GB2296972A - Capacitive movement sensor - Google Patents

Abstract

A relative movement or proximity sensing system has at least one pair of electrodes 52 variations in the capacitance between which are detected to sense movement of an object in the vicinity of the electrodes. The electrodes are connected to a frequency selective circuit 54 such as an LC or RC resonant circuit which is in turn connected to a process circuit 56 which cooperates with the frequency selective circuit to form an oscillator. The oscillation frequency of the oscillator is compared in a comparator 56 with a reference frequency produced by a reference oscillator 55, the comparator having a counter, the count in which is representative of movement or proximity of the object.

Description

TITLE: SENSING SYSTEMS This invention relates to a sensing system for sensing relative movement between the system and an object by detecting the change of capacitance associated with electrodes of the system.

There are many circumstances when it is desirable or necessary to sense or detect the approach of an object in order to, for instance, initiate stop or prevent some action. In particular circumstances it is desirable to make such a detection in a battery powered device at very low cost and in a manner that is very insensitive to the exact physical environment in which the detector is placed. Existing detecting means, such as passive infra-red devices and capacitive switches, are not desirable under such circumstances due to their significant power consumption, cost or sensitivity to effects other than the motion to be detected. For example, existing capacitive switches operate at some predetermined level of capacitance which may occur simply due to a change in the environment in which the device is used, such as a change in humidity or condensed moisture.

This invention seeks to overcome these drawbacks of existing detection means and uses a capacitive technique that is insensitive to exact physical environment, has low power consumption, low cost and is sensitive only to the rate of change of capacitance due, for example, to a body approaching in close proximity to the sensing means.

The applications to which this invention is ideaily suited include any situation where power supply is limited (for example in small battery powered devices) and where the device needs to be in a sensitive state for long periods.

According to the invention there is provided a sensing system for sensing relative movement between the system and an object, the system comprising at least one pair of electrodes, the capacitance associated with the electrodes being varied by relative movement between the object and the electrodes, a frequency selective circuit for coupling to the electrodes and having a frequency characteristic dependence on the capacitance, a process circuit which cooperates with the frequency selective circuit to form an oscillator having an oscillation frequency dependent on the capacitance, a reference oscillator for producing an oscillation at a reference frequency, comparison means for comparing the reference frequency and the oscillation frequency, the comparison means including a counter which responds to a relationship between the reference frequency and the oscillation frequency to produce a result signal variations in which are representative of the relative movement between the object and the electrodes. Normally, the sensing system will be stationary or substantially stationary and the system will detect movement of the object with respect to the electrodes. By virtue of including a counter, a sensing system according to the invention utilises a digital technique to effect the comparison between the reference frequency and the oscillation frequency.

The reference frequency may be provided by a clock providing a stable oscillation. The clock may be part of a microcontroller which preferably includes the counter. With the result signal in the microcontroller, a large number of processing or logic possibilities are available. For example, the microcontroller may have thresholding means with fixed or variable thresholds, and means for responding in different ways to result signals exceeding the thresholds.

Alternatively, the reference frequency may be derived from changes in capacitance associated with a further pair of electrodes connected to a further frequency selection circuit feeding a further process circuit which produces a signal having the reference frequency.

In one preferred embodiment the relationship between the frequencies is a ratio of the frequencies, and in this case the counter preferably has a maximum count capacity less than the count representative of the ratio so that the counter counts in a repeating cycle beyond its maximum count capacity. This technique is possible because the sensing system detects small changes in frequency so it is only the last few bits of the accumulated count that are significant. The use of a "short" counter, ie a counter with a maximum count capacity much shorter than that theoretically needed, reduces the cost of the counter and (where applicable) the microcontroller.

In another preferred embodiment the relationship between the frequencies is a difference frequency. Preferably, this is the difference between one of the reference frequency and the oscillation frequency, or a predetermined sub-division of that frequency, and a harmonic of a sub-division of the other frequency. This technique overcomes the problem of counting at the high frequency necessary for high sensitivity.

Practical applications of a sensing system according to the invention are numerous and include the following: a) A device for placing in a shop or store and operative, on sensing the approach of a shopper, to deliver a spoken or visual message, a product demonstration or other information.

b) A game of skill involving the movement of a part of the body, or some object in contact therewith.

c) A security application where the approach of a person to a protected zone, eg a door or window, is sensed, possibly to give an audible or visual warning.

d) A clock where body movement is sensed to initiate, cease or otherwise control a function such as dial illumination or alarm.

