A Study of an Artificial Telephone Line Citation Swedlund, Lloyd Edward (1928) A Study of an Artificial Telephone Line. Master's thesis, California Institute of Technology. doi:10.7907/KARN-PM22. https://resolver.caltech.edu/CaltechETD:etd-02252005-132801 Abstract No abstract. Item Type: Thesis (Master's thesis) Subject Keywords: Electrical Engineering Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Electrical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Maxstadt, Francis W. Thesis Committee: Unknown, Unknown Defense Date: 1 January 1928 Record Number: CaltechETD:etd-02252005-132801 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-02252005-132801 DOI: 10.7907/KARN-PM22 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 760 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 28 Feb 2005 Last Modified: 03 Oct 2019 23:02 Thesis Files Preview PDF (Swedlund_le_1928.pdf) - Final Version See Usage Policy. 6MB Repository Staff Only: item control page

A STUDY OF AN ARTIFICIAL TELEPHONE LINE Thesis by Lloyd E. Swedlund In Partial Fulfilment of the Requirements for the Degree of Master of Science California Institute of Technology Pasadena, California 1928 A STUDY OF AN ARTIFICIAL THLEPHONE LINE The purpose of this study was to oerform tests on an artificial televhone line to illustrate the fundamental principles of telephone transmission line phenomona. The artificial line and the equipment used with it was given to the California Institute of Technology through the generosity of the comoanies associated with the American Telephone and Telegraph Company. It was designed primar— ily to be operated as a telephone line although resistors are provided which can be short-circuited to make the line equivalent to a power circuit. The tests described in this study were taken only when the line was connected as a telephone line. The tests included measurements of the distribution of current, voltage, and impedance at a number of frequencies in the voice renge with different terminal conditions of the line. The effect of loading was studied by the use of shunt resistors. Resonance, reflection, and the Ferannti effects were also observed and measured. The value of an artificial line lies in approximating or act- ually reproducing the conditions in a rezl line, so this line wes investigated to see how nearly it behaved as 2 | real line. Since many artificial lines such as tiis heave been built and tested, their properties are well known and there are probably little of new phenomona to be observed in a line such as this one. The chief value therefore lies in Be checking transmission line calculations,in being able to observe phenomona that could not be conveniently observed on real transmission lines,and for testing new apparetus. The apparatus used for testing were, exceot for a plate voltage supoly set, built by the Western Electric Company, and are comoletely described in the manuals for the anparatus. However, it is worth while to give a orief description of the apparatus here for the manuals cover it in more detail than is required for only a general idea. The artificial line was assembled by other students previous to this study in an arrangement suggested by the Telephone Company. The Line consists of twenty-five "T# sections, each section containing a 14.6 millihenry induct- ance coil on eacn side of the line, a shunt condenser of 0.0645 microfared capacity, and a resistance on each side to bring the loop resistance of the section to approximately 83 ohms. Each section is equivalent to 7.83 miles of No. 12 NBS gauge open-wire non-loaded line, the total equivalent loop length thus being 197 miles. Since the inductance represented in the adjacent halves of two sections is com- bined into one coil in order to reduce the number of coils required and to save space, a "cut-in" unit made up of two half sections is provided to tap the line at voints at which measurements are to be made. The line might have been assembled as a pi (1) line but there is practicelly no ad- vantage to be gained by this connection for then the con- densers would have to be disconnected and be replaced by two half sized condensers when measurments were made. With api circuit all the coils can be used thus making the equivalent length of line one section Longer but extra condensers are then required. In "Laboratory Experiments in High Frequency Electrical Transmission" by ii. P. Weinbeach, an arrangement is shown in which the parte can be switched from one connection to the other quite easily. A diagram of a part of the line showing the cut-in section is given below. 