TECHNICAL FIELD The present invention relates to conveyor devices and more particularly to optical devices and techniques to detect objects thereon.

BACKGROUND ART

The retail food industry has seen significant gains in productivity in recent years. Greater volumes of food products can be passed through check-stands, thanks to the use of self bagging and double and triple belted check-stands.

Modern check-stand control systems use optical signalling to control the motion of the conveyor depending on the presence or absence of food products. In most cases, this involves the use of an emitter/receiver pair of optical units located on opposite sides of the conveyor. The optical units, along with the conveyor motors and the like, are controlled by a check-stand control.

Conventionally, these components are mounted in separate locations in the check-stand. Servicing requires that each component be dealt with separately and by skilled technicians. However, skilled technicians are often in short supply causing delays and adding to the expensive down time. It would be desirable, therefore, to provide a conveyor control device that is complete and self contained so that rectification of any fault in the conveyor control system can be achieved by any nonskilled person by simply replacing the control device.

Various photo electric sensing systems have been commonly used in the past to sense objects travelling past a monitored zone on a conveyor.

One such system is known as 'through scan' which makes use of alight emitting and receiving elements positioned on opposite sides of die conveyor. This system functions well provided that the objects to be detected are reasonably opaque and is the standard system used on modern retail check stands.

Two other systems make use of light emitting and receiving elements on the same side of the conveyor. In the 'diffuse scan' sensing system objects passing through the monitored zone are detected by light reflected off the objects. In this case, the objects must have a suitably reflective surface. The diffuse scan system is not suitable in situations where the objects to be detected and die background provide reflected optical signals of similar strength. 'Retro-reflective' photo electric sensing systems employ a light emitting element and a light receiving element positioned side by side to view a reflective target mounted on the opposite side of the monitored zone.

In all such systems, there are several factors affecting the strength of the optical signal received, including:

- the efficiency of the light emitting and receiving elements; - the reflectivity of objects intercepting the light beam as it passes from emitter to reflector to receiver;

- the distances from the light emitting and receiving elements to the objects to be detected; and

- the accuracy with which the emitted and received light beams are aligned with the reflector.

In many cases, a varying level of ambient light will also contribute to an additional variation in sensitivity, although the ambient light itself may have been eliminated, by way of signal conditioning techniques.

Thus, although the detection system may be required simply to determine the presence or absence of objects obstructing the light beam, care must be taken to minimize the deleterious effects of the above factors to obtain adequate system performance.

Employing photo electric sensing to control the flow of widely ranging objects such as customer purchases on a retail check stand is a particular challenge since some of the objects to be sensed may be highly reflective and at extremely close range, while others may be poorly reflective and at extreme range. The result is an unmanageable variation in the strengths of optical signals being received by the light receiving element.

DISCLOSURE OF THE INVENTION It is therefore an object of the present invention to obviate these difficulties.

Briefly stated, the invention involves a retail check stand device, comprising: a conveyor and a pair of side walls, each of the side walls further including a side wall section presenting an inner planar surface on a respective side of the conveyor to confine articles thereon, one of the side wall sections having a cavity adjacent the planar surface; sensing means for sensing objects on the conveyor;

an opening formed in the side wall section to provide access to the cavity; and a check stand controller responsive to the sensing means and arranged to fit through the opening and be located in the cavity.

In one embodiment, the controller includes a housing with a cap portion, and switch means including a number of switches on the cap portion to operate the controller, wherein the opening is located near an operator location and the side wall section has a top face, further comprising mounting means for removably mounting the housing in the side wall section with the cap adjacent the top face so that the switch means are within reach of an operator, the mounting means and the housing being arranged to disable the controller when the housing is disengaged from the mounting means.

In another aspect of the present invention, there is provided a binocular optic device comprising a housing to contain a light emitting element and a light receiving element, divergence control means to control the divergence angles of fight emitted by the fight emitting element and received by the fight receiving element, thereby to define a first path segment for fight emitted by the light emitting element and a second path segment for light received by the fight receiving element and to minimize overlap of the first and second path segments, the divergence control means including a plurality of optical passages to be adjacent the fight emitting element and a plurafity of optical passages to be adjacent the fight receiving element.

In one embodiment, the optical passages are formed by a plurality of vanes and, the housing is formed from a first housing portion and a

second housing portion, wherein alternating ones of the vanes are disposed on the housing portions.

In another aspect of the present invention, there is provided a controller for a retail check stand of the type having a pair of side wall sections, each with an inner planar surface on a respective side of the conveyor to confine articles thereon, one of the side wall sections further including a cavity adjacent the planar surface, the controller having a sufficiently small volume to be contained in the cavity.

In one embodiment, the controller includes a housing with a cap portion, and switch means including a number of switches on the cap portion to operate the controller, and further including within the housing, sensing means to detect the presence of objects on the conveyor, wherein the side wall section has a top face, the controller further comprising mounting means for removably mounting the housing in the side wall section with the cap adjacent the top face so that the switch means are within reach of an operator, the mounting means and the housing being arranged to disable the controller when the housing is disengaged from the mounting means; wherein the sensing means includes optical means for establishing a fight path across the conveyor; wherein the optical means includes a fight emitting element and a light receiving element, both of which are to be associated with a reflective element positioned on an opposite side wall section, wherein the fight path includes a first path segment from the fight emitting element to the reflective element, and a second path segment from the reflective element to the fight receiving element; wherein the optical means includes divergence control means to control the divergence angles of fight emitted by the fight emitting element or received by the fight receiving element and thereby to minimize overlap of the first and

second path segments, the optical passages being formed by a plurafity of vanes, the optical means includes a housing formed from a first housing portion and a second housing portion, wherein alternating ones of the vanes are disposed on the first and second housing portions to form the optical passages.

