(PDF) Institute News and Notes - DOKUMEN.TIPS (2024)

401PROCEEDINGS OF THE I.R.E.

1953 IRE National Convention ProgramWALDORF-ASTORIA HOTEL and GRAND CENTRAL

NEW YORK CITY

PALACE-MARCH 23-26

GENERAL COMMITTEEG. W. Bailey, Chairman

S. L. Bailey, Vice-ChairmanJ. W. McRae, Ex-OfficioEmily Sirjane, Secretary

W. M. Baston Bernard FarberE. W. Bemis J. M. FlynnJ. B. Buckley E. K. GannettW. C. Copp Mrs. R. F. GuyL. G. Cumming Walter HauszL. T. DeVore C. A. Mills

E. H. Ulm

BANQUET COMMITTEE

J. M. Flynn, Chairman

Rodney Chipp Orville SatherHarry Grelck Charles SingerW. A. Haerer M. C. Worster

co*ckTAIL PARTY COMMITTEEE. W. Bemis, Chairman

J. B. Blake Alden Opie

CONVENTION RECORD COMMITTEE

Walter Hausz, ChairmanE. K. Gannett, Secretary

J. J. BaruchMones BermanL. A. DeRosaJohn DoremusWilliam Fingerle, Jr.D. G. FinkL. G. FischerR. L. GarmanM. A. GilmanNorman Harvey

B. R. LesterH. J. ReichG. T. RoydenGeorge SinclairMax SpitalnyW. R. ThurstonW. N. TuttleGerhard WalterE. K. WimpyLewis Winner

EXHIBIT MANAGER

W. C. Copp

FACILITIES COMMITTEE

W. M. Baston, ChairmanS. R. Patremio, Vice-Chairman

W. R. Ahern R. P. T. JutsonG. Barton R. A. KelleyC. F. Bergman R. J. KeoghJ. B. Blake J. R. McCroryR. I. Brown D. E. PikeG. G. Bruck F. R. PikeR. M. Burkhalter L. L. PourciauA. F. Childs H. J. Reed, Jr.S. Davis Theede RystedtR. J. Fahnestock John SchallerA. C. Flackbert C. J. SimonA. A. Gallonio G. M. SmithR. W. Gast H. D. SmithG. H. Gill E. P. VehslageM. M. Haertig F. J. VisichP. E. Hogin R. C. WertmanD. B. Holmes B. M. Wojciechowski

R. J. Zabransky

FINANCE COMMITTEEJ. B. Buckley, Chairman

HOSPITALITY COMMITTEEE. H. Ulm, Chairman

Horace Atwood, Jr., Vice-ChairmanA. J. BaracketJ. T. BrothersS. S. Calafati, Jr.E. L. Crosby, Jr.A. L. DolnickH. C. GawlerH. F. Heath, Jr.H. T. HeatonT. D. HinkelmanM. J. Hughes

R. J. IversenS. S. KrinskyH. C. McBrairWilliam O'MearaH. C. PacentA. W. Parkes, Jr.J. S. ReeseJ. L. RoemischMurray SimpsonR. A. Starek

H. A. Turnbull

HOTEL ARRANGEMENTS COMMITTEE

Bernard Farber, ChairmanWilliam Adler Abraham Sawelson

INSTITUTE ACTIVITIES COMMITTEEL. G. Cumming, Chairman

R. R. Batcher R. A. HeisingBernard Farber A. G. JensenE. K. Gannett D. E. Pike

PUBLICITY COMMITTEEE. K. Gannett, Chairman

Lewis Winner, Vice-ChairmanE. L. Bragdon B. F. OsbahrW. C. Copp Harold RenneH. E. Fry Bruce RobertsonT. R. Kennedy L. G. SandsW. W. MacDonald Crump Smith

E. C. Tompson

REGISTRATION COMMITTEEC. A. Mills, Chairman

J. S. BisbyW. P. FrantzA. W. KeenR. E. Lafferty

Vincent LearnedH. C. NelsonR. C. NewhouseD. S. Rau

TECHNICAL PROGRAM COMMITTEEL. T. DeVore, ChairmanB. R. Lester, Secretary

J. J. BaruchH. J. CarlinP. S. CarterL. M. ClementR. I. ColeC. H. Doersam, Jr.R. B. DomeG. A. EspersenRudolph FeldtJ. H. FelkerJ. F. HerrickJ. J. Hopkins

M. V. Kiebert, Jr.G. R. KilgoreD. D. KingA. A. MacDonaldNathan MarchandGeorge RappaportG. T. RoydenP. C. SandrettoHarald SchutzJ. R. SteenW. N. TuttleA. H. Waynick

Lewis Winner

WOMEN'S ACTIVITIES COMMITTEEMrs. R. F. Guy, Chairman

Mrs. P. B. Harkins, Vice-Chairman andTreasurer

Mrs. S. L. BaileyMrs. R. V. GouldMrs. R. W. HughesMrs. Frank MansurMrs. J. W. McRae

Mrs. Roger McSweenyMrs. Ogden PrestholdtMrs. Carl ScholzMrs. N. F. Smith, Jr.Mrs. Paul Waples

RegistrationMembers and visitors may register at

either the Waldorf-Astoria Hotel or GrandCentral Palace at the following hours:

Waldorf-A storia

Mon.9 A.M.-5:30 P.M.

Tue.9 A.M.-8 P.M.

Wed.9 A.M.-6 P.M.

Thur.9 A.M.-2:30 P.M.

Grand CentralPalace

10:30 A.M.-9 P.M.

9 A.M.-9 P.M.

9 A.M.-6 P.M.

9 A.M.A9 P.M.

Technical Sessions

Over 200 technical papers will be pre-sented in 43 sessions. A schedule of sessions

is listed on the following page. A listing ofpapers, authors, and 100-word summaries isgiven in the following pages.

Technical sessions will be held at theGrand Central Palace, 47th Street and Lex-ington Ave.; the Waldorf-Astoria Hotel,49th Street and Lexington Ave.; and theBelmont Plaza Hotel, across the streetfrom the Waldorf-Astoria.

ExhibitsThe Radio Engineering Show, featuring

405 exhibits of the most recent advances inthe radio-electronic field, will occupy fourfloors of Grand Central Palace. A list ofexhibitors and their products starts on

page IA of this issue. Exhibits will be open

during the following hours:Monday: 11:00 A.M.-9:00 P.M.Tuesday: 9:30 A.M.-9:00 P.M.

Wednesday: 9:30 A.M.-6:00 P.M.Thursday: 9:30 A.M.-9:00 P.M.

Annual MeetingIRE members will be particularly inter-

ested in the opening meeting of the conven-

tion, the Annual Meeting of the Institute, tobe held in the Grand Ballroom of theWaldorf-Astoria at 10:30 A.M. on Monday,March 23. The principal speaker will beWilliam R. Hewlett, IRE director and vice-president of the Hewlett Packard Co. Addedfeatures of the meeting will be the presenta-tion of the gavel of office to the incomingIRE president, James W. McRae, by hispredecessor, Donald B. Sinclair, and thepresentation of special pins to nine CharterMembers of the I.R.E.

(Continued on the following page)

1953

PROCEEDINGS OF THE I.R.E.

(Convention Program, cont'd)

Convention RecordPapers presented at the convention will

be published in a new publication, theCONVENTION RECORD OF THE I.R.E., whichwill be available about two months after theconvention. The CONVENTION RECORD willbe issued in ten parts, with each part con-taining papers relating to one general field.

Every member of an IRE ProfessionalGroup who has paid the Group assessmentfee by April 30, 1953, will automaticallyreceive, free of charge, that part of theCONVENTION RECORD which contains papersrelating to the field of interest of his Group.Information on how to join a ProfessionalGroup may be obtained by writing to theInstitute of Radio Engineers, Inc., 1 East 79Street, New York 21, N. Y., or by inquiringat the Professional Groups desk at the con-vention.

Full details concerning the contents andprices of CONVENTION RECORD parts will bepublished in the April issue of the PROCEED-INGS. Delivery cannot be guaranteed onorders received after April 30, 1953.

Social EventsA 'get together" co*cktail Party will be

held on the first evening of the convention,March 23, from 5:30 to 7:30 P.M. in thespacious Grand Ballroom of the Waldorf-Astoria. Tickets may be purchased fromInstitute headquarters at $4.00 each.Tickets will be on sale also at the convention.

The Annual Banquet, to be held in theGrand Ballroom on Wednesday, March 25,at 6:45 P.M., will feature an address byGeneral David Sarnoff, chairman of theboard of RCA and first recipient of thenewly established IRE Founders Award.His speech promises to be of great interest,not only to professional engineers, but to theentire electronics industry. Dr. A. M.Zarem, director of the Los Angeles divisionof Stanford Research Institute, will serve astoastmaster. President J. W. McRae willbestow the annual IRE awards to theirrecipients.

Ticket orders will be honored in theorder received. To ensure reservations,members are urged to write now to Instituteheadquarters. Tickets are available at $13.75each.

Women's ProgramAn attractive program of sightseeing

and tours has been arranged for wives ofIRE members and immediate members oftheir families. Among the events plannedwill be a "get acquainted' party, a tour ofthe United Nations and Lever Brother'sbuildings, luncheon and fashion show, and amatinee of the "King and I" or "WishYou Were Here."

SCHEDULE OF TECHNICAL SESSIONS

BELMONT-PLAZA WALDORF-ASTORIA GRAND CENTRAL PALACE

Moderne Room Grand Ballroom Astor Gallery Jade Room Gold Hall Blue Hall

Mon. Session 1 Session 2 Session 3 Session 4 Session 5 Session 6P.M. Antennas I- Television I Circuits I-Net- Electronic Com- Symposium: In- Radio Location,

General work Theory puters I strumentation I- Navigation and Air-2:30 Automatic borne Electronics

Tues. Session 7 Session 8 Session 9 Session 10 Session 11 Session 12A.M. Antennas II- Television II Circuits II-Sym- Electronic Com- Instrumentation Significant Trends

Microwave posium: Panel Dis- puters I I II-Symposium: in Airborne Equip-10:00 cussion on Wide- Transistor Meas- ment

band Amplifiers urements

Tues. Session 13 Session 14 Session 15 Session 16 Session 17 Session 18P.M. Antennas III- Symposium: Diag- Circuits III-Time Electron Devices Instrumentation Symposium:Trends

Propagation nostic Programs Domain Networks I-Transistors III-Electronics in Mobile Com-2:30 and Marginal -Delay Lines munications

Checking for LargeScale Digital Com-puters

Tues. Session 19Symposium: Elec-

8:30 tronics in Flight

Wed. Session 20 Session 21 Session 22 Session 23 Session 24 Session 25A.M. Electron Devices Circuits IV-Ac- Noise and Modu- Symposium: Tele- Quality Control Seminar: Acoustic,

II-Electron tive Networks- lation vision Broadcast- Methods Applied for the Radio EngiTubes Transistors ing to Electron Tube neer-I

10:00 and ElectronicEquipment Design

Wed. Session 26 Session 27 Session 28 Session 29 Session 30 Session 31P.M. Electron Devices Information The- Communications Symposium: Tele- Microwaves I- Seminar: Acoustic

III-Microwave ory I-Recent Ad- Systems vision Broadcast- Symposium: Man- for the Radio Eng.2:30 Tubes vances ing and UHF ufacture of Micro- neer II

wave Equipment

Starlight Roof Sert Room

Thurs. Session 32 Session 33 Session 34 Session 35 Session 36 Session 37A.M. Symposium: Information The- Medical Electron- Broadcast and Microwaves II- Radio Telemetry

Nucleonics ory II-Theoret- ics Television Receiv- Discontinuities10:00 ical ers-I and Transitions

Thurs. Session 38 Session 39 Session 40 Session 41 Session 42 Session 43P.M. Audio Engineering Information The- Broadcast and Microwaves III- Remote Control

Management ory III-Coding Television Receiv- Ferrites and De- Systems2:30 ers-II tectors

NoteConvention papers will be pub-

lished about two months after theconvention in the CONVENTION REC-ORD OF THE I.R.E. Each paid mem-ber of an IRE Professional Group(as of April 30, 1953) will receive, freeof charge, that part of the CONVEN-TION RECORD containing papers ofinterest to his group. Full details onprices will appear in the April issue ofthe PROCEEDINGS.

402 March

IRE National Convention Program

SUMMARIES OF TECHNICAL PAPERS

SESSION 1

Antennas I-General(Organized by the Professional Group on

Antennas and Propagation)Chairman, D. C. PORTS

(Jansky and Bailey, Washington, D. C.)

1.1. THE MEASUREMENT OF HIGHLYDIRECTIVE ANTENNA PATTERNSAND OVER-ALL SENSITIVITYOF A RECEIVING SYSTEMBY SOLAR AND COSMIC

NOISEJULES AARON

(Air Force Cambridge Research Center,Cambridge, Mass).

Utilizing the 100 narrow sources of radiofrequency energy and known data about theeffective diameter of the sun at variousranges of radio frequencies, a technique forplotting the directional characteristics oflarge antennas is outlined. Over-all systemsensitivies (receiver, antenna, and trans-mission lines) are checked by using valuesalready obtained for sky contours. Thegeneral receiver characteristics necessary forsuch measurements are outlined. Patternsof the sun and cosmic sources are illustratedand their analysis evaluated.

1.2. RADIATION PATTERNS FOR APER-TURE ANTENNAS WITH NONLINEAR

PHASE DISTRIBUTIONSC. C. ALLEN

(General Electric Co., Schenectady, N. Y.)A method of antenna pattern calculation

employing automatic punch-card machinesto perform numerical integrations has beenused to obtain sets of patterns for severalnonlinear phase distributions. These resultsextend previous work to the cases of quad-ratic and cubic phase distributions combinedin various proportions together with uni-form, tapered, and cosine amplitude dis-tributions. The magnitude patterns includethe first two or three side lobes, and phasepatterns are given for the main beams. Thesets of patterns are arranged to permit inter-polation which makes them useful in semi-quantitative pattern synthesis of which an

example is given. The need for arbitrarycriteria used heretofore in determining thesuitability of aperture distributions is thuseliminated.

1.3. FACTORS AFFECTING RADIATIONPATTERNS OF CORRUGATED

SURFACE ANTENNASM. EHRLICH AND L. NEWKIRK

(Hughes Aircraft Co., Culver City, Calif.)The radiation patterns of corrugated

metal-surface antennas have been computedin the past by an integration of tangentialelectric field over the surface of the antenna.Measured radiation patterns of these an-

tennas when fed by a horn, slot, or wire feedexhibit characteristic pertuibations fromthe computed theoretical value. These per-turbations are a reduced beam width, a mini-mum at the theoretical half-power point, andextremely high first-side lobes. It is conclu-

sively shown in the present study that thetotal pattern is a combination of the radia-tion pattern of the feed and metal surfaceconfiguration. The perturbations can be re-

duced to any desired degree by means of a

long transition section between feed and cor-

rugated surface array in which the feed modeis almost wholly transformed to the surfacemode; this has been demonstrated for a widevariety of aperture sizes and array lengths.Independent control of the relative phase ofthe two fields can be achieved by alteringthe length of the corrugated surface array.Variation of the ratio of the relative ampli-tudes can be achieved over a wide range bychanging the length of the transition sectionused to accomplish the mode transformationfrom feed to array mode. Thus a practicalflush-type feed producing negligible patternperturbations is now possible for corru-

gated surface or dielectric slab antennas.

1.4. A MICROWAVE ANECHOIC CHAM-BER MAKING USE OF A NEWBROAD-BAND ABSORBING

MATERIALA. J. SIMMONS AND W. H. EMERSON

(Naval Research Laboratory,Washington, D. C.)

A room lined with a broadband micro-wave absorbing material has been built inwhich radiation patterns of small antennasmay be measured at frequencies higher than1,000 mc. The material, developed at theNaval Research Laboratory, is made byapplying graphite in neoprene to a mat ofloosely spun animal hair to create a lossylow-dielectric constant medium.

The reflective properties of this materialand the evaluation of the room as an an-

tenna test site are presented.

1.5. WIDE-FREQUENCY-RANGE TUNEDCIRCUITS AND ANTENNAS

A. G. KANDOIAN AND WILLIAM SICHAK(Federal Telecommunication Laboratories,

Nutley, N. J.)Helical circuit elements, with or with-

out coaxial outer conductors, have import-ant network and antenna applications. Asresonant circuits, they are as useful at hfand vhf as conventional coaxial lines atuhf. Among their important character-istics are ability to tune over an extremelylarge frequency range and achieve high un-

loaded Q in a limited space.In antenna applications, it will be shown

that a resonant multiturn helix has moreradiation resistance and considerable lessloss than conventional small dipoles withequivalent apertures.

The discussion will include theory,fundamental properties, useful applications,and experimental data to bear out the theo-retical conclusions.

SESSION 2Television I

Chairman, 1. J. KAAR(General Electric Co., Syracuse, N. Y.)

2.1. THEORY OF SYNCHRONIZATION,APPLIED TO NTSC COLOR

TELEVISIONDONALD RICHMAN

(Hazeltine Corporation, Little Neck,L. I., N. Y.)

This paper presents the results of ananalytical evaluation of the performancecapabilities of the system used in NTSCcolor television to synchronize the color-carrier reference signal. The color syncburst appears to contain far more timingand synchronizing information than is re-quired, although previously used sync sys-tems have been inefficient in using the in-formation.

Analysis begins with determination ofthe amount of integration required for phasestability. The properties and limitations of"standard" passive and locked (APC)integrators are discussed. Integration re-quirements limit pull-in performance.

The basic principle for overcoming pre-vious limitations and obtaining the upperlimit of performance is explained, leadingto a determination of the ultimate capabili-ties permitted by the NTSC color sync sig-nals.

Simple techniques and new sync systemsfor approaching this limit are presented, anddiscussed.

The physical principles apply to syn-chronizing systems, generally.

2.2. COLOR SYNCHRONIZATION INTHE NTSC COLOR TELEVISION

RECEIVER BY MEANS OFTHE CRYSTAL FILTER

W. E. GOOD(General Electric Co., Syracuse, N. Y.)The problem of color synchronization in

the NTSC color television receiver meansgenerating a continuous-wave signal whichis in phase with the transmitted color burst.This 3.9-mc reference signal is then used toswitch the color samplers or demodulators.Of the various methods proposed for colorsynchronization, the automatic frequencycontrolled oscillator and the quartz-crystalfilter have worked out well in practice.

The crystal filter or ringing circuit is sup-plied with a gated-burst driving signal andthe resultant in-phase damped wave isamplitude-limited and used for the referencesignal. The circuit is passive in its operationand has given satisfactory performance dur-ing the color field tests. The design char-acteristics of the filter will be discussed aswill the various sources of phase errors. Theeffect of spurious modes in the crystal willbe pointed out. It will be compared with theAFC type of circuit.

2.3. APC COLOR SYNC FOR NTSCCOLOR TELEVISION

DONALD RICHMAN(Hazeltine Corporation, Little Neck,

L. I., N. Y.)This paper presents a description of the

characteristics and capabilities of a 'stand-ard" automatic-phase-control system ap-plied to NTSC color-carrier reference-phasesynchronization.

1953 403

PROCEEDINGS OF THE I.R.E.

Following a discussion of in-sync per-formance characteristics, a physical de-scription of the mechanism by which thesystem pulls into sync provides a back-ground for the relations between frequencypull-in range and time, and the in-sync char-acteristics such as noise bandwidth. Thesystem includes a nonlinear (sinusoidal)phase detector. An explanation of results ofa mathematical analysis, presented graphi-cally, emphasizes the upper limits of per-formance and how they may be obtained. Anumerical evaluation for NTSC color syncindicates over-all satisfactory performancewith this APC system.

2.4. TRANSIENT RESPONSE IN ACOLOR CARRIER CHANNELWITH VSB TRANSMISSION

J. S. S. KERR(General Electric Co., Syracuse, N. Y.)Two independent signals used to modu-

late a carrier in quadrature can be detectedwithout crosstalk only if the transfer char-acteristic of the network through which themodulation passes fulfills certain conditionsof symmetry. For vestigial sideband trans-mission-as proposed for the transmission ofchrominance information by the NationalTelevision Systems Committee-these con-ditions may not be fully met. In systems be-ing considered by the NTSC, residual cross-talk is eliminated either by CPA or by videofiltering.

Several types of vestigial-sideband trans-fer characteristics which are used in thetransmission of chrominance information areanalyzed and compared, both for minimumphase and linear phase. Their video inphaseand quadrature transfer characteristics areshown along with the transients which arisefrom a step input.

2.5. TRANSIENTS IN COLORTELEVISIONP. W. HOWELLS

(General Electric Co., Syracuse, N. Y.)A color television system transmits three

independent signals, each of which specifiesone of the three co-ordinates that determinethe location of the reproduced color in athree-dimensional color space. When a colortransient occurs, each of these signals re-sponds in a different manner determined bythe characteristics of its own channel. Thesystem response may be characterized by theresulting path along which the reproducedcolor point moves through the color spacefrom its initial to its final location. Theshape of such color transient paths as deter-mined by the individual transient responsesof the three channels is analysed, and thesubjective appearance of different transient-path shapes is discussed. Various modifica-tions of the NTSC proposals are comparedby these methods.

SESSION 3Circuits I-Network Theory(Organized by Professional Group on

Circuit Theory)Chairman, S. DARLINGTON

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.)

3.1. A GENERAL RLC SYNTHESISPROCEDURE

Louis WEINBERG(Hughes Aircraft Co., Culver City, Calif.)Any physically realizable transfer func-

tion (impedance, admittance, or dimension-less ratio) can be realized within a multi-plicative constant by the new synthesis pro-cedure presented in this paper. The form ofnetwork achieved is a lattice with the follow-ing significant features: (1) The lattice mayhave any desirable termination. (2) Thelattice contains no mutual inductance.(3) Every inductance in the network ap-pears with an associated series resistance sothat, in building the network, low-Q coilsmay be used.

In addition, the lattice arms relative toeach other are of so simple a form as torender many of the achieved lattices amen-able to reduction to an unbalanced network.For the case of a transfer admittance, more-over, reduction always can be achieved with,at most, the use of real transformers, that is,transformers with winding resistance, finitemagnetizing inductance, and a couplingcoefficient smaller than one.

3.2. A GENERAL THEORY OF WIDE-BAND MATCHING

H. J. CARLIN AND RICHARD LAROSA(Polytechnic Institute of Brooklyn,

Brooklyn, N. Y.)A general treatment is given for the de-

sign of wide-band matching networks whichmay contain dissipation. Loss is intention-ally introduced to permit a degree of matchunobtainable with lossless matching net-works. For any prescribed input reflectionmagnitude including zero (that is, perfectmatch) the method allows the most efficientnetwork configuration to be determined.

From scattering considerations, simple,easily interpreted relations are presentedwhich predict the theoretical limitations onthe two quantities of engineering interest:the magnitude of input reflection coefficient,and the power which reaches the load. Thebest two-terminal pair network from thestandpoint of minimum insertion loss (nor-malized to available generator power) isshown to contain no more than one resistor.

3.3. SYNTHESIS OFELECTRICFILTERSWITH ARBITRARY PHASE

CHARACTERISTICSB. J. BENNETT

(Stanford Research Institute,Stanford, Calif.)

Conventional synthesis methods for de-sign of electric filters generally follow thepattern of approximating a desired attenu-ation characteristic using a minimum-phasenetwork, and then if the phase characteristicis undesirable, a phase-corrective all-passnetwork is connected in tandem. In con-trast, the design theory introduced in thispaper deals directly with the over-all inser-tion-transfer function, in general, a non-minimum-phase function, which approxi-mates, at once, the attenuation and phasecharacteristics desired. The network is thenrealized as a unit on an insertion-loss basis.

For a nonminimum-phase function, at-tenuation and phase characteristics are, to alarge extent, independent of each other, and

Lee's and Bode's relations between real andimaginary components of network functionsare not violated. It is shown that if theprescribed phase characteristic is attainableby an all-pass transfer function, it may berealized exactly in a network which alsopossesses an attenuation characteristicwhich is a close approximation to a pre-scribed filter attenuation characteristic.

3.4. WIDE-BAND FILTER AMPLIFIERSAT ULTRA-HIGH FREQUENCIES

W. A. CHRISTOPHERSON(IBM Corp., San Jose, Calif.)

D. 0. PEDERSON and J. M. PETTIT(Stanford University, Stanford, Calif.)This paper describes the development of

a stagger-tuned, band-pass amplifier at ul-tra-high-frequency using grounded-grid tri-odes and having a prescribed gain magnituderesponse. For the first time, the concepts ofstagger tuning have been extended togrounded-grid cascades so that triodes de-signed for uhf operation can be effectivelyutilized to obtain amplifiers having largegain-bandwidth products and low-noise fig-ures.

The amplifier described uses high per-formance, disc-seal triodes with special four-terminal interstages. The special interstagesaccomplish the impedance transformationneeded for grounded-grid amplifiers and inaddition have a frequency characteristicwhich is approximately that of a single-tunedcircuit. Because of this frequency character-istic, stagger-tuning is employed to conservegain-bandwidth product; however, the na-ture of the characteristic requires the use ofnovel low-pass to band-pass transformationsto determine the correct interstage tuningsfor a "maximally-flat" type of response. Thefeedback effects occurring between adjacentinterstages are incorporated in the tuningprocedure.

The design and performance of an ampli-fier having a gain of 50 db, a bandwidth of 50mc, and a maximally-flat gain characteristicwill be presented.

3.5. NETWORK ANALYSIS WITH THEAID OF GENERATING

POLYNOMIALSHERBERT KURSS

(Polytechnic Institute of Brooklyn,Brooklyn, N. Y.)

The technique emphasized here is the rep-resentation of a finite matrix by a polyno-mial whose coefficients are the various minorsof the matrix. This "generating polynomial, "determined by a vanishing determinant, ef-fectively defines a network in terms of its in-put-output relations. Two networks are thenequivalent with respect to a specified class ofterminations if the respective generatingpolynomials have proportional coefficients.For simplicity, this is illustrated for micro-wave networks where the relevant form ofthe Jacobi ratio theorem, the Campbell for-mula (for the "elimination of concealed cir-cuits"), and a compound network theoremare all simply derived.

3.6. TWO NEW EQUATIONS FOR THEDESIGN OF FILTERS

M. DISHAL(Federal Telecommunication Labora-

tories, Inc., Nutley, N. J.)

404 March

IRE National Convention Program

Two basic problems in filter design havebeen the finding of element values in laddernetworks of inverse arms (or equivalentstructures) to produce Chebishev responsefor two terminal conditions: (a) loading onboth ends of the network, (b) loading ononly one end of the network.

These problems have now been solvedfor the general case of any number of ele-ments for both the Chebishev response andits limiting case of the Butterworth response.

The complete design information is con-tained in two simple equations that can re-place those of classical filter theory. Theseequations and their derivations are pre-sented.

SESSION 4Electronic Computers I

(Organized by Professional Group onElectronic Computers)Chairman, J. H. FELKER

(Bell Telephone Laboratories,Whippany, N. J.)

4.1. MULTICHANNEL ANALOG INPUT-OUTPUT CONVERSION SYSTEM

FOR DIGITAL COMPUTERP. A. ADAMSON AND M. L. MAcKNIGHT

(Hughes Aircraft Co., Culver City,Calif.)

The conversion of dc voltages to binarynumbers, and vice versa, is performed inde-pendently of computer operation, the mag-netic drum serving as storage for the mostrecently sampled data. Input voltages arecompared by a multiar-type comparatorwith an accurately linear saw-tooth wave-form, gated once per revolution of the drumto successive inputs. Clock pulses are countedfrom the start of the ramp until comparisonis reached and the resulting binary number isshifted to the memory. The ramp slope is cal-ibrated once per cycle of inputs, making thescale factor virtually independent of drumspeed, temperature, and supply-voltagechanges.

Output voltages are derived from a fil-tered rectangular waveform whose symme-try is made proportional to the output bi-nary number. One cycle of the waveform isrecorded on a drum channel during an out-put sample period, and the waveform is con-tinuously read to a regulated current switchtube and filter every revolution. Several out-puts may be time shared on a single-drumchannel. Input and output conversions havean accuracy of about ± 0.1 per cent full scale.

4.2. AN ANALOG TO DIGITAL CON-VERTER WITH AN IMPROVEDLINEAR SWEEP GENERATOR

D. W. SLAUGHTER(California Institute of Technology,

Pasadena, Calif.)This paper describes an analog-to-digital

converter with 1 part in 5,000 (.02 per cent)accuracy and excellent long-time stability.Sampling speeds of up to 100 per second athigh accuracy are readily obtainable. Thisconverter utilizes the familiar circuitry inwhich an analog voltage is represented by atrain of pulses initiated at the start of a lin-ear sweep and ending when the magnitude of

the sweep and the analog voltage coincide.The heart of the high-accuracy sweep gener-ator is an electronically gated operationalamplifier. Optional automatic self calibrat-ing can provide exceptional long-time stabil-ity. A stable circuit for detecting the mag-nitude coincidence is also presented.

