Detection Of Mixtures Of Narrow Band Photoluminescers

Abstract

Readout process and apparatus for reading coded symbols in a code in which the components are narrow band photoluminescent compositions. The symbols, on suitable substrates such as paper tape, labels on cans, bottles and the like, are illuminated by flashes from a xenon tube with an ultraviolet transmitting filter. The flashing tube is aimable so that it can be directed either on a window in front of which tape containing the symbols is passed or through another window onto containers with labels and the like. Suitable optics are moved into position so that the different locations of the substrate do not change the image path length. When the symbols are illuminated with successive ultraviolet flashes, they luminesce in the visible in the different narrow band colors corresponding to the different components and this is imaged on a filter wheel having the same number of filters as there are components: four for numerical work or six for alphanumeric work. After passing through the filter wheel, the visible beam is imaged on the cathode of a photomultiplier tube. The filter wheel also is provided with magnets which open switches connecting the photomultiplier tube to different circuits for each filter and also generate a trigger pulse which triggers the flashes of the xenon tube in synchronism with the positioning of each filter by the rotation of the wheel. The signal from the photomultiplier tube passes through amplifiers, one for each filter, and into logic circuits which provide for gates to select the particular signal from a particular filtered beam and also start readout at the beginning of a symbol and reset after the filter wheel has made a second full revolution during which no signals are received. Response of each amplifier is delayed about 20 .mu.sec. so that the extremely short-persistence fluorescence from such compounds as optical brighteners and the like has died out and response is only from the slower decaying photoluminescence of the narrow band code components. The logic circuits send out coded signals corresponding to the code to display mechanisms, such as Nixie tubes, or a permanent recording, such as a solenoid driven typewriter of the computer readout type, or both, or to a solenoid driven diverter for deflection from a conveyor system.

Description

BACKGROUND OF THE INVENTION

Photoluminescent coded inks have been described and claimed in the Freeman and Halverson U.S. Pat. No. 3,473,027, Oct. 14, 1969 and a modification in which some of the symbols are short-life fluorescers, and these are distinguished from the longer-decay-time photoluminescent materials of the other components, are described and claimed in the U.S. Pat. No. 3,412,245 of Halverson, Nov. 19, 1968.

Various readout mechanisms involving illumination of the coded symbol are described in the above patents. There is, however, a demand for automatic apparatus and process for the rapid readout of symbols on various substrates, such as paper tapes, labels on containers, and the like.

The Freeman and Halverson and Halverson patents above referred to describe the use of narrow band fluorescent components in the coding, such as chelates of lanthanide ions having an atomic number greater than 57, lanthanide-doped inorganic fluorescers, such as vanadates, and the like. All of these narrow band fluorescers have time constants of the multimicrosecond range, such as 10 to some hundreds of .mu. seconds as contrasted to the extremely short half-life or time constant of the fluorescence of certain organic compounds, such as optical brighteners, which have a time constant of the order of a small fraction of a microsecond. As described in the Freeman and Halverson patent, codes which depend on the presence or absence of a particular photoluminescent component provide a choice of symbols represented by 2.sup.n -1, where n is the number of components. This provides 15 symbols for straight numeric work with four components, 63 symbols for alpha numeric operations with six components, and the like.

SUMMARY OF THE INVENTION

The present invention involves an improved automatic process and apparatus for the readout of photoluminescent coded symbols, such as those described in the two applications above referred to. The components to be used are the ones having a moderate half-life. This is for the reason that many substrates, such as papers, contain blue fluorescent materials which have the extremely short time constants of from 10.sup..sup.- 7 to 10.sup..sup.- 8 sec. Under certain types of nomenclature the moderately long half-life materials might be referred to as phosphorescers instead of fluorescers, but throughout the remainder of this application the more general term "photoluminescers" or "photoluminescence" will be used.

As it will appear from the more detailed description below, a wide variety of apparatus elements may be used in practicing the present invention, so that the invention may be considered both as a process for readout and as a broad apparatus. In general, from the apparatus aspect it will appear that the present invention is a combination of elements which are not individually new things.

