I) INTRODUCTION TO DIELECTRIC & MAGNETIC DISCHARGES IN ELECTRICAL WINDINGS by Eric Dollard, ©1982 II) ELECTRICAL OSCILLATIONS IN ANTENNAE AND INDUCTION COILS by John Miller, 1919 PART I INTRODUCTION TO DIELECTRIC & MAGNETIC DISCHARGES IN ELECTRICAL WINDINGS by Eric Dollard, ©1982 1. CAPACITANCE 2. CAPACITANCE INADEQUATELY EXPLAINED 3. LINES OF FORCE AS REPRESENTATION OF DIELECTRICITY 4. THE LAWS OF LINES OF FORCE 5. FARADAY'S LINES OF FORCE THEORY 6. PHYSICAL CHARACTERISTICS OF LINES OF FORCE 7. MASS ASSOCIATED WITH LINES OF FORCE IN MOTION 8. INDUCTANCE AS AN ANALOGY TO CAPACITANCE 9. MECHANISM OF STORING ENERGY MAGNETICALLY 10. THE LIMITS OF ZERO AND INFINITY 11. INSTANT ENERGY RELEASE AS INFINITY 12. ANOTHER FORM OF ENERGY APPEARS 13. ENERGY STORAGE SPATIALLY DIFFERENT THAN MAGNETIC ENERGY STORAGE 14. VOTAGE IS TO DIELECTRICITY AS CURRENT IS TO MAGNETISM 15. AGAIN THE LIMITS OF ZERO AND INFINITY 16. INSTANT ENERGY RELEASE AS INFINITY 17. ENERGY RETURNS TO MAGNETIC FORM 18. CHARACTERISTIC IMPEDANCE AS A REPRESENTATION OF PULSATION OF ENERGY 19. ENERGY INTO MATTER 20. MISCONCEPTION OF PRESENT THEORY OF CAPACITANCE 21. FREE SPACE INDUCTANCE IS INFINITE 22. WORK OF TESLA, STEINMETZ, AND FARADAY 23. QUESTION AS TO THE VELOCITY OF DIELECTRIC FLUX APPENDIX I 0) Table of Units, Symbols & Dimensions 1) Table of Magnetic & Dielectric Relations 2) Table of Magnetic, Dielectric & Electronic Relations PART II ELECTRICAL OSCILLATIONS IN ANTENNAE & INDUCTION COILS J.M. Miller Proceedings, Institute of Radio Engineers. 1919 1) CAPACITANCE The phenomena of capacitance is a type of electrical energy storage in the form of a field in an enclosed space. This space is typically bounded by two parallel metallic plates or two metallic foils on an interviening insulator or dielectric. A nearly infinite variety of more complex structures can exhibit capacity, as long as a difference in electric potential exists between various areas of the structure. The oscillating coil represents one possibility as to a capacitor of more complex form, and will be presented here. 2) CAPACITANCE INADEQUATELY EXPLAINED The perception of capacitance as used today is wholly inadequate for the proper understanding of this effect. Steinmetz mentions this in his introductory book ''Electirc Discharges, Waves and impulses''. To quote, ''Unfortunately, to a large extent in dealing with dielectric fields the prehistoric conception of the electro- static charge (electron) on the conductor still exists, and by its use destroys the analogy between the two components of the electric field, the magnetic and the dielectric, and makes the consideration of dielectric fields unnecessarily complicated.'' 3) LINES OF FORCE AS REPRESENTATION OF DIELECTRICITY Steinmetz continues, ''There is obviously no more sense in thinking of the capacity current as current which charges the conductor with a quantity of electricity, than there is of speaking of the inductance voltage as charging the conductor with a quantity of magnetism. But the latter conception, together with the notion of a quantity of magnetism, etc., has vanished since Faraday's representation of the magnetic field by lines of force." 4) THE LAWS OF LINES OF FORCE All the lines of magnetic force are closed upon themselves, all dielectric lines of force terminate on conductors, but may form closed loops in electromagnetic radiation. These represent the basic laws of lines of force. It can be seen from these laws that any line of force cannot just end in Space. 5) FARADAY AND LINES OF FORCE THEORY Farady felt strongly that action at a distance is not possible thru empty space, or in other words, "matter cannot act where it is not." He considered space pervaided with lines of force. Almost everyone is familiar with the patterns formed by iron filings around a magnet. These filings act as numerous tiny compasses and orientate themselves along the lines of force existing around the poles of the magnet. Experiment has indicated that a magnetic field does possess a fiberous construct. By passing a coil of wire thru a strong magnetic field and listening to the coil output in headphones, the experimenter will notice a scraping noise. J. J. Thompson performed further experiments involving the ionization of gases that indicate the field is not continuous but fiberous (Electricity and Matter, 1906). 6) PHYSICAL CHARACTERISTICS OF LINES OF FORCE Consider the space between poles of a magnet or capacitor as full of lines of electric force. See Fig. 1. These lines of force act as a quantity of stretched and mutually repellent springs. Anyone who has pushed together the like poles of two magnets has felt this springy mass. Observe Fig. 2. Notice the lines of force are more dense along AB in between poles, and that more lines on A are facing B than are projecting outwards to infinity. Consider the effect of the lines of force on A. These lines are in a state of tension and pull on A. Because more are pulling on A towards B than those pulling on A away from B, we have the phenomena of physical attraction. Now observe Fig. 3. Notice now that the poles are like rather than unlike, more or all lines pull A away from B; the phenomena of physical repulsion. 7) MASS ASSOCIATED WITH LINES OF FORCE IN MOTION The line of force can be more clearly understood by representing it as a tube of force or a long thin cylinder. Maxwell presented the idea that the tension of a tube of force is representative of electric force (volts/inch), and in addition to this tension, there is a medium through which these tubes pass. There exists a hydrostatic pressure against this media or ether. The value of this pressure is one half the product of dielectric and magnetic density. Then there is a pressure at right angles to an electric tube of force. If through the growth of a field the tubes of force spread sideways or in width, the broadside drag through the medium represents the magnetic reaction to growth in intensity of an electric current. However, if a tube of force is caused to move endwise, it will glide through the medium with little or no drag as little surface is offered. This possibly explains why no magnetic field is associated with certain experiments performed by Tesla involving the movement of energy with no accompanying magnetic field. 8) INDUCTANCE AS AN ANALOGY TO CAPACITY Much of the mystery surrounding the workings of capacity can be cleared by close examination of inductance and how it can give rise to dielectric phenomena. Inductance represents energy storage in space as a magnetic field. The lines of force orientate themselves in closed loops surrounding the axis of current flow that has given rise to them. The larger the space between this current and its images or reflections, the more energy that can be stored in the resulting field. 9) MECHANISM OF STORING ENERGY MAGNETICALLY The process of pushing these lines or loops outward, causing them to stretch, represents storing energy as in a rubber band. A given current strength will hold a loop of force at a given distance from conductor passing current hence no energy movement. If the flow of current increases, energy is absorbed by the field as the loops are then pushed outward at a corresponding velocity. Because energy is in motion an E.M.F. must accompany the current flow in order for it to represent power. The magnitude of this E.M.F. exactly corresponds to the velocity of the field. Then if the current ceases changing in magnitude thereby becoming constant, no E.M.F. accompanies it, as no power is being absorbed. However, if the current decreases it represents then a negative velocity of field as the loops contract. Because the E.M.F. corresponds exactly to velocity it reverses polarity and thereby reverses power so it now moves out of the field and into the current. Since no power is required to maintain a field, only current, the static or stationary field, represents stored energy. 