The present invention relates generally to an optical sensor and a method of operation thereof and in particular to a method of enhancing sensor accuracy.
Optical sensors are commonly used for a variety of functions including detecting skewed or double picked notes within the note transport mechanism of an Automated Teller Machine.
A variety of different prior art detectors have been utilized to detect note skew in ATMs. These include both electromechanical and optical detectors. However, they all have certain features in common. In particular, they all rely on a pair of sensors, each of which is located at a predetermined position along the transport path within the ATM. Also as the detector is arranged to determine skew perpendicular to the direction of travel along the transport path, both sensors and light sources must be located within the transport path, thus making assembly and serviceability of the detectors difficult. For example, cables must be laid into both sides of the transport path to connect to the sensors.
In addition, changes in LED power and sensor sensitivity throughout the lifetime of a sensor have also caused problems when attempting to use optical sensors for note detection in an ATM.
A further problem with the use of optical sensors is the large variation in the opacity of notes used today. Also, some bank notes have relatively transparent windows. With prior art optical sensors these windows are seen as holes.
It is an object of the present invention to provide an optical sensor that ameliorates the aforementioned problems.
It is a further object of the present invention to produce an improved note skew detector.
It is a still further object of the present invention to provide an optical sensor that can operate accurately while utilizing a relatively inexpensive phototransistor-as opposed to an expensive photo-diode.
According to a first aspect of the present invention there is provided an optical detector adapted to measure the opacity of media, comprising a light means and a light sensor, arranged so as to have a media path there between, the light source having a drive means which is actively adjustable, during use, for detecting media of different opacities, so as to maintain a substantially constant sensor output.
Preferably, the optical sensor is a single optical sensor.
Most preferably, the light source and optical sensor are optically coupled via two distinct optical paths, which are formed in part by optical light guides.
Preferably the detector comprises a control means arranged to make determinations as to the degree of skew of a note based on the signal produced from the sensor.
Preferably, when in use, the detector is arranged such that the sensor receives light via each optical path, the output of the sensor being dependent on whether or not a note is present in either or both optical paths.
According to a second aspect of the present invention there is provided an Automated Teller Machine (ATM) having an optical detector as described above.
According to a third aspect of the present invention there is provided a method of detecting the opacity of media utilizing a detector comprising a sensor, a light source and associated drive means arranged to provide a media path therebetween, the method comprising
According to a fourth aspect of the present invention there is provided a method of detecting skew in a bank note, being transported along the transport path of a note transport mechanism, utilizing an optical detector comprising a light source and an optical sensor, which are optically coupled via light guides arranged to transmit light from the source to the sensor via two distinct optical paths, comprising detecting the actively adjustable input to the light source, required during use, for media of different opacities, so as to maintain a substantially constant sensor output an output at the sensor corresponding to both the first and second optical paths.
According to a fifth aspect of the present invention there is provided a method of detecting double picked bank notes in an ATM transport mechanism, utilizing an optical detector comprising a light source and an optical sensor, which are optically coupled via light guides arranged to transmit light from the source to the sensor via two distinct optical paths, comprising detecting the actively adjustable input to the light source, required during use, for media of different opacities, so as to maintain a substantially constant sensor output an output at the sensor corresponding to both the first and second optical paths.
An embodiment of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
a is an illustration of the output of a sensor in accordance with the present invention when a single note is detected; and
b is an illustration of the output of a sensor in accordance with the present invention when two notes are detected.
Each picked note is passed through the sensing station 12 by the feed rollers 40 and by further feed rollers 42. If a multiple note is detected by the optical system 10, in a manner to be described in more detail below, then a divert gate 44 diverts the multiple note via rollers 46 into a reject bin 48, in a manner known to a skilled person.
If a single note is detected then the note passes on to a stacking wheel 50 to be loaded on to stationary belt means 56. The stacking wheel 50 comprises a plurality of stacking plates 52 spaced apart in parallel relationship along the shaft 51 of the stacking wheel 50. When the required number of notes have been loaded on to the belt means 56, the belt means 56 transports the notes to a cash delivery slot (not shown), again in a manner known to a skilled person, which will not therefore be described further herein.
The detector 10 is positioned within the transport mechanism 14, such that the first and second wave-guides 20A, 20B lie on opposite sides of the transport path. Thus one or more bank notes being transported by the mechanism will pass through the air gap 22 between the wave-guides 20A, 20B. As the source 16 and sensor 18 are arranged at the same side of the transport path all necessary wiring can be located at the one side making assembly and repair considerably easier than in prior art detectors. Hence there is no need to feed wiring into the body of the transport mechanism, as with prior art skew and double pick detectors.