Essentially in the present invention a substrate having a series of coded symbols, such as a paper tape or sheet or labels on containers, such as bottles or cans with labels, or printing of the photoluminescent coded symbols directly on their surfaces are passed in front of the reading mechanism, either continuously or semicontinuously. In the case of labels on containers, such as bottles and cans, the container may move intermittently and then be rotated for reasons which will appear below.

A flash tube of short-wave radiation, such as a xenon tube with an ultraviolet transmitting filter which does not transmit the visible is triggered to flash at predetermined short intervals. The flash is directed to illuminate successive small marking areas on the particular substrate from which the symbol is to be read, a marking area being defined as a small area in which a symbol in the different photoluminescent components is imprinted. A symbol may be in the shape of a number, letter or a small area of no special shape having the coded components. Each flash, which is of short duration, for example of the order of 2 or 3 .mu. sec., is quite intense and causes the components of the symbol in the marking area to photoluminesce in their particular different wavelengths, which can be in the visible or in some cases in the near infrared. Since the components have half-lives of a number of .mu. sec., they continue to luminesce for a number of .mu. secs. after illumination has ceased. The photoluminescence is optically imaged in a beam passing through successive filters, preferably on a filter wheel rotating at reasonable speed, for example of the order of 1,000 r.p.m. or a few thousand r.p.m. There is provided one filter for each component which passes only the photoluminescence from that particular component. The preferred components which contain lanthanide ions of atomic number greater than 57 are all relatively narrow band photoluminescers and, therefore, the filters can be quite narrow in their wavelength band response, interference filters being a useful type. After passing through the filter, the radiation is imaged on the cathode of a photomultiplier tube or other sensitive detector of radiation in the visible and/or the near infrared.

The filter wheel carries magnets which perform two functions: one is to produce a signal pulse which, if necessary after suitable amplification, triggers a flash from the xenon tube when a particular filter associated with the magnet is in the light beam from the substrate back to the photodetector; the magnets also perform an additional function in the preferred form of the invention, that is to say, they switch in amplifiers for a number of channels corresponding to the number of components. Alternatively inputs can be switched into a single amplifier or a smaller number of amplifiers. The magnets also produce signals for timing and initiating readout. Another magnet which actuates a switch only once during a revolution of the filter wheel starts the readout cycle so that it makes no difference which particular filter first receives visible photoluminescence. The actuation of these devices or those performing a similar function need not be from elements physically mounted on the filter wheel, as they can be from any source which is synchronously driven therewith. However, because of the great simplicity of magnets and magnet operated switches associated with the filter wheel itself, this modification is preferred and presents many practical operating and constructional advantages.

The output from the anode of the photomultiplier tube or other photodetector contains the information from some one of the components. The cutting in of a single channel when a particular filter is in the beam demultiplexes the signal due to that particular component. This is preferably effected by suitable gate circuits which receive timing pulses.

A very suitable form of gate circuits which lends itself to compact integrated circuit structure, involves disabling the circuit output until the gating pulse is received. These types of gates will be referred to as disabling gates, abbreviated DG. These are only typical ways of cutting in a particular channel when a corresponding filter is in place. Following the gating, a symbol is examined for its threshold level to eliminate spurious responses and this circuitry also preferably shapes the resulting pulses. Each pulse is then divided into a positive and an inverted or a negative pulse, one polarity being used for selection of a particular mark and the other for character storage. In the circuitry there is also included a suitable delay so that sampling of the photomultiplier output occurs only after a certain time interval from illumination has passed such that short time constant photoluminescence, such as for example from optical brighteners in paper, has died out. Delays may be of the order of around 20 .mu. sec. and can provide for the channel being kept open for a suitable readout time, for example of the order of 50 .mu. sec. The delay mechanism can be a one-shot, (monostable multivibrator), which will be abbreviated OS, and this is preferred.

When there has been a full revolution of the filter wheel after the start of readout, the pulses in the various channels, depending on whether there was a component present or not, are stored, for example using flip-flop circuits, which will be abbreviated FF. After a full revolution a gate pulse causes the flip-flops to transmit their stored signals, after suitable manipulation if desired in buffer amplifiers, to a decoding matrix of conventional design which can put out outputs corresponding to the number of symbols, for example 15 in the case of a four component code or 63 in the case of a six component code. These outputs can then be displayed, for example by means of Nixie tubes, or may operate solenoids on an electric typewriter to produce printout in the conventional manner used with printout typewriters for computers. Both modes can, of course, be used at the same time depending on the selection by suitable switching.