726mh 7é mh. WV 06495 -- 0645 mtd = fo ‘meter—> ot. are, ‘Cut-in” Half Sections cn gaya One of the difficult problems in connection with the {I dl use of alternating current lines in the bast has been that of conveniently and quickly measuring the smell voltages and currents which are used, particularly in artificial telephone lines. This problem has been satisfactorily solved by the development of the vacuum tube voltmeter- millianmeter. The Western Electric type D-79017 Voltmeter- . Williemmeter supplied with the artificial line is a very carefully designed anc a flexible instrument for this purpose. Ite voltage range is from 3.16 to .O001 r.m.a. volts anc the current range is from 31.6 to .02 r.m.s. Milliamperes for frequencies between 100 and 5,900 cycles. Between 200 and 3,000 cycles the accuracy of measurement is about @ oer cent. For frequencies outside of this range the accuracy is. not as great but should normally be better than 5 per cent. . The crinciple of the circuit is based on a comparison method of measuring. The unknown voltage is connected across & high impedance transformer (or for current measurement across & low resistance shunt) anc the voltage is then fed into a three stage impedance coupled amplifier whose ampli- fication is variable over a wide range by means of cutout Switches and potentiometers in the grid circuits. This an- plifier is connected to a detector tube, which has a sensitive d.c. milliawmeter in its plate circuit. The amplifier is first adjusted to give a convenient deflection on the meter. Then the input of the amplifier is switched to a potentiometer which has across it a known voltage at the test frequency. When this potentiometer is adjusted to give the same deflection on the meter as before, the voltage applied to the amplifier is the same. The scale of the potentiometer is so graduated that by means of switches and a resistance network, it reads directly in volts and milliamperes or in convenient multiples of them. Voltage is supplied to the voltmeter-milliamneter by the oscillator and is adjusted to the prover value for the calibrated potentiometer by means of the d.c. meter anc @ sensitive thermo-couple. Since the applied voltage is 3.16 volts, this cen be done with the ordinary thermo-couples that are available. A sending circuit which is adjusted to send one milli-watt of testing power into a 600 ohm circuit is also connected to the calibrated circuit. This system of measuring obviously overcomes the disadvantages of errors due to variation of battery voltage, tube characteristics, etc. It is, however, necessary to check the voltage on the calibrated circuit for each reading to get good accuracy. ' This is particularly true when there are variations in the voltage of the oscillator or if the load on the sending circuit varies. The whole apparatus is substantially assen- bled in a strong box and weighs about 135 pounds without batteries so that it can be readily moved about. In order to be certain of the accuracy of the voltmeter- milliammeter its calibration and accuracy was checked before and during tests. The thermocouple calibration was checked using a Weston Standard Laboratory Voltmeter by the method described in the A. T. & T. Manual D 79017, Article 4.12, which was sent with the meter. Only a slight readjustment wes found necessary. The accuracy of the voltmeter-—millian- meter wes checked by the method described in article 4.4 of the same manual. A Western Electrio tyne 8-A Oscillator supolies testing power. It is cavable of supplying from 25 to 35 milliamperes into a 600 ohm load in a frequency range of from 100 to 50,000 cycles. The frequency variations due to temperature changes are less than one per cent and a nearly pure sine wave is generated up to the maximum out put. It is satis- factory for most purposes without the use of a "CO" battery 6. but the wave form is still further improved if a "0" battery is used, The oscillating circuit is a conventional vacuum tube generator and is made of inductance coils and condensers which can be cut in ané out of the circuit by means of dial switches to vary the frequency. The calibration chart fur- nished with the oscillator gives the diel settings only for frequencies twenty cycles apart. For some work it is desir-— able to get nearly the exact frequency so graphs, whichare now in the manuel for the oscillator, were preparec. It was found that on the lower frequencies the variations in con- denser settings (which are divided in tenths) could be plotted against freauency for a given coil setting. This is not quite as accurate as determining each setting by the use of an impedance bridge as described in the oscillator manual but the fact that the curves were smooth and continuous indicated that the settings taken from the curve are nearly as accurate as those taken from the chart. The oscillator, like the voltmeter-—milliammeter, is substantially assembled in a strong box so that it is easily portable. The oscillator requires twenty-four volts for the filament supply and 130 volts for the plate voltage supply. Since the plate current is about 60 milliamperes, the life of dry batteries would be short, so a plate supply rectifier to operate from the a.c. lighting circuit was built by Mr. Ingersoll. Oonsiderable difficulty was encountered this year in using it however, for while it would no doubt be quite satisfactory as a plate voltage supply for e radio set, certain characteristics of the voltmeter—milliammeter make 7. its use less satisfactory than the use of batteries. Due to the fact thet slow variations of the a.c. voltage cannot be filtered out, the d.c. voltage varied. These were suffi- cient to make readings on the voltmeter—milliameter impossible. However, by supplying the voltmeter—-milliammeter with « battery and the oscillator only with the rectified voltage readings could be taken. The varying suoply voltage, of course, varied the