4.3. DYNAMIC BINARY COUNTER WITHANALOG READ-OUT

LEROY PACKER(Columbia University, New York, N. Y.)A binary counter using circulating pulse

techniques is described. The counting rate isone mc and the carry time per stage is ap-proximately 0.02 its. Information can be readout of the counter every ,us since the counteris not stopped when a sampling pulse ap-pears. The read-out mechanism consists of aparallel register, associated relays, and an acfeedback amplifier which provide both paral-lel binary and analog outputs equal to thetime between the start of the counting cycleand the appearance of a sampling pulse.

4.4. LIFE AND RELIABILITY EXPERI-ENCE WITH TRANSISTORS IN A

HIGH SPEED DIGITALCOMPUTERJ. J. SCANLON

(Bell Telephone Laboratories,Whippany, N. J.)

This experience is based on the use of ap-proximately 500 semiconductor diodes aslogic elements and 78 transistors as the onlyactive elements in a binary multiplier. Themachine operates at a pulse repetition rate ofone megacycle and can multiply two 16 digitbinary numbers in 272 Asec. The maximumpotential used is 8 volts with a resultant to-tal battery drain of less than 5 watts. Ap-proximately the same amount of masterclock power is also required.

Approximately 3,000 hours of continuousoperation (234,000 transistor hours) havebeen logged with only one transistor failureat 1,430 hours.

Freedom from random error has beendemonstrated by circulating binary informa-tion, without error, in closed loops within themachine for hundreds of hours. In one case a700 hour run was terminated by accidentalpower failure. For each hour of operationeach transistor must make 3.6X 109 success-ful discriminations.

Periodic voltage margin checks have dis-closed no aging effects.

4.5. ENGINEERING EXPERIENCE INTHE DESIGN AND OPERATION OF

A LARGE SCALE ELECTRO-STATIC MEMORY

J. LOGUE, A. BRENNEMANN, ANDA. KOELSCH

(IBM Engineering Laboratory,Poughkeepsie, N. Y.)

The IBM Type-701 Electronic DataProcessing Machine uses an electrostaticstorage unit. This paper will describe the en-gineering considerations encountered in thedesign and operation of the circuits involved.The video amplifier will be described andfactors involved in setting the amplitude-discriminating level of the video amplifierwill be treated in some detail. The deflectioncircuits will be discussed, together with the

special requirements of electrostatic storageand how these requirements were met. Theproblems encountered when a read-aroundratio of 200 to 1, or larger is required, will beincluded in the discussion.

SESSION 5: SYMPOSIUMInstrumentation I-

Automatic(Organized by Professional Group

on Instrumentation)Chairman, ERNST WEBER

(Polytechnic Institute of Brooklyn,Brooklyn, N. Y.)

5.1. A NEW METHOD FOR MEASURINGNOISE FIGURE AND GAIN OF A

RADAR RECEIVERR. J. PARENT AND V. C. RIDEOUT

(University of Wisconsin, Madison, Wis.)A new method has been developed for the

measurement of the gain-and-noise figure ofa radar receiver. In this method a pulsed gas-discharge noise source at the input and a co-herent detector at the output are used. Theuse of a gas-discharge noise source has theusual advantages over a small monochro-matic-signal source that output level andfrequency need not be adjusted. The sensi-tivity of the coherent detector overcomes thedifficulty ordinarily encountered with thelow-output level of this noise source. The useof the coherent detector requires that thenoise source be pulsed, which in turn givesother advantages.

The scheme described is adaptable toeither quantitative measurement or a go-no-go indication based on some set limits.

An adaptation of this method to themeasurement of noise figure and gain of aradar receiver while the radar is in normaloperation has also had successful prelimi-nary tests.

5.2. AUTOMATIC INSTRUMENTATIONFOR CONTINUOUS MONITORINGOF SYSTEMS PERFORMANCEM. V. RATYNSKI, MILTON KANT,

AND HAYWOOD WEBB(Rome Air Development Center,

Rome, N. Y.)The utilization of automatic instrumen-

tation to facilitate the measurement of per-formance parameters of electronic systems isdiscussed. A comparison is made between ex-isting measurement methods and recentlydeveloped techniques.

Techniques are described for continuousand automatic measurement of (1) receivernoise figure and gain-bandwidth product,(2) transmitter power output, (3) constantvelocity servosystem performance, (4) trans-mission line vswr. These measurements aremade while the system whose performance isbeing monitored is in operation, and advan-tages of incorporating these items in typicalsystems are demonstrated.

5.3. AUTOMATIC ONE-SHOT METHODSFOR BANDWIDTH MEASUREMENT

J. B. WOODFORD, JR., ANDE. M. WILLIAMS

(Carnegie Institute of Technology,Pittsburgh, Pa.)

1953 405

406PROCEEDINGS OF THE I.R.E.

The problem of determining a simplebandwidth acceptability criterion which canbe reduced to a number or set of numberswhich adequately emphasizing band width-related performance, is surveyed. Typicalone-shot measurement systems for go-no-goindication are summarized; results of ex-perimental tests on prototype and operatingradar and communication systems are de-scribed.

5.4. MICROWAVE POWER METERWITH AUTOMATIC ZEROSETTING AND TELE-

METERINGL. A. ROSENTHAL AND G. M. BADOYANNIS(Rutgers University, New Brunswick, N. J.)

An improved self-balancing bridge type'of microwave-power meter capable of auto-matically setting its zero, to compensate fortemperature variations, and telemeteringthe measured power is described. Automaticzeroing is accomplished by shutting off theRF power for a regular short interval duringwhich time a servomechanism adjusts thepower level so that the meter reads zero.Frequency-modulation telemetering pro-vides a direct power indication at a removedposition. The system will allow for the re-mote monitoring and recording of powerlevel in a microwave installation.

Operating principles and design of indi-vidual units are described and performancedata for the complete system presented.

5.S. MONITORING OF ERRORS INSYNCHRO SERVO SYSTEMS

GIORGIO QUAZZA(Polytechnic Institute of Brooklyn,

Brooklyn, N. Y.)In synchro-servo systems transmitting

positional information at constant or slowlyvarying speed of the input member, the con-trol transformer-output voltage depends onthe speed, and therefore is not a direct meas-ure of the system positional error. Differentpassive networks, to be inserted in serieswith the ct stator windings, are proposed tocompensate for the system output voltagevariations due to speed and permit monitor-ing of the system error by direct measure-ment of the ct output voltage. Formulas fortheir design are derived and analytical andexperimental results given, to describe theirrelative merits and the effect of temperaturevariations on their performance.

SESSION 6Radio Location, Navigation and

Airborne Electronics(Organized by Professional Group on

Airborne Electronics)Chairman, P. C. SANDRETTO

(Federal Telecommunications Laboratories,New York, N. Y.)

6.1. THIE TECHNIQUE OF MONO-PULSE RADAR

W. HAUSZ(General Electric Co., Syracuse, N. Y.)The monopulse technique in radar con-

sists in deriving sufficient information on asingle pulse by multiple simultaneously act-ing receiving channels to determine com-pletely both the angular position and therange of a target or targets. Two commonly

used variants are phase comparison and am-plitude comparison. The informational as-pects of both of these, and of monopulsemore generally compared with scanning tech-niques for angular determination, is given.

6.2. REDUCING SKY WAVE ERRORS INCW TRACKING SYSTEMS

M. S. FRIEDLAND(Air Force Missile Test Center, Patrick

Air Force Base, Fla.)NATHAN MARCHAND

(Electronics Laboratory, Greenwich, Conn.)A cw tracking system is very susceptible

to sky-wave interference and errors. Once areceived cw signal is contaminated by skywave, it is impossible to separate out the er-ror. When the possibility of contaminationexists, it is felt that the system should bemodified, and the basic theory underlying amodification by pulsing the transmitted fre-quency is given. In a system employed at theAir Force Missile Test Center. the transmit-ter is airborne. The necessary pulse lengthsand repetition rates, as well as other designdata, are determined by the ranges to be cov-ered, frequencies to be employed, and theionospheric layer heights. The interdepend-ence of these factors are detailed, modifica-tions applied to a Raydist system, and dis-cussed.

6.3. AN APPLICATION OF INTEGRATORTYPE SIGNAL ENHANCER TO

DIRECTION FINDINGEQUIPMENTS

C. A. STROM AND J. A. FANTONI(Rome Air Development Center,

Rome, N. Y.)This paper summarizes the results ob-

tained in applying integrating type of signalenhancers to certain radio direction-findingequipments and to other repetitive type sig-nals such as may be displayed on an "A"scope. The integrators used include systemsemploying magnetic storage units such as atape recorder, delay lines techniques such asmagnesium delay lines, and capacitor stor-age elements. The paper shows that improve-ments in signal-to-noise ratio of 12 db is eas-ily obtained and that improvements up to 20db or more are obtainable by use of positivefeedback around the integrator.

6.4. A THEORY OF TARGET GLINT ORANGULAR SCINTILLATION IN

RADAR TRACKINGR. H. DELANO

(Hughes Aircraft Co., Culver City, Calif.)A theory is presented to describe the sta-

tistical aspects of tracking a complex iso-lated structure, such as an aircraft or navalvessei, by radar. The results are expressiblein simplest form when the target subtends anangle small compared with the beamwidth.When the angle subtended by the target issmall, a single description applies to all radartracking systems. An instantaneous and aneffective target displacement from the meanare defined and their statistical propertiesderived. Special treatment is gi'ven to addi-tional noise arising in conical scanning due toamplitude fluctuations as such. The theoryprovides information relating to the spectraas well as to the probability densities andrms values of the pertinent quantities.

6.5. AUTOMATIC DEAD RECKONINGNAVIGATION COMPUTERS FOR

AIRCRAFTJ. L. DENNIS

(Wright Air Development Center,Dayton, Ohio)

Two airborne navigation computersbased on the principles of dead reckoningare described. The first, the Type A-1Ground Position Indicator, is a computerwhich operates from inputs of true airspeedand magnetic heading. Wind can be insertedmanually in the form of magnitude and di-rection. Position in latitude and longitudecan be set initially or reset upon sighting acheck point, but is maintained current ex-cept for the errors in input data.

The second computer, Ground PositionComputer AN/APA-58, has additional fea-tures. A set of counters are provided for des-tination latitude and longitude. Further-more, a range and azimuth mark generatorbased on signals from the difference in lati-tude and longitude of the present positionand destination are provided to a search ra-dar set, such as the AN/APS-42. These crosshairs appear on the radar indicator. If pres-ent position is correct the cross hairs will bein coincidence with the position of the radartarget representing the destination.

SESSION 7Antennas II-Microwave(Organized by Professional Group on

Antennas and Propagation)Chairman, L. C. VAN ATTA

(Hughes Aircraft Co., Culver City, Calif.)

7.1. ARRAYS OF FLUSH MOUNTEDTRAVELING WAVE ANTENNAS

J. N. HINES, V. H. RUMSEY AND T. E. TIcE(Ohio State University, Columbus, Ohio)The object of this paper is to present and

discuss the results of an investigation ofproblems that arise in the design of arraysof tapered-depth traveling-wave slot anten-nas.

Past experience with arrays of slots showsthat the conventional method of array de-sign has sometimes failed to predict the pat-tern of the array with sufficient accuracy forpractical purposes. An absolutely correct butmore complicated design technique based onthe principle of superposition has been devel-oped and tested.

The results of experimental measure-ments showing the effect of coupling be-tween adjacent elements are presented andarray patterns based on both design proce-dures are compared.7.2. TRANSIENT BUILD-UP OF THIEANTENNA PATTERN IN END-

FED LINEAR ARRAYSN. H. ENENSTEIN

(Hughes Aircraft Co., Culver City, Calif.)In linear antenna arrays that are fed

from one end, there is a finite transit time forthe wave group traveling from the first ra-diator to the last radiator of the array. As aresult the transient antenna pattern may bc.distorted from the pattern determined on acontinuous-wave basis. If the array is used ina communication system, this distortion be-comes appreciable when the reciprocal of the

406 March

IRE National Convention Program

signal bandwidth and the transit time arecomparable. If it is used in a radar system,the effect on the pattern is appreciable whenthe system pulse length and array transittime are comparable. Approaches to theproblem in the time and frequency domainsare developed in this paper. Examples of theeffect on angular discrimination of radar sys-tems due to this phenomena are given.

7.3. A NEW MICROWAVE REFLECTORK. S. KELLEHER

(Naval Research Laboratory,Washington, D. C.)

A reflecting surface is discussed which isformed by the rotation of an arc of a parab-ola about a line parallel to the latus rectum.It is shown to be superior to the parabo-loidal and spherical reflector for applicationswhere plane motion of feed horn is desired.

Besides the obvious application of thisreflector as a scanning antenna, it is shownto be useful as a "beam bender" in a micro-wave-relay link. Other possible applicationsinclude its use in marine radar for reductionin vertical-antenna beamwidth and conse-quently in transmitted power.

7.4. CROSSTALK IN RADIO RELAYSYSTEMS CAUSED BY FORE-GROUND REFLECTIONS

H. W. EVANS(Bell Telephone Laboratories, Inc.,

New York, N. Y.)Measurement of 19 horn-lens antennas of

the New York-Chicago TD-2 radio relaysystem indicated that the rms front-to-backratio was 10 db poorer than that of a proto-type antenna previously measured. Radar-type measurements showed these degrada-tions were caused by reflections from trees,buildings and hills, to a degree which had notbeen anticipated. Geographical studies con-firmed the reflecting media. These results seta limit to transmission improvement fromhigher power and lower noise figure in sys-tems using two-frequency allocations, andraise new limitations on the selection of radiorelay repeater sites.

7.5. LOW SIDE LOBES IN PENCIL-BEAM ANTENNAS

E. M. T. JONES(Stanford Research Institute,

Stanford, Calif.)A theoretical and experimental investi-

gation has been performed on the parabo-loid-reflector antenna and the isotropic-dielectric hyperboloid-lens antenna to deter-mine the depth of principal-polarization sidelobes and the depth of cross-polarized lobesthat can be obtained in practice. The analy-sis determines the aperture fields of thesetwo aplanatic devices when they are excitedat their foci by short electric dipoles, shortmagnetic dipoles and plane-wave sources. Itis found that no cross-polarized aperturefields are obtained for the reflector when it isexcited by a plane-wave source, and none areobtained for the lens when it is excited by ashort electric dipole. The experimental re-sults show that a reduction of at least 10 dbin cross-polarized radiation field of theparaboloid reflector can be obtained when itis excited by a horn, which approximates aplane-wave source, instead of by a Cutler

feed, which approximates an array of mag-netic dipoles. Principal-plane side lobes of-39 db and cross-polarized lobes of -28 dbhave been obtained with the hyperboloid-lens antenna.

SESSION 8Television II

Chairman, A. V. LOUGHREN(Hazeltine Electronics Corporation,

Little Neck, N. Y.)8.1. PROBABILITY DISTRIBUTION

MEASUREMENTS OF TELE-VISION SIGNALSW. F. SCHREIBER

(Harvard University, Cambridge, Mass.)A device was constructed which, in ef-

fect, produces two video signals from a pic-ture in a flying-spot scanner, the two signalsbeing derived from points separated an arbi-trary amount in any direction. Another de-vice was constructed which displays thejoint-amplitude probability density of thetwo signals as a two-dimensional brightnesspattern on a cathode-ray tube. Qualitativeeffects can be studied by direct observationor photography of this pattern. Quantitativeresults, which include a computation of thesecond-order approximation to the entropyof the television signal, have been made bymeasuring the brightness pattern with ascanning photometer.

8.2. A PRECISION LINE SELECTORFOR TELEVISION USE

I. C. ABRAHAMS AND R. C. THOR(General Electric Co., Syracuse, N. Y.)This paper describes the design and

operation of a versatile measuring instru-ment for use in television laboratories andstations. This precision-line selector has nu-merable useful purposes which are described.

The instrument consists of a precision-phase shifter, operating at ten times thestandard television line-repetition fre-quency, that is, 157.5 kc. This is then di-vided down to 30 cycles. Hence, variablephase shift or delay is obtained at 30 cycles,having the absolute accuracy and stabilityof that obtained at 157.5 kc.

By proper use of the phase shifter alone,the instrument may also be used to measuretime delays to accuracies of ±0.001 h. Inaddition, the frequency divider chain hasmany uses, and descriptions of their applica-tion will be given.

8.3. COLORIMETRIC PROPERTIES OFGAMMA-CORRECTED COLOR

TELEVISION SYSTEMSD. C. LIVINGSTON

(Sylvania Electric Products Inc.,Bayside, N. Y.)

Through the use of a set of system pa-rameters which measure individual proper-ties of a color television system as a functionof the chromaticity of the color viewed bythe camera, there will be presented analysesof the performance of the NTSC color tele-vision system in its present form and in sev-eral related forms. Particular attention willbe given to color fidelity and adherence tothe constant-luminance principle, includingsusceptibility of the displayed luminance on

a color kinescope to spurious signals in thechrominance channel. These analyses willtake into account several recently proposedsystem modifications.

8.4. PHASE MEASUREMENTS AT SUB-CARRIER FREQUENCY IN COLOR

TELEVISIONA. P. STERN

(General Electric Co., Syracuse, N. Y.)For reliable adjustment and checking of

the transmitter and receiver in the NTSCcolor television system, the possibility of ac-curate phase measurements at subcarrier fre-quency is of primary importance. This paperdescribes the principles and operation ofphase measuring equipment recently built inthe electronics laboratory of the electronicsdivision of the General Electric Company.

Very accurate measurements can be ob-tained by phase shifting at low frequencyand heterodyning to subcarrier frequency.The accuracy is essentially limited by insta-bilities in the equipment. The over-all erroris estimated to be less than 1.00.

Some methods employed in using the in-strument to measure subcarrier-phase accu-racy of a color signal generator are described.

8.5. A MONITORING SYSTEM FORNTSC COLOR TELEVISION

SIGNALSC. E. PAGE

(Hazeltine Corporation, Little Neck, N. Y.)The advent of the NTSC color television

signal on a commercial basis will introduce anew problem in signal monitoring. The nor-mal television monitor which displays sig-nal amplitude versus time provides only afraction of the information required forchecking the chrominance portion of the sig-nal. This paper describes an equipmentwhich displays on a cathode-ray oscilloscopethe phasor diagram of the chrominancecomponent of NTSC color television signal.This type of display permits rapid visualchecking of the chrominance portion of thesignal and is equally suitable for signalmonitoring service at the transmitter,studio, or color receiver production line. Inaddition the visual display greatly facilitatesthe correct alignment of NTSC encodingequipments.

The equipment consists basically of apair of quadrature demodulators whose out-puts are fed respectively to the horizontaland vertical plates of an oscilloscope. Theequipment described in this paper includesrefinements which make it largely self-checking and facilitate rapid operation.

SESSION 9: SYMPOSIUMCircuits II-Panel Discussionon Wide-Band Amplifiers(Organized by Professioqal Group on

Circuit Theory)Chairman, H. A. WHEELER(Wheeler Laboratories, Inc.,

Great Neck, N. Y.)

9.1. CONVENTIONAL WIDE-BANDAMPLIFIERSW. E. BRADLEY

(Philco Corporation, Philadelphia, Pa.)

1953 407

PROCEEDINGS OF THE I.R.E.

Progress in the design of conventionalwide-band amplifiers has continued alongthree principal lines: (1) Tubes are now

available with improved figure of merit, in-ternal shielding and reliability; (2) Usingnew components, layouts are available com-bining mechanical ruggedness and repro-ducibility with clean electrical character-istics; (3) Extensive development and wide-spread use of complex-frequency plane-design methods has led to high performanceand design flexibility combined with circuitsimplicity.

9.2. BROAD-BAND FEEDBACKAMPLIFIERS

H. N. BEVERIDGE(Raytheon Manufacturing Co.,

Newton, Mass.)Broad-band video and IFamplifiers using

conventional tubes and employing resistivefeedback from plate to grid will be con-sidered.

A qualitative analysis of the broad-band-ing effects of this type of feedback will bepresented. The simplest case, a two-tubefeedback pair, will be considered first, andthe discussion extended to feedback triples,infinite chain, and double tuned. Practicalproblems in the feedback path due tocapacity and effective transit angle will bediscussed.

The results of measurements of gainbandwidth product will be presented andcompared with other types of amplifiers.

9.3. TRANSISTOR AMPLIFIERSR. L. WALLACE, JR.

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.)

Mr. Wallace will discuss some of theproperties of transistors which result inlimitation of bandwidth and will indicatethe order of performance which has beenachieved.

9.4. WIDE-BAND DISTRIBUTEDAMPLIFIERS

W. G. TULLER AND E. H. BRADLEY(Melpar, Inc., Alexandria, Va.)

Distributed amplifiers using availablemultigrid tubes are limited to operation be-low approximately 400 mc. However, dis-tributed amplifiers containing triode pairshave been developed having a 3-db band-width of approximately 1,000 mc. Theaparaphase" distributed amplifier, as thenew triode circuit is called, utilizes the im-proved high-frequency characteristics of tri-odes.

This paper discusses the achievable gainand bandwidth characteristics for both typesof amplifier. The limitations encountered inthe development of these distributed sys-tems are described and evaluated. The futuretrend in the development of distributedamplifiers also is considered.

9.5. TRAVELING-WAVE ANDRELATED TUBES

L. M. FIELD(Stanford University, Stanford, Calif.)Circuit, electronic, and matching limita-

tions on the wide-band amplifying proper-ties of helix-type tubes in the range of fre-

quencies from 50 mc through the microwaverange will be reviewed. Bandwidths of fromone to several octaves in this range havebeen obtained and typical examples and de-sign criteria will be given. Bandwidth andbandtuning of several related types of micro-wave tubes using other than a helix for wavepropagation or nonpropagating structureswill also be discussed.

SESSION 10

Electronic Computers II

(Organized by Professional Group onElectronic Computers)Chairman, J. R. WEINER

(Eckert-Mauchly Computer Co.,Philadelphia, Pa.)

10.1. ANALOG COMPUTING WITHMAGNETIC AMPLIFIERS USINGMULTIPHASE AC VOLTAGES

J. E. RIcHARDSON(Hughes Aircraft Co., Culver City, Calif.)By the use of multiphase-ac voltages,

simple methods are explained by whicharithmetic operations, such as multiplica-tion and division, are achieved. Also many

algebraic and transcendental functions, suchas the general second-degree equation andtrigonometric, hyperbolic, and exponentialfunctions, are generated. The general prob-lem of two-dimensional co-ordinates istreated with relation to transfer and rotationof co-ordinate axes, as well as the resolutionof an arbitrary vector with respect to a

selected co-ordinate system. The instrumen-tation involves two basic computing unitswhich contain no vacuum tubes, being in-strumented with magnetic-core components.

10.2. SOME RECENT DEVELOPMENTSIN LOGICAL "OR-AND-OR" PYRA-MIDS FOR DIGITAL COMPUTERS

CORNELIUS LEONDES(Moore School of Electrical Engineering,

Philadelphia, Pa.)AND

MORRIS RUBINOFF(University of Pennsylvania,

Philadelphia, Pa.)This paper first reviews, briefly, the

design methods commonly employed inlogical "or-and-or" pyramids for digitalcomputers. It then describes some recentdevelopments in the design of these circuits.

The following important advantages arethen seen to result from the new design tech-niques. (1) Fewer Germanium diodes areneeded. (2) The largest voltages needed maybe reduced by factors of as much as two.(3) The power needed in the pyramid may

be reduced as much as one-half or less. As adirect result the sizes of the resistors neededare also reduced. (4) The resulting rise-and-fall times are speeded up by an appreciablefactor. (5) The impedance levels at the vari-ous inputs to the pyramid go up in somecases by factors of as much as two or more.This makes it possible to drive more circuitswith any given pulse power amplifier.(6) The clock pulse power needed to driveany input may be reduced as much as one-half or less. The paper concludes with an

illustrative example.

10.3. MAGNETIC CORE SWITCHES ASLOGICAL ELEMENTS IN

COMPUTERSE. A. SANDS

(Magnetics Research Co.,Chappaqua, N. Y.)

The use of rectangular hysteresis-loopmagnetic cores as general purpose logicalelements in computers will be discussed.Advantage is taken of the amplitude limit-ing properties of the magnetic hysteresisloop to make operation of the cores ex-

tremely uncritical with respect to variationsin input currents and voltages. Severalalternative ways of generating logical "or"(mixer) and "and" (gating) functions will beshown. Methods of designing cores to oper-

ate at a given speed and into a given loadwill be pointed out. An all magnetic com-puter using drums for low-speed storage,magnetic memory elements for high-speedstorage, magnetic shift registers for input-output buffer storage, and magnetic core

switches for arithmetic and selection units isa realizable possibility.

10.4. MAGNETIC-SHIFT REGISTERUSING ONE CORE PER BIT

S. RUHMAN, W. D. WOO, AND R. D. KODIS(Raytheon Manufacturing Co.,

Waltham, Mass.)Conventional magnetic-shift registers

use two cores and two to four diodes per bit,and require two-shift pulse sources displacedin time. A new circuit utilizing a condenserfor temporary energy storage between cores

permits the use of a single core, and a singlediode per binary digit requires only one-shift pulse source, and provides greaterstability of operation. The principle ofoperation is described, an approximateanalysis of the circuit is given, and experi-mental results are presented.

10.5. A SIMPLE COMPUTER FORAUTOMATICALLY PLOTTINGCORRELATION FUNCTIONS

A. H. SCHOOLEY(Naval Research Laboratory,

Washington, D. C.)A simple analog computer is described

which has proved useful as a research tool inevaluating the autocorrelation and cross-correlation functions of various analyticaland experimental time functions. The inputfunctions to be correlated are fed into thecomputer as two loops made from 35-mm.motion-picture film. Provision is made forautomatically varying the delay of one loopwith respect to the other. The output cor-relation function is automatically plotted ona commercial paper recorder. Correlationfunctions for several analytical and experi-mental functions are given.

SESSION 11: SYMPOSIUMInstrumentation-Transistor

Measurements(Organized by Professional Group

on Instrumentation)Chairman, G. M. ROSE, JR.

(Radio Corporation of America,Harrison, N. J.)

408 March

IRE National Convention Program

11.1. TRANSISTOR METROLOGYD. A. ALSBERG

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.)

Existing transistor test methods, condi-tions, accuracies, and apparatus will be re-

viewed. These will be evaluated in termsof the needs of the transistor measurementclientele. Principles will be indicated whichmay lead to the reduction of the presentgreat variety of methods and apparatus, bystressing absolute measurement and thoseparameters and methods which best satisfythe practical needs of the vast majority.

11.2. MEASUREMENT OF TRANSISTORPARAMETERS BY CRO AND

OTHER METHODSW. E. MORROW, JR.

(Massachusetts Institute of Technology,Cambridge, Mass.)

Equipment has been designed for thepresentation of large signal transistor char-acteristics. The collector (Z22) and base(Z12) characteristics are automaticallyplotted on the face of a cathode-ray oscillo-scope. A device for the measurement of thesmall signal parameters has also been com-pleted. The device measures a, 1-a, r,, rP,and r, directly on a meter. The equipmenthas been designed to operate with bothpoint-contact and junction transistors.

11.3. TRANSISTOR STATIC CHARAC-TERISTICS OBTAINED BY PULSE

TECHNIQUESD. R. FEWER

(Bell Telephone Laboratories, Inc.Murray Hill, N. J.)

It is desirable to know the static char-acteristics of transistors in the region whereelectrode dissipations exceed safe values.These characteristics are impossible to ob-tain by direct-current methods without dam-age to the transistor. Under these conditionsit is necessary to employ pulse methods inwhich the transistor passes current for shortintervals of such duration and recurrencerate that the unit is not damaged.

Point contact transistors have beenexamined by pulse methods in regionsgreatly in excess of rated dissipations. Thestatic characteristics in these regions are dis-cussed and the effects of various pulse shapesand widths are examined.

11.4. BRIDGES FOR MEASURING JUNC-TION TRANSISTOR ADMITTANCE

PARAMETERSL. J. GIACOLETTO

(RCA Laboratories, Princeton, N. J.)The small signal operation of a transistor

is accurately specified by means of our com-plex parameters having both a real and re-

active component. Therefore, eight quanti-ties must be measured, and since thesequantities in a fixed environment are po-tentially a function of operating voltage,current, and frequency, the measurementequipment must have considerable flexibil-ity. This talk will consider in detail the de-sign, construction, and operation of specialequipments operating on the bridge prin-ciple for measuring admittance parametersof junction transistors. These bridge equip-ments operate in the frequency range of ap-

proximately 1 kc to 1 mc, although bysuitable modifications, the operating fre-quency range can be extended.An important feature of the operation of

these bridge equipments is the use of a multi-frequency-test signal such as a square wave,pulse, or swept-frequency-test signal. Withthis mode of operation, multi-element equi-valent-circuit representations can be ob-tained which are valid over a wide range offrequencies so that a relatively complexmeasurement task is considerably simplified.