In still another aspect of the present invention, there is provided a device for processing an optical signal comprising; receiving means for receiving an optical signal; conversion means for converting the received optical signal into a proportional electrical signal; capacitor means to receive the electrical signal, the capacitor means including an output to convey a capacitor output signal whose rate of change varies in proportion to the magnitude of the electrical signal; comparison means for comparing the capacitor output signal with a threshold value, and conveying a comparator output signal when the capacitor output signal equals the threshold value; and counter means for counting increments of time from the optical signal to the comparator output signal, wherein the count is a time based measure of the magnitude of the received optical signal.

In one embodiment, the device further comprises generation means for generating a first optical signal; adjustment means for varying the intensity of the first optical signal according to variations in the magnitude of the received optical signal; compensation means to compensate for an ambient fight component in the received optical signal, wherein the compensation means includes a constant current sink to isolate and nulfify the ambient fight component, and wherein the compensation means includes generation means for generating a second

optical signal to simulate a minimum level of ambient light in the received optical signal.

In another aspect of the present invention, there is provided a technique for processing an optical signal comprising the steps of: receiving an optical signal; converting the optical signal into an electrical signal; providing a first output signal whose rate of change varies in proportion to the magnitude of the electrical signal; comparing the first output signal with a threshold value, and conveying a second output signal when the first output signal equals the threshold value; and counting increments of time from the optical signal to the second output signal, wherein the count is a time based measure of the magnitude of the optical signal.

In one embodiment, the technique further comprises the steps of: providing a generated optical signal. sampling the optical signal over a plurafity of successive intervals of time to record changes to the strength of the signal; initiating the sampling in such a manner that a central interval corresponds to an optical signal of a given magnitude; varying the onset of the step of sampling according to variations in the magnitude of the optical signal; providing a generated optical signal includes the step of varying the intensity of the generated optical signal according to variations in the magnitude of the optical signal; compensating for an ambient light component in the optical signal wherein the step of compensating includes the steps of:

providing a constant current sink to isolate and nulfify the ambient fight component; and generating a second optical signal to simulate a minimum level of ambient fight in the optical signal. In yet another aspect of the present invention, there is provided an optical device comprising an optically transparent housing, a light receiving element and a light emitting element contained within the housing; the fight emitting element being arranged to deliver fight to the fight receiving element while being positioned so as minimize the obstruction thereof to light from outside the housing.

In one embodiment, the fight emitting element is positioned to one side of the fight receiving element. In an alternative embodiment, the fight emitting element is a fight emitting diode and the light receiving element is a phototransistor formed in a T 1-3/4 moulded package.

In still another aspect of the present invention, there is provided a technique for enhancing detection of an optical signal in the presence of ambient light; the technique including the steps of: providing a light receiving element; positioning the fight receiving element so as to receive a first optical signal, wherein the first optical signal includes an ambient light component; providing a light emitting element; aiming the fight emitting element so as to deliver a second optical signal to the light receiving element, wherein the second optical signal is sufficient to overcome variations in the ambient fight component.

In one embodiment, the technique further comprises the steps of: monitoring the level of the ambient light component; and

adjusting the second optical signal according to the level of the ambient fight component, so that the fight receiving element receives a substantially constant level of ambient light.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred embodiments of the present invention are illustrated, by way of example only, in the appended drawings, in which:

Figure 1 is a fragmentary perspective view of a check-stand; Figure 2 is another fragmentary perspective view of the check- stand illustrated in Figure 1;

Figure 3a is a fragmentary perspective assembly view of a portion of the check-stand illustrated in Figure 1;

Figure 3b is another fragmentary perspective assembly view of a portion of the check-stand illustrated in Figure 1; Figure 4 is a fragmentary exploded view of a portion of the check-stand illustrated in Figure 3a;

Figure 5 is a fragmentary side view taken on arrow 5 of Figure l;

Figure 6 is a perspective view of a binocular optic device; Figure 7 is a sectional view taken on fine 7-7 of Figure 6;

Figure 8 is a side view of a first component of the device illustrated in Figure 6;

Figure 9 is another side view of the component illustrated in

Figure 8; Figure 10 is a sectional view taken on line 10-10 of Figure 8

Figure 11 is a sectional view taken on line 11-11 of Figure 8 Figure 12 is a sectional view taken on fine 12-12 of Figure 8 Figure 13 is an end view of the component illustrated in Figure

8;

Figure 14 is a side view of a second component of the device illustrated in Figure 6;

Figure 15 is another side view of the component illustrated in Figure 14; Figure 16 is a sectional view taken on line 16-16 of Figure 14;

Figure 17 is a sectional view taken on line 17-17 of Figure 14; Figure 18 is a sectional view taken on line 18-18 of Figure 14; Figure 19 is an end view of the component illustrated in Figure 14; Figure 20a is a sectional view of the device illustrated in Figure

6;

Figure 20b is an optical diagram for light waves produced by the device illustrated in Figure 6;

Figure 21 is a schematic sectional view of the check-stand illustrated in Figure 1;

Figure 22 is a fragmentary schematic view of a portion of the check-stand illustrated in Figure 1;

Figure 23 is a schematic view of a control module portion of the component illustrated in Figure 4; Figure 23a is a schematic view of a motor controller segment of the portion illustrated in Figure 23;

Figures 24a to 24c are sketches representative of time versus capacitor charge;

Figure 24d is a sketch representative of a Sensitivity Band Figure 25 is a graph representative of time versus capacitor charge;

Figure 26a to 26d are flow diagrams illustrating an operational mode of the check-stand illustrated in Figure 1; and

Figure 27 is a magnified side view of an optical element.

BEST MODE FOR CARRYING OUT THE INVENTION

Referring to the Figures, there is provided a check-stand 10 of the type found in supermarkets. The check-stand 10 has a housing 12 which supports a front conveyor 14a and two rear conveyors 14b and 14c. The housing 12 includes a pair of front side walls 16a and 16b beside the conveyor 14a and three rear side walls 16c, 16d and 16e beside the rear conveyors 14b and 14c.