10) THE LIMITS OF ZERO AND INFINITY Many interesting features of inductance manifest themselves in the two limiting cases of trapping the energy or releasing it instantly. Since the power supply driving the current has resistance, when it is switched off the inductance drains its energy into this resistance that converts it into the form of heat. We will assume a perfect inductor that has no self resistance. If we remove the current supply by shorting the terminals of the inductor we have isolated it without interrupting any current. Since the collapse of field produces E.M.F. this E.M.F. will tend to manifest. However, a short circuit will not allow an E.M.F. to develop across it as it is zero resistance by definition. No E.M.F. can combine with current to form power, therefore, the energy will remain in the field. Any attempt to collapse forces increased current which pushes it right back out. This is one form of storage of energy. 11) INSTANT ENERGY RELEASE AS INFINITY Very interesting (and dangerous) phenomena manifest themselves when the current path is interrupted, thereby causing infinite resistance to appear. In this case resistance is best represented by its inverse, conductance. The conductance is then zero. Because the current vanished instantly the field collapses at a velocity approaching that of light. As E.M.F. is directly released to velocity of flux, it tends towards infinity. Very powerful effects are produced because the field is attempting to maintain current by producing whatever E.M.F. required. If a considerable amount of energy exists, say several kilowatt hours* (250 KWH for lightning stroke), the ensuing discharge can produce most profound effects and can completely destroy inadequately protected apparatus. * The energy utilized by an average household in the course of one day. 12) ANOTHER FORM OF ENERGY APPEARS Through the rapid discharge of inductance a new force field appears that reduces the rate of inductive E.M.F. formation. This field is also represented by lines of force but these are of a different nature than those of magnetism. These lines of force are not a manifestation of current flow but of an electric compression or tension. This tension is termed voltage or potential difference. 13) DIELECTRIC ENERGY STORAGE SPATIALLY DIFFERENT THAN MAGNETIC ENERGY STORAGE Unlike magnetism the energy is forced or compressed inwards rather than outwards. Dielectric lines of force push inward into internal space and along axis, rather than pushed outward broadside to axis as in the magnetic field. Because the lines are mutually repellent certain amounts of broadside or transverse motion can be expected but the phenomena is basically longitudinal. This gives rise to an interesting paradox that will be noticed with capacity. This is that the smaller the space bounded by the conducting structure the more energy that can be stored. This is the exact opposite of magnetism. With magnetism, the units volumes of energy can be thought of as working in parallel but the unit volumes of energy in association with dielectricity can be thought of as working in series. 14) VOLTAGE IS TO DIELECTRICITY AS CURRENT IS TO MAGNETISM With inductance the reaction to change of field is the production of voltage. The current is proportionate to the field strength only and not velocity of field. With capacity the field is produced not by current but voltage. This voltage must be accompanied by current in order for power to exist. The reaction of capacitance to change of applied force is the production of current. The current is directly proportional to the velocity of field strength. When voltage increases a reaction current flows into capacitance and thereby energy accumulates. If voltage does not change no current flows and the capacitance stores the energy which produced the field. If the voltage decreases then the reaction current reverses and energy flows out of the dielectric field. As the voltage is withdrawn the compression within the bounded space is relieved. When the energy is fully dissipated the lines of force vanish. 15) AGAIN THE LIMITS ZERO AND INFINITY Because the power supply which provides charging voltage has internal conductance, after it is switched off the current leaking through conductance drains the dielectric energy and converts it to heat. We will assume a perfect capacitance having no leak conductance. If we completely disconnect the voltage supply by open circuiting the terminals of the capacitor, no path for current flow exists by definition of an open circuit. If the field tends to expand it will tend towards the production of current. However, an open circuit will not allow the flow of current as it has zero conductance. Then any attempt towards field expansion raises the voltage which pushes the field back inwards. Therefore, energy will remain stored in the field. This energy can be drawn for use at any time. This is another form of energy storage. 16) INSTANT ENERGY RELEASE AR INFINITY Phenomena of enormous magnitude manifest themselves when the criteria for voltage or potential difference is instantly disrupted, as with a short circuit. The effect is analogous with the open circuit of inductive current. Because the forcing voltage is instantly withdrawn the field explodes against the bounding conductors with a velocity that may exceed light. Because the current is directly related to the velocity of field it jumps to infinity in its attempt to produce finite voltage across zero resistance. If considerable energy had resided in the dielectric force field, again let us say several K.W.H. the resulting explosion has almost inconceivable violence and can vaporize a conductor of substantial thickness instantly. Dielectric discharges of great speed and energy represent one of the most unpleasant experiences the electrical engineer encounters in practice. 17) ENERGY RETURNS TO MAGNETIC FORM The powerful currents produced by the sudden expansion of a dielectric field naturally give rise to magnetic energy. The inertia of the magnetic field limits the rise of current to a realistic value. The capacitance dumps all its energy back into the magnetic field and the whole process starts over again. The inverse of the product of magnetic storage capacity and dielectric storage capacity represents the frequency or pitch at which this energy interchange occurs. This pitch may or may not contain overtones depending on the extent of conductors bounding the energies. 18) CHARACTERISTIC IMPEDANCE AS REPRESENTATION OF PULSATION OF ENERGY FIELD The ratio of magnetic storage ability to that of the dielectric is called the characteristic impedance. This gives the ratio of maximum voltage to maximum current in the oscillatory structure. However, as the magnetic energy storage is outward and the dielectric storage is inward the total or double energy field pulsates in shape or size. The axis of this pulsation of force is the impedance of the system displaying oscillations and pulsation occurs at the frequency of oscillation. 19) ENERGY INTO MATTER As the voltage or impedance is increased the emphasis is on the inward flux. If the impedance is high and rate of change is fast enough (perfect overtone series), it would seem possible the compression of the energy would transform it into matter and the reconversion of this matter into energy may or may not synchronize with the circle of oscillation. This is what may be considered supercapacitance, that is, stable longterm conversion into matter. 