11.5. A TRANSISTOR ALPHASWEEPER

H. G. FOLLINGSTAD(Bell Telephone Laboratories, Inc.,

Murray Hill, N. J.)A new measuring tool has been designed

for exploring the alpha variation of transist-ors with emitter current. The new instru-ment, which displays the alpha-versus-emitter current characteristic on an oscillo-scope, has an alpha measuring range of 0 to100, and an emitter current range of 0 to10,000 ,ua in both polarities. By varying thecollector voltage in discrete steps, completefamilies of characteristics can be photo-graphically recorded. When utilizing theresolution obtainable from a commercialoscilloscope the measurement accuracy is

5 per cent.

11.6. RAPID TRACING OF TRANSISTORCHARACTERISTICS BY OSCILLO-

GRAPHIC METHODSVERNON MATHIS AND J. S. SCHAFFNER(General Electric Co., Syracuse, N. Y.)An instrument is described that permits

a rapid evaluation of junction transistors. Itwill give the peak-inverse voltage and ap-proximate values for "a," rt, and I¢o as a

function of collector voltage and hence per-mit selection of a desirable operating point.

SESSION 12Significant Trends in Air-

borne Equipment(Organized by Professional Group on

Airborne Electronics)Chairman, J. A. MARSH

(North American Aviation, Inc.,Downey, Calif.)

12.1. SOME SYSTEMS CONSIDERA-TIONS IN FLIGHT CONTROLSERVOMECHANISM DESIGN

R. J. BIBBERO AND R. GRANDGENT(Republic Aviation Corporation,

New York, N. Y.)Flight control servomechanisms are de-

fined as the power amplifying link betweenthe steering intelligence and the stabilizingor directional aerodynamic controls. Toovercome aerodynamic forces, hydraulic,pneumatic, or electrical machines must becoupled to the stabilizing and sensing elec-tronics. The electronics engineer engaged inautopilot design must have knowledge of theaircraft transfer function (that is, its aero-

dynamic parameters under a given set offlight conditions), together with the transferfunctions of the mechanical elements of theservo to produce an optimum systems de-sign. This paper considers the specification

of servo requirements through steering-loopanalysis, the control-power requirements,linearization of hydraulic elements, andapproaches to servosystems synthesis.

12.2. FAIRED-IN ADF ANTENNASL. E. RABURN

(Electronics Research, Inc.,Evansville, Ind.)

A preliminary study was made of theADF bearing errors encountered at differentlocations along the center line of an SNB-2and an R4D aircraft. It was found that thebearing errors were greater, and the dis-persal of the individual error curves withfrequency became greater as the loop wasmoved farther back from the nose of the air-craft. When an installation was made in thetop of the SNB-2, a form of electrical com-pensation was discovered which employsinductive loops. Optimum-inductor com-pensation was obtained at a single frequencyby adjusting the size and location of the in-ductor elements.

A novel ring-type sense antenna wasevolved which can be adapted for many flushinstallations. It is as effective as the con-ventional sense stubs and plates, is light inweight, and can be installed in the sameopening necessary for the flush-mountedloop without interfering with the loop.

The results show that it is not necessaryto sacrifice electrical performance by anyappreciable amount to achieve a zero-draginstallation in the case of (top, nose, orbelly) locations near the center line of theaircraft. In some cases, however, it is neces-sary to employ electrical compensation.Furthermore, it may not always be possibleto achieve good sense-antenna performanceat a nose location unless the sense antennacan be placed a sufficient distance above orbelow the electrical field neutral plane of theaircraft to prevent phase reversals.

12.3. MAGNETIC AMPLIFIERS FORAIRBORNE APPLICATIONS

J. K. McKENDRY(General Precision Laboratory, Inc.,

Pleasantville, N. Y.)In the present state of magnetic-amplifier

development one of the most attractive ap-plications is in the output stages of instru-ment servo-amplifiers, particularly for use inairborne equipments. This paper discussestwo such applications, a position servo anda rate servo, with particular attention to theinfluence of the magnetic-amplifier char-acteristics on the obtainable performance.

The possibilities of improvement inmagnetic-servo-amplifier performance byutilization of more recent magnetic-amplifierdevelopments are considered, together withsome of the principal problems foreseeable.12.4. AIRCRAFT ELECTRICAL POWERJ. C. DIEFFENDERFER AND G. W. SHERMAN

(Wright Air Development Center,Dayton, Ohio)

Complex aircraft electrical and electronicsystems demand that an engineer associatedwith either system have a working knowi-edge of the other if the over-all weaponssystem is to perform its mission effectively.No longer can the electrical power engineerassume that his job is completed when hesimply energizes the airplane bus, nor can

4091953

410PROCEEDINGS OF THE I.R.E.

the electronic engineer assume that he needonly to connect the equipment to the busfor satisfactory operation. The power gen-eration and control system in military air-craft is not an infinite source of power nor isit practical to provide the power to classicaltextbook limits on voltage and frequencyregulation, waveform, or harmonic content.It is possible, however, that a fully inte-grated electrical and electronic system per-mitting optimum operation of electronicequipment and without unduly compromis-ing either system may be realized if theelectronic and electrical engineers approachtheir problems from the broad-systems con-cept.

12.5. THE EFFECTS OF ELECTRONICEQUIPMENT STANDARDIZATIONON AIRCRAFT PERFORMANCE

G. C. SUMNER(Consolidated Vultee Aircraft Corporation,

Ft. Worth, Tex.)The increasing amount of airborne-

electronic equipment on modern aircraft andincreasing performance requirements on theaircraft make it mandatory that the aircraftperformance cost of carrying electronicequipment be minimized. This seems incon-sistent with electronic equipment standard-ization. Present concepts of standardizationregarding environmental conditioning areexamined. It is shown that if conditioning isprovided as a part of the particular airframerather than as an integral part of the equip-ment, greater efficiency can be obtained. Inthis way equipments can be standardizedand performance cost to the aircraft be re-duced to a minimum.

SESSION 13Antennas III-Propagation(Organized by Professional Group on

Antennas and Propagation)Chairman, NEWBERN SMITH

(National Bureau of Standards,Washington, D. C.)

13.1. NOTES ON PROPAGATIONL. A. BYAM, JR.

(Western Union Telegraph Co.,New York, N. Y.)

A summary is given of results obtainedfrom a microwave propagation experimentinvolving an overland path 42 miles long. Acw type magnetron, operated at 4,000 mcand 10 watts output power, was used.Spaced diversity reception was employed.As a much higher carrier-to-noise ratio wasobtained, compared with an earlier similartest, fades of greater depth were recorded.Results are depicted graphically by distribu-tion curves and graphs, followed by a briefdiscussion of diversity action. These resultsgenerally support and also supplement tosome extent information contained in anearlier paper.

13.2. TROPOSPHERIC PROPAGATIONIN HORIZONTALLY STRATIFIEDMEDIA OVER ROUGH TERRAINH. M. SWARM, R. N. GROSE, AND

G. H. KEITEL(University of Washington, Seattle, Wash.)

Rough terrain along the propagationpath introduces considerable difficulty incomputing the field intensity at the diffrac-tion zone for vhf and uhf waves. In thispaper, the probable field intensities in thediffraction zone are calculated for varioustypes of atmospheric structures. The fieldsare calculated from a solution of the Hert-zian-vector wave equation with suitableboundary conditions to account for therough terrain along the propagation path.Numerical computations are made to studythe effect of different types of atmosphericstructures for various transmitter heightsand distances.

13.3. RADIO WAVE SCATTERING INTROPOSPHERIC PROPAGATIONJ. W. HERBSTREIT, K. A. NORTON,P. L. RICE, AND G. E. SCHAFER(National Bureau of Standards,

Boulder, Colo.)The scattering theory of Booker and

Gordon has been developed, assuming thecorrelation function C(r) = C(O) exp (-r/1),so as to be suitable for easy numerical cal-culation of the transmission loss expectedwith this mode of transmission; C(O) denotesthe variance with time of the refractive in-dex of the atmosphere and I denotes the scaleof turbulence. In this development theparameter, [C(O)/I], emerges as a directmeasure of the radio-wave power trans-mitted by this mode of propagation. Assum-ing our theory to be valid, the use of exten-sive radio transmission-loss measurementson 100 mc for transmission paths through-out the United States yields an estimate ofthe variation of [C(O)/I] as a function ofheight above the surface of the earth. Thisestimate is found to be in qualitative agree-ment with the rather meager meteorologicaldata which are now available for estimatingthis parameter. The validity of the scatter-ing theory is further established by using it,in conjunction with an extrapolation of ourradio estimate of [C(O)/I], to predict thetransmission loss to be expected on the trans-mission paths involved in the National Bu-reau of Standards' Cheyenne Mountain ex-periment which cover thq transmitting an-tenna-height range from 30 to 7,800 feet,distance range 223 to 628 miles, and fre-quency range 100-1,046 mc.

The results obtained in this paper ap-parently provide a useful means for explain-ing and extrapolating the results of tropo-spheric-propagation transmission-loss meas-urements, in particular their dependence ondistance, antenna height, antenna gain (thetheory predicts a loss of free-space gain atlarge distances), and radio frequency.

It is believed that the component of sig-nal power received via the scattered mode oftransmission is the principal component atlarge distances beyond the horizon and isresponsible for the short-period fading ob-served in tropospheric propagation at alldistances.

13.4. EXTENDED-RANGE RADIOTRANSMISSION BY OBLIQUEREFLECTION FROM METE-

ORIC IONIZATION0. G. VILLARD, JR., A. M. PETERSON, L. A.

MANNING AND V. R. ESHLE3MAN(Stanford University, Stanford, Calif.)

It has been found that radio communica-tion between relatively low-power stationsoperating at 14 mc and separated by dis-tances of roughly 1,200 km may be main-tained at times when no layer transmissionto any point on the earth's surface can bedemonstrated to be present. The signal ob-tained is subject to considerable fading, butsome signal is nearly always detectable. Thecontribution of overlapping oblique-incid-ence meteor reflections to the observed sig-nal is considered in the light of some pre-liminary theoretical and experimental find-ings. It is clearly important to assess themeteoric contribution with care, since thepossibility that meteoric reflections alonecould account for the signal does not seemunreasonable. Suggestions for further in-vestigation are given.

13.5. AN INTERPRETATION OF VERTI-CAL INCIDENCE EQUIVALENTHEIGHT VERSUS TIME RE-

CORDINGS ON 150 KCRUNE LINDQUIST

(Pennsylvania State College,State College, Pa.)

Results of virtual height versus timevertical-incidence pulse recordings, obtainedon 150 kc, are presented and discussed.Monthly median values of the reflectionheights are shown in a series of graphs. Thecoupling echo, predicted by current-wavetheory, is definitely shown to exist. The re-sults of measurements during undisturbedand disturbed days are discussed. It is con-cluded that one form of echo regularlynoticed during magnetically disturbed nightsmust be due to one type of sporadic E. Re-corded group and phase heights are com-pared and the differences checked againstthose predicted theoretically. Finally, theresults are given of a preliminary investiga-tion of the effects of solar flares.

SESSION 14: SYMPOSIUMDiagnostic Programs andMarginal Checking forLarge Scale Digital

Computers(Organized by Professional Group on

Electronic Computers)Chairman, NATHANIEL ROCHESTER

(IBM Corporation, Poughkeepsie, N. Y.)

14.1. DIAGNOSTIC PROGRAMS ANDMARGINAL CHECKING IN THEWHIRLWIND I COMPUTERN. L. DAGGETT AND E. S. RICH

(Massachusetts Institute of Technology,Cambridge, Mass.)

In the Whirlwind I computer, con-structed at MIT under Office of Naval Re-search sponsorship and presently operatedunder Joint Services support, it has beenfound that marginal checking vastly reducesthe machine failure rate. A series of testprograms each of which thoroughly exer-cises a different section of the machine isused in the marginal checking procedure.Marginal checking cannot prevent intermit-tent and total failures caused by shorts andopens. These are isolated by methods com-bining built-in checking features, diagnostic

410 March

IRE National Convention Program

programming, signal tracing, and operatorexperience and ingenuity. These methods aregreatly facilitated by a special program con-trol which allows a periodically repeatedtest program to be stopped at an arbitrarypoint to study indicator lights and signalwaveforms.

14.2. RELIABILITY AND DUIGNOSTICPROGRAM TECHNIQUES FOR THE

IBM TYPE 701 EDPML. R. WALTERS

(IBM Corporation, Poughkeepsie, N. Y.)Reliability of a complex machine is at-

tained by replacing components before theywear out. A calculator, such as the 701,executing diagnostic programs under non-standard supply-voltage conditions is capa-ble of aiding the engineer in this preventivemaintenance.

A large-scale calculator can perform adiagnosis on itself faster and more thor-oughly than the most capable engineer. Di-agnostic programs replace expensive testequipment and provide greater versatility.

High-speed printing is the method bywhich results of a diagnosis are presented tothe engineer. By this means, data can becompiled concerning even highly intermit-tent failures.

A severe spill test for the 701's electro-static memory is given as an example ofdiagnostic programming in current use.

14.3. DIAGNOSIS AND PREDICTION OFMALFUNCTIONS IN THE COMPUT-ING MACHINE AT THE INSTI-TUTE FORPADVANCED STUDY

G. ESTRIN(Institute for Advanced Study,

Princeton, N. J.)The original design of the Institute

machine sought to minimize the need fordiagnosis of malfunctions and to permitvariation of parameters common to thegroup of elements taking part in any parallelmachine operation.

A routine maintenance program useslimit test techniques to observe the execu-tion of basic machine processes and to pre-dict malfunction of machine elements.

In the event of malfunction a set of diag-nostic codes carry out much of the logicalpartitioning necessary to the discovery of asingle faulty element.

Experiences during a year of operationwill be evaluated and other possible meansof increasing error-free running time of thistype of machine will be projected.

14.4. CHECKING CIRCUITS ANDDIAGNOSTIC ROUTINES

J. P. ECKERT, JR.(Remington Rand, Inc., Philadelphia, Pa.)

The design of the UNIVAC System isbased upon the use of checking circuits as ameans for minimizing the need for diagnosticroutines. In a complex system such asUNIVAC the principal requirement oftrouble-shooting is to isolate the offendingelements. Coupled with the very-high speedsof operation, an error producing element canin a matter of split seconds propagate anerror throughout a major part of the com-puter unless error or checking circuits,operating in synchronism with the internal

computing operations, can detect the errone-ous operation during the cycle in which itoccurs and stop further operation.

UNIVAC operation depends primarilyupon checking circuits but also uses twoprincipal programmed routines for checkingpurposes. One of these routines causes thecomputer to perform nearly every availableoperation while the other routine primarilytests the correct operation of the input-output system, each by means of pro-grammed comparisons. However, in eachcase dependence is placed upon the check-ing circuits to help isolate faulty elementsduring the performance of the routines ifthey should occur.

Finally, routine scheduled engineeringmaintained procedures are regulated so as tominimize the unscheduled maintenance time.Although diagnostic routines may eliminatethe cost of built-in checking circuits, com-puter time today and for some years to comeis of sufficient value that time spent perform-ing diagnostic routines may in the long runcost as much as, or more than, the checkingcircuits would have cost.

14.5. EXPERIENCE WITH MARGINALCHECKING AND AUTOMATICROUTINING OF THE EDSAC

M. V. WILKES(University of Cambridge,Cambridge, England)

The paper will describe a system of mar-ginal checking depending on the use of pulseattenuators which has been fitted to theEdsac, and also a routining device whichputs the machine through a prescribed seriesof tests automatically. An account will begiven of the experience which has been ob-tained up to date, and of the tentative con-clusions which have been drawn about theprinciples on which marginal checkingshould be applied to an existing machineand to a new machine.

SESSION 15Circuits II-Time DomainNetworks-Delay Lines

(Organized by Professional Group onCircuit Theory)

Chairman, J. G. BRAINERD(Moore School of Electrical Engineering,

Philadelphia, Pa.)

15.1. CONTINUOUSLY VARLBLEDELAY LINECARL BERKLEY

(Allen B. DuMont Laboratories, Inc.,Clifton, N. J.)

A need exists for a continuously-variablemethod of delaying pulses or signals whilestill retaining their waveshapes. This hasbeen previously accomplished with tappeddelay lines with sliding contactors. Thesesuffer from the usual deficiencies of slidingcontacts and the discontinuities due to com-mutation. A method is proposed and will bedemonstrated which uses a continuouslywound delay line similar to standard linesbut with a magnetic core intended to in-crease the delay per unit length. Saturationof a portion of the core with an external mag-netic field changes the characteristic imped-ance in the saturated region and results in

reflections or changes in delay which can becontinuously adjusted either mechanically orelectrically. The specialized requirements forthe core material are considered and a num-ber of possible applications described, in-cluding: (1) a linear delay, (2) an oscillatorwith linear-frequency calibration, (3) dop-pler-effect-pulse stretcher, (4) pt modula-tion, (5) transient synthesizer.

15.2. GENERAL TRANSMISSION THE-ORY OF DISTRIBUTED HELICALDELAY LINES WITH BRIDG-

ING CAPACITANCEM. J. DI TORO

(Allen B. DuMont Laboratories, Inc.,Passaic, N. J.)

Helical delay lines are dispersive trans-mission systems because their group velocityof propagation increases with increasing fre-quency. The use of bridging capacitance toreduce this dispersion is known, but no gen-eral analysis has been presented previouslyfor distributed (that is, nonlumped) lines.This paper derives the general equations forthe propagation constant and the impedanceof such distributed helical delay lines withbridging capacitance. The problem requiresthe solution of integral equations, which iseffected using Fourier transforms. The gen-eral conditions for linear phase propagationare derived, and design data are given for thebridging capacitance geometry, along withother delay-line parameters as delay, usablebandwidth, length, diameter, and the line'simpulse response overshoot or echo

15.3. DISTRIBUTED CONSTANT DELAYLINES WITH CHARACTERISTICIMPEDANCES HIGHER THAN

5000 OHMSW. S. CARLEY

(Naval Ordnance Laboratory,Silver Spring, Md.)

Artificial delay lines with characteristicimpedances of greater than 5,000 ohms havebeen developed for use with fractional micro-second pulses. These lines have delaysgreater than 0.35 ,us per axial inch. The at-tenuation of a 1 ,s pulse may be as low as0.3 db/,us of delay. Comparison of risetimes, attenuations, time delays, and char-acteristic impedances for various formex in-sulated wire sizes will be given. A com-parison will be made between lines woundwith A.W.G. #46 formex insulated wire andteflon insulated wire. Photographs of thepulse response of these lines to 0.2, 0.3, 0.5,and 1 ,s will be shown.

15.4. HELICAL WINDING EXPONEN-TIAL-LINE PULSE TRANSFORM-ERS FOR MILLIMICROSECOND

SERVICEJ. KUKEL AND E. M. WILLIAMS

(Carnegie Institute of Technology,Pittsburgh, Pa.)

The slow-wave transmission structurecomprising a helical inside winding andshielding outer cylindrical shell has higherimpedance level and more compact physicaldimensions for a given electrical length thanlinear transmission-line sections.

Design of millimicrosecond-pulse trans-formers using helical slow-wave structuresis described and illustrated with a typicaltransformer. This transformer, designed for

1953 411

PROCEEDINGS OF THE I.R.E.

ten millimicrosecond pulses and suitable formagnetron pulsing service has a physicallength 8 per cent of that required with earlierlinear types. Methods of calculating per-formance are described in detail.

15.5. TIME DOMAIN APPROXIMATIONBY USE OF PADE APPROXIMANTS

R. D. TEASDALE(Radio Corporation of America,

Camden, N. J.)It is often desired to approximate a

complicated transfer function as a ratio ofrational polynomials in such a way that theoriginal function can be physically realizedwith a network of lumped elements byBrune's method or otherwise. It is furtherdesired that the approximation be effected insuch a way that the error in the time domainis small and predictable.

Such a desired method of approximationwas first developed by Pad6 and has sincebeen extended by others. It is a useful fea-ture of PadW's method that one can specify inadvance the relative degree of the numeratorand denominator of the rational fractionwhich is the approximant.

In this paper the basic theory is pre-sented and is then used to develop successivePad6 approximants for several functions use-ful in network theory, such as ex and J, (x) Ix.The results are summarized in tables, andthe accuracy of approximation is illustratedby plots. The corresponding error in the timedomain is computed.

The necessity for further work is empha-sized.

15.6. FREQUENCY TRANSIENTS INIDEALIZED LINEAR SYSTEMS

BEN GOLD(Hughes Aircraft Co., Culver City, Calif.)

The work of Salinger on FM transients isextended by considering (a) more complexnetworks, (b) more complicated modulationwave forms. A general technique is pre-sented for finding the response of morecomplicated networks when the response tothe simple rectangular filter is known. Thismethod is applied to problems involvingfrequency transients. Response to a fre-quency pulse is also examined and the effectof the nonlinear element (the limiter) onresulting transient response discussed.

SESSION 16Electron Devices I-

Transistors(Organized by Professional Group on

Electron Devices)Chairman, H. L. OWENS

(Evans Signal Laboratories, Belmar, N. J.)16.1. THE NEGATIVE RESISTANCE

DIODEI. A. LESK AND V. P. MATHIS

(General Electric Co., Syracuse, N. Y.)By properly biasing a p-n junction, a

large negative-resistance region may bemade to appear in the V-I characteristicof the junction. This negative-resistancecharacteristic may be utilized in variousoscillator and multivibrator circuits. A saw-tooth oscillator circuit using the device is

presented, also calculations of frequency andlinearity. Linearity of the sawtooth wave-form over a large percentage of the operatingcycle suggests linear sweep applications.

16.2. RELIABILITY OF CURRENTTRANSISTORS

R. M. RYDER AND W. R. SITTNER(Bell Telephone Laboratories, Inc.,

Murray Hill, N. J.)Under ordinary conditions of operation,

transistors encased in protective waxes andplastics will give long service. However,under conditions of combined high tempera-ture and humidity or under some types ofshelf aging there may be deterioration. Re-cent test results will be described.

16.3. CHARACTERISTICS OF M-1768TRANSISTORL. B. VALDES

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.)

Intended for economical operation in re-mote service, the M-1768 is a point-contacttransistor which attains about 30 per centefficiency at 6 volts with only 30 mw input,with some sacrifice in frequency response.Its properties are described in some detail.

16.4. A DEVELOPMENTAL GERMA-NIUM N-P-N ALLOY-JUNCTION

TRANSISTORD. A. JENNY

(RCA Laboratories Division,Princeton, N. J.)

The problems encountered in the de-velopment of a germanium n-p-n alloy-junction transistor are discussed and theresults are compared with a p-n-p alloy-junction transistor. The most serious prob-lem arises from the differential expansionstrains between the donor impurity elementand the germaniumwhichareintroduced dur-ing the cooling and solidification process dueto the relatively high melting points and thelack of ductility of many bulk impurity ele-ments. One method of overcoming thesestrains is to use a ductile low melting pointalloy containing the impurity element as onecomponent. This technique has been used toproduce transistors with essentially planarjunctions with power gains of over 50 db and"alpha' over 0.999. High "alpha" is main-tained up to very high current densities,which is at least partially due to the parallel-ism of the junctions.

16.5. BEHAVIOR OF GERMANIUMJUNCTION TRANSISTORS ATELEVATED TEMPERATUREAND POWER TRANSISTOR

DESIGNL. D. ARMSTRONG

(Radio Corporation of America,Princeton, N. J.)

The behavior of germanium junctiontransistors at elevated temperatures is exam-ined with regard to the performance of low-power devices in high-ambient temperaturesand the capabilities of power transistors withvarious means of cooling. In the case of low-power devices, values of various prametersof developmental germanium junction trans-istors are given as a function of ambient

temperature. In the case of power tran-sistors, the design and operation character-istics of laboratory units capable of severalwatts dissipation are described.

SESSION 17Instrumentation III-

Electronics(Organized by Professional Group on

on Instrumentation)Chairman, I. G. EASTON

(General Radio Co., Cambridge, Mass.)

17.1. THE RESPONSE OF A PANO-RAMIC RECEIVER TO CWAND PULSE SIGNALS

H. W. BATTEN, R. A. JORGENSEN, A. B.MACNEE, AND W. W. PETERSON

(University of Michigan, Ann Arbor, Mich.)The results of an analysis of the response

of a panoramic receiver to cw and pulse sig-nals are given. The receiver's response isstudied quantitatively as a function of theparameters: signal pulse length and fre-quency, receiver bandwidth, sweep rate, andtype of IF amplifier. The effect of theseparameters on the relative output ampli-tude, output pulse width, and apparentbandwidth is emphasized. Two specific casesare considered. Theoretically the response ofa receiver with a Gaussian-shaped IF pass-band to pulses having Gaussian envelopes isderived. This answer is given in closed form.The response of a receiver with a single-tuned IF amplifier to pulses having rec-tangular envelopes has been studied with anelectronic differential analyzer. The agree-ment between these two cases justifies ap-plication of the Gaussian case to most prac-tical design problems.

17.2. A VHF IMPEDANCE METERJ. H. MENNIE

(Boonton Radio Corporation,Boonton, N. J.)

A self-contained impedance measuringinstrument that operates from 0.5 to 250 mcwill be described. The wide-frequency rangeis the result of a recent development thatgreatly simplifies the problem of connectingoscillator and detector to the corners of abridge network. The application of this prin-ciple to a Schering Bridge has resulted in awide range instrument that will measure re-sistance directly from 15 ohms to 100,000ohms over its entire frequency range. Equiv-alent parallel reactance or susceptance ismeasured in micro-microfarads by a capaci-tance substitution method. Special tech-niques enable low inductance values to bemeasured directly with a readability of0.0001 ,uh.

17.3. SIMPLIFIED MEASUREMENT OFINCREMENTAL PULSE TIME

JITTERW. T. POPE

(Griffiss Air Force Base, Rome, N. Y.)A method of measuring incremental time

jitter such as may be introduced on a traiaof pulses passing through a stage or a seriesof stages is described. A measurement of thistype may be required for testing of modu-

412 Marci-

IRE National Convention Program

lators and other equipment using hydrogenthyratrons. Measurement of incrementaljitter to 0.001 ,us is shown to be practical.The measurement is made by observing acathode-ray oscilloscope screen. Excessivesweep speeds or unusually wide-band videocircuits are not required. The circuits usedare described and some of the difficulties en-countered are discussed. Typical measure-ments are also tabulated.

17.4. WIDE-BAND WAVE ANALYZER0. KUMMER

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.)

This paper describes a wave analyzercovering the frequency range of 100 kc to20 mc. The analyzer indicates directly, thelevel of any signal in the range -130 dbmto +30 dbm, and the frequency of the sig-nal with an accuracy of +2 kc. The analyzeris flat over the entire spectrum to within+0.5 db. The effective bandwidth is con-stant at 2 kc. Measurements in the presenceof other signals as much as 80 db higher thanthe desired signal do not produce detectableerrors in the measurement.

17.5. ULTRA-LOW-FREQUENCY,THREE-PHASE OSCILLATOR

GILBERT SMILEY(General Radio Co., Cambridge, Mass.)This paper describes the development of

a resistance-capacity oscillator using "Millereffect" multiplication of capacity values toachieve very-low frequencies with practicalvalues of capacitors and resistance. Becausethree resistance-capacity networks are usedto secure a one hundred-eighty degree phaseshift, each network is assigned its own'Miller effect" amplifier, which, in practice,results in a three-phase network, wye con-nected to the power supply. This, in turn,results in an oscillator that is relatively in-dependent of disturbances in the commonneutral supply. Furthermore, the three-phase output furnishes, by suitable connec-tions, a source voltage adjustable as tophase and magnitude, independent of fre-quency. The theory by which unwantedresiduals are exactly offset over a wide oper-ating range is also expounded.

SESSION 18: SYMPOSIUMTrends in Mobile Com-

munications(Organized by Professional Group on

Vehicular Communications)Chairman, A. B. BUCHANAN

(Detroit Edison Co., Detroit, Mich.)

18.1. THE EFFECTS OF SELECTIVITY,SENSITIVITY, AND LINEARITY INRADIO CIRCUITS ON COMMUN-

ICATIONS RELIABILITYAND COVERAGE

J. G. SCHERMERHORN(Rome Air Development Center,

Rome, N. Y.)A eeneral representation of the multi-

frequency, mobile receiver and transmitter-broadcast coverage problem common tocivilian and military applications is given,

and the communications reliability andcoverage is estimated from the isolation tointerference obtainable with various radiocircuit designs. The selectivity, linearity,and sensitivity of receiver RF circuits areexamined in particular, and an analysisdemonstrates the effects of such equipmentdesign parameters on interference isolation.Charts of useable service areas provide aphysical picture of circuit-design results.Design information is furnished that relatethe above parameters to recommendationsfor optimum sensitivity, frequency alloca-tion, and power requirements.

18.2. SINGLE SIDEBAND FORMOBILE COMMUNICATIONS

ADAMANT BROWN AND R. H. LEVINE(Signal Corps Engineering Laboratories,

Fort Monmouth, N. J.)The problems encountered in the use of

single-sideband communication in mobileinstallations where size, weight, and sim-plicity of operation are of prime importanceare discussed. A review of progress to datetowards such operation is made. A com-parison is made of the advantages or dis-advantages arising in the generation of asingle-sideband signal at high level or lowlevel, and at the operating frequency or ata fixed low frequency. Methods of providingsuitable automatic-frequency control of thereceived signal, during push-to-talk opera-tion as necessary in mobile communications,are discussed. Photographs of the spectrumof the single-sideband signal resulting fromvarious systems are shown.

