DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] The present invention provides an ultrasonic pest deterrent. Various operating conditions can be selected depending on the type of power source available (e.g. line power or battery operation), and the ultrasonic deterrent signal can be combined with a flashing light for enhanced deterrent effect. In one embodiment, the ultrasonic signal is initiated when a pest, such as a deer or dog, is detected. A selected period of time lapses before the flash is discharged, thus ultrasound draws the attention of the pest in the direction of the device, and the flash startles the pest. In a particular embodiment, the ultrasonic sound is generated with a microprocessor and the low-frequency flash oscillator AM modulates the ultrasound signal to produce sidebands of ultrasonic signal down to some frequencies within the range of normal human hearing but at much lower levels. This allows an operator to verify that the device is activated and the speakers are on.

[0021] I. An Exemplary Pest Deterrent Device

[0022] FIG. 1 is a simplified block diagram of an ultrasonic pest deterrent device 10 according to an embodiment of the present invention. A sensor 12 , such as a passive infrared (“PIR) sensor is aimed at the location where a pest is expected. Alternative sensing devices, such as Doppler radar devices, could be used. The output of the sensor 12 is provided to a microcontroller 14 , which generates the ultrasonic signal that drives the speakers 16 and triggers the strobe or flash unit 19 , if selected, which includes the bulb 18 and drive circuitry 21 . Power to the unit can be supplied by either batteries 13 or line power 15 . The power source can be automatically or manually selected with a switch 11 . Line power is stepped down using conventional power supply techniques.

[0023] Sensors typically have low output levels that are amplified with high-gain amplifiers 17 . These amplifiers can pick up signals that are not generated by the sensor, such as noise or fluctuations in the voltage supplied to the sensor or circuit, and trigger device operation. In particular, when the ultrasonic output or strobe is activated, supply voltage fluctuations, and ground current feedback can cause unintended triggering. A lockout circuit 20 prevents the device from continuously triggering if the pest dwells in the target area or if false triggering events occur, and allows the amplifiers to settle in anticipation of another signal from the sensor.

[0024] FIG. 2 is an exemplary circuit diagram for a pest deterrent device 24 according to an embodiment of the present invention. Those of skill in the art will appreciate that many alternative circuits and implementations of circuit functions are possible. The sensor 12 is a dual-element PIR sensor that produces positive and negative pulses on the output 13 when motion is detected. The PIR is decoupled from the regulated 5 V power IC 2 by a 20 k-ohm series resistor R 50 and 1000 micro-Farad filter capacitor C 41 .

[0025] This decoupling and filtering stabilizes the regulated voltage to the PIR and allows the PIR to function even in the event the batteries are producing a lowered voltage, even as low as about 3 volts in some instances. The voltage regulator IC 2 provides the unregulated input voltage 29 to the output 27 if the input voltage drops below the regulated voltage. There is only about a 0.1 V drop through the regulator, so if the battery voltage drops below about 5.1 volts, the PIR can still function. This is to maintain functions with low batteries.

[0026] The pulse train from the sensor is inputted to the amplifier IC 1 C, which inverts and amplifies the signal. A 0.1 micro-Farad capacitor C 16 in combination with a 1 M ohm resistor R 27 sets the gain and the low-frequency bandwidth limit for the detection circuit. The PIR has a relatively low-frequency response, so the amplifier bandwidth is set to about 0.25-10 Hz. This keeps undesirable higher-frequency signals, such as might arise from turning a load on or off, from triggering the circuit.

[0027] The output from the first amplifier IC 1 C drives the second stage amplifier IC 1 B, which again inverts and further amplifies the signal from the PIR and outputs to a comparator IC 1 A. Though the output of the second stage amplifier IC 1 B swings both positive and negative, only the positive excursion (which is subsequently inverted in IC 1 A) is detected for triggering purposes. This is achieved according to the programming of the microcontroller 14 , which is programmed to only activate the load (sound and flash) if an appropriate negative voltage is provided to the trigger input 25 . This frees the negative output from IC 1 B to provide a lock-out function. A quad operational amplifier, such as a model LP324M™ available from NATIONAL SEMICONDUCTOR is used to implement the comparators IC 1 A, and the amplifiers IC 1 C and IC 1 B.

