US2895073A - Circuit system for electronic memory tubes - Google Patents

Description

July 14-, 1959 w E, w, JACOB 2,895,073

v CIRCUIT SYSTEM FOR ELECTRONIC MEMORY TUBES Filed Feb. 25, 1954 3 Sheets-Sheet 1 MA Ha flrromvEY' M w W W" y 4, 1959 w. E. w. JACOB 2,895,073 CIRCUIT ,SYSTEM FOR ELECTRONIC MEMORY TUBES Filed Feb. 25, 1954 :5 Sheets-Sheet 2 [/VVE/VTOR M94727? 49m Maya/1 Jkcoa ZMZ. H

July 14, 1959 w, w, JACOB 2,895,073

CIRCUIT SYSTEM FOR ELECTRONIC MEMORY TUBES Filed Feb. 25. 1954 3 Sheets-Sheet 3 I Br Lk CIRCUIT SYSTEM FOR ELECTRONIC MEMORY TUBES Walter Emil Wilhelm .la'cob, Hagersten, Sweden, assignor to Telefonaktiebolaget L M Ericsson, Stockholm, Sweden, a corporation of Sweden Application February 25, 1954, Serial No. 412,599 Claims priority, application Sweden February 27, 1953 1 Claims. (Cl. 315-12 The present invention is a circuit system for storing information in electronic memory tubes. The storing process in question is so arranged that the storage elements in such a tube are given a marking in the form of a stable potential, in general either one of two values,

whereby the marking can be regarded as expressing a --binary function.

emission currents and the secondary collector currents.

All the storage elements are, in a tube of this type, electrically directly accessible from the outside of the: tube.

The invention for storing information in such a tube, is in principle characterized by the fact that a marking potential is applied to the respective storage element through its outer impedance, which potential may have two different values, and that at the same time or before switching on the primary electron current of the valve, the cathode is brought to a certain potential, differing from the normal condition, for example in that a pulse is applied.

The invention will hereinafter be described with reference to the drawing:

Fig. 1 a schematic diagram showing a tube according to the invention.

Fig. 2 voltage current characteristic for'a storage element of the type shown in Figure 1.

Fig. 3 "shows a circuit for connection of a storage tube.

Fig. 4 shows a current voltage characteristic ofan element as used in a circuit shown in Fig. 3.

Fig. 5 shows a circuit arrangement using a double gas discharge triode.

Fig. 6 is a diagram showing the grid and cathode potentials in the circuit shown in Fig.5.

tat s Patent The method of operation of a storage tube according to the invention will first be described.

upon they impinge partly on the collector electrode g and partly on the storage elements 2.

As long as these elements are maintained at cathode potential the electrons are unable to reach them, but as soonas one of these elements receives a positive potential an electron current will, start towards this element, which current Patented July 1959 ice 2 will initially increase with the applied potential. Due to the starting of secondary emission this current'will not increase as a linear function of the potential but will pass through a maximum and then decrease. \At'a certain potential where the secondary emission factor of the element is equal to 1, the current will again decrease to 0, and will become negative at values above this potential. After the current has passed a minimum on having the potential increased further, it will once again pass through the 0 point at a potential, which'is about the same as that of the collector electrode.

Fig. 2 shows the characteristic of such an element, which is negative within a certain range. If such an element is connected to a sufiiciently high positive potential V via a large resistance Ra, which is so large that the a resistance line for Ra intersects the element characteristic in three points S S and S it will be obvious from the known conditions of stability that only theintercepts S and S are stable, while S is an unstable equilibrium position. In such an arrangement it 'is, therefore, possible for an element to remain at only two stable positions, either a higher position and higher value for the point S in the neighbourhood of thec'ollector electrode potential, or a lower value, corresponding to S near the cathode potential. -Hence,if a mark is produced on a number of elements or a single element, with a potential corresponding either to the higher or the lower limit, the tube will store informationof a binary nature. As a consequence theexpressions 'positively marked or negatively marked element, imply that the elements after a marking procedure be locked to the potentials S or S respectively which are,