18.3. MAJOR FACTORS IN MOBILEEQUIPMENT DESIGN WITHEMPHASIS ON 460 MCMOBILE EQUIPMENTCHARACTERISTICS

J. F. BYRNE AND A. A. MACDONALD(Motorola, Inc., Chicago, Ill.)

This paper will review the design con-siderations pertinent to the development ofmobile equipment for operation in the 450-470-mc band. Receiving system require-ments will be outlined and the means selectedfor achieving these requirements will be de-scribed. Similarly the evolution of the trans-mitting component and its specifications willbe shown, and finally the complete systemperformance as a mobile package, opera-tionally consistent with lower-frequency sys-tems will be presented.

18.4. FIELD EXPERIMENTS WITH450 MC MOBILE SYSTEMS

P. H. BELLINGHAM AND J. Q. MONTRESS(Bell-Mont Communications Service

Corporation, Englewood, N. J.)Information will be presented relative to

comparison tests between 150-mc and 450-mc equipment in various cities of differenttypographical situations along with a de-scription using slides and maps of two exist-ing proven 450-mc mobile systems. Experi-ences encountered with various antennasavailable to the authors will also be de-scribed.

A brief summary of the maintenancetechniques utilized in these new frequencieswill be outlined.

SESSION 19: SYMPOSIUMElectronics in Flight

Chairman, C. S. DRAPER(Massachusetts Institute of Technology,

Cambridge, Mass.)A discussion by a panel of distinguished

authorities.SESSION 20

Electron Devices II-Electron Tubes

(Organized by Professional Group onElectron Devices)

Chairman, G. R. KILGORE(Evans Signal Laboratories, Belmar, N. J.)

20.1. GAS PRESSURE EFFECTS ON ION-IZATION PHENOMENA IN HIGH-

SPEED HYDROGEN THYRA-TRONS

W. C. DEAN(Gulf Research and Development Co.,

Pittsburgh, Pa.)G. W. PENNY AND J. B. WOODFORD, JR.

(Carnegie Institute of Technology,Pittsburgh, Pa.)

Initiation of conduction and rise of anodecurrent in highly shielded hydrogen thyra-trons involve a markedly different processfrom the ion migration phenomenon occur-ring in conventional "line of sight" thyra-tron structures. A simplified ionizationmodel for well-shielded thyratrons is givenfor analysis of pressure effects on both thetrigger delay and the anode-current com-mutation intervals and an optimum pressureis described. Analytical and experimentalresults are given.

20.2. LOW NOISE, HOT CATHODE,GAS TUBES

E. O. JOHNSON, W. M. WEBSTER, ANDJ. B. ZIRKER

(RCA Laboratories, Princeton, N. J.)For many purposes, gas tubes are more

attractive than vacuum tubes because theirmuch lower tube drop permits high-circuitefficiency. However, gas tubes used in ornear sensitive equipment require extensiveshielding and filtering because of the noisethey generate. Such noise consists of fluctua-tions in the tube drop during conductionhaving, frequently, an amplitude of manyvolts. Different types of noise having differ-ent origins and frequency bands have beenobserved. For example, a low-frequency re-laxation oscillation, usually quite incoherent,often occurs which is caused by an instabilityof the region within which ionization is tak-ing place. At much higher frequencies (50-500 mc), noise has been detected from tubeswhich have a negative-resistance character-istic. In the neighborhood of 500 kc, stillanother type of noise has been observedwhen the tube current approaches half thesaturated cathode emission.

20.3. NEW DISPENSER TYPETHERMIONIC CATHODE

R. LEVI(Philips Laboratories, Inc., Irvington-

on-Hudson, N. Y.)In recent years dispenser type cathode

known as the "L cathode" was developed in

1953 413

414PROCEEDINGS OF THE I.R.E.

Holland. Work undertaken at Philips Labo-ratories, Inc., Irvington, N. Y., for the pur-pose of improving fabrication techniqueshas resulted in a variant of the L cathode,which is being called the Philips' "impreg-nated" cathode. Interest in this new cathodehas developed to such a degree that it ap-pears desirable to present details about itsstructure and some of its inherent advan-tages.

In the new cathode, the alkaline earthmaterial is dispersed within the pores of thetungsten body thus eliminating the need fora large reservoir cavity. This has resulted ina radical simplification in cathode construc-tion and has made possible fabrication ofcathodes of practically any shape and dimen-sion within extremely close tolerance. Thishas been brought about by a special tech-nique also developed by the author whichmakes possible the machining of tungsten.Among the advantages of this new cathodederived from the elimination of the largecavity are a more hom*ogeneous temperaturedistribution across the emitting area, an im-proved thermal efficiency, and additionalheater space which permits the use of largerheaters.

20.4. MULTI OUTPUT BEAM SWITCH-ING TUBES FOR COMPUTERSAND GENERAL PURPOSE USE

S. KUCHINSKY(Burroughs Adding Machine Co.,

Philadelphia, Pa.)The prototype development of two

unique tube types has opened up a new fieldin reliable high-speed switching. One type isa small coaxial "coding tube," using crossedelectric and magnetic fields, with ten stablebeam positions and four parallel binarycoded outputs. The second type is a comple-mentary ribbon-beam cathode-ray 'selectortube" with deflection-plate inputs and tenindividual outputs. Details of design, con-struction, and operational characteristicsare given. Versatility of these tubes has beenexperimentally verified as a decimal tobinary converter, a high-speed reversiblebinary counter, a binary adder, and a codingsystem for PCM. Further designs and ap-plications are predicted.

20.5. AN EQUIVALENCE PRINCIPLEIN HIGH FREQUENCY TUBES

R. ADLER(Zenith Radio Corporation, Chicago, 111.)

A close analogy exists between beam-deflection tubes and velocity-modulation de-vices such as klystrons or traveling-wavetubes. For beam-deflection tubes in whichan inductive pick-up system is used insteadof the usual intercepting anodes, the equiva-lence is rather complete. Qualitative simi-larities between the two types of tubes aredeveloped, and quantitative differences, aswell as characteristic advantages and dis-advantages of each, are discussed and ex-plained in some detail. The analogy permitsa transfer of experience between the twofields, and several applications are discussedwhich may be valuable in the uhf range.

20.6 THE INTERNAL MAGNETIC FOCUSTUBE, ITS THEORY, PERFORMANCE

AND APPLICATIONR. B. GETHMANN AND L. E. HUYLER(General Electric Co., Syracuse, N. Y.)The "Internal Magnetic Focus Tube"

(IMF) is described and its attractive fea-tures are pointed out. The basic elements in-volved, the focus lens structure, and the iontrap are discussed in detail together with theelectron optics of the combined unit. Themagnetization procedure is described andthe performance characteristics of the tubeare discussed.

This high quality permanent-magnetfocus lens, with trimmer-shunt adjustment,and a preset ion trap combine to provide anunusually attractive picture-focus systemfor all television sets.

SESSION 21Circuits IV-Active Net-

works-Transistors(Organized by Professional Group on

Circuit Theory)Chairman, R. F. SHEA

(General Electric Co., Syracuse, N. Y.)

21.1. TRANSIENT ANALYSIS OFJUNCTION TRANSISTOR

AMPLIFIERSW. F. CHOW AND J. J. SURAN

(General Electric Co., Syracuse, N. Y.)In the calculation of transient response

problems involving junction transistors it isdesirable to use an equivalent circuit thatcan be handled readily with standard circuittechniques. Transient analysis based uponthe diffusion equation gives results which aremathematically unwieldy and which are notexplicitly related to the parameters of thelow-frequency equivalent circuit.

An approximate equivalent circuit is de-veloped which exhibits both a frequency andtransient voltage transform in good agree-ment with experimental results. The trans-form of the equivalent circuit is compara-tively simple and thus provides a rapidmeans of calculating transient response oftransistor circuits.

21.2. THE GROUNDED-COLLECTORTRANSISTOR AMPLIFIER ATCARRIER FREQUENCIES

F. R. STANSEL(Bell Telephone Laboratories, Inc.,

Murray Hill, N. J.)Expressions are derived for input resist-

ance, output resistance, and ratio of input-to-output voltage and current at low fre-quencies for transmission in both the base-to-emitter and the emitter-to-base direc-tions. These expressions are extended to thecarrier-frequency range (up to approxi-mately twice the alpha cutoff frequency) byconsidering the effect of the variation ofalpha with frequency, of collector capaci-tance and of load capacitance. Experimentalevidence is presented which verifies theequations obtained and indicates that themethod of computing the effect of frequencymay be applied to other transistor circuits.

21.3. SYMMETRICAL PROPERTIES OFTRANSISTORS AND THEIR

APPLICATIONG. C. SziKLAI

(RCA Laboratories Division,Princeton, N. J.)

There are certain transistor character-istics which are not present in vacuum tubes.Some of these characteristics may be bestclassified as symmetrical properties. Thefirst kind of symmetry may be found in thecomplementary characteristics of the n-p-nand p-n-p junction transistors. Circuits usingboth kind of transistors in combinationprovide advantages in efficiency, reductionof components, and other circuit simplifica-tions.

A second kind of symmetry is displayedby specially-constructed single units inwhich the emitter and collector may beinterchanged. This symmetry permits a cur-rent flow of either direction controlled alikeby the base current. This basic property isuseful in switching circuits for clamping,phase and frequency comparison, modula-tion, etc. A high-efficiency deflection-currentcircuit for television was developed usingthis principle.

21.4. A STUDY OF TRANSISTORCIRCUITS FOR TELEVISION

G. C. SZIKLAI, R. D. LOHMAN, ANDG. B. HERZOG

(RCA Laboratories Division,Princeton, N. J.)

A study was made to cover the variousportions of a television receiver to explorethe possibility of using transistors. Thepaper will be presented in two parts: Part I-the signal channel and Part 2-the scan-ning channel.

21.5. CONDUCTANCE CURVE DESIGNOF RELAXATION CIRCUITS

K. A. PULLEN(Ballistic Research Labs., Aberdeen

Proving Ground, Md.)Design of nonlinear repetitive circuits

using electron tubes requires data not read-ily available on static-tube-characteristiccurves. A technique for use of the recentlydeveloped conductance curves to this designproblem has been developed.

The application of the technique to thedesign of multivibrators anid blocking oscil-lators requires knowledge of the dynamic-loop gain, the plate-voltage swings, thedynamic-tube conductances, and the static-circuit characteristics. Determination of theswitching time and the initiation bias areconsidered. The effect of the conductanceof the positive grid and the effects of thetube conductances are studied.

Several examples of multivibrator andblocking oscillator designs illustrating use ofthe method are presented. Experimentalconfirming data are included. The agreementwith the theory is examined.

21.6. TRANSISTOR RELAXATIONOSCILLATORSS. I. KRAMER

(Fairchild Guided Missiles Division,Wyandanch, N. Y.)

414 March

IRE National Convention Program

Several novel relaxation oscillators havebeen developed using point-contact tran-sistors. One of these generates a rectangularwaveform but minus the rather drastic slopeof the more conventional version. The otheris a triggered- or free-running pulse gener-ator using capacitive feedback which can

supply a fast, high-energy, 1-,us pulse withvirtually no overshoot. Schematics andwaveforms are supplied together with an

explanation of the operation and designcriteria. Some data on reproducibility is alsoincluded.

These circuits provide some of the funda-mental building blocks for the transistoriza-tion of electronic equipment. Two versionsof the rectangular-wave generator have beenused. Both of these make use of resistance inseries with the timing capacitor. One is a

grounded-collector circuit with the outputtaken from the base and the other uses a

collector load to provide two out-of-phasewaveforms. The pulse generator provides a

l-,us pulse with a rise and fall time of 0.1 to0.2 /is. While such a waveform is obtainablewith a blocking oscillator, this circuit hasthe advantage of using only capacitivecoupling. Quantitative data for 19 transis-tors is given.

SESSION 22Noise and Modulation

(Organized by Professional Group on

Information Theory)Chairman, J. B. WIESNER

(Massachusetts Institute of Technology,Cambridge, Mass.)

22.1. NOISE PROBLEMS OF THEORETI-CAL AND PRACTICAL INTERESTG. 0. YOUNG AND BERNARD GOLD

(Hughes Aircraft Co., Culver City, Calif.)

The effect of a communications receiveron a signal which has unwanted noise addedis a problem which is far from having beencompletely solved. This paper touches uponsome problems of interest, indicates and dis-cusses the solutions to the solved problems,and points out some of the difficulties andsuggests possible approaches to the unsolvedproblems.

The questions dealt with are: (1) func-tional representations of noise, (2) categor-ization of noise, (3) noise in linear systems(solved problems), (4) noise in linear sys-

tems (unsolved problems), (5) noise tran-sients.

22.2. A NOTE ON RECEIVERS FORUSE IN STUDIES OF SIGNAL

STATISTICSRALPH DEUTSCH AND H. V. HANCE

(Hughes Aircraft Co., Culver City, Calif.)

The characteristics of linear, lin-log, andlogarithmic receivers are described and theeffect of their transfer response on randomsignals is obtained. The limited dynamicrange of an amplifier system is shown toproduce calculable errors in the measure-

mnent of signal statistics. Curves have beenderived which relate the agc circuits to a

predetermined measurement error.

22.3. AMPLITUDE MODULATION BYPLATE MODULATION OF CW

MAGNETRONSJ. S. DONAL, JR. AND K. K. N. CHANG

(RCA Laboratories Division,Princeton, N. J.)

Using a plate-modulated cw magnetron,means have been devised for the obtainingof high system efficiency, combined withgood performance as regards linearity, depthof modulation, and bandwidth. Magnetronpushing has been studied as a function ofmodulation frequency and loading. Themodulation impedance of the magnetron hasbeen correlated with pushing and envelopeamplitude. The performance of phase-locking systems can be predicted if the char-acteristics of the magnetron are known.Phase locking, used primarily to overcomepushing, may in turn alter the linearity,depth of modulation, and bandwidth of asystem.

22.4. COMPARISON OF MODULATIONMETHODSR. M. PAGE

(Naval Research Laboratory,Washington D. C.)

The statistical theory of communicationis interpreted in familiar radio-engineeringterms. Tuller's and Shannon's equations arecompared, and reasons are given for choos-ing Shannon's equation. The informationtransformer concept is exploited, with input-output relationships expressed in terms ofthe fundamental information equation, tak-ing into account -the relationship betweennoise and bandwidth. A distinction is madebetween coding and modulation. Reductionof equivocation by ideal coding and by'power margin are compared quantitatively.Using Shannon's definition of an ideal sys-tem as a reference standard, a quantitativecomparison is made 'between single- anddouble-sideband amplitude modulation,standard broadcast frequency modulation,and binary pulse code modulation for moder-ately high-quality audio, 32-level teletype,and a generalized bilevel function. A sum-mary of the best types of modulation foreach of several different conditions of opera-tion is followed by an evaluation of the im-pact of the statistical approach to commun-ication theory on radio methods of com-munication. The paper is predominantlytutorial in nature, although significant mate-rial is included which has not previously beenpublished.

22.5. A TECHNIQUE OF INTER-MODULATION INTERFERENCE

DETERMINATIONA. J. BEAUCHAMP

(Rome Air Development Center,Rome, N. Y.)

The' increased use of multifrequencycommunications equipment in both civilianand military applications has aggravatedthe severity of intermodulation interferencebetween radio circuits. As a result, there is aneed for a tool as an aid in the sorting proc-ess in determining potential intermodula-tion-interference products that may resultfrom spurious radiations of transmitters andspurious responses of receivers. This paper

describes such a tool that has been de-veloped. Determination of the frequencycomponents of odd-order intermodulationproducts for any number of frequencies isshown, with particular emphasis placed onthird-order products, by a graphical tech-nique that reduces this determination to amechanical level. A method of quickly deter-mining interference frequencies for any givennumber of frequencies is also indicated.

SESSION 23: SYMPOSIUMTelevision Broadcasting

(Organized by Professional Group onBroadcast Transmission Systems)

Chairman, E. M. JOHNSON(Mutual Broadcasting Co.,

New York, N. Y.)23.1. THE DESIGN OF SPEECH INPUT

CONSOLES FOR TELEVISIONR. H. TANNER

(Northern Electric Co., Ltd.,Belleville, Canada)

This paper describes two types of TVaudio consoles both specifically designed forthe particular requirements of the CanadianRadio and TV setup. The first type is in-tended for large studio installations, and isused in the Montreal and Toronto TVstudios of the Canadian Broadcasting Cor-poration. It possesses certain unique fea-tures which facilitate the production of asound component well matched to the pic-ture.

The second design is suitable for smallercenters, either in its basic form, or with twosimilar units integrated mechanically andelectrically to form a highly flexible, yetcompact, double-channel console.

In the interests of'standardization, bothdesigns are also eminently suitable for high-grade sound broadcasting use.

23.2. BUILDING TV BROADCAST FACIL-ITIES FOR GROWTH, FLEXI-

BILITY AND ECONOMYA. R. KRAMER AND E. R. KRAMER(Kramer, Winner and Kramer,

New York, N. Y.)A demonstration of the techniques which

can be used to insure the organized growthof TV broadcasting stations, utilizing thor-ough planning and unique structural design.

A basic unit designed for network relay,film, and a minimum of live broadcasting isanalyzed and used as a first-stage proto-type. Growth is charted with plans and dia-grams as the basic units expand to house-production studios, control rooms, offices,public spaces, and ielated facilities. Build-ing problems encountered at each stage aredescribed and solutions presented.

The planning and construction prin-ciples proposed provide not only efficientgrowth but a construction system flexibleenough to allow plan changes and adjust-ments to new developments other than addi-tions.

Practical construction details necessaryto achieve expansion and flexibility witheconomy are illustrated.

23.3. FASHIONS IN TV TRANS-MITTING ANTENNAS

4151953

416 PROCEEDINGS OF THE I.R.E.

F. G. KEAR(Kear and Kennedy, Washington, D. C.)

ANDJ. G. PRESTON

(American Broadcasting Co.,New York, N. Y.)

23.4. HIGH GAIN AMPLIFIERS FORHIGH POWER TELEVISION

TRANSMITTERSJOHN RUSTON

(Allen B. Du Mont Laboratories, Inc.,Clifton, N. J.)

It is shown that the availability of suita-ble high-power tetrode tubes has made pos-sible unusually high-power gain in the finalbroadband linear amplifier of a high-powervhf television transmitter. For a specific 20-kw tetrode, a power gain exceeding 100 iscomputed at a power-output level of 30 kwand a bandwidth of 5 mc. Some sacrificesof power gain enables the "lower-sidebandreinsertion" inherent in such an amplifier tobe reduced enough to permit the inclusionof the vestigial-sideband filter in the low-power driver stage.

A practical application is illustrated by abrief description of a commercial 25-kw low-band amplifier having a power gain of 50.

23.5. OPTIMUM UTILIZATION OF THERADIO FREQUENCY CHANNEL

FOR COLOR TVR. D. KELL AND A. C. SCHROEDER(RCA Laboratories, Princeton, N. J.)To produce a simultaneous television

image in color, three communication chan-nels must be available. The first of thesemay be used to transmit the scene bright-ness, the second the degree of color satura-tion, and the third the hue or color. Forcompatibility the brightness is transmittedas amplitude modulation in the usual way.A subcarrier is introduced to carry the othertwo pieces of information as amplitude andphase modulations. The optimum loading ofthese two auxiliary communication channelsis the major consideration of this paper.

SESSION 24Quality Control Methods Ap-plied to Electron Tube and

Electronic EquipmentDesign

(Organized by Professional Group onQuality Control)

Chairman J. R. STEEN(Sylvania Electric Products Inc.,

New York, N. Y.)24.1. USE OF STATISTICAL TOLER-ANCES TO OBTAIN WIDER LIMITS

ON TUBE COMPONENTDIMENSIONSE. V. SPACE

Radio Corporation of America,Harrison, N. J.

A statistical quality-control technique isused to determine the tolerances required ontube components in order to maintain plate-current within functionally imposed limits.

Production tolerances of the component di-mensions are evaluated consistent with theabsolute and natural tolerances of the cath-ode-plate spacing. A comparison of the pre-dicted values of the mean and standard de-viation of the cathode-plate spacing with ac-tual data shows close agreement.

24.2. TOLERANCE CONSIDERATIONSIN ELECTRONIC PRODUCT

DESIGNR. C. MILES

(Airborne Instruments Laboratory, Inc.,Mineola, N. Y.)

Casual selection of electronic-componenttolerances may result in a design which isunnecessarily expensive to manufacture orwhich lacks the necessary reproducible per-formance. Elementary probability theorycan provide a partial solution to the prob-lem since many components have either arectangular or modified normal distribution.The probability concept can be extended tothe problem of cumulative tolerances in sucha manner as to improve the quality of elec-tronic-product designs.

24.3. DISTRIBUTION PATTERNS FORTHE ATTRIBUTES OF ELECTRONIC

CIRCUITRYR. F. ROLLMAN AND E. D. KARMIOL(Allen B. DuMont Laboratories, Inc.,

East Paterson, N. J.)This paper presents the findings of exten-

sive studies into the distribution patterns ofthe major attributes for varied types of elec-tronic circuitry. It is shown that these pat-terns can be readily computed from data ob-tained by conventional production measur-ing techniques on small quantities in the of-der of fifty units. Very marked correlationwas found between parameter tolerances,production techniques, and the distributionpatterns. It was concluded that parametertolerances and production techniques mustbe given major consideration by the elec-tronics engineers if engineering design effortsare to be successful.

24.4. THEAPPLICATION OF STATISTICSTO FIELD SURVEILLANCE OFPRODUCT PERFORMANCE

R. HERD(Aeronautical Radio Inc., Wash-

ington, D. C.)This paper presents some of the statisti-

cal techniques applicable to field surveillanceof the performance and operation of any se-lected product. A planned experiment, uti-lized by AR Inc. in its surveillance activitiesto compare tube types, standard tubes ver-sus their improved versions, manufacturers,etc., is described, and methods of analyzingthe experiment are investigated. The prob-lems involved in planning an experiment, in-cluding the definition of terms, the purposeof the experiment, and field operations arediscussed.

24.5. RELIABILITY OF ELECTRONTUBES IN MILITARY

APPLICATIONSE. F. JAHR

(Aeronautical Radio Inc., Wash-ington, D. C.)

This paper is based upon the AR Inc mil-itary tube project and discusses electron-tube reliability. Findings are analyzed bytube type, time and nature of failures, andenvironmental conditions. Specific causes offailures are noted as well as required im-provements and expected gains.

The problem of predicting reliability anda suggested approach to this problem are dis-cussed together with an analysis of the pres-ent status of electron-tube reliability and aprojected goal.

24.6. DYNAMIC ENVIRONMENTTESTING

D. T. GEISER(Boeing Airplane Co., Wichita, Kan.)Investigation by analogy showed equip-

ment failure was caused not only by environ-ment, but also by the rate of change of en-vironment. Tests verified the usefulness ofthis concept as a design tool, thus furnishinga concise method of comparing airplane andequipment in the planning and design stage.

A proposal is advanced for a universalmethod of component and equipment envi-ronment specification. Examples are given ofthe use of both the experimental and Max-wellian analysis in air frame and missilework.

SESSION 25: SEMINARAcoustics for the Radio

Engineer-I(Organized by the Professional Group

on Audio)Chairman, J. J. BARUCH

(Massachusetts Institute of Technology,Cambridge, Mass.)

25.1. FUNDAMENTAL THEORYL. L. BERANEK

(Massachusetts Institute of Technology,Cambridge, Mass.)

25.2. MICROPHONEH. F. OLSON

(Radio Corporation of America,Princeton, N. J.)

25.3. LOUDSPEAKERSH. S. KNOWLES

(Industrial Research Products, Inc.,Franklin Park, Ill.)

This seminar will present the engineeringaspects of the science of acoustics and thosefundamental principles which have a directbearing on acoustical engineering in termswhich the Radio Engineer can understand,and which will assist him in his daily work.Leading experts in the field will discuss fun-damental theory with emphasis on equiva-lent electrical circuits, the engineering use ofmicrophones, loudspeakers, and the charac-teristics which are of importance to theirusers.

SESSION 26Electron Devices III-Microwave Tubes

(Organized by Professional Group onElectron Devices)

416 March

IRE National Convention Program

Chairman, J. H. BRYANT(Federal Telecommunication Laboratories,

Inc., Nutley, N. J.)

26.1. HIGH-POWER TRAVELING-WAVE-TUBE AMPLIFIERS

M. ETTENBERG(Sperry Gyroscope Co., Great Neck, N. Y.)

Recent experimental work has shownthat the traveling-wave tube may be usedsuccessfully for high-power amplificationover a large bandwidth. Efficiency and gainare comparable with good klystron amplifierperformance and the bandwidth is inherentlymuch larger. The magnetic focusing and thebeam-dissipation requirements are similar inboth types.

This paper will present the design andperformance parameters of high-power trav-eling-wave amplifiers. The construction andperformance of several experimental tubesboth pulsed and cw will be described.

26.2. OPERATION OF THE TRAVELING-WAVE TUBE IN THE DISPERSIVE

REGIONL. A. ROBERTS AND S. F. KAISEL

(Stanford University, Stanford, Calif.)The majority of the published informa-

tion on the traveling-wave tube has beenconcerned with operation of the tube as awide-band amplifier, in a frequency rangewhere the helix is a nondispersive structure.In many applications where very wide band-width is not required, the traveling-wavetube amplifier still offers attractive advan-tages over other types of amplifiers as re-gards gain, bandwidth, and simplicity ofoperation. Where great bandwidth is not re-quired, operation in a region where the helixis a dispersive transmission line can be con-sidered. This paper will present a compari-son of tube characteristics in the dispersiveand nondispersive regions and the effects ongain and bandwidth of this choice. Experi-mental results will be presented for tubesbuilt for operation in the dispersive region.

26.3. A TRAVELING-WAVE ELECTRONBUNCHERR. B. NEAL

(Stanford University, Stanford, Calif.)This device supplies axially bunched

electrons to the Stanford Mark III linearaccelerator. The buncher consists of a disc-loaded circular wave-guide structure, 32inches long, tapered so that the phase veloc-ity increases from 0.5 to 1 times the velocityof light in its length, while the axial fieldstrength increases 10 times. Electrons, in-jected at 80 kv, emerge with energies of 4maev, bunched within 200 in each cycle. Peakcurrents up to 170 ma are obtained. RFpower is supplied by a 10.5-cm, 10-mw kly-stron amplifier delivering 1.4,s, pulses at60 cps.

26.4. SOME PROPERTIES OF PERIODI-CALLY LOADED STRUCTURES SUIT-ABLE FOR PULISED TRAVELING

WAVE TUBE OPERATIONM. CHODOROW AND E. J. NALOS

(Stanford University, Stanford, Calif.)

This paper describes some of the factorsinvolved in the selection of structures suita-ble for high-power pulsed traveling-wavetubes, and the experimental evaluation oftheir properties, such as bandwidth, gainparameter, and space-harmonic content.The properties of periodically loaded struc-tures can be determined by measurementson a capped-off section comprising only afew periods. From such measurements, thebandwidth, group, and phase velocities canbe deduced. By perturbation methods, at agiven resonance of such a cavity, the ratio offield strength to energy storage can be made,which is essentially R0h0gt/Q of this cavity.From bead measurements along the path ofthe electron beam, the relative spatial de-pendence of the electric field can be obtainedby measuring the variation of cavity reso-nance as a function of bead position. Fromthis, Pierce's gain parameter for the struc-ture, as well as the space-harmonic compo-nents can be determined experimentally. Theabove experiments bear out the usefulness oflumped-equivalent circuits in predictingqualitative behavior of such structures.

26.5. EXPERIMENTS ON MILLIMETERWAVE AND LIGHT GENERATION

H. MOTZ, W. THON, ANDR. N. WHITEHURST

(Stanford University, Stanford, Calif.)It can be shown that electromagnetic ra-

diation of very short wavelength may be ob-tained from electron beams accelerated torelativistic velocities passing through suita-ble magnetic-field configuration. Experi-ments with electrons passing through ar-rangement of magnetic fields which we callan undulator were carried out at the micro-wave laboratory at Stanford University. Forthe first experiments, a 100-mev beam fromthe Mark III linear accelerator was used togenerate visible light. In other experimentsa beam of 3-mev electrons, obtained from asmall accelerator with good bunching action,was used to generate radiation in a band ofabout one-millimeter wavelength at a peakpower level of approximately one watt.

SESSION 27Information Theory I-

Recent Advances(Organized by Professional Group on

Information Theory)Chairman, L. A. DE ROSA

(Federal Telecommunications Laboratories,Inc., Nutley, N. J.)

27.1. RECENT ADVANCES ININFORMATION THEORY

L. A. DE ROSA(Federal Telecommunications Laboratories,

Inc.)