[0028] When a positive excursion from the second stage amplifier IC 1 B is detected, the comparator IC 1 A swings negative to turn on the load (i.e. to activate the ultrasound and strobe, if selected). This negative swing is fed back through a 2.2 micro-Farad feedback capacitor C 20 through a series resistor R 32 to the summing junction 26 , which is the negative input of the second stage amplifier IC 1 B. This negative signal provided to the negative input pulls the output of the second stage amplifier IC 1 B further up towards saturation, providing positive feedback. The saturation condition is held until the summing junction voltage is compensated through the feedback resistor R 31 (i.e. until the feedback capacitor C 20 is sufficiently charged through R 31 to pull the second stage amplifier back from saturation).

[0029] After the feedback capacitor C 20 is fully charged, the summing junction returns to a quiescent state, which switches the output of the second comparator IC 1 A positive. This positive swing is again fed back through the feedback capacitor C 20 and the series resistor R 32 to the summing junction 26 , which causes the output of the second stage amplifier IC 1 B to swing to negative saturation because of the positive feedback, essentially disabling the triggering operation. The load is now off and the second stage amplifier IC 1 B cannot trigger the load to come on again until the feedback capacitor C 20 is discharged through the feedback resistor R 31 . Thus, the circuit acts as a timer for determining how long triggering is locked out after the load is turned off, and provides a selected, positive, load on gate, trigger signal to the microcontroller 14 .

[0030] In one embodiment, the time the load is kept on is controlled according to the microprocessor programming. In other words, upon receiving a trigger input of the proper polarity, the microcontroller turns on the load for a selected period of time, such as about three seconds. In another embodiment, the microcontroller or other circuitry activates the load as long as the trigger signal has the proper polarity and is above a threshold voltage, for example, as long at the trigger signal is a negative voltage. In one embodiment a three-second-time constant is chosen for the lock out period; however, the lock out period does not have to equal or even be close to the load-on period. It is generally desirable that the lock out period hold off the triggering circuit long enough after the load is turned off so that fluctuations in the supply voltage, ground potential, or even light from the strobe or other non-pest input to the sensor, does not create a false trigger. If the load-on period is determined by the duration of the trigger input signal (i e. R 31 -C 20 time constant), a different lock-out period could be obtained by placing a diode and resistor in parallel to each other, the parallel combination being placed in series in the feedback path. As described above, the output of the second stage amplifier can swing both positive and negative, so the feedback capacitor C 20 should be a non-polarized capacitor.

[0031] The lock out or hold off time after the load goes off insures that the power supplies and the motion sensor signal amplifiers have all settled before motion detection is turned back on. The lock out period also eliminates continuous motion detection and extends battery-operated lifetime. The 10 k-ohm series resistor R 32 , which is placed close to the summing junction 26 on the printed circuit board, isolates the summing junction from noise that might be picked up on the traces between the series resistor R 32 and the trigger input 25 . This resistor has little effect on timing because it is much smaller than the 470 k-ohm feedback resistor R 31 .

[0032] The pest deterrent device can operate in a variety of modes. A power switch SW 1 A-SW 1 D allows the user to select between a high-power mode or a low-power mode. A bayonet plug connector V 1 further selects between battery operation or line-voltage operation using an 18 V DC adaptor input from a 115 V line supply. Plugging in the male bayonet plug automatically switches out the batteries and provides a control signal to the microprocessor 14 to switch from battery-operated mode to line power mode.

[0033] The unit operates at 100% load duty cycle in high power mode, enabling the load only when motion is detected (for the 3-second on, 3-second lockout periods). In high-power battery operated mode, the six-volt batteries B 1 , B 2 are connected in series to provide twelve volts to the circuit. When high-power mode is selected and the device is running on battery power, the unit operates at 100% duty cycle, only enabling the load when motion is detected. In high power mode using the power adaptor from line power, the unit continuously emits ultrasound, with a change in the ultrasound when motion is detected. In low-power battery operation, the unit operates at 50% duty cycle, with the two batteries connected in parallel to provide 6V to the circuit. This improves battery lifetime, but produces lower ultrasonic power. The load is cycled on and off at a frequency of about 20 Hz. In low-power operation running on line power, the unit operates at 50% duty cycle emitting a continuous noise, with a change in the ultrasound when motion is detected.

