ionosphere antenna 672180

Upload: ayodele-oyedele

Post on 03-Jun-2018

224 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/12/2019 Ionosphere Antenna 672180

    1/27

    Ac - Al~ -- r-----~--r'- ~ ~ -AA~ -nAAAA A-A ~ A- AA~

    A. 2 K) 0 A -

    /A-..4? A,~ - -

    4 I AC

    *0-A jt Irnpedance~ of LOFTI IRA VeryebowaFwtuenoy

    ti.anrza*TflVSO lru .tflflflflflntA~tJa AK0 - I A IV'~ AC. A - *2

    -- AA (A A- NAVAL E SEPt11 LAIOMIOJIY - A A AA A Wg~ng1XC, A AA A (A A A

    'A-

    -At - 12I4 dcatmentahac* afl~wg4 IarwtQs, nYse. ~ M3~ Ltit4t~4jtdIztZ4A C - 4

  • 8/12/2019 Ionosphere Antenna 672180

    2/27

    CONTENTSAbstract iiiProblem Status iiiAuthorization iiiINTRODUCTION 1

    * ANTENNA INSTRUMENTATION 1ATTITUDE INSTRUMENTATION 1SPACECRAFT ORBIT 2ANTENNA ORIENTATION 2ADMITTANCE OF THE MAGNETIC-DIPOLE ANTENNA 3EFFECT OF ENVIRONMENT ON LOOP-CIRCUITCHARACTERISTICS 3EQUIVALENT IMPEDANCE OF THE ELECTRIC-DIPOLEANTENNA 4DIPOLE-ANTENNA COUPLING 6CALIBRATION OF THE DIPOLE-ANTENNA COUPLING CIRCUITS 6SIGNAL INTERFERENCE DURING ANTENNA ADMITTANCEMEASUREMENTS 8DAYTIME ELECTRIC-DIPOLE ADMITTANCE 8NIGHTTIME ELECTRIC-DIPOLE ADMITTANCE 8EFFECT OF ORIENTATION IN THE GEOMAGNETIC FIELD 10CHANGE OF ADMITTANCE WITH ALTITUDE 11WAVE IMPEDANCE AND E/H RATIO IN SPACE 11COMPARISON OF LOOP AND WHIP-DIPOLE-ANTENNA-SYSTEM SIGNAL OUTPUT 11TYPICAL DAYllVT DAA 12COMPARISON OF DAY AND NIGHT ADMITTANCE OF THEWHIP DIPOLE 13

    F, _

  • 8/12/2019 Ionosphere Antenna 672180

    3/27

    VARIATJON OF VLF DIPOLE ADMITTANCE WITHTHE SPIN AXIS PERPENDICULAR TO TH EGEOMAGNETIC FIELD (ML = 0) 13

    VARIATION OF VLF DIPOLE ADMITTANCE WITH TH ESPIN AXIS PARALLEL TO THE GEOMAGNETICFIELD (MT = 0) 16

    VARIATION OF APPARENT CAPACITANCEWITH ALTITUDE 16CONCLUSIONS 17RECOMMENDATION 19ACKNOWLEDGMENTS 19REFERENCES 20BIBLIOGRAPHY 20

    1*

  • 8/12/2019 Ionosphere Antenna 672180

    4/27

    ABSTRACTThe spacecraft of the LOFTI IIA t r a n s i o n o s p h e r i c very-low-

    frequency (vlf) receiving experiment was fitted with relatively simpleaultmatic instrumentation for periodic indication of vlf antenna admit-tance in the 10 to 18 kHz range. Analysis of part of 'he resulting datahas shown the following:

    1. The admittance of the vlf magnetic dipole (a D-shaped, shieldedloop approximately equivalent in capture area to a 14-in.-diatneter cir-cular coil) was essentially unaffected by the change in environment ofthe spacecraft from the earth s surface to the ionosphere. Variationsof local electron density in the ionosphere and change of antenna ori-entation rolative to the g e o m a g n e t i c field had no discernible effect.

    2. The admittance of the vlf electric dipole (two 20-ft-long op-posed whips) remained capacitive, but the apparent capacitance variedmarkedly as the spacecraft moved along its orbital path. Ac much as10 to 20 times free-space value was indicated at a; udes shown bypublisbed typical data as likely regions of greatest electron density. Athigh ele:tron-density levels, a Iwo-to-one cyclic variation of capaci-tance was evident with change of dipole orientation relative to 'he geo-magnetic field as the spacecraft rotated on its spin axis At altitudesof likely low electron density, variation with spin decreased and thecapacitance approached that expected in free space.

    PROBLEM STATUSThis is a final report on one phase of the problem; work continues

    on other phases.

    AUTHORIZATIONNRL Problem R01-34

    Project RF 006-02-41-4353Manuscript submitted February 16, 1968.;

    ii

  • 8/12/2019 Ionosphere Antenna 672180

    5/27

    IMPEDANCE OF LOFTI IIA VERY-LOW-FREQUENCYANTE1INAS IN THE IONOSPHERE

    INTRODUCTIONPart of the instrumentation of the LOFTI IIA vlf receiving-satellite experiment of

    June and July 1963 was intendsd for determining the effects of the ionosphere on the ad-mrttance of the vlf antenas mounted on the spacecraft. A general description of the ex-periment as a whole and its instrumentation is given by Ref. 1, and much informationdeveloped in the subsequent data processing appears in Ref. 2. This report will treatonly the vif antenna-admittance itstrumentation of the experiment and the most importantcharacteristics obtained from processing the pertinevt portions of the telemetry records.