After a full rotation of the filter wheel at a blank location between inked areas where there is no symbol and therefore no signals in any of the channels, a reset pulse is generated which resets the flip-flops and the other circuits for the readout of the next marking area. It will be noted that while a single full revolution of the filter wheel will have sampled all of the channels corresponding to the number of components, it is by no means necessary that this should be limited to a single revolution. Thus, for example, if the filter wheel makes more than one revolution, it simply repeats the character storage, making possible redundancy with increased reliability at the expense, however, of the number of symbols in the marking areas which can be read in unit time. For most purposes the increased reliability provided by redundance is not essential, and in such preferred cases a single revolution or a single symbol storage is preferred.

In the case of paper substrates, such as tape, it will be noted that the spacing between symbols is what produces resetting, because in the space, which must be of a size comparable to the dimensions of a symbol marking area or at least as large as the geometrical area sensed by the optical system, there are no components, and this is what actuates resetting. If printout is chosen, the typewriter may print continuously on tape or, if it is desired to print on an ordinary sheet, an additional suitable symbol for carriage return may actuate the corresponding solenoid.

Where a series of labeled containers are moved, the container is generally rotated after being brought into position in the image plane of the flashes. At this setting of the apparatus, there will be a different readout and/or display if there is a wrong label. The present invention really ceases with readout, but of course where a wrong response is given, conventional comparator circuits can give a signal, such as an alarm, for throwing a container out of the normal path of the properly labeled containers or for any other purpose. The design of such operations is not changed by the present invention and therefore forms no part thereof and will not be described in detail below. However, it should be pointed out here that the nature of the reader which can be used to produce such additional results is an advantageous additional factor of versatility or flexibility. Needless to say, other functions may also be preformed by the signals where desired.

The speed of operation of the present invention depends both on the time constants of the components and on the reliable operating speeds of the elements used. In general the mechanical elements, such as the rotating filter wheel, will prove to be the limiting factors as the electronic circuitry is capable of responses orders of magnitude more rapid. With reasonable operating speeds, for example a filter wheel turning at 1,200 r.p.m., symbols can be read at the rate of 10 per second, and if it is desired to have visual display, this is sufficiently fast so that the persistence of vision permits a satisfactory visual readout. In the case of labels, for example a label having a 12 -symbol message, which is frequently encountered in pharmaceutical labels, it is possible to read labels in 1 or 2 seconds, which permits quite rapid examination of labeled containers. The operation is entirely automatic and the reliability is high.

It should be noted that a rapid readout of purely photoluminescent coded symbols is obtained and the symbols can be entirely secret as the photoluminescent components are colorless under visible light. This has a number of advantages in the case of labeling. However, for certain purposes, for example for bank checks, there is an advantage in being able to read specially shaped symbols visually as well as by photoluminescence. In such a case the symbols may be imprinted in inks which contain a pigment so that the symbol can be read visually as well as under ultraviolet illumination. This added advantage, where it is desired, is not unique with the present invention but is fully described in the applications above referred to. It is, however, an advantage of the present invention that the rapid automatic readouts and/or printouts can be obtained without any sacrifice in the advantageous versatility of photoluminescent coding.

In the Freeman and Halverson patent referred to above, it is brought out that in order to get the maximum of reliability, the quantum efficiency of the photoluminescent components should be high so that a strong, narrow band signal is produced that minimizes interference due to other fluorescent or phosphorescent materials in the substrate. In the Freeman and Halverson patent the choice of particular photoluminescent materials is given in terms of a figure of merit. The same considerations apply when the present invention is used, but it is noted that the intensity of the ultraviolet flash from the xenon tube is very high and, therefore, there is usually available sufficient energy so that in some cases photoluminescent components may be used which are not quite as efficient as might be needed in the Freeman and Halverson code, where lower energy ultraviolet readouts may sometimes be encountered. Nevertheless, there is still an advantage in choosing photoluminescent components having good efficiencies or figures of merit.