Specific embodiments of the invention will now be described by way example with reference to the accompanying diagrams in which: Fig. 1 is a diagram of the major functions of a sensing system according to the invention, Fig. 2 is the an outline circuit for a possible embodiment of a frequency selective circuit of the system of Figure 1, Fig. 3 is an outline circuit for an alternative embodiment of a frequency selective circuit of the system of Figure 1, Fig. 4 is an outline circuit for a frequency selective circuit with connection to multiple electrodes with optional difference of sensitivity between the electrodes, Fig. 5 is a diagram of a sensing system according to the invention and having one set of sensing electrodes, Fig. 6 shows a possible arrangement of the electrodes of a sensing system where two sets of electrodes are used, Fig. 7 shows decision means of a sensing system according to the invention, Fig. 8 shows comparison means of a sensing system according to the invention, Fig. 9 shows the mechanical arrangement of an automatic message play unit, Fig. 10 shows the major functional elements within the apparatus shown in Fig. 9, Fig. 11 is a sectional view through a toy or novelty employing the sensing system to provide a skill game, Fig. 13 shows an alternative embodiment of the comparison means, Fig. 14 shows mixing means of a sensing system according to the invention, Fig. 15 shows an embodiment of the sensing system using a microcontroller, Figs. 16 and 17 show security devices using the sensing system, Fig. 18 shows a clock using the sensing system, Fig. 19 is a timing diagram of the signals within the system of Fig. 8, Fig. 20 is a timing diagram of the signals within the system of Fig. 13, and Fig.21 is a frequency spectrum diagram of the signals within Fig. 13.

Referring to Fig. 1, two frequency selective circuits 3 and 4 are connected to respective sets of electrodes 1 and 2 so that frequency dependent characteristics of 3 and 4 are altered by the capacitances between the electrodes in the sets 1 and 2.

The circuits 3, 4 are respectively connected to process circuits 5, 6 the outputs of which are compared in comparison means 7 which feeds decision means 9 providing an output 10.

Figs. 2 and 3 illustrate two preferred embodiments for circuits 3 and 4. Fig. 2 shows an inductor/capacitor resonant circuit coupled via capacitor 22 to electrode 20. Electrodes 20 and 21 are members of sets 1 or 2. Capacitance of electrode 20 to 21 and to other electrodes and to conductive bodies (such as the earth connection of mains electricity) affects the resonant frequency of the circuit. 25 is one of the process circuits 5 or 6. Capacitor 22 isolates the process circuit 25 from interfering signals (DC or low frequency) that may be coupled to electrode 20. Fig. 3 shows a alternative frequency selective circuit (ie circuit 3 or 4) with a resistor 35 and capacitors in which the time constant is affected by the capacitance of electrode 31 to electrode 32 and other bodies.Fig. 4 shows multiple electrodes connected to a frequency selective circuit similar to that of Fig. 2. Hence, the circuit of Figure 4 may replace components 1 and 3 in Figure 1 and components 2 and 4 in Figure 1.

Electrodes 41, 42 and 43 connect to the inductor 50 of the resonant circuit by combinations of capacitors (45, 46 and 48, 47 and 49 respectively). By choosing the values of these capacitors alterations may be made to the relationship between changes of capacitance of the electrodes and the effect those changes have on the resonant circuit. For example, if capacitor 48 has a low value but capacitor 46 has a high value then changes in the capacitance of electrode 42 will strongly affect the resonant frequency. Conversely, if capacitor 47 has a high value but capacitor 49 has a low value then changes of capacitance of electrode 43 will have a weak effect on the resonant frequency. Capacitors such as 47 and 46 may be formed, in part, by the capacitance of the electrodes themselves.Likewise, series capacitors 45 and 48 may be formed from by the proximity of conducting bodies rather than as separate electronic components. In this way, the sensitivity of the frequency characteristics of the frequency selective circuit may be made high to the capacitance of some electrodes but low to the capacitance of others. Further, the sensing function may be physically extended through conducting bodies placed in proximity to the principal sensing electrode(s). These additional bodies need not be connected directly or through separate electronic components, they may be coupled through the inherent capacitance between them and the principal electrodes.

Process circuits 5 and 6 connect to circuits 3 and 4 and produce outputs that respond strongly to the frequency characteristics of circuits 3 and 4. For example, if the resonant circuit of Fig. 2 is used and an amplifier with feedback is built into the respective process circuit then an oscillator can be made whose oscillation frequency is the resonant frequency of the frequency selective circuit. In this way an oscillation may be produced whose frequency varies with the capacitance of connected electrode(s). With the circuit of Fig. 3, a process circuit may be connected that includes amplification and hysteresis so that a relaxation oscillator is formed. A characteristic period of its oscillation will then change with changes in capacitance of connected electrode(s).Alternatively, the process circuit may apply a signal to the frequency selective circuit and a resulting signal is produced that varies with capacitance of connected electrode(s). For example, if an oscillation is applied, via an impedance, to the resonant circuit of Fig. 2 at a frequency close to the resonant frequency of the circuit then the amplitude and phase of the voltage across the inductor 23 will vary with the capacitance of connected electrode(s).