A particular feature of the present invention is the use of a pair of conveyor control devices, one shown at 18 to control the flow of consumer items along the front conveyor and the other shown at 19 to control the flow of consumer items along the rear conveyors.

Referring to Figure 3a, the front conveyor control device 18 has a housing 20 which is shaped complementary to a recess 22 formed in the side wall 16b. Located along the bottom of the recess 22 is a connector unit 24 which is shaped to fit a number of corresponding connectors on the front conveyor control device. The connectors permit the front conveyor control device 18 to be connected with a second electrical circuit, namely that for the front conveyor, the rear conveyor control device and power supply and to be disconnected therefrom simply by removing the conveyor control device 18 from the recess 22.

Referring to Figure 4, the housing has a back wall 26, a cap 28 and a shroud 30. Attached to the back wall is a first electrical circuit including a control module 32 which controls: i) the photo-electric sensing of objects on the conveyor; ii) conveyor run timing, and; iii) energizing of conveyor motors.

The control module 32 has a circuit board 32a and a number of connectors 34 along its lower edge to connect with the connector unit 24. The control module 32 also has a sensing means for sensing objects on the conveyor, in this case an optical means for generating the optical signal to estabfish a light path across the conveyor as well as for receiving the optical signal. The optical means is in the form of a binocular optic device 36 for generating and receiving fight pulses along a light path 18a across the conveyor as shown in Figure 1. This area of the conveyor will be referred to below as the 'monitored zone' thereof. The shroud 30 is formed from bent material and has a lower flange 30a with a number of recesses to expose the connectors 34 to the connector unit 24. The cap 28 includes a number of switches, indicator fights and the like as shown collectively at 33 which will be further described below. Both the shroud 30 and the sidewall 16b have aligned apertures 30b, 16f respectively to expose a binocular optic device 36.

The control module 32 is arranged with a vertically oriented cavity along both ends of the circuit board as shown at 32b for locating and varying the position of the binocular optic device as desired, as illustrated in Figure 5. This allows the module to adapt to a particular conveyor application and to a range of sidewall heights. For check- stands in particular, it is preferred that the binocular optic device be positioned immediately above the conveyor surface.

The sidewall 16b is further provided with mounting apertures 31a to receive a mounting pin 31b on each end of the cap 28. The apertures and pins 31a, 31b provide secure yet releasable mounting of the housing. Of course, other methods can be used to secure the housing in place.

Referring to Figure 2, the rear conveyor control device 19 also includes a pair of binocular optic devices, as shown generally at 19a and

19b. In this case, the binocular optic devices 19a and 19b are aligned with passages on opposite sides of the rear side wall 16d to form fight pathways 19c, 19d across the two rear belts 14b and 14c.

Referring to Figure 3b, as with the front conveyor control device 18, the rear conveyor control device 19 has a housing that fits within a recess in the side wall 16d. Unlike the front conveyor control device 18, however, the rear conveyor control device 19 has two binocular optic devices 19a, 19b, each directed toward a respective rear conveyor. In addition, the rear conveyor control device 19 has a pair of push button switches 19e, 19f each for manual control of a respective rear conveyor, as will be described.

Referring to Figures 6 through 20, the binocular optic device includes a housing which has two adjacent cavities 36a, 36b, the first 36a to receive a fight emitting element in the form of a fight emitting diode (LED) 38, the second 36b to receive a fight receiving element in the form of a photo-transistor 40. A particular feature of the binocular optic device is the use of a means to control the divergence angles of fight emitted by the light emitting element or received by the fight receiving element and thereby to define the boundaries of said first and second path segments. This divergence control means is in the form of an number of optical passages which are provided by an arrangement of vanes 36c. The vanes 36c which control the direction of fight emitted by the LED and received by the photo-transistor, thereby improving the accuracy of the photo-electric sensing circuitry. Another particular feature of the housing is the manner in which it is made.

Referring to Figure 20a, the vanes are spaced to provide pathways with a width 'a' of preferably 0.030 inch and a length 'b' of preferably from about 0.450 to about 0.650 inch. By keeping the pathways relatively narrow and long, these dimensions reduce the scatter of the fight passing through the housing to an angle θ of about 5.28 degrees.

The cavities 36a, 36b are formed in a housing 36d having two portions 36e, 36f. Each portion 36e, 36f includes alternating ones of the vanes 36c as shown the Figure 6. In order to permit the housing to be ejected from a plastic injection mould, each of the vanes has a thin tapered cross section. Furthermore, the angle of the opposing faces on each vane is selected to match that of the adjacent faces when the portions 36e and 36f are nested as shown in Figure 7 to form fight pathways 36g.

It will be seen that, while the pathways 36g are inclined in the end view, they are nonetheless parallel in the side view and therefore are effective in reducing scatter.

The housings join along a pair of edge regions 36h, 36i that are shaped to prevent fight leakage, by a 'V shaped projection on edge region 36h and a 'V shaped groove on edge region 36i. The cavities 36a and 36b are also separated by a central wall 36j, that fits within a recess 36k to prevent light leakage from one cavity to the other. In addition, a tight fit is provided between the central wall 36j and the recess 36k to provide a force fit to hold the housing portions together.

The binocular unit is preferably aimed at a reflective surface in the form of a piece of reflective tape of the type sold under the tradename 'Scotchfite' manufactured by 3M as illustrated at 37 in

Figures 21 and 22. The tape 37 is believed to have a prism-like particle surface which causes incident light to be reflected back to the source as shown in Figure 20b.

The vanes function to form a first path segment originating from the LED to the reflective surface and a second path segment originating from the reflective surface to the photo-detector, and to otherwise minimize overlap of the outgoing optical signal from the LED on the first path segment and the reflected incoming signal on the second path segment, while permitting each path segment access to the reflective surface. For check-stands in particular, it is preferred that the binocular optic device be positioned immediately above the conveyor surface and with the vanes of the binocular optic device parallel to the conveyor surface.