It will be noted that since readout is by flashes with only one filter interposed at a time, only a single detector is used, and a highly sensitive photomultiplier tube can economically be employed. This is a distinct advantage as any system involving digital electronic circuits always operates with maximum effectiveness when it has strong inputs, and this is true of the amplifiers and their electronic elements that are used in the present invention and which operate with great reliability as a result of the relatively strong electrical signal produced by the highly sensitive photomultiplier tube. It is, of course, possible to use other photodetectors, provided they have the required sensitivity and low noise, but since there is plenty of room for photomultiplier tubes, the compactness of other photodetectors, such as solid-state photodetectors, is not required; and, therefore, the use of highly sensitive photomultiplier tubes constitutes a preferred modification of the present invention. In this connection, it should be pointed out that there are certain narrow band photoluminescent compositions which luminesce in the near infrared and for which some of the modern special infrared-sensitive photomultiplier tubes, such as those employing cesium in the cathodes, are desirable. The infrared photomultiplier tube is no better than one which only responds to the visible or ultraviolet, but the possibility of having a wider choice of coding components is a very real advantage, especially where six or more are used, as are needed in alphanumeric systems.

In the Freeman and Halverson patent it is pointed out that it is possible to have the code represented by absence or presence in more than one different signal level. The formula for the number of symbols would then be represented by X.sup.n -1, where X is one more than the number of levels of coded component; in the case of two levels, this is, of course, 3.sup.n -1. The precision and accuracy are not as high as when there is a presence and absence situation, and for most purposes, where sufficient components are available, this is preferred. Exactly the same situation, of course, applies in the present invention. It can be used for reading out more than one level. In the case of two levels, this would mean that in the storage part of the memory there would have to be a pair of signal amplitude responsive elements with different thresholds adjusted for the anticipated levels of response of each component in the symbols.

Throughout this specification and claims the term "time constant" or "decay half-life" will be used in a strict sense as meaning the time between points of one-half the peak luminescence. These points are passed, of course, in the rise of luminescence after illumination and then in its decay. The terms will be used in no other sense.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a horizontal section through the optics of a typical readout mechanism;

FIG. 2 is an elevation of a filter wheel;

FIG. 3 is a cross section along the line 3-3 of FIG. 2;

FIG. 4A is a side view of the mask and mounting plate for reed switches;

FIG. 4B is a pair section along the line 4B-4B of FIG. 4A;

FIG. 5 is a pair of curves of luminescent intensities of a short-time constant fluorescer, such as an optical brightener, and a longer time constant, narrow band photoluminescer corresponding to a component of the code;

FIG. 6 is a block diagram showing production of positive and negative pulses from the readout;

FIG. 7 is a block diagram of mark selection, character storage, display and printout utilizing the pulses from FIG. 6;

FIG. 8 is a schematic of an integrated circuit;

FIG. 9 is a plan view looking up at the integrated circuit;

FIG. 10 shows connections of FIG. 9 as a DG;

FIG. 11 shows connections of FIG. 9 as a FF;

FIG. 12 shows connections of FIG. 9 as an OS or delay element;

FIG. 13 is a schematic of an inverter stage;

FIG. 14 is a schematic of connections to the flash tube;

FIG. 15 is a schematic of a trigger pulse generator;

FIG. 16 is a block diagram with a partial schematic, constituting the timing pulse generator portion of the circuits of FIG. 6;

FIG. 17 is a schematic of one of a series of channel amplifiers;

FIG. 18 is a schematic of character storage reading gate circuits;

FIG. 19 is a schematic of one of the threshold level and pulse shapers for the channels;

FIG. 20 is a block diagram in more detail of a portion of the circuits of FIG. 6;

FIG. 21 is a block diagram of a sequence timer;

FIG. 22 is a block diagram in more detail of the character storage and buffer amplifiers of FIG. 6.

FIG. 23 is a block diagram of a decoding matrix.