The sensing systems preferably have large but controllable sensing range to bodies moving in the proximity of the electrodes. The electrode sets 1 and 2 are supplied with alternating voltages so that an alternating electric field is produced in their vicinity. Conductive or dielectric bodies entering this field affect the field so that the current from the electrode(s) that creates the field is affected. The capacitance of an electrode is this current normalised for voltage and frequency so the effect of a body entering the field is to affect the capacitance of the electrode(s). The change in capacitance as a body approaches the electrode(s) is dependent on the size and nature of the body and the size, shape and orientation of the electrode(s) and any other connected conductive bodies.It is desirable to create a large change in capacitance as a target body approaches so the effect of the body can be discriminated from that of thermal or other interfering variations in the sensing system.

Electrodes are, therefore, preferably large and separated across the desired sensing volume.

However, large electrodes are inconvenient in many applications of such a sensor so it is an objective to provide a sensing system that can reliably detect small changes in capacitance of the electrode(s). The geometry of the electrodes is preferably arranged to maximise the electric field strength in the desired sensing volume resulting from application of a given potential difference across the electrodes; This maximises the capacitance change resulting from a given disturbing body. When it is not possible to position electrodes in the optimum position on either side of the desired sensing volume then the electrodes should be shaped and positioned so as each to have comparable exposure in terms of projected area of the electrode to the volume and distance of the electrode from the volume.In some circumstances it may be desirable to forego this optimum configuration for other benefits.

For example, if objects need to be placed in the sensing volume then in the optimum configuration these could cause excessive variation in the nominal capacitance between the electrodes. In these circumstances, it is preferable to devise an electrode structure (e.g. as in Fig. 9) such that objects may be placed so that the nominal capacitance is not strongly affected yet adequate change of capacitance occurs under typical desired sensing situations.

Small changes in capacitance will give only small changes in the frequency characteristics of the frequency selective circuits so means are provided to detect these preferentially and to avoid the effect of other variations due, for example, to thermal changes, local interfering electric fields, effects of changes in power supplies to the system etc. One method is illustrated in Fig. 5. Electrodes 52 are connected to a frequency selective circuit 53. Process circuit 54 supplies a signal to comparison means 56. A reference is supplied from circuit 55.

Decision means 57 responds to the result from the comparison means to give a sensing result, typically a binary signal indicating the presence or approach of a body. The comparison means is highly sensitive to differences between the output from the process circuit and that from the reference so that small changes due to the approach of a body are accurately reflected in the output from the comparison means. The comparison could be of any signal characteristics that strongly vary with electrode capacitance. Supposing the process circuit 54 produces a varying frequency signal then the reference could give a fixed frequency signal and the comparison means would then produce the difference (or "beat") frequency. The decision means would then be responsive to this difference frequency.Alternatively, a varying frequency signal from the process circuit could be applied, through the comparison means, to a frequency selective circuit in the reference circuit 55. The reference circuit would then return a signal of variable phase and/or amplitude to the comparison means, the output of the comparison means then being responsive to phase and/or amplitude of the signal from the process circuit 54 relative to that from the reference circuit 55.

If the reference has fixed characteristics then the system will be sensitive to variations in the frequency selective circuit 53 or the process circuit 54 that are not caused by the presence of bodies. Disturbances due to thermal effects, ageing, changes in supply power etc. may cause unreliable operation when large sensing range is attempted. If, however, the reference circuit is made in a similar way and from similar components to those in the circuits 53 and 54 then they may be expected to vary in similar ways to such disturbances, so that their effects will be nulled by the comparison means. Referring to Fig. 1, this nulling principle may be extended to disturbances in the locality of the electrodes. Similar electrode sets, frequency selective and process circuits (1+3+5, 2+4+6) provide signals to the comparison means 7.The comparison result will then respond to effects on one electrode set that is different to the effect on the other set. For example, Fig 6 shows an array of electrodes (60 to 63) moving in the direction of the arrow across a plane 64. electrodes 60 and 61 would form one set (e.g. 1 of Fig. 1) and 62 and 63 would form the other set (e.g. 2 of Fig. 1).

The height of the electrodes above the plane could have a very strong effect on the capacitance of the electrodes. However, since the sensing system compares the outputs from the two sets, this effect is nulled, allowing the system to detect small objects such as 65.

In Fig. 5 the decision means 57 derives the useful sensing result from the output of the comparison means 56. The characteristics of the decision means are preferably adapted according to any particular application.