voltage outout of the oscillator which in turn »ro- duced variations on the meter and changed its calibration. To partielly compensate for this the rectifier was rewired to give better voltage regulation and to permit "C" voltage for the oscillator to be taken from it. An increase in voltage then made the grid more negative and tended to lower the oscillator output voltage, and vice versa. A storage battery would have given better results and the dry battery now necessary could have been dispensed with. Another piece of anneratus furnished with the artificiel line is a Western Electric 1-B Impedance Bridge. It isa standard tyne bridge for measuring the resistance and react- ance of telephone apparatus over the voice frequency range (aporoximately 100 to 3000cycles). It is used in connection with the oscillator and a receiver. The bridge has a range of from + 560 to ~560 millihenries, the "+" sign representing inductive reactance and the "-" sign capacitive reactance. The theory of transmission lines has been quite comolete- ly developed and is presented in several textbooks. Pract- ically all of these texts use the hyperbolic method of calculation so it is not necessary to repeat the development 8. of the formulae here. The notation and equations given in ‘Principles of Transmission in Telephony" by ii. P. Weinbach - will be used in the discussion of the results. The equations required will be introduced as required. RESULTS The graphs which follow are referred to by plate numbers and are grouped together. Following each plate there is a4 data sheet from which the greph was olotted. The graphs showing the distribution of current, voltage, and impedance egeinst line length are plotted in per cent of the sending end values since this makes the curves easier to interpret and the scales are arbitrary anyhow. It happens that each small division on the graph paper is one eighth of a section length and since this is 7.88 miles, it is easy to read approximate. distances in miles. Distances are plotted in section units,however, since it would be bothersome to olot them exactly into miles. Plate I shows the distribution of current, voltage and impedance along a line terminating in a load equal to its Characteristic imoedence, at a frequency of 796.4 cycles (amf = 5,000). This frequency has been chosen by the tele- phone companies as @ good averege value. If a line is of infinite length, removing some finite length of line, say one unit, should not change its characteristic impedance. That is, for finite distances from the sending end the characteristic impedance does not vary. Thus if a line of finite length terminates in an impedance equal to its characteristic impedance it should behave like a infinite line. The current and voltage will then decay exponentially according equations (6) and (7) given in Weinbach. E=E,¢"" (6) | I=1,€°" (7) The data sheet gives the calculated values es well as the measured velues. Due to the fact that the load used was not exactly ecual to the characteristic impedance these values do not check as closely as they might. The character- istic impedence was capacitive and it was not possible to get the exact value desired, however it wes within ten ver cent so the error dve to this should be small. Since the impedance remains constant along the line the phase angle between current and voltage does not change. This as well as the exponential decay is also shown on plate II, which shows a polar diagram for this condition. The length of the vector fepresents the real part of the voltage or cur- rent and the anguler displacement the imaginary part. That the attentuation increases linearly with distance, is shown by the curve on plate I which gives the loss in transmission units at each section along the line. The transmission unit as chosen by the American Telephone and Telegraph Company is defined by the equation, Loss in T.U.'s = 10 logig Egls Erly This gives the loss in units which are nearly equal in value to the former unit of miles of standard cable and are more convenivent to use. The condition thet a line act like one 1. "Principles of Telephony in Transmission" by M.P.Weinbach. 10. of infinite length is that there are no reflections and thus no phase shift between the current end voltage. Since the characteristic impedence varies with frequency, it is not possible to have a line of finite length act as an in- finite line at all frequencies, but it will be seen later from plates 15 and 16 that the value calculated for 796.4 cycles is not much different from the others except for frequencies below about 500 cycles. Plate 3 shows the current, voltage, and impedance distribution at 796.4 cycles with the end of the line oven. The current here falls to zero at the end of the line and is reflected back along the line producing stending waves in- stead of the normal exponential decay. The rapid decrease of current at the end of the line generates a higher voltage there due to the rapid decrease in flux associated with the current. The voltage and current curves clearly show that both quantities decay exponentially with the distance from the sending end and also vary.as sine and cosine functions of the distance. By observing the distances between two successive intersections of the