as described above, the higher and lower tentials respectively. a a

This invention is concerned with a novel system -for applying'these markings. In order to clarify this system some of the known systems will now be related. A positive marking is for example'most easily obtained by increasing the outer positive marking potential V applied externally through the resistance Ra (see Fig. 2) to such an extent that the points S and S coincide or in fact might not appear. 81 will then also be unstable and the element potential increases instantaneously to the higher stable value. In order to mark the element negatively, the opposite arrangement should be adopted, in other words, the application of such a large negative stablepomarking potential that the intersection 8;, will be the the elements marked with the same number 'n, connected over the individual resistances Ra Rag Ra are connected to common markingvoltage sources V0 V V The marking procedure already mentioned would in all tubes give the same marking'which is naturally not desirable. If the marking process is to be accomplished in only one of these m tubes, this is only possible through another'arrangement in connection with an additional local switching function applied to this tube. In such arrangement the marking potentials V are only limited in such a way that there are always three points ofintersection with the element characteristic. For a positive marking the potential V lies .above the unstable positions of equilibrium S and forlainegaltive marking (V =0). below. 8' In order to'raccom- S Fig. 2 shows, an example with two given; potentials of V for positive marking above S and iY' tor Tnega;

tive ma k (V'oEQ) belo .S'Z' In. o d r t ac mplish the storage of the information in a certain tube, which corresponds to the applied marking potentials, the

rrim rysles m surrsn s upp sed for nr lm n in t i tubs.- he em nts ave the e e heir r pe v sta le Po en als a he a n potentials pe ve y Yu r applie xt ly A ter he t e cu r has s artc s u he l m n P t co e p nd n to their position above or below the point of unstable equilibrium S S' respectively, are instantaneously locked in their new stable potentials and remain there for some time, positively or negatively marked, even after the suppression or alteration of the marking potentials.

Such arrangement has however many disadvantages. Eor example it is not possible to mark two different tubes simultaneously when using opposite marking processes. Eurther, it is not possible to mark a number of elements positively in one tube and. to mark simultaneously i one or more other tubes all the elements negatively, a case which in practice is very common, if a negative marking of all elements in one tube is synonymous with an erasure of the information stored therein. The reverse case is, of

course, also possible.

Another disadvantage of this arrangement is that limited tolerances within the element characteristics must be maintained, For example, it may happen that for a certain tube the positive marking potential V for a certain element n still is not sufiiciently high, i.e. so that V barely lies above S while the same potential for the same element of another tube is too high, for example so that S; and S coincide or that no points of intersection S 01.. S exist.

In. this case. the element in this tube would instantly get the positively marked condition. This difficulty is particularly pronounced in a. special method of reading,

which case will be described shortly below.

Generally speaking the methods of reading are based on. the fact that the inner resistance between the collector electrode and the positively marked elements is low and that thereby, by means of a modulation of the collector potential, an output signal can be obtained in the different element circuits, while if the inner resistance. between the collector and the negatively marked elements is high, a. modulation. ofthe. collector potential has no influence on the elements. Fig. 3 shows a reading arrangement, by means, of which the collector potential is held constant and the tube current is, modulated by pulses. The tubes are then usually non-conducting and are pulsed periodically in a certain sequence by a. multiplex system, so. that one tube at a time is rendered conductive during its pulse time. Thecorresponding elements it are interconnected by means of condensers (3,, with common output resistances Ra The potentials of the elements are no longer. fixed in this circuit. A positively marked: element tends to shift from its potential in the pulse pauses. By choosing a suitable. time constant C,, Ra in. relation to the pulse frequency it is assumed, that the, element shifts from its potential only to. an extent such, that during the pulse; time the original potential. can be. regained. The rercharging of the condensers C causes potential alterations of the pertinent output resistances Ra which alterations aroused as an output signal.