20) MISCONCEPTIONS OF PRESENT THEORY OF CAPACITANCE The misconception that capacitance is the result of accumulating electrons has seriously distorted our view of dielectric phenomena. Also the theory of the velocity of light as a limit of energy flow, while adequate for magnetic force and material velocity, limits our ability to visualize or understand certain possibilities in electric phenomena. The true workings of free space capacitance can be best illustrated by the following example. It has been previously stated that dielectric lines of force must terminate on conductors. No line of force can end in space. If we take any conductor and remove it to the most remote portion of the universe, no lines of force can extend from this electrode to other conductors. It can have no free space capacity, regardless of the size of the electrode, therefore it can store no energy. This indicates that the free space capacitance of an object is the sum mutual capacity of it to all the conducting objects of the universe. 21) FREE SPACE INDUCTANCE IS INFINITE Steinmetiz in his book on the general or unified behavior of electricity ''The Theory and Calculation of Transient Electric Phenomena and Oscillation," points out that the inductance of any unit length of an isolated filamentary conductor must be infinite. Because no image currents exist to contain the magnetic field it can grow to infinite size. This large quantity of energy cannot be quickly retrieved due to the finite velocity of propagation of the magnetic field. This gives a non reactive or energy component to the inductance which is called electromagnetic radiation. 22) WORK OF TESLA, STEINMETZ AND FARADAY In the aforementioned books of Steinmetz he develops some rather unique equations for capacity. Tesla devoted an enormous portion of his efforts to dielectric phenomena and made numerous remarkable discovers in this area. Much of this work is yet to be fully uncovered. It is my contention that the phenomena of dielectricity is wide open for profound discovery. It is ironic that we have abandoned the lines of force concept associated with a phenomena measured in the units called farads after Farady, whose insight into forces and fields has led to the possibility of visualization of the electrical phenomena. 23) QUESTION AS TO THE VELOCITY OF DIELECTRIC FLUX It has been stated that all magnetic lines of force must be closed upon themselves, and that all dielectric lines of force must terminate upon a conducting surface. It can be infered from these two basic laws that no line of force can terminate in free space. This creates an interesting question as to the state of dielectric flux lines before the field has had time to propagate to the neutral conductor. During this time it would seem that the lines of force, not having reached the distant neutral conductor would end in space at their advancing wave front. It could be concluded that either the lines of force propagate instantly or always exist and are modified by the electric force, or voltage. It is possible that additional or conjugate space exists within the same boundaries as ordinary space. The properties of lines of force within this conjugate space may not obey the laws of normally conceived space. IMPORTANT REFERENCE MATERIAL 1. "Electricity and Matter," J. J. Thompson New York, 1906, Scribner's Sons, and I904, Yale University 2. "Elementary Lectures on Electric Discharges, Waves, and Impulses and other Transients." C. P. Steinmetz, second edition, 1914, McGraw-Hill 3. "Theory and Calculation of Transient Electric Phenomena and Oscillations," C. P. Steinmetz, third edition, 1920, McGraw-Hill. Section III Transients in Space, Chapter VIII, Velocity of Propagation of Electric Field Table I Magnetic Field Dielectric Field Magnetic flux: φ = Li 108 lines of magnetic force. Dielectric flux: ψ = Ce lines of dielectric force. Inductance voltage: d −8 di e ' =n 10 =L volts. dt dt Capacity current: d di i '=n =C amperes. dt dt Magnetic energy: L i2 joules. w= 2 Dielectric energy: C e2 joules. w= 2 Magnetomotive force: F = ni ampere turns. Electromotive force: e = volts. Magnetizing force: F f= ampere turns per cm. l Electrifying force or voltage gradient: e G= volts per cm. l Magnetic-field intensity: H = 4πf 10-1 lines of magnetic force per cm2. Dielectric-field intensity: G 2 K= 2 lines of dielectric force per cm . 4 v Magnetic density: B = μH lines of magnetic force per cm2. Dielectric density: D = kK lines of dielectric force per cm2. Permeability: μ Permittivity or specific capacity: κ Magnetic flux: φ = AB lines of magnetic force. Dielectric flux: ψ = AD lines of dielectric force. v = 3 X 1010 = velocity of light Table II Magnetic Circuit Dielectric Circuit Magnetic Flux (magnetic current): φ = lines of magnetic force. Dielectric flux (dielectric current): ψ = lines of dielectric force. Magnetomotive force: F = ni ampere turns. Electromotive force: e = volts. Electric Circuit Electric current: i = electric current. Voltage: e = volts. Permeance: M= . 4 F Permittance or capacity: C= Inductance: 4 v2 farads. e Conductance: g= i mhos. e n2 −8 n −8 L= 10 = 10 F i henry. (Elastance ? ): Reluctance: 2 Li F −8 = 10 joules. 2 2 w= Ce2 e joules. = 2 2 Dielectric density: Magnetic density: B = = μH lines per cm2. A D= = κK lines per cm2. A Magnetizing force: Dielectric gradient: F f= ampere turns per cm. l G= e volts per cm. l Magnetic-field intensity: H = .4πf Dielectric field-intensity: Permeability: Permittivity or specific capacity: K= I= i = γG amperes per cm2. A Electric gradient: G= e volts per cm. l Conductivity: = 1 K = . D I mhos-cm. G 2 Resistivity: ρ= 1 G = ohms-cm. I Specific dielectric energy: Specific magnetic energy: per cm3. Electric-current density: (Elastivity ? ): Reluctivity: ρ=f/B .4 f w0 = 2 Electric power: p = ri2 = ge2 = ei watts. G 2 4 v D = K μ=B /H e ohms. i r= Dielectric energy: Magnetic energy: w= Resistance: 1 e = . C 4 v 2 F R= . G2 = fB / 2 10 joules w0 = 4 v 2 = cm3. -8 GD joules per 2 Specific power: p0 = ρI2 = G2 = GI watts per cm3. Table of Units, Symbols, and Dimensions Quantity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Length Area Volume Mass Time Velocity Acceleration Force Energy Power Charge Dielectric constant of free space Dielectric constant relative Charge density volume surface line Electric intensity Electric flux density Electric flux Electric potential EMF Capacitance Current Current density Resistance Resistivity Conductance Conductivity Electric polarization Electric susceptibility Electric dipole moment Electric energy density Symbol mks Unit Rationalized Defining Equation L A v M, m T, t v q F W P Q, q m m2 m3 kilogram second m / sec m / sec2 newton joule watt coulomb ε0 ε εr farad / m farad / m numeric ε0 = 1 / (μ0c2) ρ ρs ρl E D ψ V Vg C I, i J R ρ G σ P xe me ωe coulomb / m3 coulomb / m2 coulomb / m volt / m coulomb / m2 coulomb volt volt farad ampere ampere / m2 ohm ohm-m mho mho / m coulomb / m2 farad / m coulomb-m joule / m3 ρ=Q/v ρs = Q / A ρl = Q / L E=F/Q=-V/L D = εE = ψ / A ψ = DA V = - EL Vg = - dφ / dt C=Q/V I=Q/T J=I/A R=V/I ρ = RA / L G=1/R σ=1/ρ=J/E P = D – ε0E = ρL xe = P / E = ε0 (εr -1) me = QL ωe = DE / 2 2 A=L v = L3 v=L/T a = L / T2 F = Ma W = FL P=W/T F = Q2 / (4πε0L2) εr = ε / ε0 Dimensional Formula Exponents of L M T Q 1 2 3 0 0 1 1 1 2 2 0 cgs emu No. of emu No. of mks cgs esu No. of esu No. of mks No. of esu No. of emu cm cm2 cm3 gram second cm / sec cm / sec2 dyne erg erg / sec abcoulomb 102 104 106 103 1 102 102 105 107 107 10-1 cm cm2 cm3 gram second cm / sec cm / sec2 dyne erg erg / sec statcoulomb 102 104 106 103 1 10% 102 105 107 107 10c 1 1 1 1 1 1 1 1 1 1 100c 1 4πc2 / 107 0 0 0 1 0 0 0 1 1 1 0 0 0 0 0 1 -1 -2 -2 -2 -3 0 0 0 0 0 0 0 0 0 0 0 1 -3 -1 -3 -1 0 0 2 2 0 2 2 0 -3 -2 -1 1 -2 0 2 2 -2 0 -2 2 3 -2 -3 -2 -3 1 -1 0 0 0 -2 0 0 -2 -2 2 -1 -1 -1 -1 1 1 0 2 0 -2 1 abcoulomb / cm3 1 abcoulomb / cm2 1 abcoulomb / cm -1 abvolt / cm 1 1 -1 abvolt -1 