1956(57, No. 8 243

A NEW KIND OF TUNING INDICATOR TUBE

Tuning indicators ("magic eyes") are employedin radio receivers and also, in bridge and balancecircuits, as null-indicators for voltage comparison.The tuning indicator tubes hitherto available havethe disadvantage that, generally speaking, they onlyshow whether a voltage difference is present or not;they do not give its sign and only give a very roughidea of its magnitude.

These shortcomings have been eliminated in theE 82 M tuning indicator tube, developed in thelaboratory of the Valvo radio tube factory inHamburg 1). In this tube an electron beam of rec-tangular section produces a rectangular spot on afluorescent screen. Unlike other indicator tubes,the E 82 M has two symmetrically mounted elec-trodes which act as deflector plates. The voltagesto be compared are applied to these electrodes eitherdirectly or via amplification stages. If the twovoltages are not of equal magnitude there is de-flection of the spot to one side or the other; further,the width of the spot when not deflected (i.e. whenthe voltages are equal) gives some indication of thevalue of the two voltages.

In a tube like this there must be some sort ofdatum to mark the position of no deflection, i.e.the central zero. The datum for the E 82M is onthe outside of the tube, in order to facilitate zerocalibration, in connection with possible departuresfrom symmetry in the electrode system or the cir-cnit. Another reason for having the marker on theoutside is to make the tube better adaptable topurposes which may require special forms of indi-cation. The external datum mark necessitates achange in the arrangement of the fluorescent screen:this is viewed, not from the side on which the elec-trons strike it (as in the conventional tuning indi-cator), but from the side opposite to the beam. Thescreen is in fact coated on the inside wall ofthe tube,to avoid parallax errors. The datum mark may takethe form of a pattern painted on the outside of thetube; alternatively, it may be a thin metal mask,cut out in any desired pattern, fitted on to thecylindrical envelope. One pattern that is suitablefor many purposes appears in jig. 1; jig. 2a showsthe complete tube, and fig. 2b a tube with the maskholder removed and four alternative mask patterns(numbered 1-4).

1) See also H. te Gude and E. Schaaff, Abstimmanzeige-röhren fur die Messtechnik, Elektronische Rundschau 9,184-189, 1955.

621.385.832:621.317.7

Fluorescent screens on glass are nothing new;they have long been used in oscillograph and tele-vision picture tubes. In such tubes the electronaccelerating potential is of the order of severalthousand volts. A tuning indicator, however, must

89314

Fig. 1. Mask for the E 82M tuning indicator. Left: appearancewhen tuning or adjustment is out of balance. Right: appearancewhen balanced. The breadth B of the spot (and hence alsothe height of the luminous triangles, when balanced) is arelative measure of the mean value of the two applied voltages.

rely on the supply already available in the receiveror measuring set; thus the acceleration voltage shouldnot be higher than, say, 250 V. At voltages as lowas this the normal type of fluorescent screen acquiresa considerable charge, so that further electrons areprevented from striking it, and thus it remainsdark; at higher voltages, increasing secondaryemission drains off the charge quickly enough forthis phenomenon to be avoided 2). Where the screenis on a metal base, as in conventional tuning indi-cators (where, it will be remembered, the screen isviewed from the electron beam side, not by trans-mission as in the present tube) this base serves toconduct the charge away, even at low voltages.In the present tube the charge is drained off bymeans of a very thin layer of tin oxide coated on theinside wall prior to the application of the fluorescentscreen'; the tin oxide is practically transparent, butit has good conducting properties, and it is to thislayer that the positive acceleration potenrial isapplied. In this way the voltage below which chargewould accumulate on the screen has been reducedto about 100 V. The phosphor used is zinc oxide,which fluoresces with a bluish-greenlight and which,after further activation by electron bombardment,exhibits a slower decrease of brightness with agethan the usual willemite.

2) For a detailed discussion of these phenomena, see for exam-ple, J. de Gier, A. C. Kleisma and J, Peper, Secondaryemission from the screen of a picture-tube, Philips tech.Rev. 16, 26-32, 1954/55.

244 PHILlPS TECHNICAL REVIEW VOLUME 18

a b

Fig. 2. a) E 82 M indicator tube fitted with mask in holder. b) Tube, mask holder and (1-4) various types of masks.