A negative marked element tends during the. pulse pauses to have the marking potentials applied externally. This is, however, not desirable, unless the potential displacements of the negatively marked elements also are to be. used as output signal, which in this case would have theopposita signas opposed to. that obtained from the positively marked elements. Thereis a danger, however,

that the potential of such an element during the pulse interval obtains a value, which. lies above the unstable point ofiequilibrum S ,('see Fig 2) and that thereby the elements during the following pulses instantly moveto the positively marked condition. The relativelysmall positive current maxima of the element characteristics and the relatively small distance between the points S and S favour this tendency. If one also must consider the spread of the characteristics, it is evident from above, that a correct design of such a circuit is almost impossible.

The disadvantages mentioned above are avoided by this invention as will be described in the following. The invention provides that for both marking processes two different manipulations are used so that it is possible to perform in a circuit according to Fig. 3, for example, simultaneously in one tube a positive marking of separate elements and in one or more other tubes a negative marking of all elements. Further combinations are possible but need not be described separately.

The circuit system of the invention is further characterized by the fact that the marking potentials are placed in the negative area V 60 volt with reference to the normal cathode potential. Thereby, the negatively marked potentials always lie on a potential. where they carry no current and in accordance therewith this potential does not change, whether the tube is conductive or not. For bringing the elements, which have now a permanent fixed potential for marking purposes, to a point above or below the unstable equilibrium point of their characteristics, the cathode potential is shifted while the tube is blocked, in such a manner that, when the tube is conductive again, a difierent characteristic appears, which is shifted with respect to the original characteristic, and thus the conditions for a marking process are fulfilled in accordance with what is already known.

Further in order to clarify the function, an example with reference to Figs. 3 and 4 will be described below. 'On theelements 1-,, of all tubes the marking potentials V and V have arbitrarily been applied which marking potentials differ from each other by a certain constant value. V will for example be 0 volt and V will have a. negative value with reference to cathode potential. If now in a certain tube in a mark process is to be performed in order to mark the elements, corresponding to the applied potentials V or V' respectively, positively or negatively, respectively, the cathode potential is displaced, after the suppression of the tube current, so far into. the negative area that after the current flow or current pulses, respectively have started again, a characteristic will appear with an appearance approximitely of the same type as shown in 'Fig. 4 (dashed line), whereby in relation to this the elements, which are marked with the marking potential V will lie above the unstable point of equilibrum (respectively above the positive maximum of the characteristic-because this method works also if there is no point of intersection in the first positive current area) and thereby suddenly increase to the positively marked condition corresponding to the point of intersection S The elements supplied with a negative potential are not influenced if they have not before been positively marked, because their potential then suddenly decreases from S' to V when the tube current is suppressed. After the marking process the cathode potential may naturally return to its normal value.

A negative marking of all elements in a. tubeis successful in this arrangement if the tube current is suppressed for a short time, irrespective of which of the two marking potentials V5 and V' is applied to the elements in this case. For this reason it is possible to apply marking processes of different nature in several tubes simultaneously.

One may generalize. these conditions if one from the beginning assumes twomarking potentials with different values and displaces the cathode potential with reference hereto, so that dependingon the requirements the corresponding elements. are locked by the displacement of the cathode: potential and are marked in one or other sense. lfjone makes' the negative displacement so large that-alsothe elements: with the external-ly applied lower potential V' are included for a positive marking it is possible to apply simultaneously three marking processes differing from each other in diiferent tubes with the same externally applied marking potentials, namely: (1) positive marking of a certain number of elements in one tube (2) positive marking of all elements in one or several other tubes (3) negative marking of all elements in one or several other tubes. In a similar way it is also possible to change the conditions by means of positive marking potentials which will cause similar effects with cathode displacements in a positive sense.

