AD-A138028 SCATTERINGEXPERMNTSAT THE IPSWICHEECROMAGNETI C /PEASUREME TS FACI U )ROME AIR DE V EOPMENT CENTER

GRIFFS AFB NY R AMC AHAN AUG63 RADC- R- 83

UNC ASS FED A/ 179EhEomhhmhhEEEE IEELmomomo

I& L3.2 2

w, a L2W2

1__ A- u5.0

11111.5 LAI 1 . 6

I MICROCOPY RESOLUTION TEST CHARTONAWAIM FURFALI Of SIANDARDU 1963 A

I

2,7

!iEli

*; RADC-TR-83-1S1In-House ReportAugust 1963

00

4 SCA TTERING EXPERIMENTS A T' THE IPSWICH ELECTROMAGNETIC

MEASUREMENTS FACILITY:CALIBRATION WITH PERFECTLYCONDUC TING SPHERES

Robert V. McGahan

APPROVED FOR PUBLIC RELEAS, DISTRIBUTION UNIMNITED ~ 11.CTF

i. . FES 1 7 s84"

ROME AIR DEVELOPMENT CENTERAir Force Systems Command

Griffiss Air Force Base, NY 13441

4 2 17 046

This report has been reviewed by the RADC Public Affairs Office (PA) andis releasable to the National Technical Information Service (NTIS). At NTISit will be releasable to the general public, including foreign nations.

RADC-TR-83-181 has been reviewed and is approved for publication.

APPROVED: ) JJ '

3. LEON POIRIERActing Chief, EM Techniques BranchElectromagnetic Sciences Division

APPROVED:

ALLAN C. SCHELLChief, Electromagnetic Sciences Division

FOR THE COMMANDER:

JOHN P. HUSS

Acting Chief, Plans Office

kr

If your address has changed or if you wish to be removed from the RADCsmailing list, or if the addressee is no longer employed by your organization,

please notify RADC (EECT) Hanscom AFB 4A 01731. This will assist us in

maintaining a current mailing list.

Do not return copies of this report unless contractual obligations or noticeson a specific document requires that it be returned.

UnclassifiedSECURITY CLASSIFICATION OF TIlS PAGE (When. Del. E. .. d) __________________

4. TITLE (.,d S..btifl.) S. TYPE OF REPORT II PERIOD COVERED

7. AUTOR(.) . CONTRACT OR GRANT NUMBER(.)

Robert V. McGahan

9 PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT. TASKRome Air Development Center (RECT) AREA 6 WORK UNIT NUMBERS

Hanscom AF"B 61 102FMassachusetts 01731 2305J404

It. CONTROLLING OFFICE NAME AND ADDRESS 12 REPORT DATE

Rome Air Development Center (EECT) August 1983Hanscom AFB3 13. NUMBER OF PAGES

Massachusetts 01731 271 4 MONITORING AGENCY NAME A ADDRESSOI differnt from. Controing Office) I15 SECURITY CLASS. (of th~is report)

UnclassifiedIS.. DECLASSIFICATION DOWNGRADING

SCHEDU LE

16 DISTRIBUTION STATEMENT (ofthis1 Report)

Approved for public release, distribution unlimited.

IT. DISTRIBUTION STATEMENT (ol he .b.t,.ct .. te,.d In Block 20. it differentI-on Report)

lS8 SUPPLEMENTARY NOTES

III KEY WORDS (Continue - v.- si,.de ,ly1n, -d Id..Iify by block n--be,)

Radar cross sectionRadar scatteringRCS measurements

20 ABSTRACT (Coannu on reverse. sde itnecnn..,y and idenify by block numnb-)

A new electromagnetic scattering facility located at Ipswich, Massa-chusetts is described. A 20 X 20 X 40 ft anechoic chamber was evaluated as towall reflection coefficient at 10. 35. 98, and 140 GHz. The frequency stability'-f the measurement equipment was evaluated and was determined not to be alimiting factor in mak~ing radar cross section measurements. The backscatterradar cross sections of 24 metal spheres were measured at 10 GHz and com-pared with theorv. The spheres ranged in size from 0. 21810o8. 00 inches indiameter, corresponding to a ka range of 0. 581 to 21. 293. The Measured

DID "1,11 1473 EDITIONl OF I NOV 69is OBSOLETE UnclassifiedSEC.,RIY CLASSIFICATION OF TIS PAGE (Wh-. DJOt Enonod

UnclassifiedSECURITY CLASSIFICATION OP THIS PAG[(WheO De Rned)c.

20. (Contd)

'radar cross sections differed from theory by less than 0. 6 dB. Radar crosssections as low as -46 dBsm were measured. The calibrated facility willbe used to measure canonical shapes for both metallic and non-metallictargets. The non-metallic results will be used for the Cruise MissileSurveillance Program.

I?

ai

Unclassified

SE UPITY CLASSIFICATION OF PS E W*. n Dae Entered)

'I

Preface

The author would like to recognize the contributions of Mr. David Gaunt. Mr.

Gaunt performed most of the measurements described herein and was instrumental

in the construction of the monostatic radar cross section measurement system.