2 3 4

In order that the tube should have greater de-flection sensitivity, a stage of arnplificat.ion for eachof the two voltages to be compared is incorporatedwithin the same envelope. Each of the voltages isapplied to the grid of a triode that acts as a voltageamplifier; the anodes of these two triodc systcmsat the same time function as deflcctor plates for thcindicator section of the tube. Thc actual electrodearrangement is shown in Jig. 3.

T

___ D

(WY,,\

89155

Fig. 3. Cross-section of the E 82 M tube. KL: cathode of indi-cator section providing electrons for rectangular beam. Lfluorescent screen. D and D': deflector plates, which also formthe anodes of the triode sections T anel T'.

The characteristics of the two triode systems must,of course, hc exactl he samc within the operatingrange; differences of slope due to changes in theworking point would have the same effcct as adifference between the voltages to he compared.To ensure exact symmetry in this sense, the charac-

teris tics of the triode sections of the E 82 M aremade as straight as possible. It would hardly bepossible to manufacture the tube without trouble-some departures from symmctry if its characteris-tics were curved, as they are in many conventionaltuning indicators, which have variable-mu charac-teristics in order to extend the rangc of voltagesthat may he compared. The E 82M can also hegiven a wide measuring range, however, in spite ofits straight-line characteristics, and this is achievedin the following way. Thc two triode systems areeach provided with a separate anode resistor, whilethey share a common cathode resistor. The cathoderesistor and the cathode bias voltage are bothrelatively large so that the negative bias then cf-fectivc on thc grid of each triode changes in thesame sense and by approximately tc same amountas changes in the mean value of the two voltagesto he compared. This produces a beam-width versusmean-voltage characteristic, the working region ofwhich, though roughly straight, has a smaller slopethan it would have in the absence of the cathoderesistor. The beam width thus gives a less sensitive,indication of the mean value of the two appliedgrid voltages, which means that the operating rangeis increased and moreover this range can be variedduring operation. Whatever the magnitude (withincertain limits) of the applied voltages, their differenceis always indicated in full. In order to permit therange adjustment facility for the mean voltage, thecathodes of thc two triode sections are taken to ascparatc pin from that of the cathode of the indi-cator section.Apart from its obvious uses, particularly as a

null indicator in bridge circuits, where indication

1956/57, No. 8 TUNING INDICATOR TUBE -E 82 M 245

I· '

, of the. sign of the unbala~ce is of great valu~ espe-cially in the choice of the correct measuring range,the E 82M can be used for a number of differentpurposes. The two trio des can be made to function.as a multivibrator in a flip-flop (scale of two)counter circuit, a suitable mask pattern (3 in fig. 2)being fitted to give indication of the counting state.Fitted with another mask pattern such as 2or 4 in fig. 2, the tube becomes suitable (as areconventional tuning indicators) for giving a check

69156

,~~~"r::

Of---~"'h---'+..;::

Fig. 4. Null-balance indicating circuit using the E 82 M, thetube being operated entirely on A.C.

on modulation depth in tape-recorders and the like.Mentio'n mayalso be made of the use of the tube(with mask 1 in fig. 2) as a tuning indicator forFM receivers, and for voltage' measurement witha suppressed zero level (alarm circuits set off by thedeviation of the unknown voltage from a standardcomparison voltage).In these and ~ther applications it is sometimes a

great advantage that the E 82 M may be operatedentirelyon alternating voltages. Periodical variationof the voltage on the deflector electrodes does noharm, provided only that this voltage remáins pro-portional to that on the fluorescent screen, and thisis automatically the case when the ampli:fying sec-tions are fed from the s~me alternating voltage asthe fluorescent screen anode. It should be noted thatwhen using the range adjustment facility describedabove, jhe bias voltage on the common cathoderesistor Rk must be in opposite phase to the voltageon the screen (see fig. 4).

H: te- GUDE*) and E. SCHAAFF*).

*) Development laboratory, Valvo G.m.b.H. Radiorëhren-fabrik, Hamburg.

.'