In the case of the check-stand 10, the binocular optic device is directed at the reflective surface mounted on the opposite sidewall, for example, a 1 inch (high) X 2 inch (wide) patch of reflective tape 37. As shown in Figures 21 and 22, the vanes are conveniently arranged to limit the vertical field for the photo-transistor to the face of the side wall. The vertical field will vary according to the spacing between the binocular optic device and its target or the dimensions of the fight pathways 36g. While the horizontal field of view is not restricted significantly by the vanes, the sidewall will reflect a substantially less degree of fight than the reflective tape.

Because the electrical functions are entirely controlled by the self contained circuity of the front conveyor control device, the wiring harnesses are needed only to join the motors, the power supply and the front and rear conveyor control devices together.

By combining the switching control for the conveyors in the front and rear conveyor control devices, the check-stand 10 substantially reduces the complexity of conventional check-stands. Should repair be necessary, service technicians can simply replace the front conveyor control device, or in the unlikely event that a conveyor motor needs servicing, installation is simplified since the number of connections and wiring is greatly reduced.

The control module 32 will now be described with reference to Figures 23 and 23a. Referring to Figure 23, the control module 32 includes a calibration unit shown at 50 to calibrate the control module on a continuous basis and a motor controller 64 to control the timing and travel of the conveyors. The calibration unit 50 has an ambient fight compensator 55 which receives a photo-current signal from the photo- transistor 40 on conductive path 55a. The ambient fight compensator 55 includes a constant current sink and, in a manner to be described, compensates for an ambient fight component in the signal and conveys the signal to the attenuator 51 along conductive path 55b. The attenuator 51 sets the scaling of the signal to one of two or more preset ranges and then conveys the scaled signal to a capacitor 52 via conductive path 52a, the magnitude of the attenuator output affecting the slope, or rate of change, of the capacitor charge. A comparator 54 is coupled to the capacitor 52 to receive a signal representative of the capacitor charge. Comparator 54 also receives a time varying bias threshold level (as will be described) on conductive path 54a from a threshold generator 53.

As an alternative to attenuating the received signal in the control module of the conveyor control device, the LED fight pulse may instead be varied, by way of a level control means for controlling the intensity

of the optical pulse. This may be done through the use of a resistor arrangement, shown in dashed fines at 59, to vary the level of power being delivered to the LED.

The comparator 54 generates an output on conductive path 54b which will switch as the capacitor voltage passes, the threshold value. A counter 56 receives the comparator output on path 54b and a signal from a pulse generator 58 along path 58a. The pulse generator 58 interrogates the LED 38, to cause a pulse of fight to be emitted from the binocular optic device.

Another particular feature of the control module 32 is the means by which it accommodates the speed of response and sensitivity of the phototransistor both of which are diminished in conditions of low ambient light and which may vary with fluctuations of ambient fight. To that end, the fight pulse from the binocular optic device is in fact a two stage pulse. The first stage, in effect, simulates a minimum level of ambient fight to the phototransistor, which enhances the sensitivity of the phototransistor. The second stage for which the onset is delayed from the first, is the interrogation pulse itself.

Except for the period of time during which the second stage of the signal occurs, the ambient fight compensation unit continually adjusts its constant current sink to the current generated by the photo-transistor. During the second stage of the signal, the constant current sink maintains a current at the level determined immediately prior to the second stage of the signal. This enables the constant current sink to isolate and nulfify the ambient component of the signal and to permit the pulse to pass through the attenuator and on to the capacitor on conductive path 52a.

The counter 56 communicates with a calibration controller 60 which monitors the condition of the cafibration unit 50 to make minor changes to the cafibration of the control module as can be necessary due to fluctuations in power feed, ambient fight levels, temperature and the like. The calibration controller also communicates with a photo-counter 62 via conductive path 60a and a validity counter 68 via conductive path 60b to control the duration of the cafibration sequence.

The calibration controller further communicates with an increment counter 67 via conductive path 67a and a decrement counter 69 via conductive path 69a, to maintain the photo-electric system in cafibration throughout an operating period as will be described below.

The cafibration controller 60 also communicates with a zero crossing detector 61 which issues a synchronizing pulse on conductive path 61a. This synchronizing pulse is also conveyed to the pulse generator 58, the threshold generator 53 and the motor controller 64.

The motor controller 64 is activated by the signals received from: i) the photo-counter 62 on conductive path 62a and associated with front conveyor 14a; ii) a front push button switch 33a; iii) a rear push button switch 33b; and iv) a conveyor 'select' switch 33c.

In addition, the motor controller 64 receives two signals, each on conductive paths 71e and 71f from a pair of rear cafibration units 71a and 71b, which in turn receive signals on conductive paths 71c and 71d from the two binocular optic devices 19a and 19b. The rear cafibration units 71a and 71b operate in a similar fashion to cafibration unit 50, in

that the incoming signals from the binocular optic devices are calibrated, compared and counted. The signals on conductive paths 71e and 71f are similar to those on conductive path 62a.

Referring to Figures 23 and 23a, the motor controller 64 includes timers 64a, 65a and 65b, motor control switch 64c, and motor actuator 64d which energizes motor 66. Rear controller 70 contains motor actuators 70a and 70b connected to motors 72 and 74 respectively.

The timer 64a receives an 'initializing' signal from the photo- counter on conductive path 62a and sends a corresponding signal to the switch 64c along conductive path 64b. This signal is routed to motor actuator 64d via conductive path 64e and then via path 66a to energize the front conveyor motor 66.

The timers 64a, 65a and 65b also receive the synchronizing pulse on conductive path 61a from the zero crossing detector 61 to advance the timers in a manner to be described.