current and voltage, the wave length at this frequency can be easily determined. It is, according to the curves 218 miles, by calculation it is 220 miles. The impedance given by the ratio of volts to amperes, varies along the line beeause of the phase shift, becoming infinite at the end of the line. The attentuation in transmission units is plotted against line length. It is interesting to observe thet the rate of attenuation is about. the same in the first half of the line as it is for the infinite line given in plate I. This is because lines ll. that are long electrically show little effect of the termin- al conditions on the sending end impedance, particularly “when the line losses are great. The values of current and voltage were also calculated for this value of frequency and terminal conditions anc show a@ good agreement with the measured values. There are occasional points that are clearly in error and these may have been due to errors in reading or to poor connections. These errors could be eli-~ minated by taking the measurements over and in fact this was done for many of the tests when there were too many errors in the curve. The voltage and current distributions always come out as smooth curves unless there is some kind of dis- continuity in the line. In this case it is very easy to see where this discontinuity is from the graph and it can then be located on the line. This is not a vractical method, however, for use on actual line since it would be very diffi- cult to measure the current and voltage distribution along a real line. On most of the curves it can be seen thet every other value of current is slightly in error. This is due in some way to the method of using the half sections for measure- ments. Since they are reversed every time it appeared that this might be due to the fact that they are not symetrical but measurements of resistance showed that this was not true. It may be due to the fact that at every other section the coils of the section are connected at one side to the sending end and the other times to the receiving end and may draw a small charging current due to their distributed capacity. This error is small, however, and is not very objectionable. 12. It is quite tedious to make calculations for the current and voltage at all points along the line so after it was -found that these values checked well with the measured, there would be very little purpose of making any more. A polar diagram of the open circuited line is also given on plate 2. Plate 4 which gives the curves for the same conditions 2s plate 3, exceot that the end is short circuited, can be seen to be the same in general character but the voltage and current curves are now interchanged and the imoedance curve for the short circuit condition is the reciprocal of that for the open circuited line. This is indicated by the eque- tions for the current and voltage along the line under the two different conditions and the two quantities are here also interchanged. This shows then that there is no particular purpose in measuring the distribution under each of these conditions when one of them is known. In the cese of the short circuited line the voltege wave is reflected. Plate 5 shows the curves for the distribution of current voltage and impedance at 100 cycles. Due to the fact that the weve length of this line is long (1725 miles) and the damping is great,there is little reflection. One hundred cycles is below the frequency used for ordinary telephone conversation but it is desirable to transmit this frequency when high quality such as for radio broadcasting is desired. Plate 6 shows the distribution of current, voltage and impedence at 2280 cycles, which shows more pronounced reflec- tion than at 100 cycles. This is true because here the wave 13. length is about 1016 miles, thus a quarter wave length is equal to 256 miles and at this wave length the reflection on open circuit would be greatest if the line length were 256 miles. Since it is near this length the reflection is large. Plate 7 shows the distribution of current, voltage end impedance at 1000 cycles. Measurements of the distances be- tween the intersections of the current and voltage waves gives @ weve lengtn of 173 miles, which is less than the length of line. It is evident that at every odd quarter wave length the voltage is & maximum and the current is a minimun, while the reverse is true at distances equal to even quarter wave lengths. Plate 8 also shows the distribution at 1000 cycles but with the volteges and currents plotted in volts and milliamperes. The purpose of this is for comparison with the »pre- ceding curve to show the advantage of the scales used there. Plate 2 shows the cistribution of current, voltage and impedance with the line short circuited, at a frequency of 3,000 cycles. The wave length as measured from the graph is about 57 miles. The phenomona mentioned for the other curves are also evident here. This is considered the upoer limit of frequency necessery for good intelligibility in conversation, but for good quality a cut off frequency of about 5,000 cycles may be desirable. Plate 10 shows the distribution of current, voltage and impedance at 5,000 cycles which is about the limit at which ea good distribution curve