The decision means may simply be sensitive to a difference value above (or below) a pre-set threshold value. In this instance the system would be sensitive to the presence or absence of a body. For high sensitivity the threshold value must be set accurately but this may be difficult because of other interfering effects such as thermal drifts in component values. A way of overcoming this is to reset the threshold value to compensate. The threshold value is provided by a means which, under command of a reset signal, is capable of adjusting the threshold level to a value suitable for the desired sensing action then holding this value for a period of time. Preferably the means would set the threshold to a value equal to the output of the comparison means offset by a predetermined amount. This offset determines the sensing sensitivity.The reset signal may be triggered manually or by automatic action. In the latter case, it could be on application of power to the sensing system (optionally after a fixed delay to allow for initial stabilisation). This will compensate for variations prior to application of power but not for subsequent variations. To compensate for these a periodic reset signal may be used. The periodic signal is preferably slow in comparison to the rate at which it is desired to sense the arrival and departure of bodies in the sensing volume.

An alternative embodiment of the decision means is illustrated in Fig. 7. Filtering means 70 processes the output of the comparison means 56. Action means 72 produces the useful sensing result from the output of the filtering means 70 optionally in comparison to a threshold value from threshold means 71.The filtering means is sensitive to the time variation of its input, for example as a conventional wave filter circuit. Preferably, it includes a characteristic sensitive to changes in input value, for example differential-with-time or high frequency pass. The time characteristic is set so that slow changes in the output of the comparison means 56 (for example due to thermal drifts in component values) produce a small output from the filtering means whereas faster changes due to moements of target bodies in the sensing volume give large outputs.A consequence of such a filtering characteristic is that small but fast noise superimposed on the input to the filtering means is accentuated. Such noise may result from pickup of external interfering fields by the sensing electrodes or by sources of noise within the sensing system or other coupled systems. It is preferable to include an additional low frequency pass function in the filtering means or elsewhere in the sensing system to reduce such noise. The time characteristic (or cutoff frequency) of such a function is preferably set between the rate of the interfering noise and the rate of the target movements so that the noise is preferentially reduced in comparison to the signal resulting from the target movements. The high pass (or derivative) and low pass characteristics could be combined into a single band pass function.

Where the decision means derives the time variation of the comparison result the system would respond to the approach and/or withdrawal of a body, rather than the mere presence.

This is advantageous in some uses where the immediate environment of the electrodes is not predictable. Using this technique, the presence of fixed bodies would be nulled after an initial stabilising period so that the approach of other bodies can be determined largely independent of the presence of other, fixed bodies.

The action means 72 derives the useful sensing result according to the requirements of the application. By adapting the action means to have different characterisitcs according to the sign of its input from the filtering means 70, the sensing system can be made to respond differently to bodies in approach from bodies in withdrawal. In the extreme, by making the means insensitive to inputs of one sign, then the system can be made to ignore either approaches or withdrawals. Conversely, with identical characteristics for inputs of either sign then such movements can be sensed identically. A simple method of achieving this within the action means is to derive the modulus of its input before further processing. In applications where there is significant continual noise superimposed on the input then a threshold may be applied. Signals below the threshold cause no output whereas signals exceeding the threshold can be arranged to cause either a progressive output or full output, whichever is appropriate to the application. Where noise is of an impulsive nature or where external influences cause spurious but infrequent fluctuations in the sensor system, the action means would preferably include a nonlinear averaging function such as a peak detector with predetermined attack and decay times. These would be set so that the unwanted signals caused only low net result but a target movement caused a large net result, consequent on its magnitude, duration and recurrence rate. In yet further applications where the sensing system is disturbed by interference from coupled systems and the interference cannot otherwise be suppressed then the action means preferably includes an inhibit function.This is triggered in synchronism with the interference by direct connection and inhibits the sensing output whenever it could be distrubed by the interference. For example, if the sensing system triggers an action that draws a heavy load from the shared power source which, in turn suffers transient noise then the sensing output could be inhibited for a fixed period while the power supply and sensing system recover from the effects of the heavy load. After processing by these means a sensing signal result is obtained which could be analogue or multivalued. Where the sensing output may be desired to be a binary signal then this would be derived using a further binary threshold means. In some applications it may be desired to have an output representative of the degree of detection representing, for example, rate of approach or withdrawal.In these circumstances the output could be a signal whose amplitude (represented digitally or as an analogue signal), frequency or other time characteristic were adapted to vary with the sensed value.

If the environment local to the electrode(s) can vary greatly then the corresponding range of capacitance of the electrodes may be wider than that conveniently covered by the process circuits or comparison means. A means to overcome this is to use a feedback signal (e.g. 11 and/or 12 in Fig. 1) from the comparison means to correct for environmental changes to the sensing system. Such a signal would be filtered so that it varied only slowly and faster changes due to the approach or withdrawal of bodies would still be detected. Typically the signal(s) would be used to adjust the capacitance of variable capacitance diode(s) built into the frequency selective circuits so as to counteract the effect of capacitance changes due to changes in environment local to the electrode(s).