    ANTENNA INSTRUMENTATIONFigure 1 shows an exterior view of the LOFTI HA spacecraft. The magnetic-dipoleantenna (vlf loop) consisted of a 36-turn coil of 180/36 litzendraht (litz) inside a D-shapedaluminum tube, which served as the antenna's electrostatic shield. This coil was equiva-

    lent in aperture or pickup capability to a 14-in.-diameter circular loop. The electric-dipole antenna (vlf whips) consisted of two thin-wall hollow tabes formed by the curling offlat, prestressed beryllium tape u.,wound in orbit from two separate reels, each to a max-imum length of 20 ft. This report will treat the data obtained with the whips at maximumextension, i.e., the 40-ft-dipole case.The loop and the whip dipole were each provided with its own, separate vlf receiverand admittaice-determination circuitry. The variations in apparent admittance of theloop in orbit could be determined by internal injection of cw calibration input of 10.2 and18.0 kHz frn'quency. Variations of whip-dipole admittance could be determined similarlyby internal cw injection at 16.917* and 18.0 kHz.

    ATTITUDE INSTRUMENTATIONTwo small squibs mounted atop the loop antenna, which were fired after the space-craft was in orbit, initiated spacecraft spin. The spin rate, after full extension of the vifwhips, was about 15 rpm about an axis approximately parallel to the telemetry-dipoleantenna (telemetry whips). The spin helped stabifize the spacecraft in crbit and also ro-tated the vlf antennas, so that the effect of antenna attitude in the earth's magnetic fieldon apparent antenna admittance could be observed.The attitude of the spacecraft with respect to the geomagnet.c field and a line to the

    sun was indicated by onboard magnetic and solar sensors. The two magnetic sensors(magnetometers) were au, )matically extended on a boom after nose-cone separation.They were, thereby, moved away from the spacecraft, to lessen the possibility of mag-netic interference to or from the vlf instrumentation.*To economize spacecraft volume and power, the whip receiver s local os-,llator (LO)

    served as the source of the 16.917-kHz calibration signal.

    Io

  • 8/12/2019 Ionosphere Antenna 672180

    6/27

    2 C. E. YOUNG

    SINTEN Ni

    Figure 2 lists the initial orbital elements of LOFTI IIA and depicts the changing ori-entation of the vif antennas as the spacecraft moved along its crbital path while rotatingaround its spin axis, The lifetime of the spacecraft was about 32 days.ANTENNA ORIENTATION

    Figure 3 depicts the contours of the geomagnetic field in ,cross-section and showsthe variations of the spacecraft's magnetic-sensor indications as they might appear dur-

    ing one orbital period. When the spin axis (essentially the axis of the telemetry antenna,whips 2 and 4) was paralle'l to the local geomagnetic field, the output of the longitudinalsensor ML would be maximum and that of the transverse sensor MT practically zero.When the spin axis was essentially perpendicular to the geomagntic field, the MT varia-tion would be maxiu and tiat oi the IVL ciose to zero. -herefore, whenever the ML indication in the telemetry data was very small or aboutzron its sn assmedthat the axis of the v hip dipole was then spinning from prac-tically parallel to perpendicular to the geomagnetic field. Conversely, when ML was max-imum and MT close to zero, it could be assumed that the axis of the vtl whip dipole wasessentially perpendicular to the field irrespective of spin.ls t eo

  • 8/12/2019 Ionosphere Antenna 672180

    7/27

    NRL REPORT 6712 3N

    4

    EOUATOR~

    4\ 4..-MAGNETOMETERSWHIPS 1 8 3 - VLF DIPOLE SWHIPS 2 8 4 - TELEMETRY DIPOLE

    ANTENNA

    Fig. 2 - Variation of LOFTI IIA antenna orientation along the orbital path. Theinitial orbital elements of LOFTI IIA were an inclination of 70 degrees, an apogeeof 925 km a perigee of 170 kin, and a period of 96 min.

    ADMITTANCE OF THE MAGNETIC-DIPOLE ANTENNAThe loop inductance (943ju H) was part of a two-pole four-element network (Fig. 4)simultaneously resonant at 10.2 and 18.0 kHz. In-orbit loop admittance at these two fre-quencies was determined by observation of the two receiver outputs during periodic inter-vals of 10.2 or 18.0 kHz cw calibration input into the antenna circuit.

    EFFECT OF ENVIRONMENT ON LOOP-CIRCUITCHARACTERISTICSA loop winding and a conductive (electrostatic) shield can be considered, in effect,as the secondary and primary, respectively, of a tightly coupled transformer. The sec-ondary is usually tuned to resonance at a desired frequency by a capacitor of appropriatevalue. The shield (primary) cannot short-circuit the secondary despite the close mag-netic coupling between them, because of an insulated gap. If the loop were immersed in

    a conductive medium, such as the ionosphere, this gap would be shunted by the ionizedplasma. Since the effective admittance of such a medium can be much l--rger than t. t offree space, both the resonant frequency and the Q of the tuned secondary could be ex-pected to change to a degree dependent on the characteristics of the plasma in theparticular region.

    The effect of shield-gap shunting was experimentally determined in the laboratorybefore the LOFTI IIA spacecraft was launched. Capacitors and resistors of a wide rangeof values were connected in turn across the gap in the shield. As shown by Fig. 5, a shuntsusceptance of 0.5 mho (2-ohm capacitive reactance) or a conductance of 0.2 mho (5-ohmresistance) decreased the amplitude of loop-circuit response (i.e., the voltage appearil.at the loop-coupling-circuit output terminals with calibration input) by 10( .