• - i - j t r i b u t I o " / -. _ '

NT I

ip,1

0 A1rOSb IY Codeg

3-- 'al.hi~e

.tild

Contents

1. INTRODUCTION 7

2. DESCRIPTION OF FACILITIES 9

3. CALCULATIONS AND MEASUREMENTS 20

. 4. DISCUSSION 24

REFERENCES 27

Illustrations

1. The Ipswich Electromagnetic Measurements Facility 9

2. The Radar Cross Section Measurement Chamber at the IpswichElectromagnetic Measurement Facility 10

3. Experimental Arrangement for the RCS Measurements ChamberAbsorber Evaluation 11

4. Plot of Return From Chamber Walls vs Azimuth Angle Comparedto a Flat Metal Plate, Frequency = 35 GHz 13

5. Plot of Return From Chamber Walls vs Azimuth Angle Comparedto a Flat Metal Plate. Frequency = 98 GHz 14

6. Plot of Return From Chamber Walls vs Azimuth Angle Comparedto a Flat Metal Plate, Frequency = 140 GHz 14

7. Block Diagram of theX-Band Monostatic RCS Measurement System 15

5

Illustrations

8. Plot Showink4 the Signal Source F'requ, ncv InstabilitV Spci fication

Superimposed on the Svstern Null 18

9. Average Values of Hj(US Measurements on Small Spheres forSystem Sensitivity Tests 19

10. TarLet Mounting -rran.Cen t tor \ Hand R('S Mea2uremnts2Showing Positioner Shielfiim2

11. Measured IlWS \ sl ies o! Metal Sphe res Comipared to Theorv,

N ormalized RCS vs ka 24

12. Meas.rel HCS \ alues of Metal Sph.,res (oni pare,l to Theorv,

cm- vs ka

Tables

1. Results of Four Sets of RCS Measurements on Five Small

Metal Spheres 19

2. Theoretical and Measured Radar ('ross Sections for FiveSmall Spheres

3. Measured Radar C ross Sections for "lwentv-four Metal Spheres 23

6i

A,,

Scattering Experiments at the Ipswich

Electromagnetic Measurements Facility:Calibration With Perfectly Conducting Spheres

I. INTROI)UCTION

The lrnie Air I )evelopm ent Center, maintains a level of technical leadership

Ihat "ill provide suoerirr" C3I capabilities for the U.S. Air Force. A number of

internal renters ,F e.pertise were established, in selected areas, to strengthen

ltA)Cs technoloL~v base. One of these areas is that of scattering from terrain and

targets.

A scattering effort had been maintained at the Ipswich, lassachusetts Electro-

magnetic Measurements I- acility since about 1950, when scientists from the Cambridge

Research Station, under Dr. Roy Spencer, began radar cross sections IRCS) mea-

s urements o)n nietalli ,'anonical shapes. This program progressed to measure-

ments on birds for the i)1':%% line radar and ultimately encompassed scale model

teasuremnnts on the Meruurv Space capsule and Atlas and Titan missiles. With

!he retirement and reassignment of key personnel the scattering effort was steadily

de-emphasized until, in 1979. the outdoor ground plane was dismantled and the

mieasurement system was disassembled.

At this time a new building was approved for the Ipswich site, under the Military

Construction Plrogram. A scattering chamber was specified and designed into the

(Heceived for publicati:)n 15 August 1983)

7

il i ttd - 1qii1. liv 0 .)cola I I) 1 i. aI)s li)er ina s IStal Itcr it! he .... c

R.\IC i*:c 5-t2cI l 't ic IRrjS mealsuremotts pro"ral cfluiOi-w

.neta I II(i f Ind nol n metali I a _,etS. Initiall!., simiplIe -a noni ca1 I h lwI,, t~ , t at

sot'bji2 'raterial wiHI be !n asuroer. The t,-sults of the mieasuremients on ( anonlial

shapes .re expected to benefit the C ruist S l issi Ic Surveillance Proc rain direct iv,

sne .tes,, ta rzers can he cos e Lv -siniulated w i a coirbi nation o)f canoni cal shapes.

The pcresent work descibes the new scatte ring~ chamber at Ipswich. Th e

hariher .%as eva Iuate.I at 10, :35, '18 and 140 Gl lz and the frequency stabi litv ot the

:nteas-uritU eqtui pmnt was evaluatedi as to its possible e f'tect on n easu rem ent accu -

I* acy, . A nt onostatw svs! em totr niasurinE4 radar cross :sections at N -band, using

he ('\% cartce blat ion eitd is de scribed. Results of measurements on a se ries

*if metal sphe res are cmi ~pared to :heoretical values corn puted us inc an exact solution

)the Alie summ1nation.

The radart cross section (T is a mecasure of the s catte ring prope rties of a body

and is defined as

(T lint 4 -.R 2 H H. (1

t7 lint 4u1 R- Fv F. (2)R- 4 s I

where R{ is the distance from the observation point to the bodY, 11 1 and Hi are

the magnitudes of the incident and scattered magnetic fields, and I .! and !E sa re

the magnitudes of the incident and scattered electric fields. 3The cross section o

as defined in Eqs. (1) and 12) has units of length squared but is often normalized to

either the projected area of the bodY or to wavelength squared. Radar cross sec-

tions are also commonly expressed in decibels relative to a square mneter.

l.TdBsrnl = 10 logl1 0 t0), where CIO is in square meters.

In this report radar cross sections are normalized to the projected area

(A -, ta 2, where a = radius of sphere) unless otherwise noted. A convenient parameter

1. Mack, R. B. ( 198 1) Basic Design Principles of E12letoanicSaernMeasurement Facilities, KAtDC-TR-81-4U, AU %W#94.ntcSaerg

2. Hancock, J. H. , and Livingston, P.MA. ( 1974) Program For Calculating MlieScattering For Spheres, Using Kerker's Formulation, O-ver a Speci ledPt-7icle Size Distribution, -. R4L Report 7808. -

3. Mlethods of Radar Cross Section Analysis (1968) Crispin, J. , and Siegel. K..,Eds., Academic Press, New Ytork, p 5

8

&klit I ci to p Io t I i( ,- is k,, . \ he re k A.-~ tht. 1)u....rn con -t -flt w

flreqLrrencv of opel-Mion. rhe q uantitY ka is a imeasure of the nurmber of:'elel

in trw ci rcumrfe rence of a particular sphe re.