FIG. 24 is a block diagram of display circuits, and

FIG. 25 is a schematic of typewriter printout circuits.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

FIGS. 1 to 5 illustrate the optics of the present invention. In the illustrative example a four-component code is shown, which permits 15 different symbols and is sufficient for numerical readout. For many purposes an alphanumeric system is more desirable with six components and 63 symbols, but as the additional components merely multiply elements in the drawings, the simpler four-component system will be described in order to avoid confusion and unnecessary repetition.

FIG. 5 represents curves of a fast decaying photoluminescent material, such as an optical brightener, which is in solid lines, and a component which has a typical longer time constant, in dashed lines. This corresponds to a component having terbium as the lanthanide ion determining the green photoluminescence.

FIG. 1 is a cross section through the optical readout box. A xenon strobe light 11 provided with a focusing lens 61 and a filter 12, both of which pass ultraviolet light but the filter does not pass visible light, is mounted on a bracket 13 which is hinged to turn into either of two positions. The bracket has an extension 14 on which is mounted an auxiliary lens 10. Beam imaging lenses are present 15. The bracket 13 can be turned to either of two positions, one of which, shown in full lines on FIG. 1, directs the beam out through a window 60 adjacent to which can be mounted a tape transporting mechanism to move a printed tape. As the device is of simple well-known construction, the details of which, as such, form no part of the present invention, it is not shown, but a piece of the tape is shown at 62 which is struck by the beam from the flash of the tube 11 whenever it occurs. When the bracket 13 is moved into its left-hand position, the beam is projected out through window 59, the beam imaging lens shown in dashed lines 15 being used. At the same time, the auxiliary lens 10 is brought into the beam coming through the window 60. This lens compensates for the change of path length, because when the beam goes out through the window 59 it impinges on a container, (not shown), which is moved past it at a considerable distance, for example 8 to 10 inches. The position of the container is such that the photoluminescence from its coded symbols passes through the window 60. Since this path is considerably longer the auxiliary lens 10 is thrown into the beam and compensates for the change in path length.

The photoluminescence coming through window 60, regardless of whether originating from the substrate 62, or in the other position of the bracket 13 coming in from a container at some distance and passing through the auxiliary lens 10, then passes through an imaging lens 18 and another 9, the beam going on through a window 29 in a mask 17 and thence through a filter wheel 25. In the position shown in the cross section in FIG. 1, this is a portion of the filter wheel which is solid, as will be described in conjunction with FIG. 2. The filter wheel is turned at 1,200 r.p.m. by the motor 28 and successively brings in data relating to .alpha., .beta., .gamma., and .delta. to correspond with the code in which the four components are similarly labeled. This code is represented by the following diagram, where P represents presence and-- represents absence of the indicated components: ##SPC1##

The four components are the following compounds:

.alpha. dysprosium (dipivaloylmethide).sub.3 (trioctylphosphine oxide).sub.2

.beta. terbium (trifluoracetylacetonate).sub.3 (trioctylphosphine oxide).sub.2

.gamma. samarium (trifluoracetylacetonate).sub.3 (trioctylphosphine oxide).sub.2

.delta. europium (trifluoroacetylacetonate).sub.3 (trioctylphosphine oxide).sub.2

It will be noted that for the 10 digits only 10 of the available symbols are used, those corresponding to 11 to 15 being used for various other functions, such as a carriage return in the case of a typewriter printing out on a page, X, hyphen, period, and equal mark.

The filter wheel 25 carries four magnets 30 which are spaced around the periphery in particular spatial relations to the filters. There is also provided another magnet 31 on the face of the wheel. These are more clearly shown in FIG. 3. Let us assume that the filter .alpha. in front of the window 29 of the mask 17. It will be noted that as the cross section line in FIG. 2 goes through the filter wheel at its center in the relative position where it is in FIG. 2, no filter shows. However, when the wheel has turned so that filter .alpha. is in the beam from the lens 9, as is shown in FIG. 1, luminescence corresponding to component .alpha. passes through, and is reflected by the mirror 27 onto the cathode of the photomultiplier tube 26. Accordingly, the output of the photomultiplier tube will contain a signal corresponding to this component. If the bracket 13 had been turned into the left-hand position the same result would have occurred, but the photoluminescent beam would have come from a more distant container instead of from the substrate 62.