  • 8/12/2019 Ionosphere Antenna 672180

    8/27

    4 C.E. YOUNG

    - ~~~ -t77'-~-

    4 Jb~2. _j.~/- /~~

    A v~-:

    A. 4i- , .i

    'ML(+~' 4 : ~ ~L(~4' '

    -O E K.N' -- '

    Fig 3x"aito:f'h antmte uptwt tesaerf

    sured on thegrondthrohot the lif toneimf uwthe spacecraft

    EQUIVALENT IMPEDANCV, OF THE ELECTRIC-DIPOLEANTENFAj Figure hows the free-space-series impedance components of an idealized 40-ftelectric dipole at 18.0 kHz, calculated using formulas by Schelkunoff (3). The capacitivereactance X, (approximately 300,000 ohms) is greater than the other conipontmts of the

  • 8/12/2019 Ionosphere Antenna 672180

    9/27

    w 100-

    [ I- -- . . . . I-_ .A

    I 0 I .0 W t. O LLATO $ -

    0 1Uia--

    LOOPSHIELDI IN-SU D GAP . .-- 36 TURNS OF -AT 180 STRAND/, ,e

    < LITZ WIRE

    I *

    0.11 10 12 14 -6s 20 22 24 264/FRLOUENCY ELOFig. 4 - Frequency response of the loop system to constant -intensi~y

    radi-o field (terrestrial measurement )

    7_ I I

    2:SHUNT R ONL

    40CONOUCTA N ,W O P

    O S HU NTS I r~--5OG. MO$ (LY-J i0 o i SZ )' I I i i i , , , ,20 II , ,, _I1 11iIII001 01 I0 20

    ADMITTANCEF R OR C. MHOSSHUNTEDACROSS LOOP .ELD GAPFig. 5 - Effect of a shunt across the loop-shield gap

  • 8/12/2019 Ionosphere Antenna 672180

    10/27

    6 C. E. YOUNGto---0 ,, -1 1- to FtT -4-.

    2LV4Cr jI (I 1 22WC L .,dA L-1, ) A 4

    Fig. 6 - Computed free-space impedancecomponents of a 40-ft dipole

    impedance by six orderi. a: more. It therefore appears that around 18.0 kHz, and in freespace, such an antenna would appear to be essentially a low-loss condenser of about 27 pF.However, the actual antenna, consisting of the two diametrically opposed 20-ft whipsextending from the spacecraft, would not be electrically as simple as the idealized 40-ftdipole of Fig. 6. The approximately 2-ft-diameter body of the spacer raft separated the

    whips physically, interposing a curved common-ground plane of considerable area be-tween their nearer ends. In effect, the spacecraft carried two 20-ft-long opposed mono-poles that served as a dipole which would probably have somewhat less than 27-pF c-pac-itance in free space (4).

    DIPOLE-ANTENNA COUPLINGEach vlf whip was provided with a coupling network located inside the hull of thespacecraft (Fig. 7). in designing these networks, the likely minimum value of the mono-

    pole, or half-dipole, capacity, CA, in orbit was assumed to be 75 pF, and the likely max-imum was assumed t be 3000 pF. The coupling networks were identical and designed toresonate at 18.0 kHz, when CA was 75 pF, and at 16.917 kHz, when CA was ?%C^VF.While the spacecraft was in orbit, the whip-dipole-antenna-admittance determinationswere made periodically at 16.917 and 18.0 kHz. The response of the whip-dipole network

    to 16.917-kHz cw excitation was indicated by the rectified radio frequency (rf) output, i.e.,the direct current change of a detector. The response to the 18.0-kHz excitation was in-dicated by the 25-Hz i-f output of the receiver.

    CALIBRATION OF THE DIPOLE-ANTENNACOUPLING CIRCUITS

    Figure 8 shows the voltage output of the whip-dipole network with simultaneous shunt-ing of the whip input terminals of the coupling networks by either identical external ca-pacitors CA or resistors R A. The whips themselves were not connected in circuit. Thedeterminations were made in the laboratory prior to spacecraft launch. CA ranged invalue from 0 to 10,000 pF, and RA ranged from 10 megohms to 1000 ohms.

  • 8/12/2019 Ionosphere Antenna 672180

    11/27

    NRL REPORT 6712 7

    CAZ WHIP CAPACITANCECS SLF- CAPACITANCEFI H INDUCTORPASSBANO OF lb kHz RECEIVERSPACC~tFT__---- ZdC TO OVER 100 k34z

    CCVPLER - I COUPLER - 2

    I 33pF 9 FT4T WHIPWHIP 4OPF CS CS 4OpF

    -- I-p 318 O~ z SHUNT TT OALE H UN CAPACITANCE CA TTCA If Op

    10 OO

    SPARES

    INPUT CTNNECTORI is0 kHz RECEIVER KETROOTNE OSCIL.LATOR I6 91? &HZNPJT.AND OU PJ TO 189AC WIHIPANDRECEIVER INPUT CIRCUIT

    Fig. 7 -Vif whip-di pole -antenna coupling network

    18 0 kHz SHUNT ADMITTANCE OF CA or RA (AMOS)10 t0 too 11'

    120M6.917 kHz0A RESPONSE

    -(CIFE R-F)00 -EACH WHIP IETFE F180 kHz SHUNTED BYH , RESPONSE CA' ONLYso0 _ CA ONLY

    034~R --- =-- \4 ,ONL%\ /

    6 r U RA ONLYI- N,,,, /

    ,40. I

    - I Ii IRA 30 0 300 300 10 3 j

    CA SCALE (pF) AND RA SCALE (k)

    Fig. 8 - Effect of shut ing vlf whip terminals v.ithcapacitance or resistance

  • 8/12/2019 Ionosphere Antenna 672180

    12/27

    8 C. E. YOUNGThe solid and dashed lines represent the effect of capacitance and resistance shunt-

    ing, respectively, on antenna-network response (output voltage). It can be seen that boththe total change and the rate of change of output (measuremenreqluin) were largerwith capacitive shunting. The values at the inflection points ((,(2),(,) )are of particularinterest; these occur only with capacitive shunting. Resistive shunting (over the four-order range of the laboratory calibration) caused only a gradual decrease of response,with no sharply defined inflections. Occurrence of readings at point ,(9), or ( )in thein-orbit data would therefore, tend to indicate that the dipole admittance was then es-sentially capacitive.SIGNAL INTERFERENCE DURING ANTENNA

    ADMITTANCE MT ASUREMENTSSearch of the LCFTI data records revealed that the 18.0-kHz admittance data for thevlf whip dipole in the nighttime ionosphere were almost always obscured by relativelyvery large 18.0-kHz vif signals and noise from terrestrial sources. However, the 16.917-kHz data were not as seriously affected, probably because of larger injection voltage andlower sensitivity at that frequency.