can be drawn from measurements. This 14. is because the length of each section is nearly equal an eighth of a wave length and the curves to a certain extent _ have to be estimated. For measurements taken only at the end of the line this Limitation is not oresent, but this value is also at the limit of the range of the vacuum tube voltmeter-milliammeter, and oractically, except for carrier current frequencies which go much higher, no measurements need to be made. The Calculated cut-off frequency of the line is 7,300 cycles so measurements could not be carried much farther on this line. This cut-off frequency is based on @ resonance phenomona anc because of resistance in the circuit it is not sharp. However the correction for the readings taken at 5,000 cycles is wery small and cen be neglected, hence the line can be consicered equivalent to an actuel line up to this frequency. This is discussed in detail in “Artificial Electric Lines" by Kennely anc also in "Transmission Circuits for Telephonic Communication" by Johnson. Plate tt which shows the distribution of current, volt- age and impedance at 815 cycles, for the line open circuited, the frequency at which the line length is equal one fourth the natural wave length. This is indicated by the fect that the current and voltage waves intersect at almost the mid- point of the line. The value for the resonant frequency was calculated according to the exact formula given on page 50 of "Principles of Transmission in Telephony" by Weinbach. The measured value cheeks with the calculated value. The Value as given by the approximate formula is 236 cycles which 15. is close enough if the line resistance is low, that is, if the line is assumed to be composed only of inductance and capacitance. The graph shown in plate 12 which gives the distribution at 660 cycles at which the natural frequency should be equal to the line length. However, it does not check as well as the 215 distribution curve. The curve indicates that the frequency should be a little lower. Plate 13 shows a very interesting effect which is en- countered in lines which are short electrically, namely, the Ferannti effect. This is the effect in which the voltage is higher at the receiving end of the line than at the send- ing end. The maximum value occurs when the line length is equal to one fourth of the naturel wave length. However, if the attenuation is too great no Ferranti effect will be ob- served above thedefinite length of line, when the attenuatim constant is less than 56.09 per cent of the wave length con- stant. This is shown very well by the curves. By looking at plate 5 showing the distribution on open circuit at 100 cycles it is seen that since the receiver voltage is less then the sending voltege there will not be a Ferranti effect for the full length of line. However, if a little lower frequency were used a@ Ferannti effect might be observed, for the attenua- tion decreases with decrease of frequency although the Ferranti effect also decreases. Another interesting effect can be ob- tained by short circuiting the line at various lengths, when curves will be obtained which correspond to the Ferannti effect for voltages. | One of the distorting effects on unloaded lines is the is. une qual attenuation of the different frequencies. If the line is loaded the maximum amount, all freouencies are attenuated alike. However, lines are seldom loaded this much. The theory at first published by Heaviside, states thet for a distortionless line the product of the line re- Ssistence and shunt capacity must be equal to the product of the line inductance and the shunt conductance. In the or-~ dinary telechone line this result is attained through in- creasing the line inductance by lumped loading in the form of low resistance inductance coils. The same result could be obtained by decreasing the decreasing shunt capacitance or by increasing the shunt conductance. The effect of loading wes studied by inserting shunt resistors across the line at the section junctions. It happens that telephone lines are normally loaded at points 7.88 miles apart so it is quite easy to imitate the effect of regular loading by putting shunt resistors at the middle of each section, since the inductance in an ertificial line is already in lumps. This was accomplished by soldering clips on the conéensers to hold the resistors, which were the ordinary grid leaks such as are used in racio sets. These had to be measured and sorted to get resistors which came within 5% of an average value. | Calculations show thet in the cese of the line being studied the shunt resistors should be 25,000 ohms each at each section for distortionless transmission. This is about @ thousandth of the leakege resistance of a real line. In order to approximate an actual loading condition,resistors | having 65,700 ohms resistance were used. It would, however, 17. be quite easy to test other values of resistors in the same clips. The sending end receiving volt-amperes were measured for a range of frequencies from 100 to 2,600 cycles with the Line terminating in a load equal to the characteristic impedence at 796.4 cycles, with the resistors on the line and without the resistors. Using this deta the attenuation in transmission units was calculated end