abvolt 2 abfarad 1 abampere 1 abampere / cm2 -2 abohm -2 abohm-cm 2 abmho 2 abmho / cm 1 abcoulomb / cm2 2 1 0 erg / cm3 0 0 0 1 0 0 1 1 -1 0 0 1 1 -1 -1 0 -1 0 1 1 10-7 10-5 10-3 106 4π / 105 4π / 10 106 108 10-9 10-1 10-5 109 1011 10-9 10-11 10-5 1 statcoulomb / cm3 statcoulomb / cm2 statcoulomb / cm statvolt / cm statvolt statvolt statfarad statampere statampere / cm2 statohm statohm-cm statmho statmho / cm statcoulomb / cm2 1 statcoulomb-cm erg / cm3 c / 105 c / 103 c / 10 104 / c 4πc / 103 4π10c 106 / c 106 / c c2 / 105 10c c / 103 105 / c2 107 / c2 c2 / 105 c2 / 107 c / 103 4πc2 / 107 103c 10 100c 100c 100c 1 / (100c) 100c 100c 1 / (100c) 1 / (100c) (100c)2 100c 100c 1 / (100c)2 1 / (100c)2 (100c)2 (100c)2 100c 1 Quantity Symbol mks Unit Rationalized Defining Equation 7 Dimensional Formula Exponents of L M T No. of emu No. of mks cgs esu No. of esu No. of mks No. of esu No. of emu 105 / c2 105 / c2 1 / (100c)2 1 / (100c)2 108 / c 103 10 1 / (100c) 1 1 Q 34 Permeability of free space 35 Permeability 36 relative 37 Magnetic pole μ0 μ μr p henry / m henry / m numeric weber μ0 = 4π / 10 μ=B/H μr = μ / μ0 p = A (B - B0) 1 1 0 2 1 1 0 1 0 -2 0 -2 0 0 -1 -1 38 Magnetic moment 39 Magnetic intensity m H m = pL H = U / L or F / p 3 -1 1 0 -1 -1 -1 1 40 Magnetic flux density B weber-m ampere / m or newton / weber weber / m2 B = μH = φ / A 0 1 -1 -1 41 42 43 44 φ U F M weber ampere ampere weber / m2 φ = BA = VgT U = F = HL F=I M = B - B0 = m / L3 2 0 0 0 1 0 0 1 -1 -1 -1 1 -1 1 -1 -1 L M R v P μ Vg P ωm xm henry henry ampere / weber meter / henry weber / amp henry / meter volt watts / m2 joule / m3 henry / m Magnetic flux Magnetic potential MMF Intensity of magnetization cgs emu 107 / 4π pole = maxwell / 4π pole-cm oersted or gilbert / cm gauss or maxwell / cm2 maxwell gilbert gilbert pole / cm2 or gauss / 4π 1 103 / 4π 1010 / 4π 4π / 103 104 108 4π / 10 4π / 10 104 / 4π Inductance 45 46 47 48 49 50 51 52 53 54 self mutual Reluctance Reluctivity Permeance Permittivity EMF Poynting's vector Magnetic energy density Magnetic susceptibility μ0 = 4π / 107 henrys / m. For c = 2.998 X 108 meters / sec, For c ~ 3 X 108 meters / sec, c2 = 8.988 X 1016 ~ 9 X 1016 L= φ/I M = φ / I = W / I2 R=F /φ v=1/μ P=1/R μ=1/v Vg = -dφ / dt P = EH ωm = HB / 2 xm = M / H = μ0 (μr - 1) 2 1 0 -2 2 1 0 -2 -2 -1 0 2 -1 -1 0 2 2 1 0 -2 1 1 0 -2 2 1 -2 -1 0 1 -3 0 -1 1 -2 0 1 1 0 -2 abhenry abhenry 109 109 abvolt abwatt / cm2 erg / cm3 henry / m 108 103 10 107 / 4π ε0 = 1 / μ0c2 = 107 / (4πc2) = 8.854 X 10-12 farad / meter ε0 ~ 1 / (36π109) farad / meter statvolt statwatt / cm2 erg / cm3 ELECTRICAL OSCILLATIONS IN ANTENNAS AND INDUCTANCE COILS By John M. Miller CONTENTS I. Introduction II. Circuit with uniformly distributed inductance and capacity III. The Antenna 1. Reactance of the aerial-ground portion 2. Natural frequencies of oscillation (a) Loading coil in lead-in (b) Condenser in lead-in 3. Effective resistance, inductance, and capacity 4. Equivalent circuit with lumped constants 5. Determination of static capacity and inductance 6. Determination of effective resistance, inductance, and capacity IV. The Inductance coil 1. Reactance of the coil 2. Natural frequencies of oscillation Condenser across the terminals 3. Equivalent circuit with lumped constants I. INTRODUCTION A modern radiotelegraphic antenna generally consists of two portions, a vertical portion or "lead-in" and a horizontal portion or "aerial." At the lower end of the lead-in, coils or condensers or both are inserted to modify the natural frequency of the electrical oscillations in the system. When oscillating, the current throughout the entire lead-in is nearly constant and the inductances, capacities, and resistances in this portion may be considered as localized or lumped. In the horizontal portion, however, both the strength of the current and the voltage to earth vary from point to point and the distribution of current and voltage varies with the frequency. The inductance, capacity, and resistance of this portion must therefore be considered as distributed throughout its extent and its effective inductance, capacity, and resistance will depend upon the frequency. On this account the mathematical treatment of the oscillations of an antenna is not as simple as that which applies to ordinary circuits in which all of the inductances and capacities may be considered as lumped. The theory of circuits having uniformly distributed electrical characteristics such as cables, telephone lines, and transmission lines has been applied to antennas. The results of this theory do not seem to have been clearly brought out; hazy and sometimes erroneous ideas appear to be current in the literature, textbooks, and in the radio world in general so that the methods of antenna measurements are on a dubious footing. It is hoped that this paper may clear up some of these points. No attempt has been made to show how accurately this theory applies to actual antennas. FIG 1. - Antenna represented as a line with uniform distribution of inductance and capacity The aerial-ground portion of the antenna, or aerial for short (CD in Fig. 1), will be treated as a line with uniformly distributed inductance, capacity, and resistance. As is common in the treatment of radio circuits the resistance will be considered to be so low as not to affect the frequency of the oscillations or the distribution of current and voltage. The lead-in, BC in Fig. 1 , will be considered to be free from inductance or capacity excepting as inductance coils or condensers are inserted at A to modify the oscillations. An inductance coil, particularly if a long single-layer solenoid, may also be treated from the standpoint of the transmission-line theory. The theoretical results obtained furnish an interesting explanation of certain well-known experimental results. II. CIRCUIT WITH UNIFORMLY DISTRIBUTED INDUCTANCE AND CAPACITY The theory, generally applicable to all circuits with uniformly distributed inductance and capacity, will be developed for the case of two parallel wires. The wires (Fig. 2) are of length l and of low resistance. The inductance per unit length L1, is defined by the flux of magnetic force between the wires per unit of length that there would be if a steady current of 1 ampere were flowing in opposite directions in the two wires. The capacity per unit length C1 is defined by the charge that there would be on a unit length of one of the wires if a constant emf of 1 volt were impressed between the wires. Further, the quantity L0 = l L1 would be the total inductance of the circuit if the current flow were the same at all parts. This would be the case if a constant or slowly FIG. 2 alternating voltage were applied at x = o and the far end (x = l) short-circuited. The quantity C0 = l C1 would represent the total capacity between the wires if a constant or slowly alternating voltage were applied at x =o and the far end were open. Let it be assumed, without defining the condition of the circuit at x = l, that a sinusoidal emf of periodicity ω = 2πf is impressed at x = o giving rise to a current of instantaneous value i at A and a voltage between A and D equal to v. At B the current will be i C will be v i dx and the voltage from B to x v dx . x The voltage around the rectangle ABCD will be equal to the rate of decrease of the induction through the rectangle, hence v v dx −v =− L 1 i dx x t v i =−L1 x t (1) Further, the rate of increase of the charge q on the elementary length of wire AB will be equal to the excess in the current flowing in at A over that flowing out at B. Hence q d i = C v dx =i−i dx t dt 1 x − i v =C 1 x t (2) These equations (1) and (2) determine the propagation of the current and voltage waves along the wires. In the case of sinusoidal waves, the expressions v=cos t A cos C 1 L1 