FIGS. 4a, b, show a detail of the mask 17. It will be seen that there are mounted on this mask at 90.degree. angles four pairs of reed switches 20 to 23, A and B. Each pair is actuated only when the magnet 31 approaches them, which occurs only when the particular filter is in the luminescence beam. In addition, there is a single reed switch 63 at the top which is actuated by each of the magnets 30 in turn; in other words, it is closed 4 times during a revolution of the filter wheel or every time a filter is thrown into the beam, regardless of which filter it is. This single reed switch is used to produce triggering pulses which trigger off flashes of the xenon tube. The other switches 20 to 23 are connected into separate channels, as will be described below, and are only actuated during the time a particular filter is in the beam. In other words, as the filter wheel turns, each time a filter is in position a flash is triggered off and the two particular switches of a pair 20 to 23 corresponding to that filter are closed. As these switches act as commutators, they will sometimes be referred to in further description and on the drawings as commutating switches.

It is necessary that each time a symbol is to be read from a particular marking area on the substrate 62, or on a label on a container if the bracket 13 is thrown into its left-hand position, there must be a full revolution of the filter wheel. If the symbol has a single component there will only be a signal from the photomultiplier tube when the filter corresponding to the component is in the beam. On the other hand, if a particular symbol, such as the symbol corresponding to the equal mark or 15, has all four components, there will, of course, be four signals in the photomultiplier tube as each of the filters successively comes into play.

Turning now to FIG. 6, which shows a trigger pulse generator 58, that will be described in greater detail below, it will be seen that the filter wheel generates four signals in a revolution, which cause the trigger pulse generator 58 to trigger four flashes in the flash lamp 11. Also, there is an output to a timing pulse generator 50. The schematic of the trigger pulse generator 58 is shown in FIG. 15 and is self-explanatory, the values for voltages and other components being as given. It will be noticed that there are two transistors 64 and 65 which amplify the input signals in the customary manner.

The timing pulse generator 50 is shown in partial schematic in FIG. 16, which also shows a series of circuit elements, some of them being triggered pulse generators, marked OS, and others being inverters, INV. The inverters are conventional transistor circuits shown in schematic in FIG. 13 with their particular output waves. The triggered pulse generators, OS, are one-shot multivibrators and are integrated circuits, the schematic of one of which is shown in FIG. 8 with the pin connections in FIG. 9. To produce the required delays, the pulse lengths are set by RC circuits which are provided as shown in FIG. 12. The amount of delay depends on the values of these components, and these values are shown for the various one-shot multivibrators in FIG. 16 to give the particular delays. It will be seen that the first one-shot multivibrator triggers the second, which in turn triggers the third after a delay of 20 .mu. sec. The fourth is triggered 50 .mu. sec later, and provides a 25 .mu. sec. inverted timing pulse which goes to the switches 20A to 23A. In other words, these switches can only be turned on after the delay set forth, and of course only one switch at a time is turned on when the magnet 31 on the filter wheel 25 closes the particular reed switch, as has been described above.

Somewhat before the pulse to the commutating switches A, there is a gate pulse to the gate circuit 40, which will be described below. In the meantime, the photomultiplier 26 has been receiving signals each time a filter came into the beam and the flash tube fired provided there was present the corresponding component. The output, of course, is multiplexed for all four channels and is fed through the other commutating switches 20B to 23B of the pairs. This results in demultiplexing, and the demultiplexed signals are amplified by amplifiers 56.

The schematic of one amplifier is shown in FIG. 17 for the channel corresponding to the .alpha. component, and as indicated this is duplicated in the other three. The amplifier input signal wave appears at the left of the figure, and it will be seen that it is a negative going signal which rises rapidly to a peak when the lamp flash occurs and then gradually dies off exponentially. It will be noted that the response of the switches 20B to 23B and their respective amplifiers are not delayed, in other words, the signal comes through as soon as the flash occurs. If there were present photoluminescent materials in the marking area of very short time constant, for example optical brighteners with time constants of the order of 10.sup..sup.-7 or 10.sup..sup.-8 sec., these signals would also come through. However, it is desired that there should be no final response until sufficient delay has been introduced, and this is effected by the gate pulse, shown in FIG. 16.