olotted against frequency. The results of this test were disapnpointing as the only conelueion evident was that the attenuation is greater with the resistors. The line is rather short to require loading for correcting attenuation variations and since the transmission unit is a logarithm the variations appear small on the granh. The distribution of current anc voltage along the line wes measured with the resistors on the line but the percent variations were so near those with- out the resistors that there is no need to plot it. Plates 15 and 16 show the results of measuring the characteristic line impedance at open circuit and short cir- cuit by using the impedence bridge. The resistance and in- Guctance were measured over a range of frequencies from 300 to 4,200 cycles. From this data the short circuit and open circuit impedance could be calculated. Weinbach gives the formule for characteristic impedance as, Z. e/* = /Lse e%, Z£'" The characteristic phase angle & is also plotted. It is seen that Z, represents the mean of the oven circuit anc the short circuit impedance and also that above about 800 is. cycles its value is nearly constant. The line presents capacitative reactance at all frequencies but values plotted in millihenries since these were measured on tne bridge. Their values can, however, easily be converted into microfarads. CONCLUSIONS The measured values as shown on the plates show a good agreement with the calculated values and show that this arti- ficial line, for ourposes of tests, is equivalent to a real line up to 5000 cycles. The data on the use of shunt re- sistors for decreasing distortion is not conclusive for the line was not long enough to have much distortion. The theory indicates that this method would produce the desired results, but at the expense of a large decrease in efficiency. Tnis report represents only a small portion of the possible number of tests which can be verformed with this apparetus but it illustrates the main points in the theory of electric wave transmission over lines. To take up every point in connection with the theory would require much more time than wes available. A considerable amount of time was soent in becoming familiar with the manipulation of the apperatus and in getting it to operate satisfactorily. exe * KH OH The writer wishes to express his anpreciation for the assistance given him by Mr. Hossack and Mr. Biddle during the second term. He also wishes to thenk Prof. Sorensen and Mir. Maxstact for suggestions and criticisms. 19. BIBLIOGRAPHY "Principles of Telephony in Transmission" by M. P. Weinbach. "Laboratory Experiments in High Frequency Electrical Transmission" by M. P. Weinbach. 'Flectric Power Transmission and Distribution" by Woodruff. "Artificial Electric Lines" by Kennely. "Transmission Circuits for Telephonic Communication, by Johnson. "Telephone Communication" by Wright and Puchstein. ef sf F. eB. otters titte ita Sr a mee cheancae nag | ann oman } 2 . wJata Sheet for Plaves Lew 2 Distribution of Current, Lins & te CIM FO: WO 09 eR oe © Fa) a poh ep os Oe <q [o) Lorphinetead in « fupedence, f= » e * : oa Coenet ay: Cop cre ow . * . AO SINE SCOR bg tA tM OS ii i Bip 1.170 i. ie it eo * spe Ho Pe ST Ob GD « * * enmen inom & cn e . Hs a Ce ty co ef thas CD pel & Re eth A oo A ote Cc iead fh © * oO) FR! fe SP eee . co ca . * . . @ ° * ¢ . C5 Ca Ry BS be He oA em yn a el oa ail an) . oe SOO OO Pe men IO MSO O ib onc tora bo bo bo bo 65 no Ga a uyD. La "t it ” = 20 - "7 = L756 Voitaze and iup jed ance él wes, Stony Sy we vy cat bo its Unerecteri VY CGLc& at 2 as. va 8 sx a2 C Bs Gad -LE SyO ~26 2550 ~ ot eek He 2745 - 64 i] .60E ae Pol Peatote “ite boy LCRA aL e26 647 hte 14] 2706 2494 +154 Wt GE 2GCF -LET 2684 oe = LEO 2652 AlL4 -19S WU4E 2 2 206 errerene v0 +819 L.06C 2460-251 L.elic .oS7) -E4S Lec ok kT 1.256 .891 -270 Ley ret G -BEL L,oF4 w26G H256 1.412 woe Cy L.a7é wand -OLR OOO .liss per seco Re Onz & & stic wt qd . LU. bap oe YeessOgs C ob itty ars) 4.53 oF wea) o& B04 ov we Lt ae Aeare tes ae &.68G 2G e776 6 6.0 anes ane we 6.20 ae rasta 0 (gue we Toto ma) lores mae u.ot we D.e4 20 2.6L & 10,81 oe LOLET LF LL S8 wo Li.oe wk arene cf Lele ad a ay f bor a tg io Fe pf ft f ia! 4 Ae fate fe i, woes BOE PAT) DtyS1s2ZIOADY D Ul PALOUNMM AA) PUT ‘ fd PaLVOMD UadE pug Z AY szj2ho EOIEL =f JUBIAND PUP BODLION - . 24) U2aMjag Uol{ojar aseyd 3 auf buimoys SuIDsBOI 40/0 SET “e ! vb fo i DLA ) Data sheet for Plates 2 u 3 The. Distrinution of Current, Voltage, and Iupedance along an Open Circuited Line at 796.4 cycles a 7 {(culculated)Phase @.U,° S$. > “9 LO: a/1 fis, I. Re /Ig iis I. angle. Loss G62 1. L£.600 1.000 1.000 1.000 1.000 20.c o e791 1, 927 .99C) 6986 WEG OYTO 21L5 24i 716 1, Ooi 6947) 6.851 6.056) 6987) 27 1.04 ~GELl 1, 755.920) .OLl) «W771 OT) 28.9 Leh? -614 709 ,ofe wold .715 .ct9 50.6 z.C9 5 . 655 .760 .6405 .B56 756 SE. o.i5 6856 67Ue EET) GLE SO,6 EG wiv) 876 642) 691 S92) SAB 6621 1.01C 659 .618 40.9 4.10 .55E 1.18 0 644 FAY 41,6 04K e men « €) oe 8 6 o 8 @ NOAKaannn~in “Em ot OO ten tn OO G! OAH Oe Qwnrnroa ae onro-: n iS ~J con CO GLO Re om OR Soo 4) OOonrrmwmomnnymnooorn. L Ot. . 479 1.50 608 477 O46 F.S4 ai Eke, . 450 1.46 616 .425 G6. 