[0034] A sound switch SW 3 A-SW 3 C allows selection between a partially audible (to humans) low ultrasonic mode output and a high ultrasonic mode output. In low ultrasonic mode on battery power, when motion is detected, the unit sweeps from 15 kHz to 25 kHz once per second for 3 seconds. The sweep is not a continuous sweep, but is carried out in a series of eight steps. Different sweeps or steps could be used, and the sweep could be generated by an analog circuit in alternative embodiments. Similarly, the ultrasonic signal could be produced by an external oscillator or other device that is controlled by the microcontroller or other control circuitry.

[0035] Experiments with mice were conducted to evaluate the effectiveness of a single, continuous ultrasonic tone versus the stepwise swept signal. From these observations it is believed that changes in the ultrasonic signal is a more effective deterrent than a continuous tone. The microprocessor 14 is programmed to generate the ultrasonic signal directly, which is amplified to drive the speakers. The microcontroller has sufficient bandwidth to generate ultrasonic frequencies. In one embodiment the microcontroller is a custom-masked microcontroller with 128 bytes of random access memory (“RAM) using a 4 MHz crystal-controlled oscillator for the clock. Other types of microcontrollers could be used, such as microcontrollers with integrated flash memory or programmable read-only memory (“PROM”) programmed with the operating code, or the microcontroller can interface with an external memory chip or module that has been programmed to configure the microcontroller to operate in the desired fashion. Other types of clocks or other clock frequencies compatible with the microcontroller and sufficiently high for direct generation of the highest desired ultrasonic frequency could be used if the microcontroller generates the ultrasound.

[0036] A flash select switch SW 2 A turns the flash on or off according to a control signal (in this case ground or open) supplied to the microcontroller. The flash can be enabled in any mode. A 20 kHz flash oscillator 28 charges the 2.2 micro-Farad, 250 V flash capacitor C 27 and 0.047 micro-Farad 200 V flash trigger capacitor C 26 until sufficient voltage is reached to discharge the strobe tube 30 . The strobe flashes five times in about one half second, according to the time constant of the flash trigger capacitor C 26 and 3.3 M-ohm series resistor R 40 . In low ultrasonic mode, the flash oscillator is connected to the ultrasonic output through a 2 k-ohm series resistor R 54 and 0.1 micro-Farad Farad series capacitor C 30 , and further modulates the output signal provided to the speakers by modulating the gate voltage of amplifying transistor Q 1 . The amplitude modulation produces sidebands above and below the ultrasonic frequencies being generated and some within the audible range. This audible portion of the output can be used to verify that the unit is on and operational. Furthermore, it is believed that the noise generated by the amplitude modulation makes the sound emitted by the speakers more disagreeable to pests, thus improving the effectiveness of the pest deterrent.

[0037] In low ultrasonic mode on line power, the unit continuously generates a signal switching between 17 kHz and 18 kHz at a rate of about 20 Hz. When motion is detected, the unit step-wise sweeps from 15 kHz to 25 kHz once a second for about 3 seconds. After the 3-second sweep period, the unit returns to the quiescent state of alternating between 17 kHz and 18 kHz at a rate of 20 Hz until another trigger from the sensor enables the load (i.e. another triggering event after the lock-out period).

[0038] In high ultrasonic mode running on batteries, the unit sweeps from about 25 kHz to 40 kHz once per second for three seconds. Some animals do not respond to the high end of this range, and in other embodiments, the unit steps from 24-25 kHz, or 20-30 kHz. Other ranges and step sizes may be selected. In another embodiment, using line power, the unit continuously generates a signal switching between 31 kHz and 32 kHz until motion is detected. When motion is detected, the unit sweeps from 25 kHz to 38 kHz (or other range) for three seconds, and then returns to alternating between 31 kHz and 32 kHz.

[0039] II. Operation of the Flash Circuit and Duty Cycle

[0040] When the flash is selected, the strobe flashes five times for half a second starting one second after motion is detected. Thus, as soon as motion is detected the ultrasonic signal is generated or changes from a quiescent state (e.g. alternating 17/18 kHz or 31/32 kHz), but the flash does not strobe until the sound has been on for about one second. This provides an opportunity for the pest to direct its attention at the device before the flash comes on. It is believed that this combination of sound and flashing light with an intervening period is a more effective deterrent than if the flash started concurrently with the sound because the pest would not have an opportunity to look at the device before the flash started.