The gate circuit 40 is shown in schematic in FIG. 18. Whenever the gate opens in any particular circuit, the signal in this channel goes to a threshold level and pulse shaper 57, the schematic for one channel being shown in FIG. 19 and of course duplicated for the other channels. The purpose of the threshold level is to prevent signals coming out below a certain intensity. These signals in each channel are then fed to inverters so that, as will be seen from FIG. 6, there are pairs of signals, one positive and one negative, for each channel.

FIG. 7 shows that the positive pulses from FIG. 6 go into a mark selector, shown in FIG. 20. The positive signals pass through 200 .mu.s one-shot multivibrators into dual AND gates which also receive the positive timing pulses. The mark selector develops an output pulse only after four contiguous zero signal level inputs. The output from the mark selector starts the sequence timer 34, shown in FIG. 21.

The negative signal pulses from FIG. 6 go into a set of four AND gates in the character storage section 33, which is shown in more detail in FIG. 22. The gates are only opened when there is both an inverted signal and a gate pulse from the sequence timer. Each opened gate sets a storage flip-flop 33. The mark selector output also resets the storage which takes place at the time the sequence timer is started, (FIG. 7). As soon as the sequence timer output has conditioned the gates, the flip-flops 33 receive and store any signals present. These signals are fed through buffer amplifiers 37 to the decoding matrix 36, as is shown in FIG. 7.

The mark selector operates when there has been a complete revolution of the filter wheel after the first pulse from the photomultiplier tube which is high enough to pass through the threshold level circuit 57. There is a delay in the first OS, shown on FIG. 21, of 50 .mu.s. The second OS of FIG. 21 causes a gate pulse to feed to the gates of the character storage 33 and the gates remain conditioned for 50ms. at least a full revolution of the filter wheel 25. The signals on all four flip-flops are amplified in buffer amplifiers 37. The amplified signals from the storage flip-flops then enter the decoding matrix 36, which is shown in FIG. 23. Depending on the particular settings of the flip-flops which have been fed on through the buffer amplifiers 37, this results in decoding. It will be noticed that there are 15 AND gates, which are labeled DUG on FIG. 23. This figure merely shows the gates for the symbols 1, 14 and 15, but of course there are 12 additional ones. The 15 outputs then go to Nixie drivers 38 and a Nixie display 39. The block diagram for this portion of the electronics is shown in FIG. 24, and the result is that the Nixie tubes for 10's and 1's display in the conventional manner.

It will be noted on FIG. 21 that an 80ms. gating pulse enables the solenoid drivers 35 of a typewriter 66 which are operated from the outputs of the decoding matrix in the normal manner. This is shown for one solenoid and driver in FIG. 25. It will be noted that in FIG. 7 the output from the decoding matrix passes through a switching device 67 which permits directing either to the Nixie tubes alone, the solenoid drivers alone, or both. This is a three-position switch of conventional design and, therefore, the details are not shown.

When the mark selector has received four negative responses, that is to say, the kind of signal when there is no luminescence on any of the four channels, it resets everything and there will be no further readout until another marking area comes into position. Then magnet 31 starts the sequence at a definite position of the filter wheel and the sequence above referred to is repeated. It will be noted that this requires a total of two revolutions of the filter wheel.

The specific description has been in conjunction with a typical set of circuits which can carry out the process of the present invention. Needless to say, any equivalent mechanisms may be used. All that is necessary is that the proper sequence of operations takes place. To recapitulate in process language, this means that there must be a flash of ultraviolet light on a particular marking area for each component; the multiplexed signal from the radiation detector must then pass signals in each channel, after a delay which prevents response from very short time constant fluorescers such as optical brighteners, into a storage device which stores the signals from each of the channels corresponding to the different components. Then the stored information must be decoded to produce an output which can be displayed as by means of Nixie tubes and/or printed out, or which can actuate other logical operations, such as selected counters or deflection gates. In order to prevent spurious responses, a signal, after the delay to allow extinction of short time constant fluorescers, must be above a sufficient level so that the readout is not started by a spurious signal of low level. So long as these functions are performed, the process phase of the present invention is not concerned with the particular mechanisms which produce the results; and therefore, from a process standpoint the invention is not limited to particular apparatus.