5 G6 12 6492, as) 401 1.48 566 400 16.4 6.41 lo .4hC ~5a22 ©6407 1.26 0555 .4i1 5.6 6.76 14 .402 .47e 66 .410 1.115 475 (402 -4.8 7 tO 15.6 .5ke Cl 6476 6850) 640L 464-7 «O77 7 40 16 .272 BR eOL7 .atc 66650) 6 6oce ASL HAE Seid 1? 212 57k 246 «507 6405 1246 J BCT 649 9 605 1S .164 566 1L9C 4649600 .Sol) wav) | CBOR 26.0) 10.22 19.165) £546 185) 4.475 689) W191 6400 5.2 610.45 tw La . t- cc > RS ° Ho ite & 225 6.442 6510) 625600 (644206 677.10 «640.01 06 6.590 1.740 6501 = .oeC bY. 7 .6C 046 6295 1.165 6.265) SCL 695.4 9.08 097) 64616 1.668 1.412 .£10 109.1 10.782 405 WLLE 8.87 245 107 100.0 15.10 2Sol 000 442 20 o 454 0 a9 0 a po po P39 to fo cn a GPO po * ° . oe 1 Od Co Gi bo SD pf > pp mma eee Aen ON bP Go Ga STP © R = &1.86 ohm. X= Lw : 0292 x FOCO = 14 é 6.0 Oiuis , Cw= ,C645 x 10% GCG ao SEES x 10 iuho, 2 =Re*ei Ls 167.46e%er’, cos ~ = , [90 .G82F x 1c e! -]/(4238" 2, > [# = 720.5 ¢/! _ j 75°22" 9 = ~4¥ = .2085 a zp cos 8 = .CR& uyp. rad, vrp sind = ,224¢ % " » ih a tt te & tm v sections = £20 wiles os 78 O00 wdies oer eacong Vu — = 475,000 wiies per second, - Vv gt Ze GOSH p Ss is Io sinh p 3. c¢O0so p 30 Sinh p 3o Cnet eh Ae Geom nen nlp atti OF EE eae wi £8 fe pe om a ane neage -nat i menage pata Sheet for Plate 4 Distribution of Currdnt, Voltage, and Inupedance, along a short Circuited Line at 796.4 cycles, ; B/ a = 3 8. as I. {Bs Vt. /t * Z 0 6866 1.228 1.000 1.006 1.000 1 »GEO © L110 2J6S ~O0e 1.070 ‘2 795° 1.020 24 TEL Leild 3 765 18 669 707 1.162 4 1o7 S65 2056 ~666 4.210 B 682 2old 2790 2666 L.19 6 bee | 175 746 0654 LL 7 0577 775 evans Gud 1.06 9 056 0759 BS 2627 048 LO 0400 2760 0465 2628 W748 Li 05F 0746 eile 608 | 2676 g 2042 700 2097 0bTE 695 re Gi e i ri ie e m es o -4C0 587 LTE 421 467 ~900 452 O04 1.147 | 479 516 1.515 426 297 ASE 245 2.04 495 196 £50 Rib be Gta one oe . ° . MME EO lex) mon . e * * dO) Gis on er oon ay) 3 00D 19 404 2206 2476 195 2.44 20 0075 058k «406 2c00 24595 ed Past 27 555 0074 2290 L290 Pa en BS 2220 9% 2O4E 20652 zo eLtU e452 cn 09 094 2500 24 O98 526 »LO7 408 £50 2 ate) 2646 0 2446 e Bg= 3) Sinn p 3 Gosh p 3. I. = Il —-*" & O cosn p dsr COMO TMP OYHO & tp Data Sheet for Plate 5 the Distribution of Current, Voltage and Impedance 4LlOng an’ Open Circuited Line at 100 cycles w wm a oe eee oe eel « MO Mw Oo FOTO we cn we PO ew 5 WWOG sioe 2967 09S 90 LOC 295 L000 1.000 LOL 1.04 iL.@OL 1.02 1.02 1,01 i.OL 1.02 Wave lengtu q. 2G ae 610 2060 670 200G 2725 2676 641 2601 BEE BSF 0490 442 401 2064 JOEL 267 6250 »165 124 0840 0404 0 L750 wile “A "/ By 2000 950 0954 2IL6 sO LG Porn’) 2098 2U7k 2ord ec bE 2064 2664 2o68 2066 007 & 2o71 2875 2O7F 2oo4 .Ou2 .cog 239k 2O9E 2064 0564 0008 tr o 1 3 /Ig /I- 1.0CG 1,000 0970 L,CLE e700 LOL 2927 2990 Arotane) 1,065 2004 L.OFs 2770 1.16 e7hS 1L.aue 2600 2 .€6 2040 1.066 2596 1.460 BES freer 20 0 DEE 4.660 a7) 1.686 otal 2.005 2 oOo £.c8 Aor oe hG ~006 a 89 2 6E 5.06 Data Shest for Plate 6 the Distribution of Current Voli#ge and Tupedance eiong en Coen Circuited Line at 220 cycles. . By) I. @ Sg ag | I. / Be / I, | air = 1 2070 1.00 Aretn®) vba wc +) »tTED wok WIG sad 2045 & 740 ev BU eT rere 2UEG 5 2076 osu 2Eu' Preiayi eUUL 6 PaeEele eo 7 2645 eGLG 01 Cu vi ove onan 607 2o0C 2760 j ue e726 eevee aes 754 ¥ 6545 ware Prarazs) TE “FOL Le FEE e7oC ohh 26LE8 Parmees: Lh Paerera ~66N 2EEC C4 2o64 LE 2FLS orate E87 26he 2060 Lo wos »CBU Fed Pasvac! OT: 14 ier 264 JBSb are 1.016 L5 ed EG are ok a L.cvtk 16 2bau eka eo 2441 L.ehE Li be 240d 262 2460 4.405 Liu Puater®) Prey ee) ava’ eve 1.666 Lg oot sunt Ppater's 2a L.9Lh £0 Pasrere) sete 0PoE wf 2.00 au Paarors) sees aa | sole att BE 2G »168 2615 LAY oe dl Ba ee ma ere) »6lo nenn E.07 be 097 eOEF ELE OBE Lig 25 »600 © 2615 Q oo fi tht id pee ‘1 ‘ ‘ e | Cea ae ue Me Data Sheet tor Plate 7#H8e ze, and Itpeadance along — the Distripution of Current, Volts an Open Circuited Lige et i700 cycies x . Th . ; ry ae ay Ts Bg */t, we 7 C .0GBB 20 L0G L.ece Lee A Oke — 1,09 276G 29 0S 1.065 g 0795 690 none 2685 4.127 o asta 2340 Monee) avin er®) Lebut 4 0705 2650 Arora @) 010% L.dlet 5 Alerere) 0c7e2 Amar’ L156 1.5 6 bY Areren © 0O7e 26 id odo6 7 2hO8 aera eat FEO 2040 8 abd 8350 auaaore, 2698 tid 2 e411 ofS 480 647 1&6 Le 414 2120 G06 660 2old il 486 »626 oP ak ares 29685 12 461 DEE Pasr 07 1.24 135 266 e259 F446 are, . 14 o4E7 0064 05ot ae) 1B 42 word 2296 009 1.608 16 eotl 16 456 047 L.eh4 7 ener s e278 Oe - 409d 0650 18 2218 06S 267 44% »602 19 - o1l4b 2552 L74 460 ne) 20 LET »541 LEC GEL 200k g1 1&1 ~5LO 2215 0425 ~5CR ae el B 420 0285 2060 Aacrsee) a4 Paetay | of? 460 dl o.e1 2F 2067 G okew 0 EL Sye1)99 fi ' foes i t ; i oe ee S8/9AD OGOE 4 IAD Fa a lesh Ie nr 2 ema oe en en , : ‘ : he vata gneet tor Plate 9 The Distribution cf Current, Voltage, cui lupedance slong a & Snort Oitcuited tine ot 3,000 Uyeyles rae HOWUD MOO OUR © i Fav) all oa oon =" 2 35 Be Ie a) Tit, 4/t = & 770 L.gio 1.608 1.0 1 .G00 740 L.yec oJ6h Pastore’ L088 2062 L.090 2caG -C56 1,005 25DOS 1.060 WIS mores 2DSG mane. © vid QO 7282 1.080 »b6S 2oug Tog 264 L.i00 ale 2cbS 615 2670 ote 445 280d 0575 6C2 wok eh & 2058 2617 WAG 1.22 54258 2628 eel? 495 1.045 2040 C77 445 Fed 2608 S52 Paanels) 4F7 605 oT .088 0445 B04 2o5S 1.458 Hs) eol? »£05 eGL8 oot 1. O55 4 eal? 2558 2004 245 2692 2896 0408 2007 woOL7 1.220 oly ote ei 4 6200 1.746 7 "805 445 B76 2045 tT 5 »L67 ~£56 “S17 »wobY 2605 Y walk 49 