xB sin C 1 L1 x (3) i=sin t C1 A sin C 1 L1 x −B cos C 1 L1 x (4) L1 are solutions of the above equations as may be verified by substitution. The quantities A and B are constants depending upon the terminal conditions. The velocity of propagation of the waves, at high 1 frequencies is V = . C 1 L1 III. THE ANTENNA 1. REACTANCE OF THE AERIAL-GROUND PORTION Applying equations (3) and (4) to the aerial of an antenna and assuming that x = 0 is the lead-in end while x = l is the far end which is open, we may introduce the condition that the current is zero for x = l. From (4) A =cot C 1 L1 l (5) B Now the reactance of the aerial, which includes all of the antenna but the lead-in, is given by the current and voltage at x = 0. These are. from (3), (4), and (5). V 0 =A cos t=B cot C 1 L1 l cos t i 0=− C1 B sin t L1 The current leads the voltage when the cotangent is positive and lags when the cotangent is negative. The reactance of the aerial, given by the ratio of the maximum values of v0 to i0, is X =− L1 cot C 1 L1 l C1 or in terms of C0 = lC1 and L0 = lL1 X =− L0 cot C 0 L 0 (6) C0 or since V= 1 L1C1 X =−L1 V cot C 1 L1 l as given by J. S. Stone.1 At low frequencies the reactance is negative and hence the aerial behaves as a capacity. At 1 the frequency f = 4 C 0 L0 FIG. 3 - Variation of the reactance of the aerial of an antenna with the frequency 1 J. S. Stone, Trans. Int. Congress, St. Louis, 8, p. 555; 1904. the reactance becomes zero and beyond this frequency is positive or inductive up to the frequency 1 f= at which the reactance becomes infinite. This variation of the aerial reactance with 2 C 0 L0 the frequency is shown by the cotangent curves in Fig. 3. 2. NATURAL FREQUENCIES OF OSCILLATION Those frequencies at which the reactance of the aerial, as given by equation (6), becomes equal to zero are the natural frequencies of oscillation of the antenna (or frequencies of resonance) when the lead-in is of zero reactance. They are given in Fig. 3 by the points of intersection of the cotangent curves with the axis of ordinates and by the equation f= m ; m=1, 3, 5, etc. 4 C 0 L0 The corresponding wave lengths are given by = V l 4l = = f f C 0 L0 m that is, 4/1, 4/3, 4/5, 4/7, etc., times the length of the aerial. If, however, the lead-in has a reactance XX, the natural frequencies FIG. 4 - Curves of aerial and loading coil reactance of oscillation are determined by the condition that the total reactance of lead-in plus aerial shall be zero; that is, XX + X = 0 provided that the reactances are in series with the driving emf. (a) Loading Coil in Lead-in. - The most important practical case is that in which an inductance coil is inserted in the lead—in. If the coil has an inductance L, its reactance XL + ωL. This is a positive reactance increasing linearly with the frequency and represented in Fig. 4 by a solid line. Those frequencies at which the reactance of the coil is equal numerically but opposite in sign to the reactance of the aerial, are the natural frequencies of oscillation of the loaded antenna since the total reactance XL + X = 0. Graphically, these frequencies are determined by the intersection of the straight line - XL = ωL (shown by a dash line in Fig. 4) with the cotangent curves representing X. It is evident that the frequency is lowered by the insertion of the loading coil and that the higher natural frequencies of oscillation are no longer integral multiples of the lowest frequency. The condition XL + X = 0, which determines the natural frequencies of oscillation, leads to the equation L− L L0 C 0 L0 L0 cot C 0 L0=0 . C0 TABLE 1 – Data for loaded antenna calculations 1 L C 0 L0 L 1 Difference, per cent L 0 L0 3 0.0 .1 .2 .3 .4 1.571 1.429 1.314 1.220 1.142 1.732 1.519 1.369 1.257 1.168 10.3 6.3 4.2 3.0 2.3 .5 .6 .7 .8 .9 1.077 1.021 .973 .931 .894 1.095 1.035 .984 .939 .900 1.7 1.4 1.1 .9 .7 1.0 1.1 1.2 1.3 1.4 .860 .831 .804 .779 .757 .866 .835 .808 .782 .760 .7 .5 .5 .4 .4 1.5 1.6 1.7 1.8 1.9 .736 .717 .699 .683 .668 .739 .719 .701 .685 .669 .4 .3 .3 .3 .3 2.0 2.1 2.2 2.3 2.4 .653 .640 .627 .615 .604 .655 .641 .628 .616 .605 .3 .2 .2 .2 .2 2.5 2.6 2.7 2.8 2.9 .593 .583 .574 .564 .556 .594 .584 .574 .565 .556 .2 .2 .2 .1 .1 3.0 3.1 .547 .539 .548 .540 .1 .1 1 L 1 L0 3 Difference, per cent 3.2 3.3 3.4 0.532 .524 .517 0.532 .525 .518 0.1 .1 .1 3.5 3.6 3.7 3.8 3.9 .510 .504 .4977 .4916 .4859 .511 .504 .4979 .4919 .4860 .1 .0 .0 .0 .0 4.0 4.5 5.0 5.5 6.0 .4801 .4548 .4330 .4141 .3974 .4804 .0 .4549 .0 .4330 .0 ........................ ........................ ........................ ........................ 6.5 7.0 7.5 8.0 8.5 .3826 .3693 .3574 .3465 .3366 ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ 9.0 9.5 10.0 11.0 12.0 .3275 .3189 .3111 .2972 .2850 ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ 13.0 14.0 15.0 16.0 17.0 .2741 .2644 .2556 .2476 .2402 ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ ........................ 18.0 19.0 20.0 .2338 .2277 .2219 ........................ ........................ ........................ ........................ ........................ ........................ or cot C 0 L0 C 0 L0 = L (8) L0 This equation has been given by Guyau2 and L. Cohen3. It determines the periodicity ω and hence the frequency and wave length of the possible natural modes of oscillation when the distributed capacity and inductance of the aerial and the inductance of the loading coil are known. This equation can not, however, be solved directly; it may be solved graphically, as shown in Fig. 4, or a table may be prepared indirectly which gives the values of C 0 L0 for diferent values of L from which then ω, f, λ may be determined. The second column of Table 1 gives these values L0 for the lowest natural frequency of oscillation, which is of the major importance practically. FIG. 5 - Curves of aerial and series condenser reactance (b) Condenser in Lead-in. - At times in practice a condenser is inserted in the lead-in. If the 1 capacity ofthe condenser is C, its reactance is X 0=− . This reactance is shown in Fig. 5 by C the hyperbola drawn in solid line. The intersection of the negative of this curve (drawn in dash line) with the cotangent curves representing X gives the frequencies for which X0 + X = 0, and hence the natural frequencies of oscilltion of the antenna. The frequencies are increased (the wave length decreased) by the insertion of the condenser and the oscillations of higher frequencies are not integral multiples of the lowest. The condition X0 + X = 0 is expressed by the equation − tan C 0 L0 2 A. Guyau, Lumiere Electrique, 15, p. 13; 1911. 3 L. Cohen, Electrical World, 65, p. 286; 1915 C 0 L 0 = C (9) C0 which has also been given by Guyau. Equation (9) may be solved graphically as above or a table C similar to Table 1 may be prepared giving C 0 L0 for different values of . More C0 complicated circuits may be solved in a similar manner. 