    DAYTIME ELECTRIC-DIPOLE ADMITTANCEWhip-dipole calibration data at 18.0 kHz obtained during daylight pases were not socompletc-ly obscured by external signals. The additional attenuation of tht ionized Dlayer in the sunlit part of the atmosphere was apparently sufficient to sub ;tantially de -

    crease terrestrial vlf signal-and-noise intensity at the spacecraft. The riinimum 18.0-kHz response (17C, region @ )could be seen occasionally in the data, indicating thatdipole admittance was predominantly capacitive. If the plasma that surrounded the vl fwhips had had appreciable resistive shunting effect, the response observed would likelyhave been in the 80 to 40% range. At no time was the 18.0-kHz response in the region (-)range when the spacecraft was in daylight. The 16.917-kHz calibration response occa-sionally approached regionvalues (where the whip-dipole network would resonate withCA = 3000 pF) but was usually below this value except when the spacecraft approached theaurora regions at the higher latitudes.

    NIGHTTIME ELECTRIC-DIPOLE ADMITTANCEFigure 9 shows typical data from the records of a nighttime pass of the spacecraft

    over Panama at 217 to 236 km altitude. The five time frames shown are within a 3-minperiod at about 1 a.m. local mean ,.-e. Magnetic-sensor output is shown in the upperhalf of each frame. The changing value of apparent antenna capacitance, indicated by the16.917-kHz response of the whip-dipole network, is marked in the lower half of each frame.

    In frame 01] of the sequence, the 16.917-kHz-output trace indicates a steady low valueof admittanze (response about 15%). Radio station NBA's 18.0-kHz signals and some vlfnoise can also be seen. The antenna admittance appears to be essentially constant, eventhough the magnetometer-output traces indicate continual spin of the spacecraft. The lowand invariant response, regardless of dipole orientation, may be interpreted as evidenceof a relatively low level of electron density in that part of the ionosphere.The dipole capacitance appears to be approximately 38 pF CA = 75 pF), which is about40% greater than the calculated capacitance of the 40-ft dipole in free space. The observed

  • 8/12/2019 Ionosphere Antenna 672180

    13/27

    NRL REPORT 6712

    ifi

    ~~I' IK'

    0 C

    7

    -~.

  • 8/12/2019 Ionosphere Antenna 672180

    14/27

    10 C.E. YOUNG

    response is near the minimum obtained in the 16.917-kHz laboratory calibration (Fig. 8).*Similar low readings,indicating apparent capacitance tobe not much greater than free-space value, occur frequently in the night data of the experiment whenever the spacecraftaltitude was in the vicinity of 200 km.In frame M of Fig. 9 (approximately 1 min later than frame 01 ), the response hasincreased to about 18'1 but still shows no apparent indication of dipole admittance vari-ation due to changing orientation in the geomagnetic field. This reading represents acapacitance value somewhat greater than 100 pF.In frame [] , about 1.5 min after the start of the sequence, the response has furtherincreased and is beginning to vary at twice the spin rate. The appearance of this variation,

    the larger capacitance indicated by the higher response level, and the drop in ML lead tothe conclusion that the spacecraft has moved to a region of appreciably higher electrondensity and that its oricatation relative to the geomagnetic field has changed. The mini-mum capacitance in the frame 2 interval is apparently near 155 pF, and the maximum isabout 240 pF.

    EFFECT OF ORIENTATION IN THE GEOMAGNETIC FIELDAll the data examined (day or night passes) show little effect of spacecraft spin onwhip-dipole admittance when the apparent dipole capacitance approaches the free-space

    value. In frames r and C5 of Fig. 9, where the apparent capacitance approaches 10 timesthis value, the effeL. of change of antenna orientation relative to the geomagnetic field canbe seen very clearly.From study of the bibliography, it appears that the theoretical predictions of dipole

    admittance in a magneto-ionic medium do not completely agree. Most of the sources con-clude that vlf dipole admittance should be maximum when the dipole axis is pfrpendicularto the geomagnetic field and minimum when parallel to the field, with a variation overseveral orders of magnitude, if direct contact to the medium is achieved, i.e., no plasmasheath. However, the experiment did not verify this prediction.In frames [], I , and [], the apparent capacitance of the antenna is not quite maxi-

    mum when the MT magnetometer reading reverses polarity (zero crossing) and is notquite minimum when the magnetometer reading is maximum. The time displacementbetween the magnetometer-output maximum and apparent capacitance minimum, statedin terms of relative phase, is about 25 degrees in these three frames. This effect may beascribed to a number of possible causes, for instance, in some part to a plasma sheathsuch as forms around moving objects in the ionosphere (5). This sheath may be modifiedcyclicly to some degree as the potentials induced in the antennas and hull change withspacecraft spin and translation through the geomagnetic field.