27.2. RADAR PROBLEMS ANDINFORMATION THEORY

HARRY DAVIS(Rome Air Development Center,

Rome, N. Y.)A short review of the application of infor-

mation theory to radar problems, as carriedout by North, Van Vleck and Middleton,

Woodward, Leifer, and others is presented.A comparison of several methods of detect-ing radar signals in noise is made, relatingthese methods to the theoretical analysesand showing similarities in end result.Finally, the use of the foregoing material to apractical designer is outlined, consideringthe signal-to-noise enhancement and clutterrejection problems.

27.3. ANALYSIS OF MULTIPLEXINGAND SIGNAL DETECTION BY

FUNCTION THEORYNATHAN MARCHAND

(Marchand Electronic Laboratories,Greenwich, Conn.)

A general signal, which may be any time-varying function, is analyzed in multidimen-sional space where instants in time are theco-ordinates. It is shown how it is possible totake any signal in a limited-time intervaland obtain a multiplexing set by multidi-mensional-vector transformation. The prop-erties of the set are discussed and related tothe bandwidth and noise. The detection andcontamination of any signal in the presenceof other signals and noise is shown to dependupon the orthogonality of the functions rep-resenting the signals. Linear and the so-calledasynchronous multiplexing are shown to besimilar and to fall within the same mathe-matical analysis. Matrix transformationsand their use in signal detection are illus-trated. Circuit equivalents of function opera-tion for the detection of signals are shown inblock diagram form. Correlation analysis isfound to be a special case of function-theoryanalysis. It is shown that analysis by corre-lation techniques only gives limited resultsin most practical cases,

27.4. OPTIMUM NONLINEAR FILTERSFOR THE EXTRACTION ANDDETECTION OF SIGNALS

L. A. ZADEH(Columbia University, New York, N. Y.)A system of classes of nonlinear filters

designated as F1, F2, F3, * , is considered.The system is such that Fn_1 is a subclass ofF0, and the class of linear filters is a subclassof Fl. The input-output relationship for afilter in class F,, has the form of an n-fold in-tegral of a function which depends on n-agevariables, ri, T2, ' ' T,n and the values ofthe input at the instants t-rl, t-T2, * * * ,

-Tn,. The optimization (in the least squaressense) of a filter in class F,, requires theknowledge of 2nth order probability densityfunctions for the signal and noise, and re-duces to the solution of a linear integralequation of 2nth order. The optimization offilters of class F1 and their realization is con-sidered in detail.

27.5. DETECTION OF INFORMATIONBY MOMENTS

J. J. SLADE, JR., S. FICH ANDD. A. MOLONY

(Rutgers University, New Brunswick,N. J.)

Theoretical and practical considerationspoint to the importance of time moments inrepresenting and detecting information. Atime-limited function can be represented interms of Gaussians by the canonical form of

1953 417

PROCEEDINGS OF THE I.R.E.

the Gram-Charlier series in which the lead-ing term has the same area, mean, and spreadas the given function. These three parame-

ters which are determined by the first threetime moments can be made to convey infor-mation. In practice, any pulse-modulationsequence can also be identified by moments.These moments can be computed by conven-

tional integrators without the use of multi-pliers. The required circuitry and the effectsof network distortion and noise will be dis-cussed.

SESSION 28Communications Systems(Organized by Professional Group on

Communications Systems)Chairman, G. T. ROYDEN

(Mackay Radio and Telegraph Co.,New York, N. Y.)

28.1. AUTOMATIC-TUNING COM-MUNICATIONS TRANSMITTER

M. C. DETTMAN(Federal Telecommunication Laboratories,

Nutley, N. J.)This paper describes a 100/500-watt

transmitter that was developed to fill theneed for a modern medium- and high-fre-quency shipboard transmitter. Automatictuning to any frequency within this fre-quency range without need for a multiplicityof preset controls is featured. Total tune-uptime under normal conditions is about 30seconds. Facilities for high-speed keying,facsimile, and frequency-shift operation are

provided. The equipment also includes facili-ties for the usual types of emission encoun-

tered in this operating-frequency range. Con-siderable flexibility of installation is pro-

vided by the grouping arrangement. Theequipment and individual chassis are de-scribed.

28.2. DOUBLING TRAFFIC CAPACITYOF SINGLE-SIDEBAND SYSTEMS

C. D. MAY, JR.(Office of the Chief Signal Officer,

Washington, D. C.)Radioteletypewriter service is the pri-

mary means of radio communications be-tween the United States and Overseas ArmyCommands. At the present time single-side-band radio circuits with a capability of sixsixty-word per minute teletypewriter cir-cuits are in use. Several of these circuits havereached their traffic capacity due to in-creased requirements for Overseas Armycommunications. This necessitated the de-velopment of a means of expanding the ca-

pacity of these multichannel circuits. Thepurpose of this paper is to discuss a methodused to derive additional traffic channelsfrom the existing single-sideband systems.

The system presently used maintains cir-cuit reliability by transmitting duplicatetraffic simultaneously on separate tones toovercome the effects of selective fading. Thisfrequency-division arrangement requiresdouble the bandwidth that is actually re-

quired to transmit the traffic. By eliminatingthe frequency-diversity scheme and develop-ing a space-diversity arrangement the bandformerly occupied by the frequency-diver-

sity tone can be used to transmit the trafficfrom a second set of six-channel terminalequipment.

With the addition of necessary frequencyconversion equipment the traffic from thetwo sets of six-channel terminal equipmentcan be converted in frequency to permittones from the normal receiver to be chan-neled through the normal circuits of the un-

modified terminal equipment and the tonesfrom the diversity receiver to be channeledthrough the diversity circuits of the unmodi-fied terminal equipment. Actual physicaland electrical changes to the equipment are

minor and can be accomplished on a patchbasis with minor modification.

Operational experience on a twelve-chan-nel single-sideband radio circuit indicatesthat a degree of reliability at least as good as

the original six-channel circuit can be ex-

pected.28.3. PERFORMANCE OF SPACE AND

FREQUENCY DIVERSITY RE-CEIVING SYSTEMS

M. ACKER AND R. E. LACY(Signal Corps Engineering Laboratories,

Fort Monmouth, N. J.)AND

J. L. GLASER(Bell Telephone Laboratories, Inc.,

New York, N. Y.)Information is provided concerning the

physical installation of antennas for space

diversity systems, and the frequency spacingin frequency diversity systems in order to in-crease the communication performance of a

radio system operating in the high-frequencyrange over ionospheric paths. Means are

provided for utilizing most economically thearea that may be available for antenna in-stallations. Charts are included from whichcan be predicted the increase in communica-tion performance that may be achievedthrough the installation of diversity systemsin any particular communication circuit.This information is the result of a joint in-vestigation for over four years by the SignalCorps and Washington University.

28.4. EFFECT OF HITS INTELEPHOTOGRAPHY

P. MERTZ AND K. W. PFLEGER(Bell Telephone Laboratories, Inc.,

New York, N. Y.)Brief variations in attenuation or gain of

communications links known as "hits" maycause objectionable marks in received pic-tures unless the circuits are engineered tolimit hit intensity and duration sufficiently.This study includes judging the objection-ableness of about 200 individual hits on a to-tal of 16 positive prints of received picturesby ten observers, and measuring the hit in-tensities and durations. The paper containsgraphs of these two variables for various de-grees of picture impairment. A smoothedsummary curve indicates tentative limits ofhit intensity tolerance as a function of hitduration, for individual hits.

28.5. RELIABILITY OF MILITARY ELEC-TRONIC EQUIPMENT AND OUR

ABILITY TO MAINTAIN ITFOR WARA. S. BROWN

(Stanford Research Institute,Stanford, Calif.)

This paper discusses the increased em-

phasis being placed upon improved reliabil-ity by all concerned with the developmentand production of military electronic equip-ment. The author points out that there hasbeen very little reduction in the number oftypes and makes of equipment for similaruses. Under the free enterprise system of our

country, which we would not want other-wise, we produce as many different makes oftelevision equipments as business finds prof-itable. For military use the Armed Forcesare following a similar pattern with Army,Navy, and Air Force going their separateways in many cases to come up with individ-ual designs by different manufacturers. Rec-ommendations are made for far-reachingstandardization which would increase thefeasibility of high-quality, high-speed pro-duction, and simplify the maintenance prob-lem.

SESSION 29: SYMPOSIUMTelevision Broadcasting

and UHF(Organized by Professional Group on

Broadcast Transmission Systems)Chairman, GEORGE STERLING

(Federal Communication Commission,Washington, D. C.)

29.1. A FLEXIBLE TV STUDIO INTER-COMMUNICATION SYSTEMR. D. CHIPP AND R. F. BIGWOOD(DuMont Television Network,

New York, N. Y.)The manifold requirements of a televi-

sion studio intercommunication system willbe discussed, with emphasis on the need forspeed of communication and complete flexi-bility of interconnection between the manylocations involved. Various systems used inthe past will be described, followed by de-tailed consideration of a modern systembased on the use of "cross-bar' techniques.With this type of equipment, appropriatemembers of the production and engineeringteam can select direct circuits to other loca-tions, or can tie various groups together on a

common circuit. Interlocks and light signalsare incorporated to minimize operational dif-ficulties.

29.2. CBS-TELEVISION'S HOLLYWOODTELEVISION CITY: VIDEO, AUDIO

AND COMMUNICATIONFACILITIES

RICHARD O'BRIEN, ROBERT MONROE ANDPRICE FISH

(Columbia Broadcasting System,New York, N. Y.)

CBS-Television recently inauguratedtelevision service from its new HollywoodTelevision City headquarters. Located on a

25-acre site adjacent to the famous Farmer'sMarket, Television City is constructed froma flexible and expansible master plan thatpermits ultimate expansion to 24 studio

units.The initial construction phase now com-

pleted provides two audience and two non-

418 March

1953

audience studios, each exceeding one-fourthacre in area, together with the necessarytechnical, production, scenery construction,and office facilities to support completelyCBS-Hollywood television operations.

This paper describes the philosophy un-derlying the design of the video, audio, andcommunication facilities for this project.Emphasis is placed on description of meth-ods and features that are new or novel.

29.3. AN EXPERIMENTAL STUDY OFWAVE PROPAGATION AT 850 MCJESS EPSTEIN AND D. W. PETERSON

(RCA Laboratories Division,Princeton, N. J.)

In establishing a TV broadcasting sta--tion the prediction of the service area is ofvital importance. This involves a knowledgeof the mechanism of radio propagationwhich is made exceedingly complex becauseof its dependence on numerous physical phe--nomena. In general, this would involve anevaluation of such well-known factors aswave refraction, reflection, diffraction, ab--sorption, and scattering as a function of fre--quency and time.

This paper offers a study of propagationcharacteristics at the upper edge of the ul-tra-high TV band. A further simplificationof the problem is made by limiting it tomeasurements of wave propagation out todistances of 30-40 miles which would be theapproximate optical horizon of most trans-mitting-antenna heights likely to be used.This limitation permits the variation of fieldstrengths with time to be ignored since pastexperience has shown that this functional de-pendence is relatively unimportant withinthe optical horizon.

Theoretical prediction of wave propaga--tion for even highly idealized conditions ofthe various parameters is difficult and is ex-ceedingly complicated in actual practice byextreme deviations from the idealized form.The ultimate goal of these investigations isto formulate procedures which will permitthe prediction of median field strengthsthroughout typical broadcast service areas.The achievement of such a goal will greatlydepend upon the accumulation of experi--mental data against which the theoreticalformulations can be checked.

The purpose of this paper is to describean experimental project conducted at 850 mcand aimed at obtaining some of this neededinformation. Since the properties of propa--gation at these frequencies are related toboth the height of the transmitting and re-ceiving antennas, arrangements were madeso that these factors could be varied. At thetransmitter site, antennas were installed atfour different heights on the WOR 760-foottower. The effective radiated power was ob-tained by use of high-gain nondirectional an--tennas with a narrow, vertical, and broadhorizontal beam. By employing a narrow ele--vation beam and making the antennas tilta--ble it was possible to direct the full effectiveadiated power at any receiving site. Provi-

-sions were also made so that measurementswould be along two typical radials, onesmooth and the other relatively hilly. Thedata thus obtained have been analyzed sta--tistically to obtain the trends of the medianfield strengths, for a variety of typical re-

IRE National Convention Program

ceiving locations. An effort has been made toseparate and measure the losses introducedby houses and trees as compared to thatwhich is attributable to hills. A knowledge ofthe magnitude of these two effects underknown experimental conditions is an essentialprerequisite in attempting to formulate a

theoretical basis for the calculation of wavepropagation for a known topography.

29.4. UHF POWER TUBES IN TVAPPLICATIONSD. H. PREIST

(Eitel-McCullough, Inc., San Bruno,Calif.)

29.5. HIGH-POWER UHF KLYSTRONAPPLICATIONA. E. RANKIN

(General Electric Co., Schenectady, N. Y.)This paper discusses the application of a

family of six 15-kw, three-resonator klys-trons as radio-frequency amplifiers withparticular emphasis on their performance inaural and visual television service in the ul-tra-high frequency band.

Features, designed with the equipmentdesigner and station operator in mind, andwhich facilitate handling and installation,are reviewed.

Tube ratings and characteristics are de-scribed as they relate to and govern per-

formance.The effect of stagger tuning the three

resonators to obtain the bandwidth responseneeded for visual television is described.

Equipment design considerations, in-

cluding power-supply requirements, radio-frequency circuit requirements, and tube-protective devices are discussed.

29.6. HIGH-POWER UHF KLYSTRONAMPLIFIER DESIGN

N. P. HIESTAND(Varian Associates, San Carlos, Calif.)Latest developments in the design of

multiple-resonator high-power klystron am-

plifiers that cover the uhf band from 400 to1,000 mc are described. 15-kw, 3-resonatortubes are now in production which will pro-vide a narrow-band power gain of over 33 dbwith an efficiency of almost 40 per cent atsaturation level. Tunable over an 11 per centfrequency range, these integral cavity tubesprovide full-power output over the entireband.

Advanced design work has been com-

pleted on a 75-kw, 4-resonator amplifier andprogress on this development is described insome detail. Both tubes are particularly suit-able for use in uhf television transmitters.

29.7. HIGH-POWER. UHF TELEVISIONBROADCASTING SYSTEMS

H. M. CROSBY(General Electric Co., Syracuse, N. Y.)Up to 300-kw ERP may be obtained in

the uhf-television band by using a 12-kwtransmitter and a five-bay helical antenna.

The General Electric 12-kw uhf-televi-sion transmitter is made up of a complete100-watt transmitter and separate high-power klystron amplifiers for the visual andaural signals.

419

The 100-watt transmitter features a fre-quency-control circuit which effectivelylocks together the aural and visual carrierswith a fixed separation of 4.5 mc. Power am-plification is obtained by tetrodes which plugin to cavity-type circuits.

The 12-kw amplifier with its associatedrectifier and control equipment is built infour cubicles. The klystron used in this am-plifier has many advantages over conven-tional tubes for power amplification at uhf.

The helical antenna offers high gain perbay, a minimum of feed points, and the pos-sibility of null "fill-in" adjustment in thefield.

Operating and propagation tests havenow been made at the first two high-poweruhf installations.

SESSION 30: SYMPOSIUMMicrowaves I-Manufacture

of Microwave Equipment(Organized by Professional Group onMicrowave Theory and Technique)

Chairmsan, HARALD SCHUTZ(The Glen L. Martin Co., Baltimore,

Md.)

30.1. HOW TO DESIGN MICROWAVECOMPONENTS FOR EASE OF

ASSEMBLINGFRANK NEUKIRCH

(NRK Manufacturing and Engineering Co.,Chicago, Ill.)

Microwave components have becomestandardized to a considerable extent. Elec-trical requirements necessitate manufactur-ing to a high degree of accuracy and putthem into the instrument class.

Engineers starting out to design the com-plex microwave circuits required in present-day radars and guided-missile programsshould try to avoid close tolerances wherepossible.

Methods developed in order to simplifymanufacture are: a. precision casting, b.merco-cast process, c. electroforming, d.tube bending, e. dip brazing and other braz-ing methods, f. fabricating.

If the above methods are applied intelli-gently and if the important point of keepingtolerance requirements as loose as possible isfollowed, the average machine shop will beable to produce delicate components withouttoo much difficulty.

30.2. THE DESIGN OF MICROWAVECOMPONENTS FOR PRODUCTION

H. J. RIBLET(Microwave Development Labs., Inc.,

Waltham, Mass.)This paper discusses some of the prob-

lems of fabricating microwave componentsas they affect initial design and develop-mental effort. The special advantages of lost-wax casting broaching, forging, and form-tool cutting are reviewed. Alternate proce-dures for fabricating two waveguide-to-co-axial transitions are discussed by way of ex-amples. The importance of maintaining closeliaison between engineering and manufac-turing is emphasized.

PROCEEDINGS OF THE I.R.E.

30.3. FABRICATION OF MICROWAVECOMPONENTS EMPLOYING THE

DIP BRAZING PROCESSW. |. RUDOLPH

(The Glenn L. Martin Co.,Baltimore, Md.)

A short discussion on aluminum brazingmethods, the materials and alloys employedwill be used as an introduction. Problems ofjoint design and proper fixturing of assem-

blies will be surveyed, along with variousmeans of solving these problems. Filler sizeswill be discussed along with the methodsused to hold close tolerances. The prepara-tion of the base materials prior to brazing isone of the most important factors in produc-ing a satisfactory brazed joint; therefore,precleaning, deburring, etc. will be used as

another topic. Then, flux-removal proce-dures and other items will be presented.

In closing, a brief summary will be given.

30.4. ELECTROFORMING WITHCOPPER AND NICKEL AND

OTHER METALSC. L. DUNCAN

(C. L. Duncan Co., Chamblee, Ga.)1. Purpose of Electroforming: To pro-

duce internal surfaces of various shapes toclose tolerances and fine surface finish.

2. Scope of Electroforming: Electroform-ing is a specific field and does not competewith die casting, stamping, drawing, etc. Itis not a cheap method of manufacture.

3. Research and Development Com-pared to Production: Relation of assemblyand fabrication of intricate shapes to electro-forming in compounded sections and theelimination of joining.

4. Some Qualities of ElectroformedNickel and Copper: Experiences in machin-ing, soft and hard soldering.

5. Electroforming with Aluminum, Sil-ver, Gold, and Iron: Experiences with elec-troforming these metals.

6. Open Discussion for Question and An-swer Period: Practical problems and meth-ods.

30.5. MANUFACTURING OF "MICRO-STRIP" PRINTED CIRCUITS

COMPONENTSH. F. ENGELMANN

(Federal Telecommunication Laboratories,Nutley, N. J.)

The manufacture of components utilizingthe "microstrip" transmission system maybe achieved by mass production techniques.

Among the applicable methods are thephoto-engraving, silk-screen, and vacuum-metal deposition processes. A further simpli-fication in the manufacture of completemicrowave systems is possible since theabove processes are also suited to printed-circuit and printed wiring techniques. Thusit is quite practical to simultaneously"print" microwave and low-frequency cir-cuits and circuit elements.

A general discussion of various manu-facturing techniques, with a detailed de-scription of the manufacture of a completemicrowave receiver utilizing the photo-engraving process, will be included.

SESSION 31: SEMINARAcoustics for the Radio

Engineering-II(Organized by Professional Group on

Audio)Chairman, J. J. BARUCH

(Massachusetts Institute of Technology,Cambridge, Mass.)

31.1. PHONOGRAPH REPRODUCERSB. B. BAUER

(Shure Brothers, Inc., Chicago, Ill.)

31.2. TAPE RECORDINGMARVIN CAMRAS

(Armour Research Foundation,Chicago, Ill.

31.3. STUDIO ACOUSTICSH. J. SABINE

(Celotex Co., Chicago, Ill.)The discussion will centralize around

broadcasting studios, and the engineeringuse of phonograph reproducers and magneticrecording.

Continual interplay among the membersof the panel and between the panel and theaudience will assure the focusing of attentionon the aspects of acoustics which are of in-terest and intense importance to the radioengineer.

SESSION 32: SYMPOSIUMNucleonics

(Organized by Professional Group on

Nuclear Science)Chairman, L. V. BERKNER

(Associated Universities, Inc.,New York, N. Y.)

32.1. SERVOS FOR REMOTEMANIPULATORS

R. C. GOERTZ AND F. BEVILACQUA(Argonne National Laboratory, Lemont, Ill.)

J. R. BURNETT(Purdue University, W. Lafayette, Ind.)Master-slave manipulators have become

quite popular for nonroutine general purposehandling and manipulations involving radio-active materials.

Servomechanisms which reproduce me-

chanical position and reflect the load are

being developed to replace the mechanicalconnections used in most of the presentmaster-slave manipulators. These servosmust maintain proportional position andforce correspondence between the input(masterhandle) and output (slave tongs or

tool) for all velocities, forces, and inertialoads from zero to the maximum capabilitiesof the manipulator. Several schematic ar-

rangements will be discussed which fulfillthe requirement for positional correspond-ence, force reflection, and bilateral action ofthe servos. Analysis by impedance coniceptquickly leads to some basic requirements ofthe system. Possibilities for incorporatingthese devices into robots will be discussedbriefly.

32.2. TWO NEW PHOTOMULTIPLIERSFOR SCINTILLATION COUNTINGM. H. GREENBLATT, M. W. GREEN,P. W. DAVISON, AND G. A. MORTON

(RCA Laboratories Division,Princeton, N. J.)

The present paper describes two new de-velopmental multiplier phototubes whichare designed to meet some of the recent needsof scintillation counting. These are develop-mental numbers H-5037 and H-4646.

A large phosphor crystal is desirable forgamma-ray spectroscopy and for obtainingcomplete absorption of high energy particles.A photomultiplier with a large photocathodeis necessary in order to realize the full ad-vantage of a large crystal. DevelopmentalNo. H-5037 was developed for use with largescintillation crystals.

It is also desirable to have a photomulti-plier with a gain high enough to eliminatethe need for pulse amplifiers with theirnecessarily limited frequency response. Sucha tube is very useful for studying very fastphenomena, for portable survey instrumentsand for use in cases where the photomulti-plier is necessarily in a remote location. De-velopmental No. H-4646 is suitable for suchapplications.

Some characteristics of these two tubesare given in Table I.

Multiplier Type:TABLE I

H-5037 H-4646

Dimensions:

diameter 4 inches 1 incheslength 7 inches 7j inches

Cathode:

dimensions 31 inches dia. i inch Xl inchsensitivity 30-50 ua/l 30-50 ua/Ispectral type s-9 s-9collection eff. high high

Gain:

number dynodesgainoverall voltage

10106

1,000

1610'

2,000

The H-5037 uses a cylindrical electrostat-ic lens to focus electrons from the photo-cathode onto the first stage of the multiplier.The multiplier structure is similar to theRCA 931-A. The collection efficiency of the,electron optical system used is quite good.Preliminary tests indicate that results con-

sistent with the larger photocathode area are

obtainable.The H-4646 has a much smaller photo

cathode area but has, in addition to excellentcollection efficiency, a strong electrostaticcollecting field in the vicinity of the cathode.This decreases the transit time spread andalso makes the tube less susceptible to inter-ference from external magnetic fields. Thevery high gain of the tube introduces a

number of space charge problems. These are

met by a special 16th dynode-anode struc-ture. The output is through a 125 ohm co-

axial transmission line system. Saturationoutput current is about 300 ma.

32.3. BILLION-ELECTRON-VOLTACCELERATORS

G. K. GREEN(Brookhaven National Laboratory,

Upton, Long Island, N. Y.)Accelerators producing particles with en-

ergies of millions of electron volts to hun-dreds of million electron volts have been

420 March

IRE National Convention Program

used to explore the properties of the atomicnucleus. However these machines are notadequate for detailed examination of theparticles of the nucleus. Four machines for atleast one billion electron volts are now beingbuilt and a fifth-the Brookhaven Cosmo-tron-is giving proton energies above twobillion electron volts. The Cosmotron syn-chronizes a changing radio frequency withan increasing magnetic field to accelerateprotons for a path length of some 150,000miles. New theoretical discoveries indicatethat a one hundred billion ev machine is anengineering and economic possibility.

32.5. INSTRUMENTATION DEVELOP-MENTS IN FAST NEUTRON

DOSIMETRYG. S. HURST

(Oak Ridge National Laboratory,Oak Ridge, Tenn.)

Fast neutrons are more harmful to manybiological systems than is an equal amount ofgamma radiation; thus fast neutron dosim-eters should be gamma insensitive. Twosuch dosimeters have been developed. Thefirst type uses a proportional counter whosecount rate response versus energy is thesame as the first collision tissue dose curve.This detector is simple and adapts itself toportable instruments, of which a commercialmodel is now available. The counter has adependence on the direction of the neutrons,which is advantageous in some cases.

The second type uses a proportionalcounter which is designed in accordance withthe Bragg-Gray principle and hence is non-directional. Dose is determined by addingpulse heights which may be done by a simplepulse integrator which uses two ordinary bi-nary scaling units.

SESSION 33Information Theory II-

Theoretical(Oganized by Professional Group on

Information Theory)Chairman, W. G. TULLER

(Melpar Inc., Alexandria, Va.)

33.1. ERROR PROBABILITIES OFBINARY DATA TRANSMISSION

SYSTEMS IN THE PRES-ENCE OF RANDOM

NOISES. H. REIGER

(Air Force Cambridge Research Center,Cambridge, Mass.)

In a binary data transmission system, theerror probability depends on the transmitterpower, receiver sensitivity, pulse shape,modulation method, and detection charac-teristics; however, the output S/N ratio ofthe optimum receiver is dependent only onthe signal energy of the pulse and the noisepower per unit bandwidth. In practice, forlarge S/N, the ideal case may be approachedvery closely with simple filters. For smallS/N this may be done if the transmission, iscoherent.

Minimum error probabilities on the basisof an 'ideal observer" have been computednumerically for the following systems: a. car-rier keying (non-coherent), b. carrier keying

(coherent), c. frequency shift keying, d. rfphase shift keying.

The average information value of eachpulse has been computed for the four sys-tems as a function of the S/N ratio and theyare compared with Shannon's formula.

33.2. THE STATISTICAL PROPERTIESOF THE OUTPUT OF CERTAIN

FREQUENCY SENSITIVEDEVICES

G. R. ARTHUR(Sperry Gyroscope Co., Great

Neck, N. Y.)Frequently it is desirable to know the

statistical properties of the output of fre-quency sensitive devices when some type ofrandom signal is impressed. This problem isessentially one of the statistics of the differ-ence of two dependent random quantitiesand the statistics of a filtered signal.

The problem of finding the probabilitydensity of the difference of two dependentrandom variables has not been solved in anygeneral way. This paper gives the solution ofthis problem for the case of a frequency dis-criminator excited by a narrow-band spec-trum. It is solved by expressing the output ofthe device as the difference of two independ-ent random variables which allows the use ofthe characteristic function method. Theproblem of passing a non-Gaussian randomsignal through a low-pass filter is then solvedby obtaining the predominant moments ofthe density at the filter output. This methodclearly demonstrates the approach of thatdensity to a Gaussian as the filter band ismade narrow.

33.3. CROSS-CORRELATION APPLIEDTO AUTOMATIC FREQUENCY

CONTROLM. J. STATEMAN

(Sylvania Electric Products, Inc.,Bayside, N. Y.)

Automatic frequency control requiresthat an error signal be obtained as a functionof the slaved and controlling voltages. Thiserror signal is used in a feedback loop tominimize the difference in phase between thetwo voltages. Such an error voltage must in-dicate both the direction and the magnitudeof the existing difference to be effective inobtaining and maintaining the locked condi-tion in the presence of noise. Using examplesof AFC from television citrcuitry, standardgraphical analyses are followed by correla-tion techniques which present a new view-point concerning the criteria of waveformssuitable for AFC.

33.4. APPROXIMATE PROBABILITYDENSITY FUNCTION OF FIRSTLEVEL CROSSING FOR LINE-ARLY INCREASING SIGNAL

PLUS NOISEG. PRESTON AND R. GARDNER

(Philco Corporation, Philadelphia, Pa.)The solution of many timing and syn-

chronizing problems depends upon having adevice that registers the time a signal firstcrosses a given cutoff level. Often the signalcan be considered to be a linearly increasingfunction of time with a superimposed ran-dom noise. The noise introduces a variationin the time the composite signal first crosses

the cutoff level causing jitter in the outputsignal. T'his effect is described by the prob-ability density function for the first cutoff-level crossing. In this paper an approximateexpression for the probability density func-tion is obtained for the case of normal noise.

The conditional probability-density func-tion is obtained by linear extrapolation ofthe composite signal from some convenientpoint. By successive application of this prob-ability-density function, the desired densityfunction can be plotted. The resultant dis-tribution has many of the.characteristics of anormal distribution whose rms deviation de-pends upon the ratio of the signal slope tothe rms value of the noise.

By reference to work in the literature,the assumption of normal noise can be veri-fied for a receiver with a wideband IF and anarrow-band video. One possible applica-tion, satisfying these requirements, is in thetelevision synchronization problem. Graphsare shown of the probability-density func-tion with values typical of such an applica-tion. One result is that less jitter is obtainedas the video bandwidth is extended althoughthis action decreases the conventional sig-nal-to-noise ratio.