2. I*;SCRuwrON OF' FACILITIES

The Electromagnetic Measurements 1acility is located on 60 acres at (ire-A

Neck in Ipswich, AMassachusetts. The l{(S mieasurements chamber is located in

the main building, a 6270 squa re foot structure, completed in 1 981. The mrain

building and the measurements chamber are shown in Figures I and 2.

I-iguire 1. The Ipswich Electromagnetic Measurement Facilityv

9

I liutilt. 2. M41 HI.idar r toss Setction rNleasurenient (hamber at thellps, i.i I.l'ctrtmni.tgnetic Mea surenent I.l*cilitY

* 1

I't 'hnih.r is i1 ft lh)ng. 18 ft i nde, nd 20 ft high, Ilyranmidal absorber on

thu. AII, . floor. .,nd veiling reduces the us~thl, diniensions to 34 ft 14 ft X 16 ft.

re~spectivelv. Oine end of the chamber is a wooden partition constructed of 2-in. \ 6 in.

studs and 3/4 in. plywood. This partition contains nine ports, with removable plugs,

thait allow various antenna target configurations to be installed. Access to the

chamber is through one of two sets of double doors located along the side wall of

the chamber. These doors are also covered with absorber.

The absorber used in the chamber is AAP-24. a 24-in. thick pyramidal absorber.

except for a section on the wall opposite the antenna ports. This section is 100 ft

square and is covered with AAP1-36, a 36-in. absorber. The magnitude of the re-

flection coefficient for these absorbers is specified by the manufacturer to be -40 dB

at normal incidence. 4

A series of measurements were made on the chamber walls at 35, 98, and

140 Gllz. These measurements were made using a scatterometer constructed by

4. Product informttion Guide (1979) Advanced Absorber Products, Amesburv. ILA.

10

[7

-li~1

h ,, !titrl~ttI %1( t I t'ltilf I tlt~ 1.- 1p lt 1" 1i 'k Ii. I u

a , t t : ,, : In , .I I u nln I 1, 11 t,,Ir" ,t :i I I; 1z.< [Id i. I,- t t, II ,I ,c 1 t,, ' ,II

n ,d1,t)sU .t, i

":..1) 1 V.I I o, i 1I0 I I~. ', t h I t). ' ' I I4n 2 , IHu s! T~ ',t : I'. I.' .c, 111 , .t (i

In, ' i : (vc , ll . ,I ! ..I It it , 'l ,x i[|: I ,n X ), rc cor'I I -. :[ Ltl (".'. I'llti' c S, ' tlk -. n'l c

, ., i ~ "t h, ', .l . , t ', : , :1: 1,,f 1, 1 c h ,,'t,' c III l ~' ;|lllm " :, tt c:11 I . : . 11I,. l-' t:,ti(ltn

1:"~ ~ It 13\, M,,t!l 'tL',t Wr~ I| t't I Itt t! .1 1112 % pl i 'd I(, h X . X :, i: t'

rt .1T 1" I lc :,," * ' ' , r '| , r', I' 111 il t -h ::ntk.rl :,.I s I I t *( tt' 1 1 ,tltn tI I\ . t T-u1 ,,1!t , n 2

,. id t. I ' , II:. \ l] t[ 1 c t., I I I Itc IA ..- i l , ' ' : t 11c k' I, I , : II fit - t;, l hl 'it-Ig -.til t "

t! 1 11 1! t t IM ll2 It .' ..' ' 1't l 11k . l n t I t . p ! -l t. ' - , I I I t i s I . '!' I I I I . n I. i vt-

lt I .I I -, t I t , .[ I I- (, II Ik Itt %%.ll . t s dcfitw t'( bc |t (I cli :,nd III,' iwn -I tilt." co rd", ' I' !

p" l t!,,'! 'd I! Ill,- I''[) ),! Ilit IILIJ' |t ' . 1'111. 1)1% ' . :, i z: t 'd b V ll 'l t in

!1, t 1111, 1 r)[ ll l l ' - I ) A 11 t )11- 1itw :, . (; A 1 11 d ific r' ill tilt- o._, c \ v.

'24 ABSORBER

SCATTEROMETER 36' ABSORBR-

) ROTATION

STAITING

POIN METAL PLATE

DOOR DOOR

'i 11u1 'e 3. Arxran'ignem:en .\f'i'al)itcn rfl t t'll th ( 'itv' C f;lr nlelItS ( " lit lv hruii 1 Ahst i'be " I Alt Iu ttion

. ,imme'rs, I If. \k., ;tnd Hives, 1). 1". 119801 Multipath l'roptF g ti o ( )v e Sn,wtI Niimctter \\Avelengths, HAI-TR-89i-54, Al) . T\0,77.

11

To tmea sure ..bo the' eftec tiv it\% at Igiven ftecqUenCY, thle sca tte toneter wasWet. 'Ip as dsC hd hcuadt recordjet paper cal ib rated. *rhe scatte rontete t

aat te~d 0!0 k%% i~t jstiti! thlt- tntkenna s pointed off the calibration pla te and at the

absorber-coveted a lIiTe S cAtte tomcter was then rotated counterclockwise

ipprtoximnatct. 2. n ati g thet intentna beants inl turin acrtoss the calibration plate.