The motor control switch 64c communicates with timer 64a via conductive path 64h to convey a command signal originating from the front push button switch 33a. In addition, the motor control switch 64c communicates with timers 65a and 65b via conductive paths 65e and 65f respectively, to convey command signals originating from the rear and select push button switches 33b and 33c respectively.

This arrangement allows a conveyor to be advanced while under control of its photo-detectors either: a) for a predetermined time interval, or,

b) until the fight path is again interrupted, at which point the photo-counter detects the presence of the object and sends a stopping signal on conductive path 62a to reset the timer and stop the conveyor.

In a similar manner, timers 65a and 65b, receiving signals from their respective photo-counters along conductive paths 71c and 71d, will send initializing signals to motor control switch 64c along conductive paths 65c and 65d and thereafter to their respective motor actuators 70a and 70b along conductive paths 64f and 64g. But as will be described, only one of motors 72 and 74 will be controlled by its own photo- counter and timer, while the other will respond to signals derived from the photo-counter 62.

The motor controller 64 may also receive a signal from the front push button switch 33a, the rear push button switch 33b or from the 'conveyor select' switch 33c.

Before discussing the general operation of the conveyor, it may be useful to focus on the technique by which optical signals are evaluated.

SENSΠTV ΓY BAND

The pulse generator 58 pulses LED 38 to generate an LED light pulse as shown by curve 'a' in Figure 24. The photo-transistor 40 receives the reflected pulse to generate a substantially constant current signal on path 55a, the magnitude of which is proportional to the strength of the reflected pulse of fight. The capacitor receives the signal on path 52a and accumulates a charge as illustrated in Figure 24 by curve 'b'.

Normally, optical systems such as this are designed to maximize the speed and amplitude of the signal and to detect only the presence of either a 'white' signal or a 'black' signal. The use of a capacitive load integrates the signal and in this case is particularly useful since a variation in signal amplitude can be converted to a variation in time.

In a particular installation, in which the optical alignment and selection of the LED and photo-transistor and other components has been previously established, the intensity of the fight pulse returned to the photo-transistor from the reflective target will be nearly constant, exhibiting only minor variations due to ambient fight conditions, temperature, etc. The time interval from the beginning of the LED pulse to switching of the comparator output will be similarly constant. The Sensitivity Band is a brief interval of time in the immediate vicinity of and centred upon the point in time corresponding to die time at which a signal representing an unobstructed view of die reflecting target will be issued from comparator 54, as illustrated by curve 24c. The Sensitivity Band is illustrated in Figure 24d and has a plurafity of successive sectors of time, in this case five, corresponding to successive variations in the strength of the received optical signal.

The Sensitivity Band is positioned in time so that a central sector corresponds to a received optical signal of a given magnitude. In tiiis case, one sector, namely sector 3, corresponds to die pulse generated by d e target. The position of d e Sensitivity Band may also be adjusted according to variations in die magnitude of die received optical signal, as further described in die section entitled 'Calibration'.

If die reflected pulse is relatively low in power, die slope of die capacitor charge curve is correspondingly low, tiiereby increasing die

time to reach me threshold at which the comparator switches. The counter 56 counts die time increments from d e instant d e LED emits the pulse of fight to the time the capacitor reaches its threshold causing die comparator output to switch.

It follows tiiat a reduction in reflected pulse strengtii will move the count toward sector 5 or beyond into the 'black' region. In other words, if a nonreflective or 'black' object is interrupting die fight patii (as would occur if a consumer item is passed by die conveyor into die light patii) die count would move into die 'black' region. If die 'black' object is tiien removed from d e fight patii, die target reflects a greater quantity of fight, increasing die slope of die capacitor charge and ultimately restoring die count to die 'white' region.

The direshold generator produces a varying bias voltage tihreshold level as shown by a downward chain dotted curve in Figure 24. Witii tiiis feature, an increment of time can be made to represent a larger change in signal strengtii, ti an it would if die direshold were constant. For example, as is desirable for a check-stand application, a time delay of one microsecond can represent a change in signal strengtii of approximately one percent ratiier tiian merely a fraction of a percent. Referring to Figure 25, it can be seen tiiat die time interval, between intersections b, and b ₂ representing die times of detection for two different signals using d e downwardly sloping direshold, is clearly shorter than die time interval between intersections a ₁ and a^ for die same signals against die constant direshold.

Although unlikely, certain consumer items, such as metal cans, may present a flat mirror like surface in a plane very nearly parallel to d e reflective tape. If this occurs an amount of fight exceeding that from

tiie reflective tape may be received by die photo-transistor and die count in counter 56 will be reduced still further into die 'super-white' region. For a check stand application, super-white signals are preferably considered to represent an object on die conveyor and are dierefore treated as black signals.

Thus, ratiier tiian detecting simply a white signal and a black signal, d e present technique provides a conversion of signal strengtii into a time based spectrum of signals ranging from 'super-white' to 'black' . The number of time increments for each sector may be selected as desired, depending on die application for which die calibration system is intended. For example, if mere is a high variability in ambient fight, as may occur in a check-stand application, it may be desirable to select a higher number of time increments in those sectors considered to be 'target', specifically die 'increment', 'white' and 'decrement' sectors, as will be described. The preferred technique to provide tiiis time based Sensitivity Band is to cause die counter to interrogate die comparator on a repeated basis after a predetermined delay period (referred to as a Sensitivity Band delay).

VALIDITY COUNTER

The validity counter 68 governs die duration of the cafibration sequence and is only used immediately following application of power while establishing the position (in time) of the Sensitivity Band. Each time me received signal falls witiiin die 'target' sectors 2 to 4 in die Sensitivity Band, die validity counter is decremented by one. This means tiiat after a preset number of 'target' signals is received, die validity counter is decremented to zero to end die cafibration sequence.