    The ratio of maximum-to-minimum capacitance appeared to be dependent uponcieLtron-density and geomagnetic-field intensity but did not exceed a ratio of two-to-onein either day or night data. The variation in amplitude of the alternate capacitance max-ima, shown in frames g] and j], was only evident in the nighttime data.*Because of telemetry distortion and instabilities, the data resolution of the experimentwas about Z%, which would translate into a possible capacitance range of 57 to 93 pF

    for a 15o response reading.

    ii

  • 8/12/2019 Ionosphere Antenna 672180

    15/27

    NRL REPORT 6712CHANGE OF ADMITTANCE WITH ALTITUDE

    During the applroximately 3-min sequence of Fig. 9, the spacecraft traversed a dis-tance of about 750 naut mi, and its altitude increased from 217 to 236 kin. As alreadymentioned, the large increase in apparent capacitance of the whip dipole indicates thatthe spacecraft moved, in this short time, from an environment of low electron densityinto one of relatively high electron density.Local electron concentration was not specifically determined in the LOFTI experi-ments. Typical electron-density profiles of the nighttime ionosphere show that electronconcentration increases rapidly with increase in altitude in the 215 to 240 km region. Forinstance,Ref. 6 indicates that an increase from about 103 electrons/cm3 near 200 km toa maximum of about 2(105) electrons/cm3 near 300 km might be expected. (Such a change

    is shown by the night electron density profile in a later figure.)WAVE IMPEDANCE AND E/H RATIO IN SPACE

    In a free-space environment, the field intensity of the magnetic component of a radiowave (as indicated in this case by loop signal output) is rclated to the correspondingelectric-field intensity by the wave impedance of free space (Z0 = E/H = 120W ohms); i.e.,E = ZOH =12077 H. where E is expressed in volts per meter (V/m) and H in amperes permeter (Am). If both the loop and the whip-dipole circuits are tuned to the same signalfrequency and are immersed in a medium which approximates free space, the loop-circuit-signal output voltage should equal the whip-dipole-circuit-signal output voltage,if the two antenna systems have about the same effective output impedance and the sameeffective height or aperture. If this is not the case, the two output voltages may be nor-malized or suitably equated for comparison purposes.COMPARISON OF LOOP AND WHIP-DIPOLE-ANTENNA-SYSTEM SIGNAL OUTPUT

    Although, as previously mentioned, strong terrestrial signals and noise interferedwith nighttime 18.0-kHz whip-dipole-admittance determination, these signals were usefulfor intercomparison of loop and whip-dipole performance as signal collectors in theionosphere.

    Figure 10 compares the apparent field intensity of the 18.0-kHz signals as indicatedsimultaneously by loop and whip-dipole receiver output during the time frames shownin Fig. 9. The sensitivity of the loop system overall had previously been determined inthe laboratory in terms of the microvolts per meter field intensity at the loop whichwould produce a given signal output. Because of measurement difficulty, the sensitivityof the whip-dipole system overall was computed. The apparent field intensity at the loopis stated in Fig. 10 in terms of the electric component of a free-space field, to allowdirect comparison with the whip-dipole data.In frame [] , the apparent magnetic and electric components in microvolts/ meterdiffer by only about 0.5 db. The spacecraft was moving at that time through a region inwhich the increase of vlf whip-antenna admittance over free-space value was relatively

    minor. From frame Mi to frame (a) , the loop signal increased 4 db, while the whip-dipole signal decreased 12 db. Laboratory calibrations of whip-dipole-receiver sensi-tivity to change in antenna capacitance CA, using a constant voltage signal applied inseries with CA, has indicated a maximum reduction of only 1 db in coupling-networksignal out ut with capacitance change from 75 to 103 pF, as occurred between frames[] and W of Fig. 9. Hence, the 15-db difference between loop and dipole signals inframe j of Fig. 10 can be taken as representing a drop in electric-field intensity.

    Lo

  • 8/12/2019 Ionosphere Antenna 672180

    16/27

    12 C.E. YOUNG

    NBA C4OPIGNAL.33 5dt ABOAI ,v '~4 $~TIME-S CONDS

    NBA GROUND SIGNAL E - . . R

    * NBADIPOL.ESIGNAL 34dto ABOVE ptv/m 0~ I ~ d J 14 2 b . .. .

    START 4 I MINUTE - TIME

    3 44 1-46 5 -9 5 -415d0

    - I~b - d ' A -d - -

    15db -3db 8...5db - -+ 1 MINUTES ' + 2MINUTES - - + 7 MINUTES - TIME

    Fig. 10 - Field intensity in 3-min period covered by Fig. 9, as indicated by the loop andwhip-dipole receiver output (the loop system calibrated on earth in microvolts per meter)As previously mentioned, loop-circuit tuning appeared to be substantially unaffected

    throughout the LOFTI experiment by changes in local electron concentration, therefore,the 4-db increase in loop output must represent about that much increase in intensity ofthe magnetic component of the radio field. On this basis, it may be concluded that theratio of E/H, and consequently, the wave impedance of the medium in which the two an-tennas were moving, had decreased substantially in the time period between frames rnand [] , with still greater decrease in the following time frames ( , , and .

    The salient information derived from the data of Figs. 9 and 10 is ummarized inFig. 11. A nighttime electron density profile for the altitude range 212 to 240 km asedon information in Ref. 6, is superimposed on the antenna-capacitance graph. The varia-tions in relative whip-dipole and loup signal output shown by Fig. 11 are typical. Similarvariations were observed throughout the data examined.TYPICAL DAYTIME DATA

    Figure 12 shows a sequence of whip-dipole calibratione made simultaneously at 18.0and 16.917 kHz. The spacecraft was at about a 500-km altitude, in daylight. The Da,,er of the -onos.phere, l.hch lenu then be about 450 kin below the spaec raft, proh.ihbyserved as the extra shield which allowed observation of low-level response in the cali-bration interval without excessive terrestrial vlf signal and noise interference.

    The top trace in each of the three time frames shows the 18.0-kHz response (i-f).The actual 18.0-kHz dipole-admittance-determination interval (about 1.5 sec immedi-ately follows the voltage-calibration-reference interval marked 100'-7. The 18.0-kHzresponse observed in time frame M (17 - of the reference (100 ,) response) could occuronly with essentially nonresistive (capacity) shunting of the whip dipole. The simultaneous16.917-kHz response is shown immediately below the 18.0-kH.. data. The apparent capac-itance values at the two frequencies, as derived from the Fig. 8 calibration, agree fairlywell, considering that there was serious noise interference during the 18.0-kHz in-orbitcalibration interval.