246 PHILlPS TECHNICAL REVIEW VOLUME 18

ABSTRACTS OF RECENT SCIENTIFIC PUBLICATIONS BY' THE STAFF OFN.V. PHILIPS' GLOEILAMPENFABRIEKEN

Reprints of these papers not marked with an asterisk * can be obtained free of chargeupon application M the Philips Research Laboratory, Eindhoven, Netherlands.

2357: R. Vermeulen: Music reproduction (T. Ned.Radiogenootschap 2~, 39:41, 1956).

Short description of subjective experiments on thereproduction of music and the use of artificialstereo-reverberation (see also Philips tech. Rev. 17,171-177 and 258-266, 1955/56).

2358: G. Diemer and W. Hoogenstraaten: Evi-dence for hole-mobility in CdS (Physica 22,172,,1956).

Measurements of the pb,otoconduction under local 'illumination have been carried out on single crystals'of CdS. When illuminated between graphiteelectrodes by a narrow line of ultraviolet (3650 Á,widtli about 0.5 mm, i.e, 1/20 of inter-electrodedistance), these crystals show an increase of thecurrent, at 250 volts D.C., by a factor of 20. This in-crease of the conductivity can be explained only ifthe space charge of the conducting free electrons inthe dark part of the crystal is compensated by anearly equal quantity of positive holes, which musttherefore be assumed to migrate from the illuminat-ed part through the whole inter-electrode space.

2359: D. de Nobel and D. Hofman: The dielectricconstant of CdTe (Physica 22, 252, 1956).

The dielectric constant of CdTe was determinedat 20 "K and 77 "K at frequencies varying from1 to 100 kc/s with a bridge ofthe Schering type. Thedielectric constant is found to be approximately 11.

2360*: J.Volger, J. M. Stevels and C. van Ameron-gen: Les pertes diélectriques de divers"monocristaux" de quartz aux très bassestempératures (Verres et Réfractaires 10,,3-14, 1956).

Translation into French of R 272.

2361: A. van Weel: Récepteurs de télévision àphase linéaire (Onde Electrique 36, 4.8-56,1956).

The author shows that it is possible, withoutcomplicated circuitry, to design an I.F. amplifierwith only a very small phase distortion yet fulfillingthe normal requirements for television receivers. Thesame subject is dealt with in Philips Res. Rep. (seeR 273 of these abstracts).

2362: J. Smit: Galvanomagnetic properties offerromagnetic metals and alloys (Disserta-tion, Leiden, 11 April 1956).

A number of measurements and theoretical workby the author are here collected together. In chapterI the band structure of incomplete shell electrons3d and 4s in nickel is calculated, taking into accountcorrelation. Chapter 2 gives experimental results forthe Hall effect in ferromagnetic metals and alloysand a theoretical discussion on the origin of thespontaneous part. It is' shown that in a perfectlyperiodic lattice the spontaneous Hall effect cannotoccur. The explanation is to be found in anisotropicscattering of the conduction electrons at latticefaults, due to spin-orbit interaction. This is workedout numerically with the aid of a simple model.Chapter 3 deals with the magnetoresistance offerromagneties. Resistance anisotropy is explainedon the basis of ellipsoidal charge distribution of thecore electrons caused by spin-orbit interaction. Thisis discussed quantitatively (see also these abstractsNos. 1996 and 2319).

2363: N. W. H. Addink: Bestimmung von Spuren-elementen (ohne chemische Vorbehandlung)und Analyse kleinster Materialmengenmittels spektrochemischer Methoden (Mikro-chimica Acta, No. 1-3, 1956, pp. 299-303).(Estimation of trace elements, withoutchemical pre-treatment, and analysis ofvery small amounts of material by spectro-chemical methods; in German.)

The lowest limits of .detection of several trace.elements are given for spectrochemical and micro-'chemical determinations. There is a relationshipbetween the lowest limits, of detection as determinedby spectrochemical procedures at various tempera-tur~s. Some figures regarding the minimal amountof the sample are given.

2364: A. L. Stuijts: Sintering"of ceramic per~a-nent magnetic material (Trans. Brit. CereSoc. 55, 57~74, 1956).