PHOTO-COUNTER

The photo-counter determines when die occurrence of a 'white' or 'black' signal should change die state of die conveyor. The photo- counter is advanced to an upper count limit after a given number of 'white' signal detections and to a lower count limit after a given number of 'black' signal detections. For example, if it is assumed tiiat at a particular point in time, die photo-counter has a count of 9, which represents conditions of no objects on die conveyor to block die photo- beam and conveyor running: i) two successive black (or super-white) counts reduce die photo- counter count to 7, verifying tiiat die light patii has been interrupted by an obstacle and stopping die conveyor. The photo-counter is men preset to, and returned to, a count of 1 for every succeeding 'black' signal; ii) a number of successive white counts, say 7, can be taken to verify tiiat die fight patii has become clear of obstacles, thereby advancing die photo-counter to 8 and starting die conveyor to advance die next obstacle toward die fight patii. The photo-counter will tiien be preset to, and returned to, a count of 9 for successive white signals.

OPERATION Briefly described, d e cafibration controller adjusts die Sensitivity

Band delay on each interrogation. In a simplified example, tiiis is done by: i) decrementing the Sensitivity Band delay (that is reducing the count corresponding to die onset of sector 1) if die comparator output switches in sector 2; ii) incrementing die Sensitivity Band delay if detection occurs in sector 4; or iii) leaving die Sensitivity Band delay unchanged if detection occurs in any sector otiier tiian 2 or 4.

In effect, the cafibration unit considers die detection of die signal within any of sectors 2, 3 or 4 to represent an unobstructed view of die target (white).

Signals detected in sectors 1 and 5 are classified 'grey' and assigned die same value as die previously established condition. For example: i) if me photo-counter is at a count of 8 or 9, a signal detected in sector 1 or 5 is considered to represent d e target, and classified as falling in sector 2 or 4 respectively. ii) if the photo-counter is at a count less tiian 7, a signal detected in sectors 1 or 5 is considered to represent an object on die conveyor and classified black.

For an application such as a check-stand, die grey sectors are useful to provide hysteresis and to avoid ambiguity in a 'white/black' or conveyor start/stop decision. Signals detected earlier tiian sector 1 or later tiian sector 5, referred to as super white and black respectively, are considered to result from objects on die conveyor and both are classified black.

Thus, die present technique is capable of making cafibration adjustments to the time base to accommodate minor variations as can occur in any electrical installation and die need for manual cafibration mechanisms is eliminated.

Meanwhile, die photo-counter verifies die presence or absence of objects on d e conveyor, allowing a change of state of die conveyor only if a preset number of 'white' or 'black' signals are obtained. The speed of modern electronic components allows these multiple signals to be

measured in a mere fraction of a second and thus does not impose a recognizable delay to the user. For example, the cafibration unit may be configured to interrogate die LED 60 times per second. Therefore, for a typical conveyor speed of 6.6 inches per second, each interrogation pulse is equivalent to a conveyor travel of 0.11 inch.

The control module functions as shown in die flow diagrams of Figures 26a to 26d. On power up, die attenuator is set to 'off and die calibration controller passes die 'Begin' step, after which: i) die validity counter is set to 16; ii) die cafibration controller waits for die next power line synchronizing pulse, at which point; iii) die pulse generator interrogates the LED with a pulse

(typically 300 microseconds duration); and thereafter iv) the cafibration controller waits for a count from die counter

56.

If me count from die counter is greater tiian a certain limiting value, say to correspond to a delay of 285 microseconds, die cafibration controller determines mat die signal is 'too weak' and reverts die sequence to die 'Begin' step, switching die attenuator to 'off if applicable.

If die count from die counter is below a minimum value, say to correspond to a delay of 150 microseconds, die cafibration controller determines tiiat die signal is 'too strong' and switches die attenuator to 'on' and reverts die sequence back to die 'Begin' step.

If repeated signals are eitiier too weak (as may occur for example if the photo-sensor fight beam is blocked) or too strong (as may occur

for example if the system contains defective components), die routine described above will be executed repeatedly and die conveyor will remain stopped.

If however, die signal is neitiier 'too weak' nor 'too strong' die cafibration controller derives d e Sensitivity Band delay from the counter output and passes to die 'Again' step of d e sequence. Here: i) d e pulse generator terminates die interrogation pulse of the LED; ii) die cafibration controller waits for die next synchronizing pulse; at which point, iii) d e pulse generator interrogates die LED with a pulse.

With die Sensitivity Band delay established, d e controller executes a Sensitivity Band Routine, wherein die counter waits for the comparator output to switch. The counter interrogates the comparator output six times, for the most part immediately before die end of each of die sectors of die Sensitivity Band. The cafibration controller initiates different subroutines depending on where in die Sensitivity Band d e comparator output switches.

If d e comparator output has switched prior to die first interrogation (corresponding to die super-white sector of die Sensitivity Band), d e cafibration controller advances the sequence to die Super- white subroutine, wherein: i) if die validity count is '0', tiien die cafibration controller determines that die super-white signal is to be considered 'black' and advances die sequence to die 'Black' subroutine as will be described;

ii) if die validity count is not '0', the cafibration controUer increments die validity counter by '4'; iii) if die validity count tiien becomes greater than '15', the calibration controUer reverts die sequence to die 'Begin' step; or iv) if die validity count is less tiian '16' die cafibration controUer reverts die sequence to die 'Again' step.

If die comparator output has switched by d e second interrogation (corresponding to the first sector of d e Sensitivity Band), die calibration controUer determines that die signal is 'grey' and branches to execute die 'Grey Decrement' subroutine, wherein: i) if previously detected signals were deemed equivalent to 'target', the Grey Decrement signal is also classified as 'target' and die sequence is routed to die Decrement subroutine as will be described; ii) Alternatively, if previously detected signals were 'black', die sequence is directed to die Black subroutine.

Similarly if die comparator output has switched by die third interrogation (corresponding to die second sector of die Sensitivity Band), d e cafibration controUer considers this to be a 'target' signal but determines that a Sensitivity Band delay adjustment is necessary. As a result, die cafibration controller increases die 'Decrement' counter by one count, tiien if overflow occurs, reloads die Decrement Counter to a predetermined count, decrements d e Sensitivity Band Delay and advances die sequence to die 'white' subroutine.