The decision means can be made sensitive to combinations of effects. For example, two signal paths according to Fig. 7 could be provided within the decision means. The first path includes a derivative-with-time function and second a high pass function. Each path would provide a binary thresholded result which would be combined in logic to give a single sensing output. This would only respond to stimulus that caused a fast change of capacitance at the same time as a large change. The first path detects fast changes while the second path detects large changes. Other detection combinations would be possible depending on the filtering, comparison and logical combination employed.

The above techniques may be used in embodiments suited to a wide range of applications.

Fig. 5 shows a preferred embodiment of the sensing system. Prefereably frequency selective circuit 53 and process circuit 54 act as an oscillator, as does refernce circuit 55. Comparison means 56 then is comparing the frequencies of these two sources.

Fig. 8 is an example of an embodiment of the comparison means. Inputs 81 and 84 are from the process circuit 54 and reference circuit 55, in either order. Frequency division means 85 derives a regular periodic signal to operate control means 86 which sequences the operation of elements 82, 83 and 87. Oscillations of signal 81 increment counting means 83. At a regular time interval, latch means 87 samples the state of counting means 83 whose count state is then reset. Optionally, gating means 82 can inhibit incrementing of the counting means so that its state may be sampled reliably by the latch means. Also optionally, but not shown, a frequency division means may be interposed between incoming signal 81 and counting means 83. The comparison result is signal 88. Optionally, resetting of the counting means may be omitted.Subsequent processing of the comparison result (for example by calculation of the differences of sucessive samples) corrects for the varying count state of 83 at the start of each counting period. In operation the suceesive samples in count result 88 reflects the frequency of incoming signal 81 during each regular period defined by the frequency division of incoming signal 84. In this way the count result reflects a comparison of the two frequencies. In typical applications of the sensing system, when a target movement occurs, the frequency deviation of the oscillator coupled to the sensing electrodes is a very small fraction of the nominal oscillation frequency. Consequently, the corresponding change in comparison result 88 is small. Within the regular timing period, the counting means 83 accumulates a count from the incoming signal.The counting means can optionally be arranged to have a maximum count capacity smaller than the total number of oscillations presented to it in the timing period. In this case the counting means continues counting beyond the maximum count by reverting to a consistent reset value then incrementing again towards the maximum count and repeating the cycle. Since the system need be sensitive to small changes in the count, the counting means need only have a count capacity corresponding to the maximum frequency deviation that it is desired to distinguish.

Fig. 13 is an alternative example of an embodiment of the comparison means. Signal inputs 1301 and 1313 come from oscillators as described above. Frequency division means 1303, 1304 and (optional) 1302 derive lower frequency signals from these signal inputs. Mixing means 1305 derives a heterodyne frequency difference signal from its inputs from 1302 and 1303. The frequency difference signal is counted by counting means 1310 whose state is sampled by latch means 1311 under control of control means 1308 with optional gating means 1309 in a similar way to that described in relation to Fig. 8. The periodic signal to operate this counting system is derived from frequency division means 1304 or optionally from some other source of nominally constant period.Small frequency deviations between the inputs to the mixing means are preserved in the difference frequency output. consequently the sensor detection performance is preserved. Preferably the signal from 1303 to the mixing means 1305 should have a low frequency but with a high harmonic content. Mixing means 1305 is adapted to produce the lowest difference frequency between the signal from 1302 and the nearest harmonic of the signal from 1303. The advantage of this is that circuit elements 1303, 1304, 1305, 1309 etc. may all operate at a low frequency while the frequency of incoming signal 1301 is high and the sensing performance is maintained. This is advantageous to reduce the cost and power consumption of the system.When using a difference frequency derived from a harmonic of a low frequency signal in this way it is important that the difference frequency does not become too high or too low. In general the higher frequency signal to mixing means 1305 lies between successive harmonics of the low frequency signal input. If the high frequency signal moves from just above the frequency of one harmonic then the difference frequency increases from a low frequency to a maximum of half of the low frequency input (when the high frequency signal is midway between harmonics). Further increase in frequency of the high frequency signal cause a reduction of the difference frequency. If a target movement occurs which moves the signal frequencies so that this halfway threshold is crossed then the relative frequency deviation of the oscillators is not preserved in the difference frequency and sensing may not be reliable. A converse situation occurs when the high frequency signal approaches then crosses a harmonic of the low frequency signal. It is preferable to avoid this. This can be done by examination of the difference signal in comparison to the low frequency signal. When the difference frequency becomes too high or too low it is necessary to deliberately change one or other of the oscillator frequencies or a frequency division ratio so that an acceptable difference frequency is restored.