3. EFFECTIVE RESISTANCE, INDUCTANCE, AND CAPACITY In the following the most important practical case of a loading coil in the lead-in and the natural oscillation of lowest frequency alone will be considered. The problem is to replace the antenna FIG. 6 - (a) Antenna with loading coil; (b) artificial antenna with lumped constant: to represent antenna in (a) of Fig. 6, (a), which has a loading coil L in the lead-in and an aerial with distributed characteristics, by a circuit (Fig. 6, (b)) consisting of the inductance L in series with lumped resistance Re, inductance Le, capacity Ce, which are equivalent to the aerial. It is necessary, however, to state how these effective values are to be defined. In practice the quantities which are of importance in an antenna are the resonant wave length or frequency and the current at the current maximum. The quantities Le and Ce are therefore defined as those which will give the circuit (b) the same resonant frequency as the antenna in (a). Further the three quantities Le, Ce, and Re must be such that the current in (b) will be the same as the maximum in the antenna for the same applied emf whether undamped or damped with any decrement. These conditions determine Le, Ce, and Re uniquely at any given frequency and are the proper values for an artihcial antenna which is to represent an actual antenna at a particular frequency. In the two circuits the corresponding maxima of magnetic energies and electrostatic energies and the dissipation of energy will be the same. Zenneck4 has shown how these effective values of inductance capacity and resistance can be computed when the current and voltage distributions are known. Thus, if at any point x on the oscillator the current i and the voltage v are given by i = I f (x); v = V φ (x) where I is the value of the current at the current loop and V the maximum voltage, then the differential equation of the oscillation is 4 Zenneck, Wireless Telegraphy (translated by A. E. Selig), Note 40, p. 430. I 2 2 ∫ C 1 x dx I I 2 R f x dx L f x 2 dx =0 ∫ 1 2∫ 1 t t ∫ C 1 x 2 dx where the integrals are taken over the whole oscillator. If we write 2 Re =∫ R1 f x dx (10) L e=∫ L1 f x2 dx (11) C e= ∫ C 1 x dx ∫ C 1 x2 dx 2 (12) the equation becomes 2 Re I I I Le 2 =0 t Ce t which is the diierential equation of oscillation of a simple circuit with lumped resistance, inductance, and capacity of values Re, Le, and Ce and in which the current is the same as the maximum in the distributed case. In order to evaluate these quantities, it is necessary only to determine f (x) and phi (x); that is, the functions which specify the distribution of current and voltage on the oscillator. In this connection it will be assumed that the resistance is not of importance in determining these distributions. At the far end of the aerial the current is zero; that is, for x = l; il = 0. From equations (3) and (4) for x = l v l =cos t Acos C 1 L 1 lB sin C 1 L1 l , i l =sin t C1 Asin C 1 L1 l−B cos C 1 L1 l ; L1 and since il = 0, A sin C 1 L1 l=B cos C 1 L 1 l . From (3), then, we obtain v =v l cos C 1 L1 l− C 1 L1 x . Hence x=cos C 1 L1 l− C 1 L1 x . Now for x = 0 from (4) we obtain i 0=−B C1 C1 sin t=−A tan C 1 L1 l sin t , L1 L1 whence i=i 0 sin C 1 L1 l− C 1 L1 x sin C 1 L1 l and f x = sin C 1 L1 l− C 1 L1 x sin C 1 L1 l . We can now evaluate the expressions (10), (11), and (12). From (10) sin 3 C 1 L1 l− C 1 L1 x dx Re =∫0 R1 sin2 C 1 L1 l l l sin 2 C 1 L1 l [ − ] sin C 1 L 1 l 2 4 C 1 L1 R1 = = 2 R0 cot C 0 L0 1 [ 2 − ] (13) 2 sin C 0 L0 C 0 L0 and from (11) which contains the same form of integral L e= L0 cot C 0 L0 1 [ 2 − ] (14) 2 sin C 0 L0 C 0 L0 and from (12) l C e= 2 ∫0 C 1 cos C 1 L1 l− C 1 L1 x dx l ∫0 C 1 cos2 C 1 L1 l− C 1 L1 x dx C 21 sin 2 C 1 L1 l 2 C 1 L 1 = l sin 2 C 1 L1 l C 1 2 4 C 1 L1 = C0 1 2 C 0 L0 cot C 0 L0 C 0 L0 (15) [ ] 2 2 2 sin C 0 L0 The expressions (14) and (15) should lead to the same value for the reactance X of the aerial as obtained before. It is readily shown that X = Le − L 1 =− 0 cot C 0 L 0 C e C0 agreeing with equation (6). It is of interest to investigate the values of these quantities at very low frequencies (omega = 0), frequently called the static values, and those corresponding to the natural frequency of the unloaded anterma or the so-called fundamental of the antenna. Substituting w=o in (13), (14), and (15), and evaluating the indeterminant which enters in the iirst two cases, we obtain for the lowfrequency values Re = L e= R0 3 L0 (16) 3 C e =C 0 At low frequencies the current is a maximum at the lead-in end of the aerial and falls off linearly to zero at the far end. The effective resistance and inductance are one-third of the values which would obtain if the current were the same throughout. The voltage is, however, the same at all points and hence the effective capacity is the capacity per unit length times the length, or C0. At the fundamental of the anterma, the reactance X of equation (6) becomes equal to zero and hence C 0 L0= . Substituting this value in (13), (14), and (15), 2 Re = R0 2 L e= L0 (17) 2 C e= 8 C0 . 2 Hence, in going from low frequencies up to that of the fundamental of the antenna, the resistance (neglecting radiation and skin effect) and the inductance (neglecting skin effect) increase by 50 per 2 L and cent, the capacity, however, decreases by about 20 per cent. The incorrect values 0 2 C have been frequently given and commonly used as the values of the effective inductance 0 L and capacity of the antenna at its fundamental. These lead also to the incorrect value L e= 0 for 2 the low-frequency inductance.5 5 These values are given by J.H. Morecroft in Pro. I.R.E., 5, p. 389; 1917. It may be shown that they lead to correct values for the reactance of the aerial and hence to correct values of frequency, as was verified by the experiments. They are not, however, the values which would be correct for an artificial antenna in which the current must be equal to the maximum in the actual antenna and in which the energies must also be equal to those in the antenna. The resistance values given by Prof. Morecroft agree with these requirements and with the values obtained here. The values for other frequencies may be obtained by substitution in (13), (14), (15). If the value L of the loading coil in the lead-in is given, the quantity C 0 L0 is directly obtained from Table 1. 4. EQUIVALENT CIRCUIT WITH LUMPED CONSTANTS In so far as the frequency or wave length is concerned, the aerial of the antenna may be considered to have constant values of inductance and capacity and the values of frequency or wave length for different loading coils can be computed with slight error using the simple formula applicable to circuits with lumped inductance and capacity. The values of inductance and capacity L ascribed to the aerial are the static or low frequency; that is, 0 for the inductance and C0, for the 3 L capacity. The total inductance in case the loading coil has a value L will be L 0 and the 3 frequency is given by f= 1 2 L (18) L0 C 0 3 L0 C 0 (19) 3 or the wave length in meters by =1884 L where the inductance is expressed in microhenrys and the capacity in microfarads. The accuracy with which this formula gives the wave length can be determined by comparison with the exact formula (8). In the second column of Table 1 are given the values of C 0 L0 for different L values of as computed by formula (8). Formula (18) may be written in the form L0 1 C 0 L0 = L 1 so that the values of C 0 L0 which are proportional to the frequency, L0 3 may readily be computed from this formula also. These values are given in the third column and the per cent differences in the fourth column of Table 1. It is seen that formula (18) gives values for the frequency which are correct to less than 1 per cent excepting when very close to the fundamental of the antenna; that is, for very small values of L. Under these conditions the simple formula leads to values of the frequency which are too high. Hence to the degree of accuracy shown, which is amply L sufficient in most practical cases, the aerial can be represented by its static inductance 0 with its 3 static capacity C0, in series, and the frequency of oscillation with a loading coil L in the lead-in can be computed by the ordinary formula applicable to circuits with lumped constants. In an article by L. Cohen,6 which has been copied in several other publications, it was stated that the use of the simple wave length formula would lead to very large errors when applied to the antenna with distributed constants. The large errors found by Cohen are due to his having used the L value L0, for the inductance of the aerial, instead of 0 , in applying the simple formula. 