The permanent magnetic properties of the hexag-onal compound BaFe12019 are described. To achievethe best magnetic values the product should have ahigh apparent density and consist of particles witha diameter of the order of 1 !L' Sintering of a non-

1956/57, No. 8 ABSTRACTS, OF RECENT SCIENTIFIC PUBLICATIONS 247

stoichiometrie composition is favourable for obtain-ing high densities. In order to keep the crystallitesas small as possible, à_ high rate of heating up and ashort time at the highest temperature are desirable.. In sintering, two different types of crystal growthare observed: a continuous and a discontinuous one.In the latter case, an explanation may be found inthe presence of a second. phase. A considerableimprovement in magnetic properties can he ob-tained by lining up tbe hexagonal axes of the com-ponent crystallites. In practice this can he achievedby pressing a paste of the powder in a' magneticfield. Crystal growth during sintering enhances themagnetic anisotropy of the material considerably.

2365: J. S. C. Wessels and R. van der Veen: Theaction of some derivatives of phenylurethanand of 3-phenyl-l,l-dimethylurea on theHill reaction (Biochim. et biophys. Acta 19,548-549, 1956).

The action of some derivatives of phenylurethanand of 3-phenyl-l,l-dimethylurea on the Hillreaction was investigated. 3-phenyl-l,l-dimethyl-urea derivatives inhibit the Hill reaction at lowerconcentrations than derivatives of phenylurethan.The inhibitory activity is enhanced by introduetionof electron-attracting groups into the benzene nu-cleus. A possible explanation of the. experimentalresults is presented.

2366: J. J. Balder: Verfärbung farbiger Objektedurch die Einwirkung von Tageslicht,Glühlampenlicht und Leuchtstofflampen-licht (Lichttechnik 8, 57-81, 1956). (Dis-coloration of coloured objects under theinfluence of daylight, incandescent lightand fluorescent light; in German.)

The discoloration of a number of coloured objectsby various types of light was investigated. Fourspecimens of each object were exposed to four typesof light: daylight filtered through Window glass,incandescent lamp light, and light from tubularfluorescent lamps "White-de-Iuxe" and "Warm-white-de-luxe", The dose of light administered wasin all cases 8 Mluxh. The discolorations were judgedvisually by three persons. This method could hejustified by eolorimetric measurements. Daylightthrough window glass had the worst discolouringeffect, followed by fluorescent lighting White-de-Iuxeand. then incandescent lighting, while fluorescent.lamps type Warm-white-de-Iuxe gave least dis-coloration.

2367*: J. Hermsen, A. M. J. Jaspers, P. Kraaye-veld and K. van Duuren: New Geiger tube

designs, hollow anode and parallel platecounters. (Proc. Int. Conf. Peaceful usesAtomic Energy, Geneva; Aug. 1955, Vol. 14,pp. _275-276, New York 1956).

Short notice concerning some Geiger counters ofunusual geometry. One type has a tubular anode(internal diameter 23 mm), in place of the usualwire; the sample may he placed within this tube.Another type has paraUel plate electrodes. Theproperties of these counters are given in the form ofgraphs. .

2368: J. A. Kok: Experiments with gas-filledtrio des (Appl. sci. Res. B 5, 445-453, 1956).

. In this paper experiments are described showingthe different types of electrical discharges in a gas-filled triode. The determining parameters are thefollowing: the cathode emission, the spacing ofcathode, grid and anode, the diameter of the meshesof the grid, the potentials of the grid and the anode;the gas pressure and the differential ionizationfunction of the gas. The anode voltage may beconcentrated in a space-charge sheath. Ifthis space-charge sheath is located at the grid, the anodecurrent may be modulated with moderate grid

_potentials. If not, much larger voltages are requiredfor modulation.

2369: K. H. Klaassens and C. J. Schoot: Somefluoro-suhstituted phenoxyacetic acids (Rec.trav. chim. Pays-Bas 75, 186-189, 1956).

The preparation of 2,4-dichloro-6-fluorophenoxy-acetic acid, 2,4-dichloro-3-fluorophenoxyacetic acidand 2,4-dichloro-5-fluorophenoxyacetic acid is de-scribed.

,2370: C. J. Schoot and K. H. Klaassens: 5,7-dichlorocoumaran-3-one (Rec. trav. chim.Pays-Bas 75, 190-192, 1956).

