that the energy lost in trans- .mitting a given current I, throu h. the cable, is a minimum when each stran carries the same quantity ofcurrent. If the -total true resistance of the cable is `called. R, the total alternatin current is called I, the ltrue resistance o each lstrand, r, and the current .in any strand is called i, being a variable, different in quantity for different strands, then the total heat generatedand lost inthe cable is Ezr; dividing this by I2 gives a quotient which we will call R,.which is commonly preferred arrangement of cable strands; Fig. 2 is a cross-sectional of a complete, sheathed, cable; A Fig. 3 1s a similar view with the sheaths o coverings omitted; v
' Fi 4 is a diagrammatic view showin one possi le arrangement of the strands o the cable Fig. 5 is a view of the tubular arrangement Y 4 flattened out;
of the strands shown in F' 'agrams of possiis, composed of a number of strands, th.en\
called the ap arent resistance of the cable In the drawings, 1 is a diagram of av perspective viewA ble Ways of folding and curvin the iiattened tubular arrangement shown in ig. 5;
Fig. l1 is a side view of a complete cable of the type shown in Fig. 2, with the coverings broken away for better illustration; and 55 ig. 12 is a cross-section enlarged of one. j of the cable strands. v
The problem is therefore to make all strands carry equal currents, or as nearly s0 as possible or practical. There are several M ways of accomplishing this result. Figure l is a diagram showing the requirements for complete solution of the problem. The cable is assumed to have eight strands (the small number being assumed to avoid complicat- 5 ing the drawings). Each of the eight strands is shown shifted or transposed so as to occupy successively every strand position in the cross section of the cable, for an e ual distance along the cable.v Each stran is 7 therefore subjected to the same total inductive effect as every other strand. Each strand therefore carries the same quantity of current.
The transposition can be accomplished in 75 either of several ways, for instance as shown in Figure 4. The practical objection to such a cable would be the large space which it occupies, which would render it very costly to cover the insulation and lead or armor.
In order to eliminate this objection, the cable, after being made up in the circular form shown in Figure 4, could be flattened as shown in Figure 5. In doin this it becomes necessary to introduce a slig t insulation between adjacent strands; otherwise, the flat cable would act somewhat like a flat bar of solid copper, in which the greatest current ldensit would be at the outeredges of the bar. 4 he insulation between adjacent strands need be only suiiicient to offer a contact resistance between adj aoent strands, many times as great as the resistance of one strand for a" len h of one turn in its progress around the 95 cab e, which is in the order of a few thousandths of an ohm. Therefore, a few ohms of insulation resistance between strands will be sulicient for practical purposes. Such extremely light insulation can be obtained in 10o any number of ways, such as by oxidizing the surface of the metal of the strand `or by japanning the surface or by treating the surface chemically in any manner, the mam requirement being practically to secure a coatin which is extremel thin but which will not ru ofi in contact with the adjacent strand. Referring to Figure 5, the iiat form of cable is not a convenient or economical form of cable for many purposes, especially in the case of cables requirin an insulating covering overall. To obtain amore convenient shape, the flat cable can be folded or rolled into more compact forms, as shown in Figures 6, 7, 8 and 9, 4which are diagrammatic, showing only the center line of the strands in one direction in which the strands wind as the advance along the cable.
ince rectangular cables are not usually' convenient or economical, thecable can`be made into a round form by a rolling process as shown in Figure tially round form could be obtained from the original fiat form, by folding it` into the shapes shown in Figures 8 and 9. y
The foregoing forms of cable, while they fulill the theoretically ideal requirements of complete equalization of currents in all strands still they have practical objections for eneral use; and by sacrificing an entirely negligible amount of current inequality in the strands a more practical form of cable can be obtained as shown in Figure 3. In
this form, the cable is divided into any num-y ber of identical segments, each segment consisting of a number of concentric layers or strands. 'Each strand in any one layer will carry exactly` the same quantity of current since each strand is completely transposed and is, therefore, subjected to identically the same inductive eiect as every other strand in the same layer.v Now lto compare strands in different layers, consider the outermost layer;
, each strand in it alternately occupies a position at the periphery of theround cable and then a position nearest to the center ofthe round cable. The total induction to which this strandy is subjected in the entire length of cable, is, therefore, the same as if the strand were'placed throughouty its length, in a fixed position at a point very near the center of the segment. lThe same may be said of `every layer in the segment. The current carried by every strand in the segment is the same as if the strand were located at the"center of the layer. All layers have the same center and, therefore. all strands in the segment carry the same amount of current. (It is possible that the geometrical center of a layer does not e0 exactly coincide :with its inductive center,
but this deviation must be very slight, hence the foregoing statement is correct for all practical purposes) ln view of the foregoing it will be seen that there is very little it anything to be 10. Cables of substan gained by subdividing the round cable into more than three or foursegments. rIheoretically two (semi-circular) segments would be suiiicient, but practically, three or four segments will be better, depending somewhat V upon the total cross section required in the cable. Referring to Figure 3, the four segments shown are .laid spirally around each other and therefore, make the large cable more flexible. There is also another advantage'ous effect of my construction, viz: the heat generated near the center of the cable cross section is conductedv along the strands to the` outer part of the cable cross'section thereby facilitating the dissipation ofthe heat from the cable. v
' Referring to Figure 3, it would be possible to omit the sheets a of insulation between the se ents without impairing the desired resu ts, provided that the light insulating coating applied to the individual strands is suf- `liciently durable to withstand rubbing off by the entire periphery of the segment and forms a part of the insulation of the entire cable from ground. A lead sheath b and a belt insulation cis shown in Figure 2. Obviously, other forms of covering, such as braid or an armor, can be applied, around the cable.
It will be understood, of course, that at the ends of the cable all the strands therein will be electrically connected together to form a single conductor so that throughout practically the len h of the cable the strands composing isshown 1n Figure 11, in which d designates a copper lu soldered or otherwise electrically connecte pared for making this electrical connection,
the same will bein parallel.v This to all the strand ends. To en- I able the strand ends to be conveniently premay use as insulation (e, Figure 12, which isv an enlarged cross section of one of the strands) oxide of copper, which may be readily removed by ordinar soldering iiux.
It is to be lunderstoo that by power cables I mean those conductors which are adapted to carry alternating current of the usual fre- Y quency, namely, 60 cycles per second or less, and it will be further understood that although the insulation may be slight, the resistance must be high i. e., many times as great as the resistance of one strand for a length of one turn in its progress around the cble, as stated. Also it will be understood t at it readily removed by ordinary soldering flux or other chemical, this being advantageous in that the chemical may permeate the interior of the large cable, i.` e., enter the s aces between the cable strands', thereby avoidin the necessity of spreading apart the ends o the is important that this resistance shall f be of such nature that, as stated, it may be lll vstrands and scraping each strand (which would be an impractical job on a large power cable) whenever it is necessary to solder a lug onto the end of the cable.
What is claimed is:
A cable for alternating power currents, composedof a plurality of sectors helicall arranged around the 'axis of said cable, eac of said sectors being composed of a plurality of layers of solid Wires, each layer being separate and continuous throughout the entire length of the cable, each wire continun ing in the same layer throu hout the length of the cable and being he ically arranged around the axis of its segment, all of said wires and sectors being of the same polarity and connected together electrically at each end of said cable and having suicient contactresistance between adjacent wires throughu out the length of the cable to tend to prevent current flowing between adjacent wires, for the purposes set forth.
testimony whereof I hereunto aiiix my signature. v