  • 8/12/2019 Ionosphere Antenna 672180

    17/27

    NRL REPORT 6712 13

    50 z/Sl , LE A . ,30dB

    .0>...-so

    CID 20 -210 1

    SPb-rnTALIUE kn

    06CA MAXCIMUM 0

    600 -..L400 - - ZIAL..NIU;i I ~200 III

    212 216 220 224 228 232 236SPACECR'IFT LTITUDE kin)

    Fig. 11 - Apparent 18-kHz field intensity and the whip-dipolecapacitance during period covered by Fig. 9 (3 min of night-time pass 473P).

    COMPARISON OF DAY AND NIGHT ADMITTANCE OFTHE WHIP DIPOLEFigure 13 presents a comparison of 16.917-kHz day and night data at essentially thesame altitude (-375-kin) and for the condition of the dipole spinning from parallel to per-pendicular to the geomagnetic field (ML 0). This is the orientation relative to the fieldfor which dipole-,,dmittance change should theoretically be maximum. The maximumariation observed under this condition in the entire experiment was found to be lesstl..1 2 to 1. In the upper half of the figure (day), calibration response is near the Fig. 8r-egion @ value. The apparent antenna capacitance ranges from about 600 to about930 pF.In the calibration period marked I sec in Fig. 13, the inboard end of each of the twowhips was shunted to the spacecraft hull by a 10,000-pF capacitor. If the admittance ofThe~ ~ ~ ~ ~ ~~~~pangatalobeediFi.1istpcapaTisoresult adt smallthe antenna without this shunt was in fact capacitive, the effective capacitance in the an -tenna circuit during the 4-sec shunt interval would be 10,600 to 10,930 pF, and the16.,1 7-kI-Iz response during this time should be about 90' relative to the referencevalue. The reading actually observed in Fig. 13 is typical. Ths result and the smallsuperimposed cyclic variation on it, which is synchronous with the following unshuntedinterval, affords further confirmation of the original assumption that the antenna is, ineffect, a condenser with negligibly low loss.

    VARIATION OF VLF DIPOLE ADMITTANCE WITHTHE SPIN AXIS PERPENDICULAR TO TH EGEOMAGNETIC FIELD (ML = 0)During both the day and the night period of the data of Fig. 13, the spin axis of thespacecraft was approximately perpendicular to the geomagnetic field. This orientationis indicated by the very small output of the ML magnetic sensor in the two time frames

  • 8/12/2019 Ionosphere Antenna 672180

    18/27

    7.14 C. E. YOUNG

    IREF. I CAL10 . P40- 1Wq- 18 0 kH z

    ALT. -538 km (.-700pF)LONG. -77.00 START -e.j36-LAT. 5.50'9p 50F

    16.9;7 kHz

    R-REFERENCE-b10TIMZ

    I REF. I CAL.~.W 100%18.0 kHz

    ALT. -524kmd - J ~ 17 % ('300pF)(NULL)LONG. -75.7 +15MNTS-LAT. 1O.716.917 kHz

    45 60%

    I REF. ICAL.Ioil _0, 18.0 kHz

    I ALT. -472km 0 0 pF~ NOw1IS. 1Y~ .rLONG. -71.8 -30.70pFIOISY)i-0- REFERENCE --b-j

    TIME- SECONDSFig. 12 -Comparison of simultaneous 18.0 and 16.917 kHz whip-dipole ca'ibrations(18-kHz trace: IF output of receiver and 16.917-kHz trace: rectified output)

  • 8/12/2019 Ionosphere Antenna 672180

    19/27

    NRL REPORT 6712 15

    IS ts

    a to

    110'C Cc-

    -1 14

    * 4. 'I %;41I

    .3 - A. i

    4~ntC-

  • 8/12/2019 Ionosphere Antenna 672180

    20/27

  • 8/12/2019 Ionosphere Antenna 672180

    21/27

    i

    NRL REPORT 6712 17

    o.70 -uOUTPUT OCS SORS

    PASS (3/SLOMCL TIME = /oam

    16.917/rIz LOMV. -59./oC, LIBATION 1AT. -36.0

    I I I I l I . , I I I I I I I [r/IE - ECWODS

    r} _ -- - - - . -

    PASS 38/1/AZT. 230 kmLOCAL T/,qE--JecrriLOw&. - 150.50L4T. 18.3

    _ _ _ _ ___ _ _ _ _ __ _ _ _ _ ./d7, AI,,,: _( 8-F )_____ -.RA- ' qE C--

    Fig. 14 - Comparison of 16.917-kHz calibration data, day and night,with the spin axisnearly parallel to the geomagnetic field. (Note: The ripple in the daytime data is dueto the tape recorder.)

    It is evident tnat electron density and apparent capacitance are correlated. Th elarge spread in capacitance value between max. max. and min. min. at any particularaltitude on the graph is indicative of the wide variation in ionization encountered by thespacecraft as it moved along its orbital path.

    Figure 16 is a similar graph for several hundred night passes. Here also the max. max. boundary appears to be quite closely related to the typical elect ron-density profile.'ONCLUSIONS

    From the data presented above, it can be concluded that1. The admittance of a magnetic-dipole (loop) antenna at vlf-band frequenciesis not substantially affected by change of antenna location from the terrestrial surfaceto an ionospheric environment.