1ltu thle long a%! paist otic Set of douible doorslI, Across the back wall (traversing

heC Oi- t. SOAb', ild I!lng thle othler !lng w all, passing thrtough thle point

di ctlc opposite, IIIe ci liblratieorlt.

sitick- tilt , ilibrjitiotn p!,tt %%,Is placed ait the wall, and the scatte romete t was

inl the center of til, hamtbert, thtis prtduI teallowed direct comipa rison of the

caljibration p LIte to \%it i- et oe e lt tile sa me distance, w ithout having tot

m1ove thle plate. .Me.sttelienits a eter taken a ith thle doors open and closed. Thie

,bove descrtibed proceduI: t p to \ided ai taasurti ttent of tilt 36- in .absoribe r at

norimat incidence, 41010eUgJ It Li dilffe re ttt tange than the 24- in. absorber. Addi-

tional a ttenuation w.sin se rted a ibe the sa tttcroniete t was pointed in the v icinitY

of thte nteta I pta Ic W tOa!Me tile peOak oIf thle re't tnl -signal1 to be plotted, and w.as

tenmoved as tite anltelna be. uis itioved a cf rotm tle plate. Figurtes 4 thrtongh 6

hatilt, restlts of tile's, toe sa em'Cl!lts aIt theV thrVCc frequencies. It can be seen

ftoi these p)lots that thet ha, kSCratter ftrom the :ibso thert is at least 40 d B below% thle

incident .-nertgy level ait thIe sC tit'C te tequctienies , as cla inied by the mantufac tutre r.

[liiis was trite tor tionoimail incidence ais well as grtazing angles ttp to 68 ' from no rnmal.

[hei st ightlv, iig her rta ri at I ii/ I is believe.Ld dute to aI resonance effect catused by

thle paint used to coait the aibsotbe t. I'tie ene rgy levels retutrtned from thle absorbert

at-c low enough in attv catse tliaI thley\ will tet p toltibit IW.S nteasnucements at thtese

frequencies.

A monosta tic sca tteriing mte;i sure men t system was designed and built, aind is

shown in block diagtain fot'tt itt I igui- -c . Trhe svstent uses tile c\\ cancellation

method, aind cotnsists of a signal surcte whose outtptt is fed through a bandpass

filter and an isolator, ai nitlling tietwork, an antenna and a receiver. Trhe trans-

ntitter is a Scientific-Atlanta 2150B signal source. This unit has at digitally'

ope rated control box which commands a ft eqtaencv synthesizer. The manufacturer s

lite rature specifies a frequency\ accuracY of one part in t10 and a frequency, sability\

of one patrt in 10'7 per 24 hotrs. The sYnthiesized HI- outptt power is +9 dllm at

X-band. When at higher signal level is required an lIP495A T\% T amplifier is used.

A Scientific -A tlanta 1770 teceiver is used, along with an 1833A dual rationleter

and an 1823 phase displaY, both also from Scientific-Atlanta. This receiver has

two channels, an automatic phase control (APC) channel and a signal channel.

12

The APc channel is Used to phase lock the receiver local ost illator to the trans-

mitter. The signal channel is actuallY tw~o channels, designated A and 13. 1I.ach

can be monitored by the rationieter, which displays signal levels in dii. Manual

switches allow one to mionitor channels A and Bt simultaneouslY, or channel B

and the ratio A'/13. W\hen the latter mode is used the signals A and B are fed

to an external switch which is operated sYnchronously% withi an identical switch in

the receiver. This muILltiplexing arrangement allo-ws two signals to be applied to

the signal channel of thle receiver.

20 __________

S30

- TTNUTIN20d8- - - -ATENUATION 0dB

cc 470

00 9010

AZIMUTH ANGLE (DEGREES)

I- igure 4. Plot of Return From Chamber Walls vs Azimuth Angle Comparedto a Flat Metal Plate, Frequency =35 t;llz

13

3o;

0 -E- AIENUATION 20dS .- ATTENUATION OdB-

z02 40-

so _.0 SNO SIGNAL70O-----~--- - --- T- -

000 1800SACK SIDE

AZIMUTH ANGLE (DEGREES)

ligure 5. l'lot of Return ["or Chamber Walls vs Azimuth Angle Comparedto a Flat Metal llate, FIequency= 98 6hlz

; 1 0 F- -- _________________

_20. 10d. _, _ r

-ATTENUA TION 10SATTENUATION Ode ___

0

00io

BACK SIDEAZIMUTH ANGLE (DE!OEI EB

F-igu re 6. Plot of Return From Chamber Walls vs Azimuth Angle Comparedto a Flat Metal Plate, Frequency =140 GHz

14

t iII I I IIm l-

" 1 ." " " "- : .

SIGNALSOURCE AMPLIFIER ISOLATOR

: APC

i SlG XX_

10dB ATTENUATOR

TORECEIVER MIXER

rH - - ANTENNA

SG NALCHA WAfrC

NETWORK ) ISOLATOR

Figure 7. Block Diagram of the X liand Alonostatic I(S MeasurementSystem

The RCS measurement system operates as follows. The transmit signal is

passed through a directional coupler, after the above mentioned filter and isolator.