2onT L356 £560 QO. 2256 2248 OUR eb & 1.708 4 »L0n 410 ~Lo4 Pasyous) 2415 2 2t66 wbd 220 208 Rors! 5) 2267 .083 ook? O65 F640 , 2a Ue 286 2obk eo 1.65 5 G 410 i¢) .uLe ¢) po bs More oe ee wave denote LPOu Data Sneet for Plate 1¢ she Distribution of Current, Voltase, and Inpedeance along an Open Circuited ine at 6,00C Cycyles ag 3 3; ‘is Is | a/3s Vt te ©) 2556 1.266 1.000 1.006 L.0CG 1 2570 1.170 .OFG 7G o95R a bad L,Ghvu eu 207 1.007 ) 271 »965 218 72 2d BE 4 2242 2664 Ave, 2676 £6000 5 2ool 786 536 eaer-r2) 2 5o 6 2068 2102 -bEL oh 46 1.026 7 0 56 56605 516 2oae 207 & 2004 »56E 464 455 L018 F] 0268 2540 2405 428 ov 6F LO 2a b5U 6465 O61 2076 1,012 Li 0G 20 0444 2040 dE 0958 12 2208 29S wl Pronne 1.000 Lé elyl »o59 B94 raat L.042 . Ld ~L64 arora 2250 2265 29O5 15 2165 204 hE 281 ielel L6 el EG vi 0Lv7 sade 0546 17 i44d 226 06a oil Lenk’ Lo 098 eek ele 218 1 C7 - Lo Loe .L66 .2OL elZy 1,556 Fa 2066 2252 ~L04 .L96 Bol 21 ~LE5 2106 Loi O84 Bet a2 O41 aod ~062 i686 - 006 a4 2OOLF aK 200265 Lol OCS ae elZ0 G LOO 0 oo BR bi Re Data Soeet gor Piate il suc Gigsrusation of Current, Voltase, and Tupédences along i & . an O9en Circuiled Line F = 215 Oycies (1/4 wave seagti) ke sO DR OW’ OAM nt ONE a + . YA td f | oa 3 a. i. 4 om /ig ‘fy iS eo 1.000 L.CCG L.GOC Lee 2O0 bead wk wad WE 2OE0 para eiere, 2ole 2VOU 247 ef bo 2vEC wUlL7 pnererss won 2616 2ol2 0&9 ~15E Aloka) AEsTozs) se CEE avers Arezars| 26U ef al BYE 6GoG 2004 249 Pasa) 2bG4 ore cul Dak 6666 ef TE 2616 rae) 04a e057 ole Aagrere, Ave a og of IO 27S BAG 4.050 2546 Pears) ~) WGES ones LSC 0550 2AOS Aretona ETE a.aFk Pol A aan® UT 2215 teed 2BTO 4b 2606 2S LPS 2561 wks 2616 wun? i .Ug obYOE ek 4 1600 eoGk 2.16 FIG werd 2656 oka £64 +605 Sly C47 aur owl 612 »L60 2651 eh ke 615 ei lO Auenarey 2LCk 6 ke 2G2C0 2VFe25 GBs 2049 La. G 2661 0 oe 433 se lhe + Loj" + (25 + i) Le Data Sneest for Biate 12 wJistrinution of Curren> Voltace .- nd Imoedance alon y o Fe ; an Open CGireuitel Line at 660 Cycles. “ss Ms “s tS (as Tir. ale . 20628 1.17 2 Ce 1.000 4.000 L 2JE0 1.0% iE 915 1.c4ce a 0 ES Leb ~GSO0 OF 728 7) «7 C0 1.1 »ols 0SO2 ea h 2606 1.08 virone 2ole 2G 5 2626 096 726 2020 20% 6 F268 ek 614 777 siod 7 505 205 2BOb Poe 10) 2607 8 9 496 20k 528 2700 2627 +506 69 2587 590 | 0995 2515 64 0295 0546 1,066 o5OB 61 586 2406 1.645 0497 48 0b77 410 1.410 470 042 05g 20g LFS 452 47 502 ALB 1 .2R0 587 475 448 .406 1.104 309 »5BC O59 470 765 838 2558 276 477 ATS 168 B90 195 504 087 145 FLO 166 456 ,O8C 180 B40 209 462 455 weed 460 2250 FES 276 e695 ol? 042 Oke 1.60 40 2255 2O9eO 221 1.776 2566 ol23 0&B 2b o.04 sol G ae sore) 8) a 7 fe ene pho bet t arte sea > 7 nan ces hay ik wie) Data Sheet for Plate 13 fhe Ferennti iffect at three freqyeuncies 3, fe £50 f= 860 £ = 1720 GC 1.COCc L.C0G L,000 lL 1.ces 1,062 1.162 2 1.012 - 1.121 1,875 3 LOLS 1.297 4.31 4 Leled 1.645 - £.68C0 5 1.054 22150 1.246 6 1.06<4 BedDed 205K q 1.090 Bei hC e2EE fs) Le VET L608 1.056 9 iLeblé 4.255 1.38 10 1.120 1,021 1,45 il 1.i81L ~870 1.0 2 L142 0195 Pasteas) Lo L.L-EG 0154 ol 14 1.128 1&0 A eae) 1é 1.123 739 (G2 L6 1,074 2748 ~o16 L? 4.021 0155 Pansy 1L& 2095 745 Ar op, 19 ISS 708 £68 £0 »685 2655 restora al 2005 2601 Pasrer £e 2788 0542 »5C9 23 742 0495 2490 £4 2695 0252 aan) as 2665 405 0406 php gee cope eter wht Data for Flate 14 whe Attentuation in Transwission Units with Fryeuency for the Line with end without 65,700 ohm shunt iiesistors frequency T.U. Tou. witnout with 100 %49C 3.04 200 LG.07 11.46 rele) Li eZE Le.76 400 il. o1 le .Leo 600 12260 16.66 70 12.81 -15.76 sere he 95 14,i1 900 156,00 14.12 Loce Lo.46 14.14 Lioc | 15.04 14,82 LE OO 16660 vn a 1400 13127 va LEOO LS .54 were L600 Lo.76 la .7€ L7 CC io .76 14.92 LEGO 15.94 14.96 1900 15.96 LE.O8 2000 14.10 LE.LS ALe 14,26 TRS eect ia .iF LA 37 22 vie 12.29 15.47 2400 » 14,328 LE 64 2500 14,51 15.77 £2600 14,62 |. 15,06 @.U. Loss = 10 losig sets Bel ct aay oo Ther | Sas , oO ‘“e <7 i + ALE IT ) Saha ey supe ul 2 a ie a a Data for Plates 15 & 16 Whe Vareition of Open Circuit and short Cincuit Impedance wesistance, Reactance, and Phase Angle with Prequency, it guort circuit L 2 aR L Ho e@ f O1un Ohm -mn O tin “in Ohw degree cycle o - go2 6G OSA ETE 71G00© 8EE 9CO) RAY] S41 G67 146 73 G&G fF chy 24.2 7F6 75S lly 74. Fi" 6U 746 BULL 6&7 boG 69 819 794 FO TFC Lusk 6é9 FE Sl 796 7B c& 73l 16,0 745 Vo? 88 631 GBl 46 717 14.1 782 FF SE 641s BG 25 705 LEE 7CO 67e SR BGR NR ll 697 i.e 636 616 6 FV THC iv 631 lvl 658 6F6 7 715 694 25 666 9,8 (55 75C 8 ool 619 LE G0 8B (265 711 16 638 65¢ € 677 Bl 644 €OL 15 ETH C672 6A 615 614 3 726 19 16 66600 A 669 669 1 66h 688 12 666 Gel 719 717 F603 6GE 6 662 §.F 67S 666 8 €42 642 1 687 4.6 CLS 61¢ Fook FOL S 666 4.6 E86 BE L685 647 a er ses GEG GLE G 620 él 6 ORR O49 omcrae = GCP BLE 2 iFe 4.4 59 608 C 665 bbe GC 6FL 4k Gif ble 2 Gus LTe & 656 4.0 662 862 L €85 9 619 FR 645 4. E58 68 SG Rye Od 5 6at as eo? €21 F 6 60 6 O CoS ALO Eu Fy 2 664 GCE 3 627 sal cB ben uw. Ge €LS £-68F of CFL FC 2 ENG mae 2 615 2.2 GCO FGF & 608 602 ue be ak Fed Fgh GS 65L cso 2.7 GOT ae EGO 590 GO FSG Fer. CFG ek 620 e2C LL BFE TER 47 BEG O8Y GCE ELE 2 Fel Lo Feo 68 FER BRE 1 600 1 BYY 2 FTL FTL 0 BLE £ Fin | ECE LUG EO Lo FRG FEC LoORTF Sue EYE OFS S fed ES U Fey &.F FOR ROE L fer Ber L Ree fet FER BER GPRS FFE i ere