[0041] If the unit is in low ultrasound mode, a further advantage is obtained from the amplitude modulation of the low ultrasound signal by the flash oscillator. Namely, when the flash oscillator is enabled, an additional change in the sound output occurs as the AM sidebands are produced. This produces a very disturbing sound from as low as about 4 kHz to over 30 kHz that occurs during, and as a function of, strobing.

[0042] The flash oscillator transistor Q 7 uses feedback from the transformer XF 1 . The high voltage output of the transformer XF 1 is rectified by the diode D 5 and charges the flash capacitor C 27 . As the flash capacitor charges, the flash trigger capacitor C 26 also charges through the resistor R 40 . The flash trigger capacitor C 26 is in series with the trigger transformer L 1 . When the voltage at the cathode K of the programmable uni-junction transistor (“PUT”) 32 drops low enough, the PUT turns on, discharging the series capacitor C 26 through the trigger transformer L 1 . This current creates the trigger voltage to fire the flash and discharge the flash capacitor C 27 through the strobe tube 30 . The flash capacitor C 27 immediately starts re-charging, and the cycle starts over with the resistor R 40 and series capacitor C 26 controlling the flash strobing rate. The flash oscillator 28 is turned on and off by the microcontroller 14 through a flash control line 31 .

[0043] When low-power operation is selected, the microcontroller 14 generates a low-frequency square wave at the duty cycle output 34 to drive a bipolar transistor Q 2 on and off. This transistor is in series with the ultrasonic output 36 of the microcontroller and the output amplifier Q 1 . The edges of the square wave are rolled off by 1 k-ohm resistors R 71 , R 72 in series with the base of Q 1 , and the 1 micro-Farad capacitor C 44 connected between a center node 38 of the resistors and ground 40 . The square wave edges are rolled off to prevent clicking in the speakers as the ultra sound is gated on and off.

[0044] III. Output Driver and Speakers

[0045] The output amplifier Q 1 is turned on and off by the ultrasonic output 36 from the microcontroller 14 . External resistors R 23 , R 24 allow changing the high and low frequencies generated by the microcontroller without having to re-program the microcontroller. The tapped inductor L 2 forms a resonant circuit with the speakers SP 1 , SP 2 , SP 3 , and SP 4 to ring them with a sine wave. The speakers are bi-morph ceramic speakers with a frequency response from about 4 kHz up to about 40 kHz and are connected in series-parallel. In other words, two sets of two parallel speakers are connected in series. In the low range, the speakers are connected across the entire center-tap inductor L 2 , while in the high range the speakers are connected to the tap 42 of the inductor L 2 for a lower inductance. The switch SW 3 A, SW 3 B is a mechanical two-position switch, but electronic switching of the inductance could be done to broaden the range of frequency sweeping using several inductors and taps, or variably-tunable inductors could be used.

[0046] By connecting the speakers in a series-parallel configuration higher output power is achieved. Each speaker has a capacitance of about 0.2 micro-Farads, which is twice the capacitance used in a conventional ultrasonic speakers that typically have a capacitance of about 0.05-0.1 micro-Farads. The series-parallel arrangement of the speakers provides a combined capacitance of 0.2 micro-Farads. This allows more current draw and more power to be drawn by the speakers, and more sonic power to be directed at the pest. Similarly, while most speakers in conventional ultrasonic pest deterrent devices are driven at about 20V peak-to-peak, producing about 90-100 dB total sonic power, the present circuit, utilizing a combined total 0.2 micro-Farad speaker load and achieving about 50 V peak-to-peak when in resonance with L 1 , produced over 120 dB. In another embodiment, the total ultrasonic output level was over about 110 dB.

[0047] The ultrasonic output of a unit fabricated according to the present invention generated over 120 dB measured 18 inches from the speakers at same ultrasonic frequencies. Not all the ultrasonic power could be focused on the point of measurement, so it is believed that the actual ultrasonic output power is well over 120 dB. Using either a 12 V battery or 18 V line source also increases the power available to the speakers over conventional 6 V or 12 V designs.