The major portion of the signal is fed into a hybrid (Magic Tee0 where part is

directed to the nulling network and part to the antenna, where it is radiated. The

coupled portion of the transmit signal (-20 dB) is passed through a variable attenuator,

mixed, and fed to the APC channel of the receiver. A portion of this APC signal

is coupled off (-10 dBI at the R- level, making it -30 dB with respect to the trans-

mitted signal. This signal is then passed through a variable attenuator and fed to

the reference port channel BI of the external synchronous switch. The received

signal passes through the antenna to the hybrid, where a portion of it is directed to

the matching network and a portion is fed through an isolator to the signal port

(channel A) of the synchronous switch. The ratio A/B is proportional to the radar

cross section of the target and can be displayed on the ratiometer by means of the

synchronous switching arrangement. By properly calibrating the system, radar

cross sections can be read directly off the ratiometer in dBsm.

Before making measurements, the system must be nulled to reduce, insofar

as possible, reflections from objects other than the desired target. The nulling

process consists of adjusting the matching network, which comprises a phase shifter

and two attenuators, while transmitting into an empty chamber. This adjustment,

an iterative process, introduces a signal into the hybrid that has the proper phase

and amplitude to cancel both the stray reflections due to imperfections in the hybrid

An empty chamber is considered to be the room plus target mount and necessaryabsorber, but not the target.

15

i

and any signal reflected from the empty chamber. The signal level in channel A

is then a minimum. When a target is placed in the chamber the received signal

can be presumed to be due to the target and not the chamber. This presumption

is valid only if the presence of the target does not introduce reflections, from parts

of the chamber or mount, strong enough to unbalance the system. This point should

be considered whenever a target is to be measured. In actual practice the nulling

signal and the transmitter signal drift with time in phase and amplitude. A very

small change in either signal is sufficient to cause the system to become unbalanced.

These changes can be caused by thermal gradients across the equipment or the

chamber, movement of the chamber walls due to wind, or interaction between stray

leakage signals from the equipment and people present in the equipment rooms.

These effects can be minimized by careful shielding, insulating, and shock mounting,

but cannot be completely eliminated.

Measuring a calibration target of known radar cross section, such as a sphere,

in close time proximity to the measurement of each unknown target, however,

allows the errors due to the (ffects of signal drift to be minimized. The radar

cross section o of a test target is found by means of the relationship

0 = M (C- T) , (3)

where M is the measured radar cross section of the target in question. C is the

measured IWS of the calibration target, and T is the theoretical HUS of the calibra-

* tion target. The terms in this equation are in units of decibels relative to a square

meter (dBsm). Since the ratiometer can display the I'S of a target in dB. the

calibration process shown in parenthesis in Eq. (3) can be done by nulling the system,

placing the calibration target in the chamber, and adjusting the gain of the ratiometer

until the A/B is equal to the theoretical IMCS. As long as the system remains nulled

the reading on this ratiometer is the H('S of the target being measured, in dBsm.

The calibration target need not be measured after every test target, but must

be measured often enough to ensure that the system null has not drifted. For the

present set of measurements the calibration target was measured and the empty

chamber null was checked after every test target. The system was renulled when-

ever necessary.

The chamber was evaluated at 10 Gtlz using a method reported by Buckley. 6

This method involves comparing the return from the empty chamber to that from a

known target. A signal is transmitted into the empty chamber and the system is

nulled. The output from the hybrid is then very close to zero. The nulling signal

6. Buckley, E. F. (1968) Outline of evaluation procedures for microwave AnechoicChamber, Microwave Journal.

16

'i _ _ _ _ _ _

consists of twO "ro1uipnOts. One corpr)nent raroels Ihe e(,rror due to system in) t

perfectioos and the thet' corm ponent cancels the relections 'ron1t the chamber.

This latter cortnporrit is the tine we desire to measure. l3v Byomoving the antenna a

small amo!unt we can changle the phase at' the chamber reflection by 180', while

leaving the huling siainal unaltered. The signal at the o)utput :it the hybrid is now

equal to twice the magnitude ot the return from tile emptV chaiher.

WVhen a target of known IuS is placed on the chamber, a signal of' knocwn ampli -

tude is added to the existini signal in tile hybrid. Wttatirig this target (usuallv a

S phe re) s tzilhtlv I'f- cetler on tile posititner rotates the phase of this second signal

'hrrtgh 180 . thus causing the )utput of the hybrid to, cycle from a mininlum value

of 5-2t', to a matrimunt vatlue of S % 2 where S is the theoretical I 'S of the sphere

and C is the tunknown) R(CS of tile ch:mber.

The recorded signal excursion at the receiver output can then be equated to the

ratio of tile above values,

(S-2(')t(S+2C , (4)

i Jnd ai value for the tItS of the chaniber can be calculated. An alternative value can

be calculated, based on tile possibility that tile sphere return might be smaller than

twice tile chamber return. In this case

* (2 -S)/2( +S ) (5)

htepeating the measurement for a different size sphere will yield two additional

I values for the chamber RCS. Two of these values will be nearly equal and two willbe widely divergent. The two nearly equal values will be close to the IHCS of the

chamber. The two can be averaged with additional values to obtain a value for the

I('S of the chamber.

Following this procedure, using a 2. 625- and an 8. 000-in. diameter sphere

a chamber RCS of -32. 9 dBsm was determined. Dividing this value by ten and

taking the inverse logarithm gives the RCS in square meters. Performing this

operation yields 0. 0005 m 2 for the radar cross section of the chamber.