  • 8/12/2019 Ionosphere Antenna 672180

    22/27

    18 G. E. YOUNG

    ELECTRONS PER CUBIC CENTIMETER103 104 lobt

    GO -

    SW* - MIN. MIN. .. ... CAPA.CITANCE -M..AX. MAX.'400 - - - __

    itLoZO0-., 0

    O 35 too 200 300 400 500 600AVEfAM.'OVALUE OF DIPOLE CAPACITACE (;F I

    Fig. 15 -Summary of daytime vif -whip-.dipolecapacitance data

    ELECTRONS PER CUBIC CENTIMETER103 104 109go

    Soo - 400__ 400r/A20W1 1 T

    200

    0 35 100 200 300 400 500 60AVERAF.O VALUE0 Or *OJ.E ( APACITAMCE pFj

    Fig. 16 -Summary of nighttime v'lf w.hip-dipolectpacitance data

  • 8/12/2019 Ionosphere Antenna 672180

    23/27

    NRL REPORT 6712 192. The admittance of an electrically small electric (whip-dipole) antenna oper-ating in the vicinity of 18.0 kHz in the ionosphere is predominantly capacitive.3. The apparent capacitance of the LOFTI IIA electric-dipole antenna increaseswith increase of electron density (concentration) in the environment, to as much as 10 or20 times its free-space capacitance.4. The orientation of an electric dipole relative to the earth's magnetic fieldaffects its apparent capacitance substantially when the electron density exceeds someminimum value. The maxima and minima of apparent capacitance are generally dis-placed from exact perpendicular and parallel dipole orientations, respective)y, in thegeomagnetic field, perhaps as a result of plasma-sheath interaction. The variation withchange in orientation is not more than two-to-one.

    RECOMMENDATIONThe results reported here in brief are indicative of antenna characteristics in theionosphere and should not be considered comprehensive. The antenna calibration instru-mentation in the spacecraft was necessarily rudimentary, a factor which has made dataanalysis difficult. In view of the intended use of the derived information for design of avlf transmitting experiment, the effects of geographical latitude and various other param-eters have not been isolated. It is therefore recommended that spacecraft n future vlfsatellite experiments be provided with instrumentation for direct, accurate, and (in sofar as possible) continuous in-orbit determination of all components of vlf antenna im-pedance (reactance, ohmic resistance, radiation resistance, etc.).

    ACKNOWLEDGMENTSThe writer conveys special thanks to Messrs. E.E. Koher and A.E. Showalter fortheir direct aid and diligent contribution in this study. Particular appreciation is ex-pressed to Mr. J.P. Leiphart for his guidance, encouragement, and many helpful sug-

    gestions and to Messrs. E. Toth and R.W. Zeek for their invaluable report reviews andsuggestions.

  • 8/12/2019 Ionosphere Antenna 672180

    24/27

    20 C.E. OUNG

    REFERENCESI. Zeek, R.W., "Penetration of the Ionosphere by VLF Radio Waves:Reception of 10.2and 18.0 kc/s Signals by the LOFTI IIA Satellite," NRL Report 6252 (ConfidentialReport, Unclassified Title), June 19652. Bearce, L.S., Cushing, R.E., Kohler, E.E., Leiphart, J.P., ioung, C.E., and Zeek, R.W.,

    "Atlas of LOFTI IIA Satellite Orbit Maps and Quick-Look Data," NRL Report 6455,Oct. 1966.

    3. Schelkunoff, S.A., "Electromagnetic Fields," New York:Blaisdell, pp. 195-200, 19634. Williams, R.H., and Wang, T.N.C., "Linear Antenna Symmetrically Driven with

    Respect to a Spherical Satellite," Report dated July 1965, on work done under ContractNonr-2798(01)5. Zachary, W.W., 'The Distribution of Particles Around Vehicles Moving Through the

    Ionosphere," Scientific Report NAS 585-3, Dec. 15, 1961, prepared for NASA GoddardSpace Flight Center, Greenbelt, Maryland, by Electromagnetic Research Corporation,5001 College Avenue, College Park, Maryland

    6. Johnson, F.So, ed., "Satellite Environment Handbook," 2nd ed., Stanford:StanfordUniversity Press, 1965

    BIBLIOGRAPHYThe following is a chronological list of the more important sources of pertinent

    information other than the references previously listed:Kogetnik, H., "On Electromagnetic Radiation in Magneto-Ionic Media," Journal Res.

    NBS 64D:515-523 (1960)Katzin, J.C., and Katzin, M., "The Impedance of a Cylindrical Dipole in a Homv,geneous

    Anisotropic Ionosphere," Electromagnetic Research Corporation Report NAS 585-2,Sept. 26, 1961

    Kononov, B.P., Rukhadze, A.A., and Solodukhov, G.V., "Electric Field of a Radiator in aPlasma in an External Magnetic Field," Soviet Phys.-Techn. Phyb. 6:405-510 (1961)

    Mittra, R., and Deschamps, G.A., 'Field Solution for a Dipole in an Anisotropic Medium,"in 'Electromagnetic Theory and Antennas." Proceedings of a Svmnosium held atCopenhagen, Denmark, June 1962, E.C. Jordan ed., New York:Pergamon, pp. 495-512,1963

    L iser, T.R., 'The Admittance of an Electric Dipole in a Magneto-Ionic Environment,"n"anetary Space Sci. 9:639-657 (1962)