33.5. OPTIMUM DEMODULATIONF. W. LEHAN AND R. J. PARKS

(California Institute of Technology,Pasadena, Calif.)

The problem of demodulation of a signal,modulated in a general way, is studied usingthe statistical technique of curve fitting.This technique attempts to make the bestpossible fit to an incoming noisy signal witha locally generated signal by proper choice ofcertain parameters of the local signal. Thebest fit is defined to be that which maximizesthe likelihood of the noise function which isdefined as the difference between the incom-ing signal and the locally generated signal. Apriori assumptions concerning the modula-tion are introduced implicitly by the choiceof form of the locally generated signal. Otherassumptions may be introduced explicitly bymeans of Bayes Theorem. Various types ofmodulated signals are considered and undercertain simplifying assumptions the methodis found to lead to correlation detection, lo-cal-carrier insertion, and other so-calledideal detection techniques.

SESSION 34Medical Electronics

(Organized by Professional Group onMedical Electronics)

Chairman, L. H. MONTGOMERY, JR.(Metal Products Co., Nashville, Tenn.)

34.1. ELECTRIC PHOTOGRAPHYK. S. LION

(Massachusetts Institute of Technology,Cambridge, Mass.)

A great number of problems, particularlyin the fields of medicine and biology, requirethe use of photographic methods of extremesensitivity. Such problems arise, for in-stance, in medical radiography, in X-raydiffraction pattern technique, or in low-levelspectroscopy, where even the use of the mostsensitive photographic emulsion is not satis-

1953 421

PROCEEDINGS OF THE I.R.E.

factory or leads to an overdose of radiation,and intensifier screens must be employed.

The new photographic method presentedin this paper combines the advantages of thephotographic method (simultaneous two-di-mensional, pictorial presentation, and inte-gration) with the sensitivity of the Geigercounter. The sensitivity of this method isseveral orders of magnitude higher than thatobtained with ordinary photographic meth-ods. It may be called electronic photography,although the application of this method farexceeds the field of photography.

The method consists in a parallel-plateGeiger counter whereby a photographicemulsion, on a carrier, is brought into thecounting volume. The counter is filled withan appropriate gas and quenching agent, anda voltage source is applied. Under appropri-

ate operating conditions a radiation enteringthe counter will trigger a discharge whichdoes not spread throughout the counter butwhich is strictly localized and which locallyexposes the photographic plate. Theoreticalconsiderations show a possible increase ofsensitivity of a photographic emulsion by a

factor of 1,000. Actual tests so far have re-

sulted in an increase of sensitivity by a fac-tor of 100, with a resolving power accept-able for a great number of applications.

Experimental results and applications ofthis method in the field of photography andin other fields will be discussed.

34.2. CONCERNING THE USE OF HIGHENERGY PARTICLES AND QUANTA

IN THE DETERMINATION OFTHE STRUCTURE OF LIVING

ORGANISMSR. J. MOON

(University of Chicago, Chicago, Ill.)The important basic principles for the

determination of the structure of living or-ganisms by means of high-energy quantaand particles are developed. Primary empha-sis is put upon obtaining the information rel-ative to structure of the organisms with a

minimum amount of damage to it and amaximization of the amount of informationderived. Several instruments are consideredwith reference to their ability to fulfil theseconditions. Systems which employ "thick de-tectors" and derive their information in se-

rial-time sequence seem to fulfil these condi-tions best. Experimental work with a scan-

ning X-ray system performed with regard tothese principles is described.

34.3. POSSIBLE MEDICAL AND IN-DUSTRIAL APPLICATION OF

LINEAR ELECTRONACCELERATORS

W. C. BARBER, A. L. ELDREDGE, ANDE. L. GINZTON

(Stanford University, Stanford, Calif.)A linear-electron accelerator has many

possible applications since it provides a sim-ple means of obtaining intense electronbeams in the multimillion volt-energy re-gion. Uses to be considered are the directemployment of the electron beam for sterili-zation of biological materials or for cancertherapy. Alternatively, the electron beamcan be converted to high-energy X-rayswhich can be used for cancer therapy or ra-

diography of thick sections.Analysis and experience suggests that

linear accelerators are practical for any ofthese applications, and as an introductoryventure in this field the Microwave Labora-tory is constructing a 6-mev accelerator toproduce X-rays for cancer therapy.

34.4. CAPACITY AND CONDUCTIVITYOF BODY TISSUES AT ULTRA-

HIGH FREQUENCIESH. P. SCHWAN AND KAM Li(University of Pennsylvania,

Philadelphia, Pa.)

It has been recognized recently that elec-tromagnetic-radiation operating in the fre-quency range from 300 to 600 mc is muchmore suitable for diathermy than radiationoperating above 1,000 mc. However, no de-tailed data of the dielectric properties ofvarious body tissues within the range from100 to 1,000 mc are available at present.Such data are desirable in view of the factthat they permit more quantitative deter-mination of such data as depth of penetra-tion, reflection energy at tissue interfaces,and so on. Dielectric constant and conduc-tivity of various body tissues have beenmeasured, therefore, in the range from 200 to900 mc and are presented in this paper. Thesignificance of this data with respect toproblems of diathermy is discussed.

34.5. THE PROBLEM OF THE APPLI-CATION OF ELECTRONICS TO

MEDICINER. S. SCHWAB

(General Hospital, Boston, Mass.)The task of the physician is the prompt

recognition of bodily dysfunction (pathol-ogy), its identification and location (diag-nosis), and later correction or elimination(therapy).

He is handicapped in this task by inac-cessibility of many structures, their delicacy,the presence of pain, extreme variationsfrom mean values and relationships, andurgency.

Alone with his eyes-and ears-and thetouch of his hands, the physician is limitedin dealing with the many complicated varia-bles he knows exist, and which are early andclear signs of the puzzle he must solve.For better, more exact, more comprehen-

sive measurement of countless parameters,he turns to electronics for help.

Some basic difficulties in the liaison be-tween medicine and electronics are: (1)Most physicians have very little idea of lim-its of the measurements they want to obtain.A great many doctors have no idea of themeaning of significant figures. (2) On theother hand, most electronic engineers havelittle experience with the uncertainty ofmedical knowledge, of the immense numberof variables, many that are not known evento the well-trained specialist in his particularbranch of medicine. They have little idea ofjust how their services can be used, or whatis really wanted by the physician, and par-ticularly why it is wanted.

We need, first of all, a book on electronicsfor the doctor and a book on medicine for theelectrical engineer.

Five principal applications of electronicsand fifteen examples of existing electronicdevices that are in use in medical problemsare described.

Some future challenges to spark furtherdevelopments along these lines are men-tioned.

34.6. PROGRESS REPORT ON ELEC-TRONIC MAPPING OF THE ELEC-TRICAL ACTIVITY OF THE HEART

STANFORD GOLDMAN, D. W. SPENCE, MARYRIzIKA, AND SILVAN LIDOVITCH

(Syracuse University, East Syracuse, N. Y.)Electronic mapping (or area display) is a

method of investigating the electrical activ-ity of the heart. Many common diseases ofthe heart can be recognized and distin-guished by slow-motion pictures of their areadisplays. These include left and right ven-

tricular hypertrophy, left and right bundle-branch block, posterior and anterior and an-

terolateral infarction, and auricular fibrilla-tion. The movies give an informative pictureof the pathological physiology.

Moving pictures will be shown illustrat-ing normal and abnormal types. The relationbetween electronic mapping and vector car-diography will be discussed.

SESSION 35Broadcast and Television

Receivers-I(Organized by Professional Group on

Broadcast and Television Receivers)Chairman, G. L. BEERS

(RCA Victor Division, Camden, N. J1)

35.1. GAIN STABLE MIXERS ANDAMPLIFIERS WITH CURRENT

FEEDBACKG. E. BOGGS

(National Bureau of Standards,Washington, D. C.)

Narrow-band radio-frequency amplifiersand mixers may be stabilized by negativefeedback without increasing the bandwidthexcessively. A couple of this type using cur-

rent feedback is described. This couple re-

quires only a simple resistive beta circuit andmay be designed such that the bandpasscharacteristic is largely independent of thefeedback. Consideration is also given to theproblem of input impedanice. Amplifier cou-

ples of this type with 20 db of feedback havebeen operated at 15 mc and can probably beused at higher frequencies. Current feedbackmixer couples have been used satisfactorilyat frequencies as high as 50 mc.

35.2. VIDEO AMPLIFERS WITHINSTANTANEOUS AUTOMATIC

GAIN CONTROLW. E. AYER

(Stanford University, Stanford, Calif.)Circuits are described which allow essen-

tially complete control of the output-inputamplitude characteristic of multistage videoamplifiers for both positive and negative in-put signals. The incremental gain of eachstage is determined instantaneously by thesignal current through the tube, diodes beingemployed to introduce degeneration for sig-nals above a certain amplitude. The gain re-

duction achieved in this manner lasts only as

long as a strong signal is present so that re-

covery time is not adversely affected.As a typical practical example, these cir-

422 Marcit

IRE National Convention Program

cuits may be readily utilized to provide a"logarithmic" amplifier, the output voltagebeing proportional to the logarithm of the in-put voltage.

35.3. AN AUTOMATIC LEVEL-SETTING SYNC AND AGC

SYSTEME. 0. KEIZER

(RCA Laboratories, Princeton, N. J.)AND M. G. KROGER

(Motorola, Inc., Chicago, Ill.)The failure of many commercial televi-

sion receivers to remain in stable synchro-nism in the presence of high-energy types ofinterference is often due to charging up dur-ing the noise pulses of capacitors in the syncseparator and agc circuits of the receiver. Asystem has been developed which largelyovercomes this failure. In this system, theagc is derived following a dc-coupled-syncseparator in such a manner that any depar-ture from the correct operating level for thesync separator is counteracted by a changein agc voltage and sync-separator bias. Threetube functions are required for the system.Both polarities of sync output are provided.

35.4. PACKAGED ADJACENT CHAN-NEL ATTENUATION

J. P. VAN DUYNE(Allen B. Du Mont Laboratories, Inc.,

Clifton, N. J.)The problem of adjacent-channel attenu-

ation is discussed in its relation to the alloca-tion problem, the economics of receiver de-sign, and the ultimate cost to the consumer.It is pointed out that the design of a televi-sion receiver for excellent adjacent-channelrejection is not compatible with economicaldesign for operation in the usual service area.A solution to this dilemma, in the form of

a plug-in adjacent-channel rejection filter, isproposed. The results achieved with the fil-ter described, indicate that the limitation tointerference-free reception in an adjacent-channel area is the nonlinearity of the RFtuner.

The design of a filter for such service im-poses several practical limitations on the de-signer. These limitations, and their influenceon the design problem, are discussed.

35.5. METHODS OF MATRIXING INAN NTSC COLOR TELEVISION

RECEIVERW. M. QUINN

(General Electric Co., Syracuse, N. Y.)In order to obtain the required Red,

Blue, and Green signals in an NTSC colortelevision receiver, it is necessary to combineor matrix the brightness signal with the twodetected components of the chrominancesignal.

Several methods have been employed toaccomplish this matrixing. One method con-sists of applying the Yor brightness signal tothe three grids of a tri-color kinescope andthen applying the three-color difference sig-nals (R- Y, B- Y, G- Y) to the individual cath-odes. This particular method utilizes thekinescope itself as an adder. Other methodsare resistive matrixing, summing amplifiers,and the feedback summing amplifier. All ofthese methods will be discussed with themost emphasis being given to the feedback

summing amplifier. Consideration will begiven to linearity, phase distortion, band-width, and general performance.

SESSION 36Microwaves II-Discontinui-

ties and Transitions(Organized by Professional Group onMicrowave Theory and Techniques)

Chairman, G. A. DESCHAMPS(Federal Telecommunications Laboratories,

Inc., Nutley, N. J.)36.1. RF MEASUREMENTS ONMETALLIC DELAY MEDIA

S. B. COHN(Sperry Gyroscope Co., Great Neck, N. Y.)

This paper presents RF index-of-refrac-tion data for metallic delay-line media con-taining square and circular obstacles, andcompares data for strip obstacles with theo-retical values. The measuring equipmentused is described, and the necessary correc-tion formulas are given. The test specimensconsisted of alternate layers of polyfoamspacers and thin polystyrene sheets im-printed with conducting obstacles. The RFdata was correlated with the static data fromelectrolytic-tank measurements, in order toobtain graphs of index of refraction versusfrequency suitable for design purposes. Asshown in this paper, the data may be ex-tended readily to other techniques of fabri-cation. A practical and economical methodof construction will also be described.

36.2. IMPEDANCE MEASUREMENTSIN A CIRCULAR WAVEGUIDEWITH TEol EXCITATION

L. S. SHEINGOLD(Sylvania Electric Products Inc.,

Boston, Mass.)A unique method of measuring imped-

ance in a circular waveguide supporting onlythe dominant circular-electric wave is de-scribed. Results are given of precise imped-ance measurements made on circular ob-stacles, circumferential gaps, radiatingguides, and sharp axial bends. The experi-mental results are compared with theoreticalvalues and are found to be in excellent agree-ment. It is demonstrated that application ofDeschamps' graphical method in the experi-mental determination of the scattering pa-rameters of a circumferential-gap junctionresults in a rapid determination of the perti-nent quantities; for example, the power re-flected by, power transmitted through, andpower dissipated in the junction.

36.3. EXPERIMENTAL DETERMINA-TION OF THE PROPERTIES OFMICROSTRIP COMPONENTS

M. ARDITI(Federal Telecommunications Laboratories,

Inc., Nutley, N. J.)Transmission properties of microstrip

components are readily obtainable experi-mentally due to the case of manipulation ofthe line element. The properties of a wide-band transition for connecting coaxial linesto microstrip systens are given. The scatter-ing matrix coefficients and equivalent cir-cuits are presented for a right-angle bend,

offset junction, step discontinuity, parallel-coupled junction, and for transverse posts,gaps, or slots in the strip conductor. Thesedata facilitate the design of resonant sectionsthat include such obstacles. Measurementshave verified the predicted performance ofthese components when assembled into amicrowave receiver.

36.4. A WIDE-BAND TRANSITIONBETWEEN WAVEGUIDE AND

COAXIAL LINEN. A. SPENCER AND H. A. WHEELER

(Wheeler Laboratories, Great Neck, N. Y.)In an unusual type of transition, two

parallel paths are inserted between a rec-tangular waveguide and a coaxial line. Thesetwo paths are the two halves of an H-planewaveguide ring, series connected at theguide junction and parallel connected at theline junction. The coaxial line has the maxi-mum diameter consistent with single-modepropagation and maximum pulse-power ca-pacity. The two-sided junction with the lineis conducive to wideband-impedance match-ing. The entire transition has been matchedwithin 0.3 db swr over a 10 per cent fre-quency bandwidth.

SESSION 37Radio Telemetry

(Organized by Professional Group onRadio Telemetry and Remote

Control)Chairman, M. V. KIEBERT

(Bendix Aviation Corporation, Teterboro,N. J.)

37.1. TELEMETERING REQUIREMENTSFOR UPPER AIR ROCKET RE-

SEARCH EXPERIMENTSMARCUS O'DAY

(Air Force Cambridge Research Center,Cambridge, Mass.)

Rocket-borne experiments which requiretelemetering, as well as other experimentswhich, although proposed, are restricted bydata transmission systems currently in use,will be described in this paper. The require-ments of a high-sampling rate will be con-trasted with that of high-time resolution,and some of the experiments which illus-trate the difference will be discussed. Therequirement of high accuracy for certainprojects, for example, the determination ofthe solar constant at high altitudes, will beanalyzed and possible solutions proposed. Inaddition, the speaker will discuss certainoperation restrictions, such as the inabilityto change batteries after the rocket is fueled,as well as limitations imposed by the para-chute-recovery system.

37.2. TELEMETERING-BROAD BANDON SHORT ORDER

T. F. JONES, JR.(General Electronic Laboratories, Inc.,

Boston, Mass.)The Navy required the delivery, in four

months' time, of a telemetering system hav-ing eight channels, each having a signalbandwidth of zero to ten ke. The specifica-tion on operating range dictated the use of

1p5 ? 423

PROCEEDINGS OF THE I.R.E.

an airplane-relay station. The requirementof side-by-side operation of similar systemson adjacent channels imposed stringent con-ditions on carrier stability and receiver sta-bility.

The need was met within the time limitby the development of an AM-AM systemwith the aid of experimental telemeter ele-ment designs supplied by NRL. Problemssuch as calibration, deviation limiting, sub-carrier stability, subcarrier interaction, sonicshielding, carrier stability, receiver stability,automatic control of relay-signal levels, andsubcarrier separation were adequately solved.

The basic design can meet effectively theneeds of other telemetering applications.

37.3. FLUTTER COMPENSATION FORFM-FM TELEMETERING

RECORDERJ. T. MULLIN

(Bing Crosby Enterprises, Inc.,Los Angeles, Calif.)

Records of FM-FM telemetering infor-mation reproduced from magnetic tape arerestricted in the accuracy with which infor-mation may be read because of flutter and dcdrift in the tape-transport mechanism. Amethod is described wherein the effects arereduced by electronic compensation duringplayback. A high-frequency pilot tone isadded to the FM-FM signals as they are re-corded. During playback all channels aremodulated by a high-frequency carrier. Theresulting side bands are demodulated sepa-rately for each channel by a carrier whosefrequency is controlled in an absolute man-ner by information derived from the pilottone. Demodulation thereby restores thechannel frequencies to their correct valueswith variations due to flutter and dc driftgreatly reduced.

37.4. A MAGNETIC TAPE RECORDINGSYSTEM FOR PRECISION DATA

L. L. FISHER(Ampex Electric Corporation,

Redwood City, Calif.)A new magnetic recording system has

been developed which makes it possible toaccurately record and reproduce all forms ofdata from the range of 0 to 5,000 cycles. Em-ploying an FM carrier technique, this newsystem overcomes the deficiencies of conven-tional magnetic recorders, allowing responseto dc with very-low phase shift, excellenttransient response, and complete freedomfrom the coating deficiencies and the non-hom*ogeneity of the magnetic medium. Spe-cial packaging arrangements have beenworked out to allow assembly of the equip-ment for any number of recording tracks upto 14 on 1" wide tape. Operation at a greatvariety of speeds makes possible frequencyexpansion, frequency contraction, and othertricks which allow complete flexibility inanalysis of recorded data. Use of such tech-niques on the "live" data signal allow the ul-timate amount of information to be derivedfrom the original recorded signal.

37.5. AN IMPROVED FM-FMDECOMMUTATOR GROUNiD

STATIONF. N. REYNOLDS

(The Ralph M. Parsons Co.,Pasadena, Calif.)

A new and straightforward approach tothe extraction of intelligence from variableamplitude data pulses is carried out in thesystem described herein. Theoretically, if alow-pass filter could be designed to have itscutoff frequency (fc) at one-half of the com-mutation rate (F/2) it would be possible torealize information from amplitude-modu-lated pulses up to a frequency equal to(F/2), assuming that the filter had an infi-nite rate of attenuation. In practice, such afilter cannot be built, but these characteris-tics can be approached at the higher fre-quencies by using LC-types of networks. Inthe frequency spectrum employed in theRDB-FM-FM standard telemetering equip-ment, it becomes uneconomical and wastefulin terms of space to build such low-passfilters. It is possible, though, to duplicate thetransfer function for a normal pi-section low-pass filter with 2 vacuum tube envelopes anda few resistor and condenser components. Inthis equipment two such filters are placed ina series and as an example, for a 30-cycle per-second commutator, the cutoff frequency ischosen at 10 cps, so that the maximum theo-retical rate of attenuation of 36 db per oc-tave produces an attenuation of approxi-mately 60 db at the commutation rate of 30cps. It can thus be seen that such a devicewhen driven with amplitude-modulatedpulses, essentially acts as not only a storagenetwork but also as a very reliable inte-grator, producing at the output of the filtersa voltage whose waveform is identical to thatof the input-modulation voltage up to thelimit of approximately 10 cps. The fre-quency response of this type of device isabout twice that which can be expected withthe generally used stair-step type of storagenetwork, and the voltage between samplesbecomes usable since it contains a high orderof intelligence approaching that of the input-modulation signal. The complexity of theequipment is quite a bit less than is normallyaccepted for a series-RC type of integratorand storage network, and as a consequence,the over-all reliability and operation effi-ciency is greatly increased. A much higherorder of linearity and accuracy is alsoachieved with this type of device.

37.6. SOME INDUSTRIAL APPLICA-TIONS OF TELEMETRY

H. R. HOYT(Great Lakes Pipe Line Co.,

Kansas City, Mo.)AND

J. H. VAN HORN(Midwest Research Institute,

Kansas City, Mo.)Some current techniques for telemetering

process variables in those industries havingwidely distributed operations are presented,in particular, those associated with the petro-leum industry. Telemetry systems employ-ing carrier-telephone and microwave tech-niques in the transmission of such variablesas pressure, flow rate, levels, positions, etc.and operating from a wide variety of trans-ducer elements are described.

SESSION 38Audio

(Organized by Professional Group onAudio)

Chairman, D. W. MARTIN(Baldwin Co., Cincinnati, Ohic)

38.1. SOUND SYSTEM FOR PLENARYCHAMBER OF UN GENERAL

ASSEMBLY BUILDINGC. W. GOYDER

(Telecommunication Div., UnitedNations, N. Y.)L. L. BERANEK

(Massachusetts Institute of Technology,Cambridge, Mass.)

The sound system for the plenary cham-ber of the General Assembly Building of theUnited Nations in New York was especiallydesigned to function properly in the unusualstyle of architecture of that room. The prin-cipal design requirements were intelligibil-ity, naturalness, concealment, and freedomfrom dangei of feedback even when operatedwith automatic-volume control. Satisfactoryspeech intelligibility is attained by faithfulreproduction of the frequency range between400 and 6,000 cps. Naturalness is assured byfreedom from nonlinear distortion, uniformcoverage of the hall at all frequencies, and bysuitable employment of time-delay systems.Sound-level distribution and response testshave been performed and the data are pre-sented in this paper.

38.2. DEVELOPMENT OF A VARIABLETIME DELAYKENNETH GOFF

(Massachusetts Institute of Technology,Cambridge, Mass.)

This paper describes the design, con-struction, and performance of a magnetic-re-cording drum time-delay system. The device,which has been developed for use in suchacoustic studies as noise localization, rever-beration analysis, and others amenable tocorrelation techniques, utilizes conventionalproportional recording. Two channels yielda relative time delay between their outputswhich may be varied from minus 15 ms toplus 190 ms with an accuracy of ±0.2 percent using the instrument dial.

Principal attention in the paper will bepaid to the following: 1. Obtaining a uniformlayer of magnetic material by spraying witha dispersion of iron oxide. 2. Analyzing theeffect of spacing between the magnetic ma-terial and the head upon the processes of re-cording the reproducing. 3. Developing amechanical driving system as free from flut-ter as possible.

38.3. A FLUX-SENSITIVE HEAD FORMAGNETIC RECORDING

PLAYBACKD. E. WIEGAND

(Armour Research Foundation,Chicago, Ill.)

A playback head of special constructionprovides signals proportional to the fluxfrom the recording medium, rather than itstime derivative. Power from a high-fre-quency oscillator is applied to the head. Theoutput signal appears across a separatewinding and is in the form of a modulatedcarrier, the carrier frequency being twicethat of the oscillator. Its inherently flat re-sponse extending to dc at the low-frequencyend makes the head useful in many instru-mentation applications, particularly in cases

424 March

IRE National Convention Program

where waveforms must be preserved, orwhen a recording is studied in detail by play-ing back the record at greatly reduced speed.Full signal levels are preserved when thespeed of the recording medium is reduced.In fact, the recording can be examined pointby point by taking readings with the mediumstationary. Signal levels are unusually high.This fact, and the elimination of the usuallow-frequency equalization make the newhead appear attractive in sound recording. Asimple adaptor for playback of magnetic re-

cordings through a standard broadcast re-ceiver requires, in addition to the drivemechanism, only a small oscillator feedingthe head and a small transmitting-loop an-

tenna fed by the signal winding of the head.The oscillator is tuned to one-half the fre-quency of a vacant spot on the radio dial.

38.4. UNIAXIAL MICROPHONEH. F. OLSON, JOHN PRESTON, AND

J. C. BLEAZEY(RCA Laboratories, Princeton, N. J.)A small unidirectional microphone has

been developed with the following features:maximum sensitivity along the axis of themicrophone, a high ratio of electrical outputto size, a sharper directivity pattern than acardioid, a directivity pattern that is inde-pendent of the frequency and a blast-proofvibrating system. The high discriminationwhich this microphone exhibits to soundswhich originate from the sides and rearmakes it particularly suitable for long-dis-tance-sound pickup in radio, television,sound-motion pictures and sound-reinforc-ing systems.

38.5. SOUND PRESSURE MEASURE-MENTS BETWEEN 50 AND

220 DBJ. K. HILLUARD

(Altec Lansing Corporation, Beverly Hills,Calif.)

The paper will describe applications ofa miniature-condenser high-intensity micro-phone system in jet-rocket motor and indus-trial-plant measurement. Small probe tubeswill be discussed which enable measurementsto be made at ambient pressures of 300 p.s.i.and temperatures around 20000 F such asoccur in combustion chambers. A system willbe described to monitor continuously thenoise interference of jet-engine testing cellsover an area of several square miles.

SESSION 39Engineering Management(Organized by Professional Group on

Engineering Management)Chairman, C. F. HORNE

(Department of Commerce,Washington, D. C.)

39.1. GENERAL PROBLEMS OFENGINEERING MANAGEMENTFACING THE ELECTRONICS

INDUSTRYHARADEN PRATT

(Telecommunications Advisor to the Presi-dent, Washington, D. C.)

The rapidity of the evolutionary develop-ment which the electronics industry has ex-

perienced in a relatively few years, and the

ramification in variety of uses and applica-tions which has resulted, are dealt with fromthe point of view of the growing problemsfor engineering management consideration.It is emphasized that in addition to themany direct as well as allied fields there arealso aspects such as education, training, so-cial factors, research, invention, informationdissemination, production, and marketing,all of which must be taken into account bymanagement in order to plan and operate in-telligently and efficiently.

39.2. RESEARCH AND DEVELOPMENTPROBLEMS OF ENGINEERINGMANAGEMENT IN THE ELEC-

TRONICS INDUSTRYM. J. KELLY

(Bell Telephone Laboratories, Inc.,New York, N. Y.)

The management of a research and de-velopment organization has in common withall industrial production organizations prob-lems of human relations, fiscal control, andservices auxiliary to the productive elementof the organization. The extent to whichthese are different will be discussed. The pro-gramming of the research and developmentactivities, the organization and control of theprograms, and the long-range building ofprofessional man power are areas of manage-ment in research and development that haveunique characteristics. The paper will dwellprincipally on these areas of research and de-veloping management.

39.3. PRODUCTION ASPECTS OFENGINEERING MANAGEMENT

IN THE ELECTRONICSINDUSTRY

W. A. MAcDONALD(Hazeltine Electronics Corporation,

Little Neck, N. Y.)There will be discussed the following:

Three paramount factors in the problem ofmeeting a payroll; How it is possible to pro-vide a uniform flow of products out of aplant, avoiding great peaks and valleys inproductive output; Five necessary successivesteps in a successful timing cycle; The elec-tronics industry has entered the field of bigbusiness with related financial, personnel,and organizational problems, all the concernof management; Advantages of project re-

sponsibility for complex, custom-designedelectronic equipment; Changes in the phi-losophy of business management; Impor-tance of true leadership; What money can-not buy.39.4. WHAT THE MILITARY SERVICESEXPECT FROM ENGINEERINGMANAGEMENT IN THE ELEC-

TRONICS INDUSTRYD. L. PUTT

(Air Research and Development Command,Baltimore, Md.)

The military establishment makes use ofelectronics to augment and extend the physi-cal and mental abilities of man. To maximizeindustry's contribution to defense theremust be an understanding of the conditiornsin which military equipment must function.Environment problems may be grouped intologistic, personnel, and combat operations.Arising from the military environment, thefollowing items need additional effort: a. in-

creased reliability along with mission per-formance, b. increased ruggedness achievedsimultaneously with weight and size reduc-tion, c. "built in" ease of maintenance andadjustment, d. simplicity.

The military establishment desires th?following contributions by engineering mani-agement: a. willing acceptance of militaryresearch, development, and production con-

tracts, b. design for the environment of theanticipated use, c. realistic research and de-velopment schedules, d. realistic productionschedules, e. plans for a switch to militaryproduction in the event of an emergency.

SESSION 40Information Theory III-

Coding(Organized by Professional Group on

Information Theory)Chairman, W. R. BENNETT

(Bell Telephone Laboratories, Inc.,Murray Hill, N. J.).

40.1. A NECESSARY AND SUFFICIENTCONDITION FOR UNIQUE DE-COMPOSITION OF CODED

MESSAGESA. A. SARDINAS AND G. W. PATTERSON

(Burroughs Adding Machine Co.,Philadelphia, Pa.)