An experiment was performed to determine whether the signal source stability

would be a limiting factor in the accuracy of the RCS measurements. The system

was nulled and the output of the signal channel of the receiver was applied to the pen

input of an antenna pattern recorder. The signal source was then allowed to sweep

a known amount above and below the center (nulled) frequency, f . At the same time

the chart recorder was allowed to run on its internal time base so that the paper

advanced as the frequency was swept out of and back into the null. Allowing the

17

,I

source to sweep through two successive nulls enabled calibration of the chart paper

in Hlertz division since the extremes of the frequency excursion were known. The

portion of the curve depicting the bottom of the null was then redrawn on an expanded

scale, as a function of null depth in dli vs frequency deviation about f . The manu-0facturers specification for signal source stability %vas then superimposed on this

curve. This is shown in iigure 8. The published specification for frequency

stabilit.v of one part in 10 translates to a maximum frequenc.v deviation of I kliz

at a center frequency of 10 (Glz. From this, and inspection of Figure 8. it can be

concluded that the stability of the signal source will have no effect on the accuracy

of the measurement.

Z MAXIMUM Af DUETO SIGNAL SOURCE

0 INSTABILITY. 1 kHz

0

0

0

,-

o L-

-32 -24 .16 . . 0 a 16 24 32 40

FREQUENCY DEVIATION [kHz)

Figure 8. Plot Showing the Signal Source FrequencyStability Specification Superimposed on the SystemNull. Frequency = 10 GHz

Five small metal spheres, covering a ka range from 0.282 to 0. 581, were

measured to determine the lower limit of system sensitivity. Four RCS measure-

ments were made on each sphere and the results are shown in Table 1. The aver-

ages of the four sets of measurements are also tabulated, along with the theoretical

RCS for each sphere. The average RCS values for each sphere are plotted in Fig-

ure 9 along with a curve showing theoretical RCS for the ka range in question.

18

Table 1. Results of Four Sets of tiS Aleasurenients on Five SmallMetal Spheres, and the Average of the Four Measurements for EachSphere

Diameter .\l3 ted R(S (dsn1)(inches) ka Trial I Fri, 2 "FriT il F rial 4 Ave rage

0. 106 0.282 -57.4 -16.7 -57. 9 -58.0 -57.5

0. 124 0. 330 -58.6 -656. 6 -56.4 -56. 4 -57.0

0.151 0.402 -53.6 -51. T -51. -51. 5 -5). 2

0.212 0.565 -49.4 -4 1'. 9 -48. -49. I -49. 3

0.218 0.581 -46. 5 -46.5 -47. 4 -43). 4 -46.0

l ['

2

U 1

0

Ix

U

4 1o

4

0 01 __

N

g -'

0 025 05 075 1

ke

Figure 9. Average \ alues of IW(S Measurements on SmallSpheres, for System Sensitivity 'rests. rheoryMeasured o

The measurements were made at a distance of 12 ft. and a power level of 24 dlim

at the input to the antenna. The antenna was a rectangular horn. with a 4-in. by

6-in. aperture. This antenna has a calculated gain of 19.6 dB, with an effective

aperture of 6.5 X 10- 3 m 2 . Using the radar range equation, and a figure of 81 per-8

cent for the efficiency of the horn antenna, we can calculate the power densities at

8. Silver, S. . op cit, pp. 182- 187.

19

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

the target position and, for a given target 0, the power present at the receiver

input. The receiver has a specified sensitivity of -80 dlim. Based on this sen-

sitivity, and the above mentioned antenna parameter, the system should be capable

of detecting targets having radar cross sections :is low as -52.4 dl3sm at a distance

of 12 feet.

Table 2 contains the pertinent results for these sphere measurements. The

sphere diameters, projected areas, and ka values are tabulated, along with the

measured and theoretical received powers for each. Examination of this table, in

conjunction with Figure 9, allows us to draw a conclusion about the system sensi-

tivity. We see that measurements on spheres as small a- 0.218 in. in diameter

(ka = 0. 581) are possible with this system, The nleasurements on smaller ka

value spheres are unreliable and inaccurate and, in addition, the received power

is leveling off near -80 dl3m, indicating that the m1easurement s'stem sensitivity

is limited by noise rather than by background reflections.

Table 2 shows that the RUS measurements become unreliable near -50 dlism.

This is close to the lowest lH(S theoretically measurable (-52. 4 dlisml with the

system. The discrepancy can be accounted for by small losses in the waveguide

circuitry which were not measured or accounted for in the loss budget. Since the

system is noise limited rather than background limited, radar cross sections even

lower than -46. 5 dlism can be measured by increasing transmitter power.

.1. CAILLATIONS AND MEASUREMENTS

A series of 24 metal spheres was measured ait a frequency of 10 GlIz. The

spheres ranged from 0. 218 to 8. 000 in. in diameter, corresponding to a ka range

* of 0. 58 to 21. 28. The backscatter cross sections were measured, and compared

with theoretical values obtained from a computer program which solves the Mie

summation.

The computer program used was written by ltancock and Livingston of the

Naval Research laboratorv2 for computing scattering from aerosols. The program

was designed to calculate bistatic scattering from spheres having a complex refrac-

tive index and had to be modified to handle metal spheres. Several programs were

available for the theoretical part of this task, however, the number and extent of

convergence checks in the NRI. program made it the most suitable by far of those

considered. In addition, planned work on dielectric spheres will require the use

of the program in its original form so the task of modifying it and adapting it to the

HADC computer was doubly justified. The program was also modified to handle an

extended range of ka values.