    Bramley, E.N., "The Impedance of a Short Cylindrical Dipole in the Ionosphere,"Planetary Space Sci. 9:445-454 (1962)

    i

  • 8/12/2019 Ionosphere Antenna 672180

    25/27

    NRL REPORT 6712 21Whale, H.A., "The Impedance of an Electrically Short Antenna in the Ionosphere," in Proceedings of the International Conference on the Ionosphere, held at ImperialCollege, London, July 1962; Institute of Physics and the Physical Society, London,

    pp. 472-477, 1963; Goddard Space Flight Center Report X-615-62-88, July 1962; alsoNASA TN D-1546, Jan. 1963Mlodnosky, R.F., and Garriott, O.K., The VLF Admittance of a Dipole in the LowerIonosphere," Proceedings, International Conference on "The Ionosphere," Institute ofPhysics and Physical Society, London, England, pp. 484-491, 1963Storey, L.R.O., "The Design of an Electric Dipole Antenna for VLF Reception within theIonosphere," CENTRE NATIONAL D'ETUDES des Tdldcommunications TechnicalReport 308TC, 1964Brandstatter, J., and Penico, A.J., "The Calculation of the Impedance of a CylindricalAntenna in an Anisotropic Plasma," Stanford Research Institute, Menlo Park,California, 1964, Final report by these authors has title: A Study of the Impedanceof a Cylindrical Dipole in an Anisotropic Plasma, Nov. 1964Ament, W.S., Katzin, J.C., Katzin, M., and Koo, B.Y.-C., "Impedance of a CylindricalDipole having a Sinusoidal Current Distribution in a Homogeneous AnisotropicIonosphere," Radio Sci. 68D:379-405 (1964)Weil, H., and Walsh, D., "Radiation Resistance of an Electric Dipole in a MagnetoionicMedium," IEEE Trans. on Antennas and Propagation AP-12:297-304 (1964)Blair, W.E., "The Driving-Point Impedance of an Electrically Short Cylindrical Antennain the Ionosphere," Electrical Engineering Department, University of New Mexico,

    Albuquerque, New Mexico, Report EE 109, June 1964Bolmain, D.G., "The Impedance of a Short Dipole Antenna in a Magnetoplasma,"

    Department of Electrical Engineering, Engineering Experiment Station, Universityof Illinois, Urbana, Illinois, issued under NASA Grant Ns G511, July 1, 1964

    Staras. H., The Impedance of an Electric Dipole in a Magneto-Ionic Medium," IEEE"Trans.on Antennas and Propagation AP-12:695-702 (1964)Faust, W.R., aElectrodynamics in a Magneto-Ionic Environment,' NRL Report 6163,Nov. 1964Cook, K.R., Johnson, G.L., and Edgar, B.C., "Current Distrilutions for a Cylindrical

    Dipole in an Homogeneous Anisotropic Ionosphere," School of Electrical Engineering,Oklahoma State University Progress Report 2, Feb. 1, 1964-Jan. 1, 1965Faust, W.R., "Effective Lengths of Antennas in Magneto-lonic Media," NRL Report 6190,

    Feb. 1965Ament, W.S., Katzin, M., McLaughlin, J.R., and Zachary, W.W., "Satellite Antenna

    Radiation Pioperti~s at VLF in the Ionosphere," Electromagnetic ResearchCorporation Final Report ONR-4250-1, April 30, 1965

  • 8/12/2019 Ionosphere Antenna 672180

    26/27

    St4 utl|m L. |aS I.t( JiionDOCUMENT CONTROL DATA. R & D

    (f -, , fl -J f ,,.. , ndex.. ,Inn)ttg l nut be entered when the overal r.po* 1. . MtiedlI' t A ;. AC I -poltaeaiuthI, i,,, e. REPORT SECURITY CLASSIFICATIONNaval Research LaboratoryL UnclassifiedWashington, D. C. 20390 2b GROUP

    IMPEDANCE OF LOFTI IIA VEPY-LOW-FREQUENCY ANTENNASIN THE IONOSPHERE

    4 OLSCRIPTI.VE NOTESf7 p eolreport and inclusivedates)A finai report on one phase of the problem; work is continuing.

    AU THORISj Postul name, middle initial, last name)

    C.E. Young6 REPORT DATE [.. tOTAL NO OF PAGES Tb. NO OF REFS

    June 18, 1968 27 64B CONTRACT OR GRANT NO 8C.OR GINATOR'S REPORT NUMIERIS

    NRL Problem ROI-34b PROJECT NO NRL Report 6712RF 006-02--41-4353

    R 0 h OTHER REPORT NO(S) (Any other numbete thet may be assigeedthis r portd.

    1O OISTRIBUTION STATEMENTThis document has been approved for public release and sale; its distributionis unlimited.

    II SUPPLEMENTARY NOTES I2 SPONSORING MILITARY ACTIVITYDepartment of the Navy (Office of NavalResearch), Washington, D.C. 20360

    1) ABSTRACT

    The spacecraft of the LOFTI IHA transionospheric very-low-frequency (vlif) re -ceiving experiment was fitted with relatively simple automatic instrumentation forperiodic indication of vlf antenna admittance in the 10 to 18 kHz range. Analysis ofpart of the resulting data has shown the following:

    1. The admittance of the vlf magnetic dipole (a D-shaped, shielded loop approxi-mately equivalent in capture area to a 14-in.-diameter circular coil) was essentiallyunaffected by the change in environment of the spacecraft from the earth s surface tothe ionosphere. Variations of local electron density in the ionosphere and change ofantenna orientation relative to the geomagnetic field had no discernible effect.

    2. The admittance of the vlf electric dipole (two 20-ft-long opposed whips, re -mained capacitive, but the apparent capacitance varied markedly as the spacecraftmoved along its orbital path. As much as 10 to 20 times free-space value was indi-cated at altitudes shown by published typical data as likely regions of greatest elec-tron density. At high electron-density levels, a two-to-one cyclic variation of capa-citance was evident with change of dipole orientation relative to the geomagnetic fieldas the spacecraft rotated on its spin axis. At altitudes of likely low electron density,variation with spin decreased and the capacitance approached that expected in freespace.

    DD mov.1473 23S/N 0O0I.807.-5801 Security Classificaion

  • 8/12/2019 Ionosphere Antenna 672180

    27/27

    Security Classification14 LINK A 1 LINK B INK CKErY WONOSROLE WT ROLE WY R O L E WY

    Impedance/admittanceAntennasIonosphereVery-low-frequency antennasSpacecraftElectric dipoleMagnetic dipole

    DD FORM..1473 (BACKPAGE 2) Security Clasfi(ati-rn