Claims

We claim:

1. A process for the readout of coded symbols, the code being represented by the absence or presence in at least one level of photoluminescent materials which luminesce under ultraviolet illumination and which therefore permit a choice of symbols represented by X.sup.n -1, where X is one more than the levels at which the components may be present, which comprises,

a. illuminating a marking area containing the coded symbol with brief flashes of short wave radiation,

b. optically imaging photoluminescence from the marking area onto at least one detector, transforming radiation striking the detector into an electrical signal,

c. sequentially limiting radiation to the detector to that of a single code component,

d. starting electrical readout by the occurrence of two signals, one in synchronism with sequential limitation of the whole number of different components and the second the first electrical signal from the detector above a predetermined low level,

e. amplifying the signals from the detector in separate channels sequentially, the channel amplifying only the signals corresponding to its component,

f. temporarily storing the amplified signals from each channel, and

g. decoding the stored signals after a full sequence into forms corresponding to the symbols represented by the code and of suitable nature for initiating at least one operation selected from the group consisting of visual display, printout, and other logical operations requiring the same type of electrical signal.

2. A process according to claim 1 in which the code is represented by the presence or absence of a photoluminescent component and the choice of symbols is, therefore, 2.sup.n -1.

3. A process according to claim 2 in which the components of the code have multimicrosecond decay half-lives and storage of the signals from the detectors takes place after a delay less than the decay half-life of the component but much longer than the decay half-life of organic fluorescers which have decay half-lives of a small fraction of a microsecond.

4. A process according to claim 3 in which there is a single detector and the sequential irradiation in the wavelength ranges of the photoluminescence of the different components is effected by filtering out luminescence in wavelength ranges outside that of the particular detector components the luminescence from which is to irradiate the detector.

5. A process according to claim 4 in which the photoluminescent components of the code are complexes containing different lanthanide ions having an atomic number greater than 57.

6. An apparatus for carrying out the process of claim 4 comprising in combination,

a. a short-duration, intense, gaseous-discharge flash tube provided with a filter passing ultraviolet light but substantially opaque to visible light,

b. means for imaging flashes from the flash tube onto the plane of a marking area containing a coded symbol,

c. a single detector responsive to longer wave radiation at least through the visible range,

d. means for imaging radiation from the marking area corresponding to coded components therein onto the detector,

e. a filter wheel and means for rotating it, the wheel being located to interpose successive filters in the luminescent radiation, each filter passing a wavelength band including the wavelength range of only a single component,

f. means driven in synchronism with the filter wheel to trigger flashes from the flash tube each time a filter is interposed in the luminescence beam,

g. a plurality of electronic amplifying and processing circuits receiving inputs from the photodetector, the number of channels being equal to the number of components,

h. means operated in synchronism with the filter wheel to open the electronic circuits for a single channel only each time a particular filter is interposed in the luminescence radiation,

i. a plurality of temporary electronic signal storage means equal in number to the channels and responsive to input signals of at least one level, means for connecting the output of each channel to one of the electronic signal storage means,

j. electronic decoding means for transforming stored signals into separate signals corresponding to particular coded symbols represented by one or more photoluminescent components, and

k. means for initiating actuation of a full sequence of channels representing a full revolution of the filter wheel, said means being operated in synchronism with the filter wheel and including means controlled by the first filter receiving luminescence radiation to produce a signal from the photodetector of predetermined level and means for resetting the electronic circuits actuated by a full revolution of the filter wheel during which no signals of predetermined level are produced from any component.

7. An apparatus according to claim 6 in which the electronic processing circuits include an electronic time delay less than the half-life decay of photoluminescence from any components but more than a microsecond whereby very short life luminescence from organic fluorescent material, such as optical brighteners, do not reach the electronic signal storage.

8. An apparatus according to claim 7 in which the flash tube is movable to one of at least two positions and optical means moved by the same movement to image the flashes from the flash tube onto substrates at different distances and to image luminescent radiation from these different distances onto the photodetector.

9. An apparatus according to claim 8 in which the photodetector is a photomultiplier tube.

10. An apparatus according to claim 7 in which the photodetector is a photomultiplier tube.