20

A~7 ________

-1i

• -. i- - - -i -

• - -- -. z --N

I_ I I•

i I i i I

- 21

* 1:-

- 0 0." III -• m, -- ,,,, ,.. N

.\ft, th'. L, . 444 44 III (le To4 till 4411 11' I :.\I)(' 4114441 I sl4 i'54 1 tI

c1A I I'.S. 4l t~l 144 k II I tI II ! I I I I h I I 11(1 4 14dR~ 4144i 41 414

I: -I( ILI4 k. % ,I I11 'A I44 .544444 1141o t tl tit- bi-1.4114 r'::l~kt r' 4444.. tion r, l4-L

i, , . , Il4 41 0 1 t I4 4.4 :11'.4 r 441 So I 4 If ' I It,.- :4 i4ith 14i I, I4 n 4 I 1c : I, !,I1

,1444 -X,\4k1-'k4k - 444 r4II44I4e 1 I nIII ki s I S t. \\ i ) I v. t.n Ii I14444' r-4~ 4141 h4.rl4!

I1\51 4 k 144 I 14 , 'I It II I (I I 1 -?,,ing 14 )4til- S )1 . ....4 . 11 r I11)14 511 I %%4 I I., t I , 4 I4

I tI \ 4 1 '1 94' i' 111 44k! I I ' t4' 4444 - t . PlC I*'t I1I'S 4A C 4'l f, I 11Il I I h( 1 l t, 12 44 : I SI I, II IIT I'4J-

,s !I .44 If 1 I4'~ I I I k I44 t r11 I444412 if I44., 4 II I I , 1 4 '4! 44. '1 1 I Ti, e4 .- 4 1 14'

r, I'I (I , : , r'4 , '. s I ( I s -.. II ( II ,4 i 4 I '1 4- ' , I 4 I (1.: I (1t-g0* ) 1I c

isIl i it,44441 t 4414 till, 4114141 V S1I, ill, h4 1 44 h -ki - II ,4 4114 1 4 4 .tb'

4414 ct 41 . Fllik. -;~w t 'A 111 11211'd oi 44': 'I.,4'*',141[I 1441!l 11 44it1 4' 4' !tfd

d I , - ( , ! I4444 I tti d I ,UnIlt - s t rt.-' i 4 If441 t,-c .4111 fl--4 1444' d(Iti,4 rcife . 1C I ZC of

th4' t.4~c I",4 I14''',5'1 144 114144.lt4 11444'- 41 1 tilt' po)S111 ti14-l %%I4 -.hieldt.'d b%

%,-odcn4'f 4141 o-Ur' .4(44241 with ,1ibs4 )!h'vr. N Ill. '' f , 4144144'tlid, .4 bsor'her' 44 aS

;)L.44'4'4 ht'4'4 4'4 1 w 14' l'Oof4n CO4 414114f :Id li( ' (- .41:41 I c2 4, 1c posi'(4 5tion(4. v tulrn-

t.4h14 , 45 th11s %4:4s foundl4. to b,- z, s1414441 sm4444444.! 1' li ons1411. 111,' :4 4'12t moun44tinlg

4r-1'4 4'2v t.41t is S1 ,1 Iin I 14111142 10. \\ itil t!1I, I44114! Tit Tilt- rcc,42t.'l'e j444l o4f till

* h l'jd 4444114. he isol.t'd fro4nt:11 tic I'ri511411 t 1' t 1). 14 411 1 , I --; 1I 41'tIv41i 1111111

'.44441( List t 10 444'n )I. 14)14. ( 44'4'asionflkI nlls o 41 201 dli 44444 ht.iifl4' butl tilt,\

s.444 ol44tlI'.'4.. \1.'st 144t';.544142144'11s 444.44 ln4.4d4. %%itli1 101 (1, flitllis.

TAPERED STYROFOAMCOLUMN

A ANTENNA

- PYRAMIDALABSORBER

TUNED ,.ABSORBER . . WOODEN,

4vmox

<4

POSITIONER

Figure 10. Target Mounting Arrangement for X BandRCS Measurements. Showing 1'ositione r Shielding

22

.-v

k i iL iq~ -I It n n , i k .W I I I I It ( I I t Il I' 1 1)i

- I i ,l t 2.

nii

2 I iii I. j - ), st. 'vc it . 0-f

I i~tttki (I Isut Ii

I.500 I. ';l0 -35.5 2. 2 1 0- 4

0.656 1. -15 -42.0 6.31 1 0 . :

0 . ; ;I '2. 16 I - 32. G 5. 50 10

0. ,5 2. -2 -30. !t 6. 13 10- 4

0.1t3 2.4'4 -31. .3 . 1 10

1. 000 2. 662 -3 2. A 5. ; P9 0 - 4

1. 125 2. .5 -33. 8 4. 17 10- '

I. 250 ;.32 -29. P 1.02 •10

I.3-1'5 3. 65! -2 8. 2 1.51 • 3

1.500 'i.!1!2 -3(0.4 9. 12 10 - 4

1 4.49 1 -216. 0 1. 581 10 - 3

1. 50 4. 65! -2.0 2.00 10 5

1. 875 4. -0 -20. - 2. 14 10-3

I. 9: :. .5 -27. 5 1. 7 " 10- 3

2. 000 5. 324 -28. 0 1.5 10-:3

2. 125 5.fi56 -26. 1 2.45 10-3

2. 375 6. 321 -25.5 2. 82 10- 3

2. 500 6.654 -25. - 2.69 " 10-3

3.000 7.985 -23.8 4. 17 10-3

4.000 10.647 -20.0 1.00 " 10 2

5.000 13.308 -18. 6 1.38" 10-2

6.000 15.970 -17.4 1.83 • 10-2

8.000 21.293 -14.8 3.33 10 " 2

23

IsI

2

0

- !-U

4 . .

o

0 3 6 9 12 Is iS 21 24 27ka

I Igr'tu I e. Meastied l( S \alues of Metal SpheresSoipoi'd to, wheoi, \oric' iliz d IW , vs kzi.P'lio 1.. i awel511 redo

zt0

U

0

oIg

8 .1 . . . . . . . . . . .