This report gives a rigorous formulationto the question: What is the underlying con-

dition on a set of words or codes, so that allmessages constructed with them may be de-composed uniquely? A complete answer isgiven to this problem in the form of a testwhich may be applied to any set of words or

codes contemplated for a coding scheme.

40.2. A SYSTEMATIC SURVEY OFCODERS AND DECODERS

B. LIPPEL(Signal Corps Engineering Laboratories,

Fort Monmouth, N. J.)A number of coders and decoders will be

described, and they will be organized in ac-

cordance with a classification system whichrecognizes three basic types of coders andthree basic decoders. By means of feedbacksystems or other comparison schemes, allcoders can be inverted to give decoders, andvice versa. Furthermore, basic converterscan be combined with auxiliary converters,purely analog or purely digital, to change theanalog mediums and the digital mediums be-tween which the basic devices operate.

40.3. METHOD FOR TIME ORFREQUENCY COMPRESSION-EXPANSION OF SPEECH

GRANT FAIRBANKS, W. L. EVERITT,AND R. P. JAEGER

(University of Illinois, Urbana, Ill.)A method has been developed which uti-

lizes, by sampling techniques, the inherentredundancy in verbal speech to compress itinto shorter time intervals without change inits frequency spectrum ot to transmit speechinformation in the same time over a narrowerfrequency band.

The same device can be used to expandthe time interval or frequency spectrum. Acombination of two devices, one used to com-press and the other to expand, permits trans

1953 425

PROCEEDINGS OF THE I.R.E.

mission of speech signals over channels oflimited bandwidth, and without delay.

The device can also be used to tailorbroadcast programs, initially recorded withan undesired duration, into an assigned timeinterval without change in the frequencyspectrum or observable distortion.

40.4. A NEW CODING SYSTEM FORPULSE-CODE MODULATION

A. G. FITZPATRICK(Burroughs Adding Machine Co.,

Philadelphia, Pa.)This paper describes a new coding system

for pulse-code modulation in which the oper-ations of sampling, quantizing, and codingare accomplished by means of two simplespecial-purpose tubes and a minimum of cir-cuitry. The first of these tubes is a beam-de-flection type with ten individual outputs.The second is a quantizing-coding tube inwhich crossed electric and magnetic fieldsproduce ten stable, controllable positions ofan electron beam, each with parallel binary-coded outputs. Low cost and simplicity ofthis system will be compared with presentsystems. Adaptations of this same method toother analog-to-digital functions will be dis-cussed.

40.5. COINCIDENCE DETECTORS FORBINARY PULSESCLARENCE GATES

(California Institute of Technology,Pasadena, Calif.)

The recovery of binary pulses when thesignal-to-noise ratio is poor is investigated.The technique of using redundant channelsfor improving the effective signal-to-noiseratio is analyzed and optimum criteria de-veloped for the two cases of interest; thefirst in which all channels are sampled for theprecise value of signal in each before the de-cision as to whether a pulse is present ismade, and the second in which the decision ismade first in each channel. Methods of cir-cuit synthesis for the second case are given,and a beginning towards the analysis ofhigher order pulses is made.

SESSION 41Broadcast and Television

Receivers-II(Organized by Professional Group onBroadcast and Television Receivers)

Chairman, R. M. BowIE(Sylvania Electric Products Inc.,

New York, N. Y.)

41.1 FACTORS AFFECTING THEDESIGN OF VHF-UHF TUNERS

E. H. BODEN(Sylvania Electric Products Inc.,

Emporium, Pa.)This paper reviews some of the present

arrangements for tuning the vhf and uhf tel-evision channels' performance and circuitryin the use of recently developed tubes whichmakes possible the design of a single unitthat would tune both the vhf and uhf chan-nels. Performance over the 82 televisionchannels is reviewed.

41.2. THEORY OF AFCSYNCHRONIZATION

W. J. GRUEN(General Electric Co., Syracuse, N. Y.)Automatic frequency control (afc) has

been widely used for the horizontal sweepsynchronization in television receivers, andmore recently for the color synchronizationin the proposed NTSC color system.

The paper deals with the theory of a gen-eralized system in which the phase of local-oscillator signal is compared'to the phase ofthe transmitted-reference signal in a phasediscriminator. The resulting control voltageis then passed through a control network tocontrol the frequency of the local oscillator.Systems having either a zero, single or dou-ble-time constant-control network are con-sidered. The transient response, frequencyresponse, and noise bandwidth, as well as thehold-in range and pull-in range of synchroni-zation are presented.

41.3. STANDARDIZATION OF PRI1NTEDCIRCUIT MATERIALS FOR MECHA-

NIZED RADIO ASSEMBLYW. HANNAHS, J. CAFFIAUX, AND

N. STEIN(Sylvania Electric Products Inc.,

Bayside, N. Y.)Conditions of use of "printed circuits" in

radio and TV sets are critically examinedwith the purpose of developing standards ofperformance and comparison tests for wiringprefabricates and raw materials.

Tests for etched, stamped, and flexiblybacked copper-foil circuits are described,and test results given for the various typesand makes reveal characteristic differencesaffecting their selection for various applica-tions. Some standards are proposed towardunification of the requirements which repre-sent both suppliers' and users' viewpoints.

41.4. A COLOR TELEVISION RECEIVERFOR THE NTSC SYSTEM

K. E. FARR(Westinghouse Electric Corporation,

Metuchen, N. J.)The basic elements of color television

transmission will be outlined, and the salientfeatures of the NTSC system in its presentform will be discussed. A receiver designedfor this system will be described. The re-ceiver circuits are divided into four basicgroups: 1. the monochrome or brightnesssignal channel, along with sound, deflectionsync, and agc, 2. the color decoder and videocircuits, 3. the color sync circuits, 4. the de-flection, convergence, and power supply cir-cuits.

The performance of this receiver will bediscussed, and color photographs of colorpictures taken from the picture-tube screenwill be shown in the slides, as well as circuitdetails and photographs of the receiver.

41.5. A SIMPLIFIED VIDICONTELEVISION CAMERA

V. K. ZWORYKIN, L. E. FLORY, ANDW. S. PIKE

(RCA Laboratories Division, Prince-ton, N; J.)

There are many everyday uses for aclosed-circuit television system which havenot previously been feasible even with avail-able industrial equipment due to the cost of

such units. The present paper describes asimplified camera using the vidicon pickuptube which, it is believed, will expand theusefulness in industry, commerce, and edu-cation. The camera contains the vidicon it-self and a video amplifier which provides asignal in the form of a modulated carrier.Power for operating the amplifier and de-flecting the vidicon beam are obtained froma television receiver on which the picture isviewed.

SESSION 42Microwaves III-Ferrites

and Detectors(Organized by Professional Group onMicrowave Theory and Techniques)

Chairman, W. W. MUMFORD(Bell Telephone Laboratories, Inc.,

Murray Hill, N. J.)

42.1. SPACE CHARGE DETECTORFOR MICROWAVES

A. B. BRONWELL, JOHN MAY, CHARLES NITZ,T. C. WANG, AND HILLIARD WACHOWSKI

(American Society for Engineering Educa-tion, Evanston, Ill.)

This paper presents results of theoreticaland experimental studies of the vacuum-tube detector and converter for microwaves.Cylindrical diode tubes are used as detectorsand converters at low-power levels and shortwavelengths in the microwave spectrum.The theoretical studies treat the behavior ofelectrons in a space-charge cloud with super-imposed dc and microwave fields.

Solutions are obtained for the cylindri-cal-diode space-charge equations in seriesform, based upon certain approximations.These solutions show the change in transittime and plate current resulting from super-imposing the microwave field upon the dcfield.

42.2. LOW LEVEL SYNCHRONOUSMIXING

M. E. BRODWIN, AND C. M. JOHNSON(The Johns Hopkins University,

Baltimore, Md.)AND

W. M. WATERS(Bendix Radio, Towson, Md.)

A synchronous detection system utilizingthe same oscillator to furnish power for twosignal channels has been used to obtain sen-sitivities of the order of 110 dbm at 100 kc,9 kmc, and 33 kmc. The energy from an un-modulated RF source is divided into twochannels. A portion of the energy is modu-lated in the signal channel where the desiredmeasurements take place. The rest of theunmodulated carrier is directed through thereference channel where its phase and am-plitude may be controlled. The signal andreference powers are mixed in the detectingelement and the audio component is ampli-fied and metered. The qualitative theory isarrived at by approximating the nonlinearelement with a power-series and a Fourier-conductance representation.

42.3. GUIDED WAVE PROPAGATIONTHROUGH FERRITIES AND ELEC-

TRON GASES IN MAG-NETIC FIELDS

426 March

5IRE National Convention Program

L. GOLDSTEIN, M. GILDEN, ANDJ. ETTER

(University of Illinois, Urbana, Ill.)Guided-microwave propagation through

certain anisotropic dielectrics has been theobject of recent investigations. In these ex-periments free electron gases and ferromag-netic dielectrics in magnetic fields consti-tuted the anisotropic media. The resultsshow that, in general, both polarizationtransformation and resonance phenomenaare observed. This paper reports further re-sults on the magnetic resonances and Fara-day-rotation effects in ferromagnetic dielec-trics and in free electron gases immersed inmagnetic fields.

42.4. CAVITIES WITH COMPLEXMEDIA

A. D. BERK, AND BENJAMIN LAX(Massachusetts Institute of Technology,

Cambridge, Mass.)Expressions for the input and transfer

impedances of cavities containing complexmedia, that is, media with tensor permeabil-ity, conductivity, or permittivity, are de-rived and discussed. It is shown that undercertain conditions cavities of this sort pos-sess the basic property of a gyrator. Specialapplication of these results is made to theferrites.

42.5. RESONANCE IN CAVITIES WITHCOMPLEX MEDIA

BENJAMIN LAX, AND A. D. BERK(Massachusetts Institute of Technology,

Cambridge, Mass.)It is possible to study the tensor proper-

ties of such complex media as the ferrites andmagneto-ionic gases by the use of resonant-microwave cavities. The impedance methodof analyzing a cavity with degenerate oroverlapping modes suggests the scheme formeasuring the pertinent components ofeither the permeability or the conductivitymatrix of these media. A specific illustrativeexample is treated in which rotating modesare considered.

SESSION 43Remote Control Systems(Organized by Professional Group on

Radio Telemetry and RemoteControl)

Chairman, C. H. DOERSAM, JR.(Office of Naval Research, Port

Washington, N. Y.)

42.1. THE ORGANIZATION OF ADIGITAL REAL TIME

SIMULATORH. J. GRAY, JR.

(Moore School of Electrical Engi-neering, Philadelphia, Pa.)

When a digital computer is used in a realtime simulation problem, best results are notnecessarily obtained with conventional ma-chines. A typical set of thirteen first-ordernonlinear differential equations was pro-grammed for a machine similar to the Ray-theon hurricane computer. Roughly 3,000memory positions were required and the

time required for computation in one inte-gration interval was found to be greater than0.22 second. Other studies indicated thatthe integration interval must be less thanabout one-tenth second. For various reasons,serial machines were considered over parallelmachines. Several machine organizationswere set up making use of one-address, three-address, and four-address codes, special par-titioning of the memory, use of auxiliarystorage registers, special codes, a high-speedmultiplier, a high-speed divider, more thanone arithmetic unit, use of digital-differen-tial analyzer techniques, etc. Computationtimes for the above problem were obtaiiiedand compared for the different machine or-ganizations and an "optimum" computerwas evolved which, subject to the same as-sumptions as the hurricane with regard toprogramming, required about ten ms for oneintegration interval and after further manip-ulation of the original equations, this timewas reduced to about 7 ms. Roughly 2,000one-word memory positions were required.

43.2. CONTROL SYSTEM ENGINEER-ING APPLIED TO SUSPENSION

SYSTEMSG. J. MARTIN AND R. JEsKA

(University of Michigan, Ypsilanti,Mich.)

Automotive-suspension systems are stud-ied by means of analogue-computing equip-ment, and the measured results comparedwith performance. The system is viewed as aclosed-loop control system rather than a con-ventional open-ended passive system, whichleads to new concepts of suspension synthe-sis, using the techniques of control-systemengineering. The control system is thentranslated into practical hardware.

43.3. EXPERIMENTAL EVALUATION OFCONTROL SYSTEMS BY RANDOM-

SIGNAL MEASUREMENTSW. W. SEIFERT

(Massachusetts Institute of Technology,Cambridge, Mass.)

Theoretical advances in the past decadeindicated the advantages of random-signalcalculations over sinusoidal and step-re-sponse methods in control-system design,but practical application of random-signalmeasurements to experimental evaluation ofcontrol systems was relatively limited. AtMIT the Dynamic Analysis and ControlLaboratory has developed equipment andtechniques for generating and monitoringrandom signals. One recent application ofthis technique, the expet-imental evaluationof a rate servomechanism, is presented andthe experimental results are correlated withapproximate theoretical results. The prac-tical advantages of random-signal measure-ments are reemphasized and special atten-tion given to the study of nonlinear systemsunder realistic operating conditions.

43.4. EXTENSION OF CONVENTIONALTECHNIQUES TO THE DESIGN OF

SAMPLED-DATA SYSTEMSW. K. LINVILL, AND R. W. SITTLER

(Massachusetts Institute of Technology,Cambridge, Mass.)

Techniques developed in feedback-am-plifier design and servomechanisms can beextended for design of sampled-data systems.A sampled-data system is made from onlythree kinds of linear elements all of whichcan be described either in the time or the fre-quency domain. Flow graphs (or block dia-grams) of sampled-data systems can be ma-nipulated so as to reduce any complicatedconfiguration to a simple equivalent configu-ration without feedback. A simple compen-sation procedure has been devised for sam-pled-error-data servosystems which uses er-ror coefficients and correlates transient re-sponse with positions of system poles.

43.5. GENERALIZED SERVOMECHA-NISM EVALUATION

W. P. CAYWOOD AND WILLIAM KAUFMAN(Carnegie Institute of Technology,

Pittsburgh, Pa.)Servomechanism-performance evalua-

tions for cases of statistically described sig-nals have generally been made using themean square of the error as a criterion ofperformance. There are many instances ofapplications of servos in which the meansquare of the error holds little realism, suchas in fire-control systems, and its use mayresult in a definitely inferior design or systemadjustment.

Described in this paper is an analyticmethod of evaluating the operation of anylinear system having unvarying parametersand an unvarying criterion of the importanceof the instantaneous-error magnitude. Themethod comprises expanding in a power se-ries the curve of relative importance of errorversus error and, as is shown in the paper,using the coefficients of the power series todetermine a new series, each term of whichincorporates the first or a higher-momentcorrelation function taken of the statisti-cally described signal. The highest-momentcorrelation needed is the same order as thedegree of the power series needed to satisfac-torily represent the criterion. An applicationof the calculus of variations completes themethod and allows determining the best sys-tem adjustment. A magnetic-tape type ofcorrelator to perform the higher-momentcorrelation is briefly described.

43.6. A METHOD FOR REDUCINGTHE FORCED DYNAMIC ERROROF CLOSED-LOOP SYSTEMS

L. H. KING(Massachusetts Institute of Technology,

Cambridge, Mass.)This paper illustrates a method for re-

ducing the forced-dynamic error in servo-mechanisms by design based on error coeffi-cients. After a brief review of the dependenceof the forced-dynamic error upon error co-efficients, the relationship between error co-efficients and the parameters of a servo-mechanism is derived. This relationship isthen used to show how closed-loop systemscan be modified to obtain favorable error co-efficients, which reduce the forced-dynamicerror. The method has been tested by simu-lation, and photographs of simulator re-sponse show how the effect of additional in-tegrations can be achieved by error-coeffi-cient adjustment,

1953 427

428 PROCEEDINGS OF THE I.R.E.

Institute News and Radio NotesCalendar of

COMING EVENTSIEE Symposium on Insulating Mate-

rials, London, Eng., March 16-18Optical Society of America Meeting,

Hotel Statler, New York, N. Y.,March 19-21.

1953 IRE National Convention, Wal-dorf-Astoria Hotel and GrandCentral Palace, New York, N. Y.,March 23-26

IRE New England Radio EngineeringMeeting, Storrs, Conn., April 11

9th Joint Conference of RTMA ofUnited States and Canada, Am-bassador Hotel, Los Angeles,Calif., April 16-17

IRE Seventh Annual Spring Techni-cal Conference, Cincinnati, Ohio,April 18

Symposium on Nonlinear CircuitAnalysis, Engineering SocietiesBuilding, New York, N. Y., April24-25

SMPTE Convention, Statler Hotel,Los Angeles, Calif., April 26-30

URSI-IRE Meeting, National Bu-reau of Standards, Washington,D. C., April 27-30

NARTB Convention, Biltmore Hotel,Los Angeles, Calif., April 28-May 1

Electronic Components Symposium,Shakespeare Club, Pasadena,Calif., April 29-May 1

1953 National Conference on Air-borne Electronics, Dayton, Ohio,May 11-14

1953 Electronics Parts Show, ConradHilton Hotel, Chicago, Ill., May18-21

National Electronics Conference,Hotel Sherman, Chicago, Ill.,September 28-30

1953 IRE-RTMA Radio Fall Meeting,Toronto, Ont., October 26-28

1954 Sixth Southwestern IRE Con-ference and Electronics Show,Tulsa, Okla., February 4-6

TECHNICAL COMMITTEE NOTESUnder the Chairmanship of R. J. Wise,

the Facsimile Committee met on December5, 1952. There was a discussion regardingthe definition for "facsimile"; however,agreement could not be reached and it wasdecided to carry the matter over to the nextmeeting. The Chairman suggested thateach member prepare a solution and pre-sent it at that time. Pierre Mertz proposedthat the term 'effective band" be defined asa substitute for the term "nominal band"previously deleted. The proposal was ac-

cepted with the definition to read exactlyas it is written in the IRE Standard, 42IRE 9. S. Kenneth McConnell (alternate

for A. G. Cooley) proposed a new definitionentitled "facsimile band width," which wasapproved by the Committee.

The Electron Devices Committee meton December 10, 1952, under the Chair-manship of G. D. O'Neill. ChairmanO'Neill presented to the Committee thecomments of R. S. Burnap and A. C. Rock-wood concerning the proposed modificationof the definition for "accelerating electrode"by adding the noun "(accelerator)" beforethe definition proper. The comments notedthat the use of a single noun to describe aparticular function violated IRE standardiz-ing policy, which is to use the single nounfor structural names, for example, "anode,"and to progress logically from such a funda-mental specification to particular usageswith modifying adjectives or adverbs. Itwas further pointed out that the 1950Electron Tubes Standard had not eliminatedconfusion over whether an electrode was an

anode or a grid, depending on whether ornot it drew current. Mr. Rockwood had re-

ferred to the minutes of a JETEC com-mittee on cathode-ray tubes in which thisdifficulty was recognized, and had recom-mended some joint action be taken toimprove the definitions, based upon agenerally acceptable philosophy. This mat-ter was referred to R. B. Janes for furtheraction and Chairman O'Neill suggested thata member of the JETEC committee be re-

cruited as a subcommittee member. It wasdecided that JETEC be notified of thechange in attitude of Committee 7 towardthe use of the shortened terms and that thedefinition in question is one of severalwhich will be under review shortly for pos-sible revision. Comments from the Stand-ards Committee concerning the klystrondefinitions did not require action, with theexception of electronic efficiency. However,the discussion of electronic efficiency andcircuit efficiency, revealed that these defini-tions were redundant, in the opinion of theCommittee, and it was decided that they bedeleted. Output-circuit electronic efficiencywas added to the list of klystron definitions.R. M. Ryder presented the final draft of theproposed methods of test for noise. L. S.Nergaard and W. J. Dodds presented a

complete roster of definitions recently orcurrently under consideration by this Com-mittee.

On December 12, 1952, under the Chair-manship of P. C. Sandretto the NavigationAids Committee convened. R. E. Gray whois compiling a list of all the terms definedthis year reported discrepancies in the exist-ing definitions of deviation sensitivity, lowclearance field, and true axis of the nutationfield. Corrections were made by the Com-mittee. Action on the terms crystal holder,crystal-video figure of merit, output noiseratio, and crystal current sensitivity weredeferred until the next meeting. The Com-mittee took up the second half of HarryDavis' list of terms, which were under con-sideration for the remainder of the meeting.

The Video Techniques Committee, underthe Chairmanship of W. J. Poch, met onDecember 9, 1952. A. J. Baracket reported

on the Subcommittee on Video Systems andComponents. Reports on geometric distor-tion and pickup-tube interlacing are readyfor subcommittee action. A number ofother reports are in preliminary stages ofpreparation. J. L. Jones summarized prog-

ress of the Subcommittee on Methods ofMeasurement of Video Transmission. Prom-ising results have been obtained by the use

of special signals for checking amplifierlinearity and the preparation of a reporthas been initiated. Dr. Athey reported forthe Subcommittee on Video Utilization.Several matters should come before theSubcommittee in the near future, such as arecommendation for a method of measure-ment on X-ray radiation, a re-evaluation ofL. D. Grignon's tutorial paper on videorecording, and a report on the reaction fromnetwork representatives to the question-naire submitted some time ago. It was sug-

gested that this Subcommittee review theproposed definitions in the field of videorecording and make recommendations tothe Video Techniques Committee for termswhich should be standardized.

The Radio Transmitters Committeemet on December 12, 1952, under theChairmanship of M. R. Briggs. ChairmanBriggs summarized the work done with theAnnual Review and expressed his apprecia-tion to members for their cooperation involuntarily having accepted assignmentsand for their promptness in submittingsummaries to P. J. Herbst. ChairmanBriggs read a letter from J. B. Heffelfinger,Chairman of Subcommittee 15.3, DoubleSideband AM Transmitters, who reportedthat his group is still working to produce arough draft of the Standards on DoubleSideband Transmitters: Methods of Testing.Harold Goldberg, Chairman of Subcom-mittee 15.4, Pulse-Modulated Transmitters,reported that his Subcommittee has com-pleted the final draft of its proposed Stand-ards on Methods of Measurement of PulseQuantities. A. E. Kerwien, Chairman ofSubcommittee 15.5, Single Sideband RadioCommunication Transmitters, reported thatduring two recent meetings the group hasattempted to define a number of terms ofspecial significance to the single sidebandtransmitter field. Considerable progress hasbeen made on a number of terms, which willenable the group to commence work on themethods of test soon. The remainder of themeeting was devoted to a review of theProposed Standards on Methods of Meas-urement of Pulse Quantities, as resubmittedby Subcommittee 15.4. Dr. Goldberg out-lined the work done and pointed out thatthe recommendations made last Januaryfor the most part had been incorporatedin the present draft. The comments result-ing from the discussion by the RadioTransmitters Committee will be passedalong to Dr. Goldberg's subcommittee forreview at their next meeting. It was thegeneral opinion that every effort should bemade to submit the standard to the meas-

urements coordinator at an early date.The Standards Committee did not hold

a meeting during the month of December.

March

Institute News and Radio Notes

IRE OFFICERS AND DIRECTORS AP-POINTED FOR 1953 TERM

The IRE Board of Directors, at its an-

nual meeting on January 7, 1953, New York,N. Y., appointed six officers and directors forthe year 1953.

Haraden Pratt, telecommunications ad-viser to the President, was reappointedSecretary of the Institute, a post he has heldsince 1943. W. R. G. Baker, vice presidentof the General Electric Co., Syracuse, N. Y.,was appointed Treasurer for the third suc-

cessive year. A. N. Goldsmith, consultingengineer, was appointed Editor, an office hehas held since the IRE was founded in 1912.

Appointed as directors for 1953 were

R. D. Bennett, technical director, UnitedStates Naval Ordnance Laboratory, SilverSpring, Md.; W. R. Hewlett, vice president,Hewlett Packard Co., Palo Alto, Calif.; andA. V. Loughren, vice president in charge ofresearch, Hazeltine Electronics Corp., LittleNeck, L. I., N. Y.

MIT OFFERS ELECTRONICFELLOWSHIPSA number of Graduate and Advanced

Research Fellowships is offered by theMassachusetts Institute of Technologyfor study and research in the field of elec-tronics. These "Industrial Fellowships inElectronics" are sponsored jointly by a

group of industrial organizations concernedwith the advancement of electronics and itsapplications.

Recipients of Student Fellowshipswill be awarded a stipend varying between$1,500 and $2,400, according to their experi-ence and qualifications, and in addition willbe granted a credit to meet the tuition fee.Advanced Research Fellowships will range

from $3,000 upwards, according to thequalifications of the recipient.

Applicants should communicate withthe Director, Research Laboratory of Elec-tronics, Massachusetts Institute of Tech-nology, Cambridge, Mass. Applicationshould be made at least four months priorto the intended date of entrance.

URSI-IRE MEETING SCHEDULEDFOR APRILA meeting of the USA National Com-

mittee of the International Scientific RadioUnion (URSI) and the IRE ProfessionalGroup on Antennas and Propagation is beingheld at the National Bureau of Standards,Washington, D. C., April 27-30, 1953.

Sessions of the meeting will be concernedwith the topics of radio measurement meth-ods and standards, tropospheric radio prop-agation, ionospheric radio propagation,terrestrial radio noise, radio astronomy,radio waves and circuits (including generaltheory), and electronics.

A preliminary program and advanceregistration forms will be available afterMarch 16, 1953. These and further informa-tion about the meeting may be obtainedfrom A. H. Waynick, Secretary, USA Na-tional Committee of URSI, The Pennsyl-vania State College, State College, Pa.

A certificate of recognition and pocket-size sliderule were presented to Vince Di Caudo (left), Akron,Ohio, by Irving Knapp on behalf of the Akron IRESection, for Di Caudo's scholarship, interest, andachievement at the college of engineering, Universityof Akron.

TECHNICAL CONFERENCESCHEDULED IN CINCINNATI

The Seventh Annual Spring TechnicalConference, sponsored by the CincinnatiIRE Section, will be held April 18, 1953,Cincinnati, Ohio.

The Conference, which is the only IREtechnical conference national in scope andentirely devoted to television, will be ofinterest to all executives, engineers, andtechnical personnel in television and alliedfields. Papers to be presented will include thelatest information in various phases of theart and will deal with material unpublishedto date.

The list of speakers and papers to be de-livered at the conference are as follows:

Morning Session, April 18

"Television and the Bell System," (speakerto be announced)

"A High Powered UHF-TV BroadcastSystem," F. J. Bias, General ElectricCo., Syracuse, N. Y.

"The Design of TV Receivers UtilizingNON-Synchronous Power," G. D.Hulst, DuMont Laboratories, Inc.,Clifton, N. J.

"Approach to Mechanized Assembly ofElectronic Equipment Applicable to TVReceivers," L. K. Lee, Stanford Re-search Institute, Stanford, Calif.

Afternoon Session, April 18

"The Selection and Amplification of UHFTelevision Signals," Wilson Boothroydand John Waring, Philco Corp.,Philadelphia, Pa.

"Transient Considerations in the NTSCColor System," B. S. Parmet, Motorola,Inc., Chicago, Ill.

"A Four Gun Tube for Color TelevisionReceivers," John Rennick and CharlesHeuer, Zenith Radio Corp., Chicago,Ill.

"Latest NTSC Color System," (OrangeCyan Wide Band), R. D. Kell andA. C. Schoeder, RCA Laboratories,Princeton, N. J.

All information regarding advertisingexhibits should be directed to R. H. Leh-

man, the Baldwin Company, 1801 GilbertAvenue, Cincinnati, Ohio.

Matters of advance registration for theconference, including hotel, luncheon, andbanquet reservations, should be directed to:A. C. Wahl, P.O., Box 8, Green Hills 18,Ohio. A late registration may be made atthe conference.

NONLINEAR CIRCUIT ANALYSISSYMPOSIUM ANNOUNCED

An international symposium on Non-Linear Circuit Analysis will be held onApril 23-24, 1953, at the EngineeringSocieties Building Auditorium, New York,N. Y.

The symposium, organized by the Poly-technic Institute of Brooklyn with the co-operation of the IRE Professional Group onCircuit Theory and with the co-sponsorshipof the Office of Naval Research, Air Re-search and Development Command, and theSignal Corps, will be of particular interest tothose working in the field of nonlinear sys-tems. It is intended to cover the basic ex-position of nonlinear phenomena and thefundamental mathematical methods of anal-ysis, as well as illustrative applications tononlinear electronic circuits, magnetic cir-cuits, feedback systems, and feedback-control systems. American and Europeanauthorities, who have made original con-tributions to the art, will participate.

No registration fee will be charged foradmission to the Symposium. However, allpersons interested in attending are urged toregister early. Copies of the detailed pro-gram, hotelaccommodation information, andregistration forms are available on requestto: Polytechnic Institute of Brooklyn,Microwave Research Institute, 55 JohnsonStreet, Brooklyn 1, N. Y.A 'Proceedings of the Symposium on

Nonlinear Circuit Analysis" will be pub-lished by October, 1953, at four dollars percopy. Members of the IRE ProfessionalGroup on Circuit Theory may obtaincopies at three dollars per copy. Orders forthe Proceedings, accompanied by check ormoney order made out to 'Treasurer, Non-linear Symposium," will be accepted inadvance, at the above address.