3 6 See footnote 3. 5. DETERMINATION OF STATIC CAPACITY AND INDUCTANCE In applying formula (8) to calculate the frequency of a loaded antenna, a knowledge of the L quantities of L0, and C0 is required. In applying formula (18), 0 and C0, are required. Hence 3 either formula. may be used if the static capacity and inductance values are known. We will call L these values simply the capacity Ca and inductance La of the antenna. Hence C a =C 0 , L a= 0 3 and the wave length from (19) is given by =1884 LL a C a (20) where inductance is expressed in microhenrys and capacity in microfarads, as before. The capacity and inductance of the antenna are then readily determined experimentally by the familiar method of inserting, one after the other, two loading coils of known values L1 and L2 in the lead-in and determining the frequency of oscillation or wave length for each. From the observed wave lengths λ1 and λ2 and known values of the inserted inductances, the inductance of the antenna is given by L a= L1 22−L 2 12 21− 22 (21) and the capacity of the antenna from either 1=1884 L 1L a C a 2=1884 L2 La C a (22) using, preferably, the equation corresponding to the larger valued coil. This assumes that formula (20) holds exactly. As an example, let us assume that the antenna has L0 = 50 gomicrohenrys and C0 = 0.001 microfarad and that we insert two coils of 50 and 150 microhenrys and determine the wave lengths, experimentally. We know from formula (8) and Table 1 that the wave lengths would be found to be 491 and 771 m. From the observed wave lengths and known inductances, the value of La would be found by (21) to be La = 17.8 microhenrys and from (22) Ca = 0.000999 microfarad. L0 . This accuracy 3 would ordinarily be sufficient. We can, however, by a second approximation, derive from the experimental data a more accurate value of La. For, the observed value of La furnishes rough values L1 L2 of and , which in this example come out 0.96 and 2.88, respectively. But Table 1 gives L0 L0 Ca is very close to the assumed value C0 but La differs by 7 per cent from L and shows that this formula gives a 0.7 L0 L =0.96 but no appreciable difference per cent shorter wave length than 491 m (or 488 m) for L0 L =2.88 . Recomputing La, using 488 and 771 m, gives for L0 the per cent error of formula (20) for diierent values of La = 0.0168 which is practically identical with the assumed L0 . 3 6. DETERMINATION OF EFFECTIVE RESISTANCE, INDUCTANCE, AND CAPACITY When a source of undamped oscillations in a primary circuit induces current in a secondary tuned circuit, the current in the secondary, for a given emf, depends only upon the resistance of the secondary circuit. When damped oscillations are supplied by the source in the primary, the current in the secondary, for a given emf and primary decrement, depends upon the decrement of the secondary; that is, upon the resistance and ratio of capacity to inductance. The higher the decrement of the primary circuit relative to the decrement of the secondary the more strongly does the crurent in the secondary depend upon its own decrement. This is evident from the expression for the current I in the secondary circuit. P= N E 20 4f R2 ' I ' where δ' is the decrement of the primary, δ that of the secondary, R the resistance of the secondary, f the frequency, E0 the maximum value of the emf impressed on the primary, and N the wave-train frequency. These facts suggest a method of determining the effective resistance, inductance, and capacity of an antenna at a given frequency in which all of the measurements are made at one frequency and which does not require any alteration of the antenna circuit whatsoever. The experimental circuits are arranged as shown in Fig. 7, where S represents a coil in the primary circuit which may be thrown either into the circuit of a source of undamped or of damped oscillations. The coil L is the loading coil of the antenna, which may be thrown over to the measuring circuit containing a variable inductance L', a variable condenser C', and variable resistance R'. The condenser C' should be resistance free and shielded, the shielded terminal being connected to the ground side. First, the undamped source is tuned to the antenna, and then the L'C' circuit tuned to the source. The resistance R' is then varied until the current is the same in the two positions. The resistance of the L'C' circuit is then equal to Re, the eifective resistance of the aerialground portion of the antenna and L'C' = LeCe. Next, the damped source is tuned to the antenna and the change in current noted when the connection is thrown over to the L'C' circuit. If the current increases, the value of C' is greater than Ce, and vice versa. By varying both L' and C', FIG. 7 - Circuits for determining the effective resistance, inductance, and capacity of an antenna keeping the tuning and R' unchanged, the current can be adjusted to the same value in both C ' Ce = positions. Then, since L'C' = LeCe and , the value of C' gives Ce and that of L' gives Le. L' Le Large changes in the variometer setting may result in appreciable changes in its resistance so that the measurement should be repeated after the approximate values have been found. To eliminate the resistance of the variometer in determining Re, the variometer is shorted and, using undamped oscillations, the resonance current is adjusted to equality in the two positions by varying R'. Then R' = Re. The measurement requires steady sources of feebly damped and strongly damped current. The former is readily obtained by using a vacuum-tube generator. A resonance transformer and magnesium spark gap operating at a low-spark frequency serve very satisfactorily for the latter source, or a single source of which the damping can be varied will suffice. An accuracy of 1 per cent is not difficult to obtain. IV. THE INDUCTANCE COIL The transmission-line theory can also be applied to the treatment of the effects of distributed capacity in inductance coils. In Fig. 8, (a) , is represented a single-layer solenoid connected to a variable condenser C. A and B are the terminals of the coil, D the middle, and the condensers drawn in dotted lines are supposed to represent the capacities between the different parts of the coil. In Fig. 8, (b), the same coil is represented as a line with uniformly distributed inductance and capacity. FIG. 8 - Inductance coil represented as a line with uniform distribution of inductonce and capacity These assumptions are admittedly rough but are somewhat justified by the known similarity of the oscillations in long solenoids to those in a simple antenna. 1. REACTANCE OF THE COIL Using the same notation as before, an expression for the reactance of the coil, regarded from the terminals AB (x = 0) will be determined considering the line as closed at the far end D (x = l). Equations (3) and (4) will again be applied, taking account of the new terminal condition: that is, for x = l; v = 0. Hence A cos C 1 L1 l=−B sin C 1 L1 l and for x = 0 v 0 =A cos t =−B tan C 1 L1 cos t i 0=− C1 B sin t L1 which gives for the reactance of the coil regarded from the terminals A B, X '= L1 tan C 1 L1 l C1 or X '= L0 tan C 0 L0 (23) C0 2. NATURAL FREQUENCIES OF OSCILLATION At low frequencies the reactance of the coil is very small and positive but increases with increasing frequency and becomes infinite when C 0 L0= . This represents the lowest 2 frequency of natural oscillation of the coil when the terminals are open. Above this frequency the reactance is highly negative, approaching zero at the frequency C 0 L0= . In this range of frequencies the coil behaves as a condenser and would require an inductance across the terminals to form a resonant circuit. At the frequency C 0 L0= the coil will oscillate with its terminals short-circuited. As the frequency is still further increased the reactance again becomes increasingly positive. Condenser Across the Terminals. - The natural frequencies of oscillation of the coil when connected to a condenser C are given by the condition that the total reactance of the circuit shall be zero. X' + X0 = 0 From this we have L0 1 tan C 0 L0 = C0 C or cot C 0 L0 C 0 L0 = C (24) C0 This expression is the same as (8) obtained in the case of the loaded antenna, excepting that C C0 L and shows that the frequency is decreased and wave L0 length increased by increasing the capacity across the coil in a manner entirely similar to the decrease in frequency produced by inserting loading coils in the antenna lead-in. occurs on the right-hand side instead of 3. EQUIVALENT CIRCUIT WITH LUMPED CONSTANTS It is of interest to investigate the effective values of inductance and capacity of the coil at very low frequencies. Expanding the tangent in equation (23) into a series we find X '= L0 1 2 C 0 L 0 ...... 