The Decrement counter serves the purpose of making the Sensitivity Band delay adjustment less reactive to abrupt changes to die

lighting conditions. In tiiis manner, die system is made less erratic. For example, die decrement counter may if desired have a count as high as 256, which can represent a real time delay of about 4 seconds for die counter to run its course and become 'overflowed' as wiU be referred to below.

If increasing d e Decrement Counter does not cause an overflow die sequence is advanced directly to die 'white' subroutine, wherein: i) die cafibration controUer increments d e photo-counter by '1'; and thereafter ii) if die photo-counter count is less tiian 8, the cafibration controUer reverts d e sequence to die 'Again' step; iii) if die photo-counter count is greater tiian 7, die calibration controUer advances die photo-counter count to '9'; and thereafter iv) if die validity count is '0', die motor control unit energizes d e conveyor motor to advance die conveyor and die calibration controUer reverts die sequence to die 'Again' step; or v) if die validity count is not '0', meaning tiiat die process of initializing die Sensitivity Band Delay is not fully completed, the cafibration controUer decrements die validity counter by T, die motor control unit energizes die conveyor motor to advance die conveyor and die calibration controUer reverts die sequence to die 'Again' step.

If die comparator output switches in die fourth interrogation

(corresponding to the third sector of the Sensitivity Band), die calibration controUer determines that die signal is 'white' and moves die sequence

to the 'White' subroutine as described above, effecting no adjustment to me Sensitivity Band delay.

If die comparator output switches in d e fifth interrogation, die cafibration controUer determines tiiat d e signal is falling in die fourth sector and considers tiiis to be a 'target' signal. However, die calibration controUer determines tiiat a Sensitivity Band delay adjustment is necessary. As a result, die calibration controUer increases die 'Increment' counter by one count, tiien if overflow occurs, reloads ti e Increment Counter to a predetermined count, increments die Sensitivity Band Delay and advances die sequence to die 'white' subroutine.

As with die Decrement counter, the increment counter serves the purpose of making die Sensitivity Band delay adjustment less reactive to abrupt changes to die fighting conditions.

If increasing die Increment Counter does not cause an overflow die sequence is advanced directly to die 'white' subroutine, as described above.

If die comparator output switches in die sixtii interrogation, die calibration controUer determines that the signal is falling in die fifth sector and thus is a 'grey' signal. Therefore, d e cafibration controUer diverts die sequence to die Grey Increment Subroutine. As for die Grey Decrement signal, eitiier die Increment or Black subroutine wiU then be executed.

If die comparator output switch has still not switched after die sixth interrogation, d e cafibration controUer determines tiiat the signal is 'black' and advances die sequence to the 'black' subroutine; wherein:

i) die photo-counter is decremented by '1'; and tiiereafter ii) If die photo-counter count is greater than 7, the cafibration controUer advances die sequence to die Again step; ϋi) If the photo-counter count is less than eight, die photo- counter signals die motor control unit to stop die conveyor, at which point, die photo-counter is reset to T and d e cafibration controUer reverts die sequence to die 'Again' step.

The conveyor 'select' switch enables die check-stand operator to designate one of die two rear conveyors as selected and die other non- selected. The rear conveyors operate in two control modes as foUows: i) Selected: In tiiis case, die selected rear conveyor is jogged, say for 1.25 seconds each time the front belt is stopped by die detection of an obstacle in die fight patii. Altiiough die motor control switch 64c prevents die 'selected' conveyor from responding to d e photo-sensors monitoring its own downstream end, cafibration unit 71 wiU continue to perform aU functions required to correctiy update die photo-counter and Sensitivity

Band Delay. ii) Non-selected: The non-selected conveyor operates under control of its own photo-sensors and timing control in exactly die same manner as die front conveyor.

At the conclusion of die check dirough process, die operator may operate me 'select' switch to obtain selection of die otiier rear conveyor. In this event, die previously 'selected' conveyor wiU become 'non- selected' and tiius controUed by its own photo-sensors. MeanwhUe die

previously 'non-selected' conveyor wiU become 'selected' and be caused to jog on each occasion tiiat die front light patii becomes obstructed.

For a single rear conveyor check-stand it can be arranged tiiat die same conveyor is re-selected foUowing operation of me select switch.

In this event, a fast take away action will be obtained to quickly remove die checked items to the bagging area, before die jog mode is resumed for checking die next customer's purchases.

The front and rear push button switches, 33a and 33b respectively, are a 'momentary action' type aUowing die operator to generate command signals to override automatic control of the conveyors. The switches operate in two modes: i) when a front or non-selected rear conveyor is running, briefly depressing die appropriate switch, tiiat is switch 33a for the front conveyor or switch 33b for die non-selected rear, wiU immediately halt a respective conveyor; ii) when a front or non-selected rear conveyor is not running, depressing and holding die appropriate push button switch wiU cause die conveyor to be advanced for as long as die appropriate switch is maintained, tiiat is switch 33a for die front conveyor and switch 33b for die non-selected rear conveyor. The halted conveyor will revert to automatic operation when its fight patii is interrupted.

The additional push button switches 19e and 19f associated witii left and right rear conveyors respectively may be located near the bagging area to enable a customer to manuaUy advance die non-selected conveyor.

A particular feature of die motor controUer 64 is tiiat it may be easUy adapted to check-stands having a number of different configurations, ranging from a single conveyor to as many as five or more conveyors. These include a single rear conveyor case where die 'selected' and 'non-selected' operating modes are applied to die same conveyor (rather than between two rear conveyors as described above).