The preparatien of 5,7-dichlorocoumaran-3-one isdescribed.

2371: .J. F. Klinkhainer: Reflectionless transmis-sion through 2n-terminal-pair networks(Proc. Symp. Mod. Network Synthesis,'Polytechnic Inst. Brooklyn, April 1955).

Various authors have treated 2n-poles by a gener-. alization of corresponding considerations of 2-poles.In such cases current and voltage vectors take theplace ofthe scalar currents and voltages ofthe 2-polecase, while impedance matrices take the place of thenormal scalar impedances. In a similar way, in thisarticle, by a generalization of considerations of a4-pole chain, a "mirror-impedance' and-a "trans-

• 248 PHILlPS TECHNICAL REVIEW VOLUME 18

mission constant" are defined in matrix form for achain óf 4n-poles. Both matrices, for positive valuesof the real part of the frequency parameter, areunambiguously determined. The eigen vectors of thetransmission matrix can be considered as independ-ent travelling waves, each with its own dampingand phase shift.

2372: C. Z. van Doorn and D. de Nobel: Lumines-cence, transmission and width of the energygap of CdTe single crystals (Physica 22,338-342, 1956).

Irradiation of a CdTe single crystal or biasing aCdTe p-n junction in .the forward direction lead toa luminescence with a maximum at about 8880 A.From transmission measurements and measure-ments of the long wave limit of the photo-e.m.f. of a.CdTe p-n junction as a function of the temperature,the shift of the band edge was seen to vary from2.34X10-4 eV;oK at 77 OK to 5.44x10-4 eV;oK at800 OK.

2373: J. L. Ouweltjes and W. L: Wanmaker:Contributions to the problem of nonstoi-chiometry in oxygen-dominated phosphors(J.Electrochern. Soc. 103, 160-165, 1956).

According to general experience, oxygen-dominat-ed phosphors of highest efficiencies are obtainedwith non-stoichiometric compositions. This has beenexplained by the hypothesis that crystal defectsare essential to obtain luminescence. In this paperit is postulated that the ideal, and therefore stoi-chiometric crystal should have maximum efficiency.That an excess of one of the ingredients in generalresults in higher efficiency, is explained with certainfactors arising from the mechanism of the solid-state reactions, such as the difficulty of reaching.equilibrium conditions. The incorporation of theactivator, the presence of U.V. absorbing separatephases, the problem of homogeneity of the. phos-phors, and the reactivity of the ingredients used forthe synthesis are discussed, mainly on the basis of

experimental evidence gathered in the study of thephosphors used in fluorescent lamps.

R 298: P. Zalm: The' electroluminescence of ZnStype phosphors (Philips Res. Rep. ll, 353-399, 1956).

First part of a paper dealing with various aspectsof the electroluminescence of ZnS (Destriau effect).Section I gives a brief survey of the theories of-photeluminescence -in ZnS, -in -so .far-as -these -are'-of consequence for electroluminescence. Variousmethods of preparation of electroluminescent ZnSpowders are treated in section 2. Section 3 dealswith electrical and optical measurements whichhave led to a qualitative model with which themechanism of electroluminescence can be explained.A distinction must be made between phosphorswhere excitation by the field causes ionization ofthe activators, and those where this does not happen.In the latter case the light emission is virtually inphase with the voltage; in the former case, electricalfactors cause a shift over a substantial fraction ofthe period of the field between the moments ofmaximum excitation and of maximum emission.In experiments with hexagonal single ZnS crystalsactivated by Cu it was found that the mechanism ofelectroluminescence does not differ essentially fromthat of powders. In single crystals the excitationoccurs at the internal barriers, which have rectify-ing properties. A relationship is established betweenthe orientation of the barriers at which light emis-sion occurs and that of the crystal axes. For thesecond part of this paper, see R 300.

R 299: M. J. Sparnaay: On thermal hysteresis oftransformations in solids (Philips Res. Rep.ll, 400-409, 1956).

A simple order-disorder theory for the qualita-tive understanding of hysteresis phenomena is dis-cussed. The essential feature is the introduetionof an element of asymmetry into a zeroth-ordertreatment.