In actual practice positive markings are produced by means of negative cathode pulses and negative markings by means of current suppression. Fig. 5 shows an example wherein the switching function is carried out by means of a double triode of the gas filled type. This double triode consists of a common anode a, two cathodes k and k and two trigger electrodes t and t The double tube with its common anode resistance R for both discharge gaps as also the cathode resistances R and R are so connected that only one half, in other words either the discharge gap at k or the gap at k will glow at the same time. Due to the fact that the trigger electrodes t and t are supplied with suitable striking potentials it is possible to transfer the glow discharge from one half to the other. For every changeover a charging current will flow to the condensers C and C respectively which will cause a larger fall of potential to appear across R The negative potential pulse is transmitted through condenser C to the control gride g; of the storage tube. The grid of this tube has normally such a negative bias across the resistance R; taken from a bias potential that current can flow only during the positive portions of the pulses P supplied by the pulse transformer P. If a negative impulse is now applied from the gas discharge triode the grid potential will be displaced so far into the negative area that the storage tube will be completely blocked. The time of duration of this blocking is determined usually by the time constants which are formed by R C and R ,C respectively. A negative marking in the storage tube is obtained by the ignition of the glow discharge triode at k controlled by the time constant R 0 It is natural that the time constant must be chosen in such a way that the time will be suflicient for the positively marked elements to discharge in the lower stable potential areas. In order to perform a positive marking process the discharge is shifted to the discharge gap at k during which period the grid in the storage tube will receive a negative potential pulse due to a charging current for C Simultaneously a charging current transient is transferred through the primary winding of a transformer T. The secondary winding of this transformer constitutes, together with the condenser 0., a resonant circuit, which by means of the current transient in the primary winding will initiate a damped oscillation. The polarity of the transformer circuit has been chosen in such a way that the first half cycle of the oscillation is negative. Only the first negative half cycle can be transmitted to the cathode of the storage tube. The oscillation is suppressed after the first half cycle by means of the short circuiting diode D The design of the circuit has been chosen in such a way that the current in the storage tube starts exactly when the cathode pulse passes its negative maximum. A diode combination D D is used in order to ensure that the grid cathode bias never exceeds the potential of the bias source, i.e. as soon as the negative grid impulse and the negative cathode impulse difier from one another by an amount corresponding to the grid bias, the diode D closes and the grid bias is then taken over by means of the diode D which becomes conducting at the same instant, by the condenser C and the cathode, in such a way that the grid impulse will be transmitted in parallel to the cathode impulse, which means that the storage tube will remain in the cut-01f state even during the intervals between the pulses applied to the grid by the pulse transformer.

Fig. 6 shows graphically the function of the potential of the grids and the cathodes for both manipulations. If, for example, at the time T a negative marking process is made, the negative grid potential will instantaneously be strongly negative and will go along curve 1 back to the -Vg -1ine. For a positive marking process the same thing happens in the grid circiut only with the exception that the grid potential goes back faster to the starting point, approximately in accordance with curve 2. Simultaneously the cathode pulse with the zero-line as a base appears (see curve 3). Because of the diode connection D D the grid pulse at the point where its difierence, compared with the cathode pulse, is the same as Vg deviates from curve 2 and runs in parallel to the cathode pulse corresponding to line 4 back to the -Vg -line.

As the embodiment above described for the performance of marking processes only is an example of the circuit, which is suitable for certain purposes, the invention is not limited to this circuit.

I claim:

A circuit system for storing information by marking an electronic storage tube having at least one electron emitting cathode, means for control of the electrons emitted from the cathode, several secondary emission electrodes constituting storage elements, said electrodes being interconnected with outer impedances and arranged to obtain two diiferent stable potentials within a certain potential range due to their negative current voltage characteristic and a collector electrode for collecting the primary and the secondary electron currents, all said storage elements being separately and directly electrically accessible from the outside of the tubes, said system comprising means for applying a marking potential to the respective storage element through the respective .outer impedance, said marking potential differing from the normal operating potential of the cathode, and means for discharging the primary electrons into the direction of all the storage elements simultaneously.

References Cited in the file of this patent UNITED STATES PATENTS 2,576,040 Pierce et al. Nov. 20, 1951 2,617,072 Van Gelder Nov. 4, 1952 2,695,974 Skellett Nov, 30, 1954

Images