Also, changes in operation of the light sources or sensors used in such detectors during their lifetime can cause comparable changes in output from detectors leading to false readings.
When no notes are present the output of the detector is maintained at a fixed, low level, say 300 mV by applying a current of 0.12 mA to the light source within the detector. In order to maintain the same sensor output, when a note is placed in the optical path between the light source and the sensor, the current supplied to the light source must be raised, say to 8.0 mA. If a second, superposed note is located between the light source and sensor the input must be raised again, to say 30 mA, in order to maintain the same output from the sensor.
Thus the change in input from zero to one note is almost a 7-fold increase and the increase from one to two notes is more than 4-fold. Thus these increases are much more easily determined than with prior art methods. Thus measuring the input to the light source instead of the output from the sensor provides an improved detector.
With more powerful light sources these current levels would be greater and more linear, therefore, allowing the detection of extremely opaque media.
The Compensated Opacity Schematics
The Loop Reaction Speed Depends On:
The charge current delivered from the driver circuit to the charge capacitor The efficiency of the LED. Higher efficiency demands less current and thus speeds up the charge of the charge capacitor as well as it demands less change in a given situation and thus speeds up the loop reaction.
The phototransistor load resistor. A smaller load resistor (greater load) depletes the base region of the phototransistor faster and allows for a faster turn off.
The load of the charge capacitor. The smaller the two resistors R3 and R4 are the faster the charge capacitor can be depleted.
The charge capacitor. A smaller capacitor is charged and depleted faster.
The inductor. A larger inductor increases the drive current.
Closed Loop
The LED (D4) and the phototransistor (U2) are physically positioned such that U2 receives light from D4. This light path, together with the FB input of U1, creates a closed loop. The loop balances when the voltage UFB to GND is approximately 0.252 [V].
Reduction of Light
By reducing the photo current in U2 (reduction of light received by U2) the voltage UFB is reduced. This result in a current increase delivered by U1 and thus (over time) a voltage increase across C1 which in turn results in a current increase in D2, D3, D4, R4 and R3. A current increase in D4 (white LED) gives a rise in the light produced and equilibrium is restored. As this results in a current increase in R3 the output voltage increases with the light increase.
Over Voltage Protection and Maximum Current
U1 has a built-in over voltage protection circuit, which prevents the voltage across C1 from rising beyond 27.5 [V].
The maximum current that can pass through D4 is thus given by
ID4max=(UOVP−UD2+D3+D4)/(R3+R4)=(27.5−(0.7+0.7+4))/(68+270)=65 [mA]
Maximum Output Voltage
The maximum output voltage is given by the maximum current through R3.
Uo
Avoiding closed loop oscillations
If U1 is capable of charging C1 faster than U2 can change the photo current then the feed back voltage (UFB) will change too slowly and a UC1 overshoot will be the result which in turn gives excess D4 current and thus excess light.
The rise time created by U2 and its load resistor (R2) must be so much smaller than the charging of C1 that the resultant overshoot can be accepted. The actual speed with which C1 is charged by U1 depends on a set of factors which depends on the efficiency of the boost converter formed by U1/L1. Experiments are needed to obtain these data. A good result is achieved for R2=100 k, L1=5.6 uH and C1=10 uF.
LED On Time
When a more opaque media is introduced into the light path the feed-back loop increases the LED current to compensate for the measured light loss. The LED ON time depends on the speed with which the driver can increase the drive voltage (charge the charge capacitor) and thus the LED current. This in turn depends on the maximum drive current and the size of the charge capacitor.
A larger capacitor reduces the ON time at the delivered current and vice versa.
The current being delivered depends on the inductor. A larger inductor increases the current. The driver is limited to handle inductors below 27 uH.
By using over current (70 mA versus 20 mA) the LED On Time is reduced. The total light path must be so efficient that a common bill results in a LED current of 20 [mA] or less. The light path should not permanently be obstructed as this will lead to decreased lifetime.
The higher the LED efficiency is the less current is used to create light and similarly more current is available to charge the charge capacitor.
LED Off Time
The speed with which the light output will be reduced depends on the capacity of C1 given that U1 can switch off in a few microseconds.
The C1 discharge path depends on R3 and R4 assuming that the forward voltages of the diodes are reasonably constant.
τ=R*C=(68 +270)*10 u=3.38 [ms]
This is too slow. A τ of less than 0.3 [ms] is wanted.
This can be achieved by increasing max current. A higher max current will result in smaller resistors. However a higher max current stresses the LED! This also demands a faster phototransistor/resistor pair as C1 will charge faster.
Modifications may be incorporated without departing from the scope of the present invention.
Number | Date | Country | Kind |
---|---|---|---|
0319884.3 | Aug 2003 | GB | national |