An embodiment of mixing means 1305 is illustrated in Fig. 14. Signal 141 is the higher frequency signal from optional frequency division means 1302 or direct from an oscillator and is a binary signal. Signal 142 is a low frequency binary signal from frequency division means 1302. Latch 143 (for example, a D-type flip flop) samples the incoming signal 141 according to the signal 142. Second latch 144 samples the output of latch 143, so it produces a signal equivalent to the output of 143 but delayed by one cycle of 142. Gate 145 compares the outputs of the two latches and produces a signal 146 which is active when the outputs are different. Signal 146 will advance a connected counter (as in 1310) whenever this occurs. In operation, latch 143 performs a heterodyne mixing and storage function, its output can be the difference frequency between signal 141 and the nearest harmonic of 142.For this to be true it is a necessary condition that the mark space ratio of signal 141 is close enough to 1:1.

If Tp is the period of the cycle of signal 141 and Tn is the time of the shortest state of the cycle (be it mark or space), and Fs is the frequency of 142 and Fd is the difference frequency between signal 141 and the nearest harmonic of signal 142 then it is necessary that (Fd/Fs) < (Tn/Tp). A digital frequency divider between the oscillator and latch 143 is one way of assuring even mark space ratio. Another is by design of the oscillator circuit itself or by an intervening voltage comparator automatically biassed to give a mark space ratio close to 1:1.

Subsystems as illustrated in Figs. 8, 13 or 14 are implemented using digital techniques. They may be assembled from logic elements such as gates, latches and counters or they may be implemented using programmable logic elements such as a microcontroller. In the latter case, such functions as frequency division may be implemented as number of processor clock cycles needed to execute a software loop. Within such a loop an input port may be polled.

This would be equivalent to the action of latch 143 which samples an incoming signal at a low rate. In this instance the processor clock source would be either of the oscillators of the sensing system describe in connection with Fig. 8. A limitation of using low cost, low power consumption programmable parts is that the execution rate of instructions is generally low in comparison to the frequencies desirable for high sensing sensitivity. Using combinations of the techniques described above (optionally heterodyne mixing and optionally counters with low maximum count) a sensing system may be made using a low speed microcontroller with only one additional oscillator in addition to the processor clock oscillator. This is illustrated in Fig. 15 151 are a set of sensing electrodes coupled to, for example, an LC resonant circuit 152.This forms an oscillator with circuit 153 (e.g. a conventioanl LC transistor oscillator such as in the colpitts configuration). The oscillation is connected to an input port of the microcontroller 154. Frequency selective circuit 155 works with the oscillator integrated with the microcontroller. For short term stability necessary for good sensing performance it is preferable that it is a quartz crystal or ceramic resonator. For less demanding applications, RC oscillator network may be acceptable. The microcontroller need not have any internal counters dedicated to counting external events provided the heterodyne mixing technique is used and counting is done in a software loop. Once a comparison result is obtained (as 88 in Fig. 8) that is in a digital form within the microcontroller it is a simple matter to implement the decision means within the microcontroller as well.Thresholding is done by arithmetic comparison. Filtering functions are implemented using conventional digital filtering techniques. In many applications the required filtering functions are not extensive.

A simple rate of movement detector is implemented by taking the difference of the present comparison result with the comparison result as stored some set time previously. If the difference crosses a threshold value then an active sense result is generated. Further, more complex functions can easily be envisaged. For detection of typical hand movements it is preferable that the sensing system works at a constant rate of period in the range of about 50ms to about 500ms. Within this period, counts relating to the incoming frequency are accumulated and the decision function operates on the total counts produced at this rate.

The microcontroller may also integrate other functions necessary for the application but not directly related to the sensing function so that a complete product may require few, if any, additional component parts. For example, the decision function can be adapted as described above to produce a variable pulse rate related to, for example, the rate of approach or withdrawal of a target body. This may be implemented within the microcontroller or otherwise. The pulse rate may then be used to operate a sounder or light to warn of the sensing function and to indicate the degree of movement that has been detected. Alternative to using a microcontroller, a sensing system may easily be created using an integrated digital circuit where circuit elements are dedicated to each of the subfunctions necessary for the sensing system.

Preferably, an embodiment should not cause interference to other electronic apparatus nor should it be sensitive to interfering fields. A nominal oscillation frequency in the range from about 50kHz to 1MHz is suitable to achieve this but other frequencies are possible. Within this approximate range, strong interfering fields are rare in typical environments. Also, oscillation voltages applied to the electrodes (typically in the range 0.5-10V) do not radiate substantial energy provided the electrodes are not large, for example less than 1 metre in any dimension.

Other apparatus may be combined with the sensor to yield practically useful results. For example, such a sensor could be built into an electronic toy so that it activated when approached or handled by a child. A timing device or manual operation would be used to return the product to the dormant state. Low power consumption is important as the toy preferably may be left dormant for months without significant battery consumption.

Other products may incorporate such a sensor to perform sensing functions intrinsic to the operation of the product as well as optionally activating the product from its dormant state.

A toy for a child could, for instance, use the sensor to activate it, the product making responses to the child (by sound, lights, pictures etc.) whenever the sensor is reactivated. If activations cease for a pre-set period then different response(s) may be made.