[0048] IV. An Exemplary Timeline and Methods

[0049] FIG. 3 is a representative timeline illustrating the operation of a pest deterrent device with the flash enabled according to an embodiment of the present invention. The sensor detects the presence of a pest at T 0 and generates a series of alternating positive and negative pulses. Essentially instantaneously the ultrasound is turned on, if originally off, or changed to a sweep from an alternating tone, depending on the mode of operation. After one second T 1 , the flash flashes for about one-half second until T 2 , which is at 1.5 seconds. The sweeping ultrasound remains on for an additional 1.5 seconds (three seconds total) until T 3 . The sensor is than locked out of triggering the load for a period of about 2.5 seconds, allowing the trigger sensing amplifiers to settle after the load(s) is turned off, until T 4 , which occurs at 5.5 seconds. After the lock-out period the cycle can re-start if the sensor detects another or the same pest. These times are merely exemplary.

[0050] FIG. 4 is a simplified flow chart of a pest deterrent process 400 according to an embodiment of the present invention. A sensor generates a pest detection signal (step 402 ) that initiates an ultrasonic signal (step 404 ) at a pest deterrent unit. After waiting a selected period of time (step 406 ), for example about 1 second, a light flash(es) is generated (step 408 ) at the pest deterrent unit. In a further embodiment the ultrasonic signal is amplitude modulated by the flash oscillator signal during the period the light is flashing (step 410 ). It is intended that the ultrasonic signal attracts the attention of the pest to the pest deterrent unit, and that the flash then startles the pest to drive it away.

[0051] FIG. 5 is a simplified flow chart of a pest deterrent process 500 according to another embodiment of the present invention. A pest deterrent unit produces a first ultrasonic output prior to motion being detected (step 502 ). For example, the unit could produce a continuous tone or alternating high-low tones. When motion is detected (step 504 ), the unit changes to a second ultrasonic output (step 506 ), such as a swept (frequency) ultrasonic signal, amplitude modulated ultrasonic signal, or chopped (on/off) ultrasonic signal.

[0052] FIG. 6 is a simplified flow chart of a signal triggering process 600 according to another embodiment of the present invention. When movement occurs within the range of a motion sensor, the sensor generates a trigger signal having a first polarity (i.e. positive or negative) (step 602 ). The trigger signal is provided to a saturating amplifier (step 604 ) capable of producing a positive or a negative output voltage and a load is activated (step 606 ). Positive feedback is provided to the saturating amplifier to drive the amplifier to a first saturation voltage (step 608 ). Positive feedback from the first saturation voltage is provided through an inverter and a feedback capacitor to a summing junction input of the saturating amplifier and negative feedback from the output of the saturating amplifier is provided to the summing junction through a feedback resistor (step 610 ), which discharges the feedback capacitor at a first selected rate.

[0053] After a selected period of time, the voltage at the summing junction drops below a first threshold voltage and the saturating amplifier provides a second saturation voltage having the opposite polarity from the first saturation voltage (step 612 ). The load, which is only enabled if a signal with the first polarity is detected, is de-activated (step 614 ). Positive feedback from the second saturation voltage is provided through the inverter and the feedback capacitor to the summing junction input of the saturating amplifier, and negative feedback is provided from the output of the saturation amplifier to the summing junction input through the feedback resistor to discharge the feedback capacitor at a second selected rate (step 616 ), wherein a second trigger signal will not activate the load until after the load is de-activated and the voltage at the summing junction drops below a second threshold voltage. The term “below” relates to the absolute value of the voltage(s) at the summing junction.

[0054] The rate of compensation in one embodiment is essentially equal, that is, the high and low saturation voltages have about the same magnitude and only a feedback resistor is used to couple negative feedback to the summing junction. In an alternative embodiment, the saturation voltages might not be symmetrical, and in yet other embodiments, a diode (or diode with a series resistor) can be placed in parallel between the output of the saturating amplifier and the summing junction.

[0055] Although the present invention has been described with reference to specific embodiments, modification and variation can be made without departing from the subject of the invention as defined in the following claims. For example, a sound detector, vibration detector, or radar detector might be used instead of a PIR as the pest detector. Further, the load, although described in specific embodiments as a series-parallel combination of high-capacitance ceramic speakers could be other types of speakers or ultrasonic devices. Similarly, although specific circuits with specific values of components have been described, other circuits, component values, and types of devices could be used, such as by using analog circuits to provide some of the functionality of the microprocessor in certain embodiments.