3 and neglecting higher-power terms this may be written 3 C0 X '= 3 L 0− C 0 L0 − C0 , which shows that at low 3 C0 frequencies the coil may be regarded as an inductance L0, with a capacity across the terminals 3 and therefore in parallel with the external condenser C. Since at low frequencies the current is uniform throughout the whole coil, it is self-evident that its inductance should be L0. This is the reactance of an inductance L0 in parallel with a capacity Now, the similarity between equations (24) and (8) shows that, just as accurately as in the similar case of the loaded antenna, the frequency of oscillation of a coil with any capacity C across the terminals is given by the formula f= 1 2 L 0 C C0 3 This, however, is also the expression for the frequency of a coil of pure inductance L0 with a C0 capacity across its terminals and which is in parallel with an external capacity C. Therefore, 3 in so far as frequency relations are concerned, an inductance coil with distributed capacity is closely equivalent at any frequency to a pure inductance, equal to the low-frequency inductance (neglecting skin effect), with a constant capacity across its terminals. This is a well-known result of experiment7, at least in the case of single-layer solenoids which, considering the changes in current and voltage distribution in the coil with changing frequency, is not otherwise self-evident. WASHINGTON, March 9, 1918. 7 G.W.O. Howe, Proc. Phys. Soc. London, 21 p. 251, 1912; F.A. Kolster, Proc. Inst. Radio Eng., 1, p. 19, 1913; J.C. Hubbard, Phys. Rev., 9, p. 529, 1917. FURTHER DISCUSSION* ON "ELECTRICAL OSCILLATIONS IN ANTENNAS AND INDUCTION COILS" BY JOHN M. MILLER By John H. Morecroft I was glad to see an article by Dr. Miller on the subject of oscillations in coils and antennas because of my own interest in the subject, and also because of the able manner in which Dr. Miller handles material of this kind. The paper is well worth studying. I was somewhat startled, however, to find out from the author that I was in error in some of the material presented in my paper in the Proceedings of The Institute of Radio Engineers for December, 1917, especially as I had at the time I wrote my paper thought along similar lines as does Dr. Miller in his treatment of the subject: this is shown by my treatment of the antenna resistance. As to what the effective inductance and capacity of an antenna are when it is oscillating in its fundamental mode is, it seems to me, a matter of viewpoint. Dr. Miller concedes that my treatment leads to correct predictions of the behavior of the antenna and I concede the same to him; it is a question, therefore, as to which treatment is the more logical. From the author’s deductions we must conclude that at quarter wave length oscillations L e= L0 (1) 2 and C e= 8 C 0 (2) 2 The value of L really comes from a consideration of the magnetic energy in the antenna keeping the current in the artificial antenna the same as the maximum value it had in the actual antenna, and then selecting the capacity of suitable value to give the artificial antenna the same natural period as the actual antenna. This method of procedure will, as the author states, give an artificial antenna having the·same natural frequency, magnetic energy, and electrostatic energy, as the actual antenna, keeping the current in the artificial antenna the same as the maximum current in the actual antenna. But suppose he had attacked the problem from the viewpoint of electrostatic energy instead of electromagnetic energy, and that he had obtained thc constants of his artificial antenna to satisfy these conditions (which are just as fundamental and reasonable as those he did satisfy); same natural frequency, same magnetic energy, same electrostatic energy and the same voltage across the condenser of the artificial as the maximum voltage in the actual antenna. He would then have obtained the relations L e= and *Received by the Editor, June 26, 1919. 8 L 0 (3) 2 C e= C0 2 (4) Now equations (3) and (4) are just as correct as are (1) and (2) and moreover the artificial antenna built with the constants given in (3) and (4) would duplicate the actual antenna just as well as the one built according to the relations given in (1) and (2). I had these two possibilities in mind when writing in my original article "as the electrostatic energy is a function of the potential curve and the magnetic energy is the same function of the current curve, and both these curves have the same shape, it is logical, and so on." Needless to say, I still consider it logical, and after reading this discussion I am sure Dr. Miller will see my reasons for so thinking. When applying the theory of uniform lines to coils I think a very large error is made at once, which vitiates very largely any conclusions reached. The L and C of the coil, per centimeter length, are by no means uniform, a necessary condition in the theory of uniform lines; in a long solenoid the L per centimeter near the center of the coil is nearly twice as great as the L per centimeter at the ends, a fact which follows from elementary theory, and one which has been verified in our laboratory by measuring the wave length of a high frequency wave traveling along such a solenoid. The wave length is much shorter in the center of the coil than it is near the ends. What the capacity per centimeter of a solenoid is has never been measured, I think, but it is undoubtedly greater in the center of the coil than near the ends. The conclusions he reaches from his equation (22) that even at its natural frequency the L of the coil may be regarded as equal to the low frequency value of L is valuable in so far as it enables one better to predict the behavior of the coil, but it should be kept in mind that really the value of L of the coil, when defined as does the author in the first part of his paper in terms of magnetic energy and maximum current in the coil at the high frequency, is very much less than it is at the low frequency. One point on which I differ very materially with the author is the question of the reactance of a coil and condenser, connected in parallel, and excited by a frequency the same as the natural frequency of the circuit. The author gives the reactance as infinity at this frequency, whereas it is actually zero. When the impressed frequency is slightly higher than resonant frequency there is a high capacitive reaction and at a frequency slightly lower than resonant frequency there is a high inductive reaction, but at the resonant frequency the reactance of the circuit is zero. The resistance of the circuit becomes infinite at this frequency, if the coil and condenser have no resistance, but for any value of coil resistance, the reactance of the combination is zero at resonant frequency.