1953 IAS OFFICERS ANNOUNCEDThe Institute of the Aeronautical Sciences

has announced the new officers for 1953.They are as follows.President: C. J. McCarthy, United Aircraft

Corporation.Vice-Presidents: G. W. Brady, Curtiss-

Wright Corporation; C. L. Johnson,Lockheed Aircraft Corporation; J. S.McDonnell, Jr., McDonnell AircraftCorporation; E. G. Stout, ConsolidatedVultee Aircraft Corporation.

Treasurer: P. R. Bassett, Sperry GyroscopeCompany.

Director: S. P. Johnston.Secretary: R. R. Dexter.Controller: J. J. Maitan.

The men assumed the duties of theirrespective offices at the IAS Twenty-firstAnnual Meeting, January 26-29, 1953,Hotel Astor, New York, N. Y.

1953 429

PROCEEDINGS OF THE I.R.E.

Professional GroupNews

BROADCAST TRANSMISSIONS SYSTEMSThe Boston Chapter of the Broadcast

Systems Group held a meeting recently atRadio Station WCOP, Boston, Mass., P. K.Baldwin presiding. Twelve broadcast sta-tions were represented. During the meetinga paper was presented on "The Future Pros-pects of UHF Television," by W. Y. Pan,RCA Victor Division, and an RCA film was

shown entitled, "Success Hill."COMMUNICATIONS SYSTEMS

The Professional Group on Communica-tions is planning a Symposium on RadioCommunications, June 11-12, 1953, NewYork, N. Y.

The program will include technical ses-

sions in the auditorium of the AmericanTelephone and Telegraph Long Lines Build-ing, New York City, and an inspection tripto the AT&T overseas radiotelephone trans-mitting and receiving stations at Lawrence-ville and Netcong, N. J.

The Washington Chapter of the Com-munications Group has been officially ap-

proved by the IRE Executive Committee.At the inaugural meeting held in Februarythe following officers were elected: Chair-man, C. L. Engleman, United States Navy;Vice Chairman, W. C. Boese, Federal Com-munications Commission; Secretary, J. D.Wallace, Naval Research Laboratory. Adocumentary film, "Communications Sys-tems of Operation Sandstone," was shown.

ELECTRONIC COMPUTERSThe Washington Chapter of the Electronic

Computers Group has been officially ap-proved by the IRE Executive Committee.

At the chapter's inaugural meeting heldat the PEPCO Auditorium, Washington,D. C., the following officers were elected:Chairman: C. V. L. Smith, Office of NavalResearch; Vice Chairman, D. H. Jacobs,Jacobs Instrument Company; Secretary,R. J. slu*tz, National Bureau of Stand-ards. The program included a paper on

"How the Univac Predicted the Election,"by H. F. Mitchell, Jr., Eckert, MauchlyDivision, Remington Rand, Incorporated.

The San Francisco Chapter of the Elec-tronic Computers Group also has been ap-proved by the IRE Executive Committee.T. H. Meisling, University of California, ischairman protempore.

ENGINEERING MANAGEMENT

An organizational meeting headed by F.W. Schor, Motorola, Inc., was held recentlyby the Chicago Chapter of the EngineeringManagement Group. The chapter has beenofficially approved by the IRE.

TRANSACTIONS OF IRE PROFESSIONAL GROUPSThe following issues of Transactions are available from the Institute of Radio

Engineers, Inc., 1 East 79 Street, New York 21, N. Y., at the prices listed below.

Group IRE Non-Sponsoring Publication Mem- Mem- mem-

Group bers bers bers*_ _

AirborneElectronics

Antennas andPropagation

Audio

Broadcast andTelevisionReceivers

CircuitTheory

ElectronDevices

ElectronDevices

ElectronicComputers

QualityControl

Vehicular Com-munications

PGAE-4; "The Selectivity andIntermodulation Problem in UHFand Communication Equipment"(11 pages)PGAE-5; "A Dynamic AircraftSimulator for Study of HumanResponse Characteristics" (6pages)PGAE-6; "Ground-to-Air Co-Channel Interference at 2900MC (10 pages)PGAP-4; IRE Western Conven-tion, August, 1952 (136 pages)PGA-5; "Design Interrelations ofRecords and Reproducers," byH. I. Reiskind (8 pages)PGA-6; Editorials, TechnicalPapers, and News (42 pages)PGA-7; Editorials, Technical Pa-pers, and News (48 pages)PGA-8; July 1952 Issue (40pages)PGA-9; September-October Is-sue (28 pages)PGA-10; November-DecemberIssue (28 pages)Vol. AU-1, No. 1; January-Febru-ary Issue (24 pages)PGBTR-1; Round-Table Discus-sion on UHF-TV Receiver Con-sideration Presented at 1952 IRENational Convention, March 6,1952, New York, N. Y. (12 pages)PGCT-1; IRE Western Conven-tion August, 1952 (100 pages)PGED-1; Papers from IRE Con-ference on Electron Tube Re-search and IRE-AIEE Confer-ence on Semiconductor Research,June, 1952 (32 pages)PGED-2; Papers on Electron De-vices presented at the IRE Con-ference on Electron Tube Re-search, Ottawa, Ont., Canada,June 16-17, 1952 and the IREWestern Convention, Long Beach,Calif. (84 pages)PGEC-1; Papers presented atTechnical Sessions on Elec-tronic Computers at WesternElectronic Show and Convention,August 27-29, 1952, Long Beach,Calif. (75 pages)PGQC-1; Papers presented at1951 Radio Fall Meeting, and1952 IRE National Convention,(60 pages)PGVC-2; Symposium on What'sNew in Mobile Radio (32 pages)

$0.45

.30

.30

2.20

.30

.80

.90

.80

.60

.70

.60

.50

1.60

.80

1.60

1.50

1.20

1.20

$0.65

.45

.45

3.30

.45

1.20

1.35

1.20

.90

1.05

.90

.75

2.40

1.20

2.40

2.25

1.80

1.80

$1.35

.90

.90

6.60

.90

2.40

2.70

2.40

1.80

2.10

1.80

1.SO

4.80

2.40

4.80

4.50

3.60

3.60

* Public libraries and colleges can purchase copies at IRE Member rates.

NoticelULTRASONICS PROFESSIONAL GROUPA petition to form an IRE Pro-

fessional Group on Ultrasonics hasbeen received. The proposed Groupwill hold an informal meeting at the1953 IRE National Convention,Thursday from 1:30 to 2:30 p.m. inthe Blue Room of the Grand CentralPalace. All those interested in the ac-tivities of such a professional Groupare urged to attend.

430 March

IRE PeopleNathaniel KL Zelazo (M'47-SM'52) has

been appointed vice president of the KetayManufacturing Corporation of New York

City and Los Ange-les, designers andmanufacturers ofelectronic equip-ment. Mr. Zelazo'sresponsibilities in-clude organizing in-dustrial and govern-ment sales, andco-ordinating the cor-poration's researchand development

N. K. ZELAZO work. He was for-merly with the De-

partment of Defense as acting chief of thespecial projects division in the ElectronicsProduction Resources Agency, Washington,D. C., and closely associated with the Muni-tions Board and its industry advisory com-mittees.

Mr. Zelazo was born in Lomza, Poland,and received the B.S. degree from the Col-lege of the City of New York in 1940. He didgraduate work at Columbia University andGeorge Washington University. At the be-ginning of World War II, Mr. Zelazo wasassociated with the Army Air Force in theirprocurement of aircraft instruments andlater transferred to the Army Signal CorpsRadar Laboratory at Belmar, N. J. From1942-1947, he was with the Bureau of Ships,Navy Department, as a project engineer ondevelopment of fire-control radar equip-ments. From 1947-1950, Mr. Zelazo washead of the bureau's miniaturization unitand as a result of his activity, the Researchand Development Board established a sub-panel on miniaturization for which he servedas the first secretary. Prior to leaving theNavy Department, Mr. Zelazo was systemsengineer and electronic consultant to theArmament Division, Bureau of Aero-nautics.

Arthur Albert Dyson (A'33) has been ap-pointed an Officer of the Most ExcellentOrder of the British Empire, according toHer Majesty's New Year Honours List.

Mr. Dyson began his career as worksmanager for Pritchard and Simpson of New-castle, England, in 1929. In 1932 he becameworks manager for Erie Resistor, Ltd., ofLondon. He is now managing director andchief engineer of Eire Resistor.

Gerard Lehmann (SM '44) has beenelected president of the French NationalCommittee of the International ScientificRadio Union (URSI) for the next threeyears.

Mr. Lehmann was born in Paris, France,on April 6, 1909. After receiving his degreein engineering from the lcole Centrale, in1931, he became associated with the SadirCompany as a technical director, buildingvhf communications and radio navigationequipment.

After serving with the French Army until1940, he joined the Lyon laboratory of LeMatdriel Telephonique. In 1943 he came tothe United States and worked as a researchengineer with Federal Telephone and RadioLaboratories. He later returned to LeMatdriel Telephonique in France.

In addition to research work, Mr.Lehmann hastaught at the lcole Central andwas appointed professor of direction findingand radio navigation at the lcole Superieured'tlectricit&. At present he is the scientificdirector of Laboratoire Central de Tele-communications, Paris, France.

Bernard Hecht (M'45) has entered thefield of quality control consulting, specializ-ing in the problems of electronics industry.

Mr. Hecht re-ceived the B.E.E. de-gree from the Schoolof Technology CityCollege of New York,and the M.S. degreefrom the Universityof Pennsylvania. Atthe beginning ofWorld War II, he wasa civilian officer incharge of radio in-

BERNARD HECHT spection for the Sig-nal Corps. In 1943 he

was selected to represent the United Army-Navy Specification Program for electroniccomponents, embracing the standardizationof tubes, resistors, capacitors, and trans-formers. Later, he was assigned to aid in thereconversion to peacetime production theplant operation of the International Re-sistance Company. He has managed qualitycontrol for such firms as the Starrett Tele-vision Corporation and RCA Victor.

Mr. Hecht has lectured on quality con-trol subjects at Temple, Princeton, andRutgers Universities. He is a member of TauBeta Pi, and a senior founding member ofthe American Society for Quality Control.He has served as Vice Chairman of the IREProfessional Group on Quality Control.

H. B. Steinhauser (M'51) has beennamed manufacturing engineer of the Du-mont Laboratories, Inc., instrument divi-sion.

Mr. Steinhauser was born in Norwich,N. Y., on July 25, 1914, and received theE.E. degree at Rensselaer Polytechnic In-stitute in 1934. In 1935 he became a juniorradio engineer at the General Electric Com-pany in Bridgeport, Conn., and then trans-ferred to the Western Electric Company inKearny, N. J., as test maintenance engineer.In 1938 he joined the Sperry GyroscopeCompany as a production engineer, and in1944, he became associated with the Boon-ton Radio Corporation as production mana-ger. He later became administrative assist-ant to the general manager.

Mr. Steinhauser was a senior engineerwith the Instrument Division of DumontLaboratories, Inc., before his promotion.

Earl G. Ports (A'25-M'33-SM'43),assistant technical director of FederalTelecommunication Laboratories

Inc., Nutley,N. J., died re-cently at hishome in Liv-ingston, N. J.

Mr. Portswas born inHanover, Pa.,on August 14,1901. He re-ceived the B.S.degree in elec-

E. G. PORTS . trical engineer-ing in 1923,

and the M.S. degree in physics in1925, at Gettysburg College.

Serving on the engineering staff ofthe Bell Telephone Laboratories until1929, he * joined the InternationalTelephone and Telegraph Corpora-tion as an engineer with the Inter-national Communication Labora-tories. In 1932, he became a trans-mitter engineer with the FederalTelegraph Company, Newark, N. J.,and in 1934, he was appointed chiefengineer. In 1942, Mr. Ports becamemanager of the communications prod-uct division of the Federal Telephoneand Radio Corporation. He later heldthe positions of technical director, as-sistant vice president, and chief en-gineer and assistant manager of theradio division at FTR.

In 1947, Mr. Ports received theNavy Department's Certificate ofCommendation for outstanding as-sistance in organizing and supervisingthe radio engineering staff at FTR,engaged in the development of radiotransmitters, receivers, and directionfinders for the United States Navy.

Mr. Ports served on the IRECommittees of Annual Review,Standards, and Transmitters and An-tennas. He also served on committeesof the Radio and Television Manu-facturers Association, Radio Techni-cal Commission for Aeronautics,National Electrical ManufacturersAssociation, and the American Stand-ards Association. He was a fellow ofthe American Institute of ElectricalEngineers.

Alex A. Javitz (A'48) has been namedspecial features editor of Electrical Manu-facturing. He was previously associate editorof the publication.

Mr. Javitz was born in New York, N. Y.,and attended Columbia University. He re-ceived the B.S. degree in 1918 from theCooper Unioq School of Engineering.

He is a member of the American Chem-ical Society, the Society of Plastics Engi-neers, the Inter-Society Color Council, andthe Conference on Electrical Insulation.

Institute News and Radio Notes1953 431

PROCEEDINGS OF THE I.R.E.

Chairman

I. L. Knopp628 Ecton Rd.Akron, OhioW. L. FattigBox 788Emory Univ., Ga.

C. E. McClellan1306 Tarrant Rd.Glen Burnie, Md.L. B. Cherry1418 Central Dr.Beaumont, Tex.J. H. Merchant2 Cedar St.Binghamton, N. Y.F. D. Lewis275 Massachusetts Ave.Cambridge, Mass.I. C. GrantSan Martin 379Buenos Aires, Arg.W. K. Squires1115 St. John's Ave.Kenmore, N. Y.R. M. Mitchell357 Garden Dr., S.E.Cedar Rapids, IowaA. R. BeachLoueridge Circle E.Eau Gallie, Fla.

R. M. Krueger5143 N. Neenah Ave.Chicago, Ill.

J. P. Quitter509 Missouri Ave.Cincinnati, OhioJ. L. Hunter3901 E. Antisdale Rd.S. Euclid, OhioJ. H. Jaeger361 Oakland Park Ave.Columbus, OhioJohn Merrill16 Granada Terr.New London, Conn.R. A. Arnett4073 Rochelle Dr.Dallas, Tex.J. L. Dennis3005 Shroyer Rd.Dayton, OhioW. R. Bliss1426 Market St.Denver, Colo.W. L. CassellIowa State CollegeAmes, IowaF. W. Chapman1756 Graefield Rd.Birmingham, Mich.D. E. Reynolds3709 Memphis St.El Paso, Tex.R. E. Neuber130 Willonwood CenterEmporium, Pa.

H. L. ThorsonGeneral Electric Co.Owensboro, Ky.R. B. Jones4322 Arlington Ave.Fort Wayne, Ind.Arthur AinlayRR 6, Mt. HamiltonHamilton, Ont., CanadaF. L. MasonElec. Office NavalShipyard

Pearl Harbor, Oahu, T.H.

p

9

CE

SecretaryAKRON (4) Buford Smith, Jr.

1831 Ohio Ave.Cuyahoga Falls, Ohio

ATLANTA (6) H. W. Ragsdale654 Cooledge Ave., N.E.Atlanta, Ga.

BALTDIORE (3) C. D. Pierson, Jr.1574 Waverly Rd.Baltimore, Md.

BEAUMONT- C. B. Trevey'ORT ARTHUR (6) 2555 Pierce St.

Beaumont, Tex.IINGHAMTON (4) R. F. New

654 Chenango St.Binghamton, N. Y.

BOSTON (1) A. J. PoteCyclotron LaboratoryCambridge, Mass.

BUENOS AIRES C. A. CambreOlazabal 5255Buenos Aires, Arg.

BUFFALO- R. R. ThalnerNIAGARA (4) 254 Rano St.

Buffalo, N. Y.CEDAR RAPIDS G. W. March

(5) 424 Liberty Dr.Cedar Rapids, Iowa

:NTRAL FLORIDA Hans Scharla.Nielsen(6) Radiation Inc.

P. O. Drawer QMelbourne, Fla.

CHICAGO (5) J. J. Gershon2533 N. Ashland Ave.Chicago, 111.

CINCINNATI (5) D. W. MartinBox 319-A, RR 1

Newtown, OhioCLEVELAND (4) H. R. Mull

R.F.D. 3, Elyria, Ohio

COLUMBUS (4) R. W. Masters1633 Essex Rd.Columbus, Ohio

CONNECTICUT H. E. RohloffVALLEY (1) The So. New Eng. Tel. Co.

New Haven, Conn.DALLAS-FORT J. A. GreenWORTH (6) 6815 Oriole Dr.

Dallas, Tex.DAYTON (5) A. B. Henderson

801 Hathaway Rd.Dayton, Ohio

DENVER (5) R. E. Swanson1777 Kipling St.Denver, Colo.

DES MOINES- R. E. PriceAMES (S) 1107 Lyon St.

Des Moines, Iowa

DETROIT (4) N. D. Saigeon1544 GrantLincoln Park, Mich.

EL PASO (7) J. E. HoeflingBox 72

Fort Bliss, Tex.EMPORIuM (4) E. H. Boden

Box 14Emporium, Pa.

EVANSVILLE- A. P. HaaseOWENSBORO (5) 2230 St. James Ct.

Owensboro, Ky.FORT WAYNE (5) J. J. Iffland

1008 Madison St.Fort Wayne, Ind.

HAMILTON (8) John Lucyk77 Park Row S.Haniilton, Ont.

TERRITORY OF J. W. AndersonHAWAII (7) 4035 Black Pt. Rd.

Honolulu, T. H.

* Numerals in parentheses following Section designate Region number.

ChairmanH. T. Wheeler802 N. Avenue "A"Bellaire, Tex.

F. D. Meadows5915 N. OxfordIndianapolis, Ind.D. L. Ewing108-A ByrnesChina Lake, Calif.

D. G. WilsonUniv. of KansasLawrence, Kan.

W. F. Stewart1219 Skyline Dr.N. Little Rock, Ark.R. B. Lumsden332 Hale St.London, Ont., CanadaW. G. Hodson10806 Smallwood Ave.Downey, Calif.M. C. ProbstRt. 7, Box 415Louisville, Ky.H. W. Mehrling36S La Villa Dr.Miami Springs, Fla.D. E. Mereen3260 N. 88 St.Milwaukee, Wis.N. R. OldingCanadian Broad. Corp.Montreal, P.Q., CanadaC. W. Carnahan3169-41 P1., Sandia BaseAlbuquerque, N. M.H. T. Budenbom82 Wellington Ave. W.Short Hills, N. J.V. S. CarsonN. C. State CollegeRaleigh, N. C.C. E. Harp524 E. Macy St.Norman, Olda.C. W. RookUniv. of NebraskaLincoln, Neb.E. L. R. Webb31 Dunvegan Rd.Ottawa, Ont., CanadaC. M. Sinnett103 Virginia Ave.Westmont, N. J.R. E. Samuelson1401 E. San Juan Ave.Phoenix, Ariz.J. G. O'Shea104 N. Fremont St.Pittsburgh, Pa.

E. D. Scott4424 S.W. Twombly

Ave.Portland, Ore.

W. H. Surber, Jr.Princeton Univ.Princeton, N. J.Garrard Mountjoy100 Carlson Rd.Rochester, N. Y.W. F. Koch1340-33 St.Sacramento, Calif.H. J. Hicks62 Whitehall Ct.Brentwood, Mo.

I:

L

L

L,

I

SecretaryHOUSTON (6) J. K. Hallenburg

1359 DuBarry LaneHouston, Tex.

NDIANAPOLIS (5) J. T. Watson2146 Admiral Dr.Indianapolis, Ind.

INYOKERN (7) F. S. Howell313-B Tyler St.China Lake, Calif.

CANSAS CITY (5) Mrs. G. L. CurtisRadio Industries, Inc.Kansas City, Kan.

ITTLE ROCK (5) J. E. Wylie2701 N. PierceLittle Rock, Ark.

,ONDON, ONTARIO J. D. B. Moore(8) 27 McClary Ave.

London, Ont., Canada,OS ANGELES (7) B. S. Angwin

238 N. Frederic St.Burbank, Calif.

LOUISVILLE (5) G. W. Yunk2236 Kaelin Ave.Louisville, Ky.

MIAMI (6) M. C. Scott, Jr.Station WIODMiami, Fla.

,WILWAUKEE (5) H. J. Zwarra722 N. BroadwayMilwaukee, Wis.

MONTREAL, R. W. CookeQUEBEC (8) 17 De Castelneau St.

Montreal, P. Q., CanadaIEW MExIco (7) L. E. French

107 S. WashingtonAlbuquerque, N. M.

NEW YORK (2) A. B. Giordano85-99 Livingston St.Brooklyn, N. Y.

NORTH J. G. GardinerCAROLINA- 3502 Kirby Dr.VIRGINIA (3) Greensboro, N. C.)KLAHOMA CITY E. G. Crippen

(6) 3829 N.W. 23 St.Oklahoma City, Ok1a.

)MAuA-LINcOLN V. H. Wight(5) 1411 Nemaha St.

Lincoln, Neb.OTTAWA, D. V. Carroll

ONTARIO (8) Box 527Ottawa, Ont., Canada

SILADELPHIA(3) S. C. SpielmanWalton Rd.Huntingdon Valley, Pa.

PHOENIX (7) Z. F. McFaul4242 N. 2nd Dr.Phoenix, Ariz.

PITTSBURGH (4) J. H. Greenwood530 Carlton HousePittsburgh, Pa.

PORTLAND (7) J. M. Roberts4325 N.E. 77Portland, Ore.

C

Pi

PRINCETON (3)

ROCHESTER (4)

SACRAMENTO (7)

Jerome KurshanRCA LaboratoriesPrinceton, N. J.R. N. Ferry196 Lafayette Pkwy.Rochester, N. Y.H. C. Slater1945 Bidwell WaySacramento, Calif.

ST. LOUIS (5) R. W. Benson818 S. KingshighwaySt. Louis, Mo.

432

Sections

March

NI

I

I

Institute News and Radio Notes

SectionsChairman

Stanley BensonBox 1707Salt Lake City, UtahA. H. LaGroneBox F, Univ. StationAustin, Tex.

C. R. Moe4669 E. Talmadge Dr.San Diego, Calif.W. E. Noller1229 Josephine St.Berkeley, Calif.D. E. Norgaard1908 Townsend Rd.Schenectady, N. Y.

L. A. Traub2816-31 Ave., N.Seattle, Wash.

Samuel SeelySyracuse UniversitySyracuse, N. Y.

W. M. Stringfellow136 Huron St.Toledo, Ohio

G. E. McCurdy74 York St.Toronto, Ont., Canada

C. E. BuffumBox 591Tulsa, Okla.

0. A. Schott4224 Elmer Ave.Minneapolis, Minn.

A. H. Gregory150 Robson St.Vancouver, B. C. Canada

M. W. Swanson1420 Mt. VernonMemorial H'wayAlexandria, Va.

J. H. Canning1701 Chestnut St.Williamsport, Pa.

SecretarySALT LAKE CITY M. E. Van Valkenburg

(7) Univ. of UtahSalt Lake City, Utah

SAN ANTONIO (6) Paul Tarrodaychik215 Christine Dr.San Antonio, Tex.

SAN DIEGO (7) R. A. Kirkman6306 Celia Dr.San Diego, Calif.

SAN FRANCISCO 0. J. M. Smith(7) Univ. of Calif.

Berkeley, Calif.SCHENECTADY R. L. Smith

(2) Station WRGBSchenectady, N. Y.

SEATTLE (7) G. K. Barger1229 9 Ave., W.Seattle, Wash.

SYRACUSE (4) W. H. HallGen. Elec. Co.Syracuse, N. Y.

TOLEDO (4) G. H. Eash845 W. Woodruff Ave.Toledo, Ohio

TORONTO, Clive EastwoodONTARIO (8) 658 Pharmacy Ave.

Dawes Rd. P.O.Toronto, Ont., Canada

TULSA (6) W. J. Weldon2530 E. 25 St.Tulsa, Okla.

TWIN CITIES (5)

VANCOUVER (8)

WASHINGTON (3)

WILLIAMSPORT(4)

F. S. Hird224 S. 5th St.Minneapolis, Minn.

Miles Green2226 W. 10th Ave.Vancouver, B. C. Canada

T. B. Jaco*cks777 14 St., N.W.Washington, D. C.

R. C. LepleyR.D. 2Williamsport, Pa.

SubsectionsChairman

R. F. Lee2704-31 St.Lubbock, Tex.Carl Volz160 W. Hamilton Ave.State College, Pa.F. G. McCoyRt. 4, Box 452-JCharleston, S. C.S. L. Johnston207 Edgewood Dr.Huntsville, Ala.L. B. HeadrickRCA Victor Div.Lancaster, Pa.C. J. HirschHazeltine Elec. Corp.Little Neck, L. I., N. Y.R. T. BlakelyTitusville Rd.Poughkeepsie, N. Y.S. D. RobertsonBox 107Red Bank, N. J.A. G. Richardson180 Vreeland Ave.Boonton, N. J.E. L. Michaels1167 Casa Vista Dr.Pomona, Calif.0. G. Villard, Jr.2050 Dartmouth St.Palo Alto, Calif.A. A. KunzeLee CenterNew York, N. Y.

George Weiler1429 E. MonroeSouth Bend, Ind.H. G. SwiftRte. 2Derby, Kan.R. F. Tinkler166 Portage Ave., E.Winnipeg, Canada

SecretayAMARILLO- C. M. McKinneyLUBBOCR (6) Texas Tech. College

(Dallas-Ft. Worth) Lubbock, Tex.CENTRE COUNTY R. L. Riddle

(4) Penn. State College(Emporium) State College, Pa.

CHARLESTON (6) C. B. Lax(Atlanta) Sergeant Jasper Apts.

Charleston, S. C.HUNTSVILLE (6) R. C. Haraway

(Atlanta) 603 College Hill Apts.Huntsville, Ala.

LANCASTER (3) C. G. Landis(Philadelphia) Safe Harbor, Box 6

Conestoga, Pa.LONG ISLAND (2) B. F. Tyson

(New York) 49-16 Douglaston Pkwy.Douglaston, N. Y.

MID-HUDSON (2)(New York)

MONMOUTH (2)(New York)

NORTHERN N. J.(2)

(New York)ORANGE BELT

(7) (Los Angeles)

PALO ALTO (7)(San Francisco)

G. F. Senn81 Garden Rd.Little Silver, N. J.P. S. ChristaldiBox 111

Clifton, N. J.Eli Blutman6814 Glacier Dr.Riverside, Calif.J. V. Granger772 Paul Ave.Palo Alto, Calif.

ROME (4) J. M. Thompson(Syracuse) Box 1245

Haselton Br. P.O.Rome, N. Y.

SOUTH BEND (5) A. R. O'Neil(Chicago) WSBT-WSBT-TV

South Bend, Ind.WICHITA P. A. Bunyar

(Kansas City) 1328 N. LorraineWichita, Kan.

WINNIPEG (8) H. R. Gissing(Toronto) 65 Rorie St.

Winnipeg, Canada

Professional Groups

ChairmanAIRBORNE ELECTRONICS

ANTENNAS AND PROPAGATION

AUDIO

BROADCAST AND TELEVISIONRECEIVERS

BROADCAST TRANSMISSION SYS-TEMS

CIRCUIT THHORY

COMMUNICATIONS SYSTEMS

ELECTRoN DEVICES

ELE:CTRONIC COMPUTERS

ENGINEEzRING MANAGEMENT

K. C. BlackPolytechnic Res. and Devel. Co.Brooklyn, N. Y.A. H. WaynickPennsylvania State CollegeState College, Pa.J. J. BaruchM.I.T., Cambridge, Mass.D. D. Israel111 8 Ave.New York, N. Y.Lewis Winner52 Vanderbilt Ave.New York, N. Y.R. L. Dietzold34 W. 11 St.New York, N. Y.G. T. Royden67 Broad St.New York, N. Y.George D. O'NeillSylvania Electric Products, Inc.Bayside, L. I., N. Y.M. M. AstrahanI.B.M. Plant. no. 2Poughkeepsie, N. Y.Ralph I. ColeGrlffiss A.F.B., Rome, N. Y.

INDUSTRIAL ELECTRONICS

INFORMATION THEORY

INSTRUMENTATION

MEDICAL ELECTRONICS

MICROWAVE THEORY AND TECH-NIQUES

NUCLEAR SCIENCE

QUALITY CONTROL

RADIO TELEMETRY AND REMOTECONTROL

VEHICULAR COMMUNICATIONS

harmamEugene Mittlemann549 W. Washington Blvd., Chicago, Ill.

Nathan MarchandSylvania Electric Products Inc.Bayside, L. I., N. Y.I. G. EastonGeneral Radio Co.Cambridge, Mass.L. H. Montgomery, Jr.Vanderbilt UniversityNashville, Tenn.Ben WarrinerGen. Elec. Co., Advanced Elec-tronics Center, Cornell Univ.,Ithaca, N. Y.L. R. HafstadAtomic Energy Comm.Rm. 132, 1901 Constitution Ave.Washington, D. C.Leon BassGeneral Elec. Co.Schenectady, N. Y.M. V. Kiebert, Jr.Bendix Aviation Corp.Teterboro, N. J.F. T. BudelmanBudelman Radio Corp.Stamford, Conn.

1953 433