The sensor also enables certain device in their entirety. Fig. 9 shows a first example. Box 90 (shown dotted for clarity) encloses two sensor electrodes 91, 92 and an electronic apparatus 93. Fig. 10 is the block diagram of the apparatus. Electrodes 91 and 92 connect to sensor circuit 103, an embodiment of the sensor described previously. The sensor is arranged to give a positive result when a body approaches the electrodes. Control circuit 104 is activated by the positive result from the sensor and causes message store unit 105 to replay a pre-recorded message. The message store may use conventional magnetic media or electronic means for storage. The message is played through loudspeaker 107 via amplifier 106.

Such a device may be used in retail outlets or other public places to give information. A piece of merchandise, a leaflet dispenser, or other items may be placed on or near the device; anyone who shows specific interest in the items by touching or approaching closely will trigger the device and cause the message to be played. The message is not limited to solely speech since music or even visual replay is possible. Further, it is possible to trigger an animated display or even to activate the merchandise on display.

Fig. 11 shows a sectional view through a toy or novelty that relies on the sensor. The mechanism of the toy is enclosed between the lid 110 and the base 111. The lid may be pushed down relative to the base, compressing spring 113 via plunger 112. A latch (not shown) locks the spring in a compressed state until solenoid 114 is activated whereupon its plunger 125 releases the latch and the spring pushes up the lid at speed. Beneath the lid 110 are sensor electrodes 115-119. Some of these connect to sensor electronics which activates the solenoid in response to increases in capacitance of the connected electrodes. Other electrodes of the set 115-119 may couple to the connected electrodes through the inherent capacitance between them. As part of a game, playing pieces are placed or thrown onto the lid. They may land in different orientations and in different places on the lid.An objective could be to remove as many pieces as possible without triggering the sensor. Should the sensor be triggered then the remaining pieces are thrown in the air and the game is over.

The decision means in the sensor is arranged to be principally sensitive to rate of increase of capacitance of the electrodes. It may be further enhanced by having a variable threshold of sensitivity; the threshold may adjusted manually (e.g. as a switch between "easy" or "hard") or automatically, with time (e.g. so it becomes increasingly difficult to remove without triggering as the game progresses) or some other variable.

Optionally, the pieces may be made of different sizes or materials so that some are more difficult to remove without triggering than others, so commanding a higher score. Pieces that are low in height or are difficult to grip will be more difficult to remove since the player's fingers will have to approach the electrodes more closely. As this happens, the change in capacitance of the electrodes increases for a given amount of movement. If a piece is made of conductive material then it is also difficult to remove without triggering since the piece will act to extend the electric field from the electrodes beneath the lid.

Areas of the playing surface on the lid may be made to have different sensitivities, again so that they command different scores for pieces removed from them. This may be achieved by altering the electrical coupling to respective electrodes in the way illustrated in Fig. 4.

Alternatively, the spacing between the respective electrode and the top surface of the lid may be altered. Electrode 116 is spaced further than is electrode 118 so it would be more difficult to remove a piece (without triggering) from above 118 than from above 116. This alteration of spacing may be through multiple electrodes spaced at the necessary distances from the top of the playing surface or may be done with a single, shaped or plane electrode in combination with a shaped or plane lid.

In the game described, it is the objective to remove pieces without triggering the sensor. The sensor responds to an increase in electrode capacitance as an approach to remove a piece.

If the player can approach sufficiently slowly to avoid triggering then he may remove the piece as quickly as he likes since this causes a decrease in capacitance so would not cause triggering. The decrease in capacitance may be used to add an automatic scoring feature. If the electronic system detects the decrease in capacitance then measures the peak rate of reduction of capacitance and the actual decrease of capacitance during the detection of removal then this can be used to accumulate a score. A fast removal of a "difficult" piece (i.e. one whose characteristics and location would cause a large increase in capacitance on approaching it) would give a large reduction of capacitance and large rate of reduction, so would score highly.

Fig. 16 illustrates a security device using the sensing system. Enclosure 162 contains the sensor system, power supply and a signalling system. Wire 163 is one of the sensor electrodes. The enclosure of 162 or circuits within it act as a second sensing electrode. The device is laid on, against or close to an item to be protected. Disturbance of the item is registered by the device via small changes in the capacitance of the wire 163. When a disturbance of sufficient magnitude is detected the signalling system is triggered. Optionally the signal may be a sound, light contained in the device. Optionally the signal may trigger a remote alarm system by direct connection, light, sound or radio signal. A further feature may be a pulsing sound or light within the device that starts and increases in frequency with increasing disturbance to act as a warning.This may be combined with a second alarm signal which is triggered at a higher threshold of disturbance. Fig. 17 illustrates an alternative arrangement of the security device where it is used to protect an access such as a window or doorway. The security device similar to that described above is shown as 172. This connects to two wires 173 and 174 which act as sensor electrodes. These are suspended substantially at either side of the aperture so that bodies moving through the aperture disturb the electrode capacitance and trigger the alarm.