While die conveyor control device 18 has a connector unit located along die bottom of die recess and a mating connector on die bottom of the housing, it will of course be understood that the connectors may be located on die side waU of die housing and die inner side waU of die recess. It wiU also be understood tiiat die conveyor control device may be located instead on die side face of die side waU 16b, provided of course that it is positioned near die top face and sufficiently close to die operator for easy operation. The conveyor control device may also hinged to die side waU in order to be moved into a shaUow exposed recess exposed in die side waU, as opposed to die relatively narrow deep recess as described above.

While die conveyor control device 18 disclosed above has apertures which are aligned with similar apertures in die side waU, die optical means may instead be directed through some other passage. For example, the housing of die conveyor control device may have a suitable side waU which itself may form a portion of the inner side waU of die conveyor.

WhUe me conveyor control device described above employs die binocular optic device for reflective sensing as its optical means, otiier optical arrangements may be used instead of die binocular optic device. For example, me optical arrangement may if desired include separate

light emitting and receiving assemblies plugged into die housing and mounted on opposites of the conveyor.

WhUe the present binocular optic device includes vanes to confine the fight padiways for die optical elements, die vanes may perhaps be replaced by an array of holes, an optical fibre bundle, other formed passageways, or for tiiat matter an array of alternating transparent and opaque regions in an object placed immediately in front of the optical elements, which wiU restrict die fight padiways in a simUar manner to die vanes. A suitably dimensioned structure providing a single passageway, as opposed to a number of passageways, may also be sufficient in some cases.

The binocular optic device may be used in a number of other applications beyond its use witiiin die control unit, in order to opticaUy isolate optical elements.

WhUe die cafibration technique described above to generate die sensitivity band has been restricted to use in an optical system for a conveyor, it will be understood that die technique may be suitable to a number of otiier applications. For example, die technique may be perhaps be used to detect and quantify die changes in optical signals, such as contrast meters, photometers and die like. This technique may also be suitable in die detection of colours. In addition, while die embodiments above have been described witii respect to conveyors and more particularly to check stand conveyors, it witt be understood tiiat various features of die present invention may be equaUy applied to other areas. For example, die optical features of die above embodiments may be useful in otiier areas where optical sensing is or can be used.

WhUe die optical device described above makes use of a single LED producing a two stage optical signal, it would be equaUy feasible to employ a pair of LED's, one of which provides die first stage and die otiier of which provides die second stage. To give greater precision, or if circuit, timing or otiier constraints so indicate, d e LED which provides die first stage may be continuaUy energized or may be modulated to simulate a substantiaUy constant level of ambient fight.

Given that the purpose of the first stage of die signal is to minimize the variation and to enhance die sensitivity of die phototransistor, die LED which provides this first stage may, if desired, be integraUy formed with or be located adjacent to die phototransistor. Referring to Figure 20a, this may be achieved by modifying die housing of die binocular unit to provide a third cavity 36m to receive a bias LED 41 therein. As can be seen, die bias LED 41 is positioned to expose die phototransistor 40 witii a fight level selected to overcome die effects of varying ambient light levels. In otiier words, die bias LED 41 provides a level of fight tiiat exceeds die strongest level of ambient fight.

Alternatively, die bias LED may be integraUy formed with die phototransistor. Referring again to Figure 20a, die photo-transistor 40 is schematicaUy shown with a right angled fight 40b extending therefrom and carrying a bias LED 40c thereon.

An integraUy formed bias LED is also iUustrated at 80 in Figure

27 and which combines a photo-transistor shown generaUy at 80a and a bias LED shown generaUy at 80b. The device is manufactured using the standard T 1-3/4 moulded package commonly used for botii LED's and phototransistors as single separate devices, for example sold under die tradename Motorola. The device 80 has four wire leads, namely the

coUector and emitter 82a, 82b of the phototransistor and die anode and cathode 84a, 84b of the LED respectively.

The phototransistor 80a may be constructed in die usual way, tiiat is by way of a semiconductor die mounted on one of the wire leads, say die coUector 82a, which also establishes a first connection to that region of the transistor. A second connection is made by means of a fine wire 82c from die emitter region of die die to die coUector lead 82b. In a simUar manner, die LED semiconductor die is mounted on die side of a third wire lead, say die anode 84a, and connection is made by means of a second fine wire 84c from die cathode region of the LED die to the catiiode lead 84b.

The phototransistor assembly is positioned within die moulded package which forms a substantiaUy opticaUy transparent housing such that, with die aid of a lens 86 moulded into the front of die package, die desired optical beam characteristics are obtained. The LED assembly is placed in close proximity to die phototransistor die and oriented to batiie the latter with its light, w le providing the least possible obstruction of die fight signal entering through die lens of the device.

The above embodiments make use of a sensitivity band whose position in time is changed with corresponding changes to die magnitude of die received optical signal. However, it may be equaUy feasible to instead adjust the intensity of the generated optical signal in order to maintain die received optical signal at a substantiaUy constant magnitude, bearing in mind tiiat fluctuations in die received optical signal will occur due to such tilings as ambient fight. This may be done, for example, by having die calibration controUer 60 communicate with die resistor arrangement 59 by way a conductive patii shown in dashed lines at 59a.

WhUe die above embodiments concern a retro reflective sensing system utilizing a reflective target on the opposite side of d e monitored zone, the present technique of integrating die electrical current signal may be applied to optical signals obtained from a 'diffuse scan' sensing system, tiiat is a sensing system in which die emitted fight pulses is reflected by die object which it is desired to detect. In this case, it is necessary tiiat aU objects appearing in die sensor beam provide a reflected fight signal greater tiian die background signal. The strongest signal received is used to derive a reference signal level against which signals from aU other objects are assessed.

The present technique of integrating die electrical current signal may also be applied to optical signals obtained from a 'through scan' sensing system, tiiat is a sensor system in which emitter and receiver a situated on opposite sides of die monitored zone. In this case, aU to-be- detected objects must be sufficiently opaque and die reference signal level is thus equal to die signal obtained when the monitored zone is clear of such objects.