a 3 6 9 12 IS is 21 24 27

ka

Figure 12. Measured HICS \alues of Metal SpheresCompared to Theory, cm 2 vs ka. Theorymeasured o

*,. I)IS EISSJON

The accuracy of sphere measurements is easy to ascertain because an exact

solution exists. In this case the theoretical RC(S values were provided by a com-

puter program written to calculate the scattering from dielectric spheres and

24

V -l

inuiotijd to huindle, inutil sphe recs. l'iht- l. (tt1.S( elxnvA s for the s'phil ot

(nicer 0. 2 1 B in. 10. 56 cm' I, co res po din~p to ,k1 of 11. Dich meaisured ~lid~ff~l-td from11 theor\ b.\ 0.6 (Mdl. Alea-sureintnlts o)n tis s pucr i 'Ae ,lv in

error by about thiS sameC amo1unt no tia tter ho. miuh rc -,- s Liken to insui-( e

dehc1 null i% ls prestnllt [his- SituLatlul is; iel ll thek si:, ih ( -4'i. 5 dIlisii; of this

S ) 1h -I'e. \iiich puLtS it neat ' till I-)c l V ItII.t of S S stti I I ItI- It I I a ' t ho)vfn iit

it. rein inin,2o sice ould be( ltc smu:-'cu '. irAl t':r'i' ()I lesis thu'o 0. 6dli1.

\ unb f lusue ctsA':e rdv qn ("' :i oWtc ! c. .a PI' a sin Itided in

this repit .tr A,' rsett' the'ij (o: at,\,~tiil

'Ib-Se n11noititic re-sults (on itt ):ire.t'''.'ol\ -s , be.nin-rk tor tlt

\-b~lild lherlA. u-lhi~t-S J 'Or'tia. I'l( lht' i ' el ttttt ill till, p''--'A~ llh td "

in estI4,itingi Si ittetrLn frous diclecti. sae AwIositi. !cuti slive

ben nude it inlieltcric sp~iere-S u!ld t, teubet 1t 1l!uc reporMt. In r(Idtltin,

,series )f :lttil ubes has bee2C1 1110,SUF'd In "I'tit 4ftr o'uert

Of ditie-letIeC ribtS. A bistaotie til hs been Irli .11d pr'ln, nrirlu

nients uijng it irc tindcv\%nt

[h!is report !is (h5't'ihtd the bItS I!,snrts 1 'iiiti at ll-jh .ss;a-

chusett. Vi X-rta-il nro-ros tatic 5'steir. %)l-iicis i t- . '-a d ,ail''' tiiiu;

4 Ilrc'1sorenenit %%aIs designed ind built. I'lit 61SU': st ,S I t-nsitiv iti ' of -46. -) d us .-l.

2doosat neslrnietsona ciesof r~tlIspheres a ce des. !-ihd. Ale,,Sure'(-

nient eror i, Ace shown to he less tlltan 0. 6 dII, ti r he' p 0. K7; 2 1. 2 6.

25

Ref erences

1.Aack, 1j. B). 1911 Ba sic De~jgn P'rinciples of Electromugricti c 6catte tineAleasSLUMrelzet I a Cilitie'S, l{A x tic- 1-40, AlD Al1039431.

2. Hancock, J. II. , aind 1.ivingston. 1'. Al. (19741 l'rograinz for' ClC Lla tini' AlieScatte ving 10 v Sphte res. L sine IKe kerz's I-u 1IIi ltion. ()Vel, a SpeCifiedPa rticle Size Distribution, N Ill. Reporit T,308.

3. Alethods of Rada r C ross Section Analysis I 196,80 ( rispin, J. , atnd Siegel, 1K.Eds. , Academic Press, New Yo rk, 1) 5.

4. P'roduct Information (GUidv ( 197 91 Advantced Absorber 1' roCIItS, Aniesburv, MlA.

j 5. Laniniers, L. . G. . , nd I Lives, 1). 1'. (1980) AlUltinath I'ropaigation (lye r Snow

at Millinmete r Wave len ths, HAI( '- TR-8O- 54 AD) A0817 747.

6. Buckle1ev, E. I-. ( 19681)u tline of evaluation v rocedu res for mzicrooave A nechoicChanmberz, Al iCliow ye Journal.

Silver, S. (1949) Mlic rowave Antennia Theory and Design. Mlitraw - Hill HookUomipany, Inc. , 'Newv Y rk, pp. 586-581.

8. Silve r. S. ,op cit, pp. 182- 1817.

27

M ISSION* Of

Rom Air Development CenterRAVC ptans and executes %ezea'ch, deveopmen-t, test andselected acquisiLtion p~togtom~ in suLppot..t o6 Command, Cont'LoLComuictin and Intel~tgence IC3l) activtes~. Technicatand engineevaing suppott within cvreaz o4 technicaz competenceis pwuvZded to ESO Ptogtam O6 ces (POs) and otheL ESVetementa. The pxinciZpat techica~t msson a4w..& W~e

comuniatinsetethoagnticguidance and cont~ot, AuA-veitanc ojg~ond ad a~o~aceobjfect. inteU-gence dt

Cottecton and handeting, -Lndo'mation 4tem technotogy,iono~pheAZc piropagation, 6otid s5tate sciences, rmicoukzvephy4sLca and etect~onic tetiabZLZ-ty, maintainabiLitq andcornpatibititt.

Printed byUnited Stat.s Air ForceHonscoom APS. Mass. 01731