Fig. 18 illustrates the application of the sensing system to an electronic clock. Fig. 18a is an overall view, Fig 18b is a sectional view showing the internal arrangement. The clock 181 is optionally provided with an alarm (not shown) and optionally a light 182 to illuminate the face when in darkness. The alarm could be similar to those conventionally fitted to small clocks. Once sounding it may be cancelled permanently or optionally may be cancelled for a period ('snooze"). It is preferable that such clocks are battery powered and have low power consumption. It is impractical to maintain illumination alight for long periods so a desired feature is to be able conveniently trigger the illumination. Within the clock the mechanism 184 operates the hands and the alarm function as in a conventional electronic clock.A sensor system as described above is built in as 185 and works in conjunction with sensor electrode 186. This may be any convenient conductive material such as foil or card that has been vacuum coated with a thin layer of aluminium. The sensor electronics and other connected circuits (such as the electronics of the clock mechanism) act as another sensing electrode.

The sensing system is adapted to detect hand movements around the clock. This triggers the face illumination light 182 (which is connected to the sensor system) for a preset period after which it extinguishes automatically under control of the sensor system. If the alarm is sounding when a trigger is detected then the sensor cancels the alarm via a direct connection to the clock mechanism electronics. Sensor system and clock electronics share batteries (not shown) as power source. Preferably the sensor system is built using an integrated digital subsystem (for example a microcontroller) as described above. This subsystem preferably incorporates the drive functions for the motor of the clock mechanism and the alarm function.

Referring to Figs. 8 and 19. Signal 191 represents signal 84, signal 193 represents 81. A division ratio of K is illustrated in Fig. 19. this is the division ratio of frequency divider 85.

Signal 192 is the regular periodic pulse signal from 85. 194 represents the count state of the counter 83. When the output of the frequency divider is active, the count in the counter is sampled by the latch means 87 then the counter is reset to a predetermined value (represented by R). With each pulse of 191, the counter advances by one count until the next pulse occurs in 192. 195 represents 88, the state of the latch 87, and is the output of the comparison means.

Referring to Fig. 20. 201 represents signal 1313, the input to the frequency divider 1303.

The output for this divider to the mixer 1305 is represented by 202. Signal 2-3 represents the other input to the mixer 1305. A representation of one example of the mixer is 204.

This signal is the result of signal 203 being sampled on each rising edge of 202. It has a frequency of the difference between the frequency of a harmonic of 202 and the frequency of 203. 205 represents the output of the mixer illustrated in Fig. 14. This also relies on sampling signal 203 by signal 202. A pulse is produced whenever the state of 203, as sampled by latch 143, changes.

Fig. 21 illustrates the operation of the heterodyne comparison means in the frequency domain. Narrow band signals 211 and 212 represent the frequencies of oscillation of the two process circuits. Signal 212 is applied to the frequency divider 1303 and its output, with high harmonic content, is represented by 213. The harmonic interval is an integer submultiple of the frequency of 212. Signals 211 and 213 are applied to the mixing means.

The output of the mixing means has a frequency, df, equal to the difference in frequency between that of 211 and the nearest harmonic of 213. The output of the mixer may have harmonics which are not shown.

Claims (7)

Claims

1. A sensing system for sensing relative movement between the system and an object, the system comprising at least one pair of electrodes, the capacitance associated with the electrodes being varied by relative movement between the object and the electrodes, a frequency selective circuit for coupling to the electrodes and having a frequency characteristic dependent on the capacitance, a process circuit which cooperates with the frequency selective circuit to form an oscillator having an oscillation frequency dependent on the capacitance, a reference oscillator for producing an oscillation at a reference frequency, comparison means for comparing the reference frequency and the oscillation frequency, the comparison means including a counter which responds to a relationship between the reference frequency and the oscillation frequency to produce a result signal variations in which are representative of the relative movement between the system and the electrodes.

2. A sensing system according to claim 1, wherein the relationship between the frequencies is a ratio between the frequencies.

3. A sensing system according to claim 2, wherein the counter has a maximum count capacity less than the count representative of the ratio so that the counter counts in a repeating cycle beyond its maximum count capacity.

4. A sensing system according to claim 1, wherein the relationship between the frequencies is a difference frequency.

5. A sensing system according to claim 4, wherein the difference frequency is the difference between one of the reference frequency and the oscillation frequency, or a pre-determined sub-division of that frequency, and a harmonic of a sub-division of the other frequency.

6. A sensing system according to any of the preceding claims and including decision means responsive to the comparison means, the decision means being responsive to the variations in time of the result signal to produce a signal representative of the movement.

7. A sensing system construed and arranged substantially as herein particularly described with reference to the accompanying drawings.