WO1993014559A1

WO1993014559A1 - Pulse width modulation power circuit

Info

Publication number: WO1993014559A1
Application number: PCT/US1993/000523
Authority: WO
Inventors: Arthur R. Mckendry; Ronald D. Ingraham; Mark R. Wheeler
Original assignee: Nartron Corporation
Priority date: 1992-01-21
Filing date: 1993-01-21
Publication date: 1993-07-22
Also published as: EP0576668A1; EP0576668A4

Abstract

A pulse width modulation speed control for a d.c. motor (12). A control input (26) corresponding to a motorist setting or a sensed condition is coupled to pulse width modulation circuit (16) that generates a variable duty cycle signal. The signal is used to adjust the speed of the motor. Auxiliary sensing of the motor's operating condition, such as current sensing, automatically shuts down the motor.

Description

PULSE WIDTH MODULATION POWER CIRCUIT

Field of the Invention

The present invention concerns a control circuit for regulating power supplied to a control device and more particularly concerns a circuit having a pulse width modulation control for energizing a direct current motor.

Background Art

There are applications in a motor vehicle for use of a pulse width modulation for powering control devices. Prior art systems have used a pulse width modulation control scheme to adjust the power delivered to a head lamp at a level below maximum during daylight operation of the motor vehicle. U.S.

Patent No. 4,841,198 to Wilhelm is an example of a such system.

A typical motor vehicle employs a variety of d.c. motors that are energized by low-voltage d.c. signals. Motor speed control systems adjust the speed with which the d.c. motor operates. As an example, fan speed controls allow the motorist to adjust the operating speed of a blower motor by setting the position of a dashboard mounted switch. Heating, ventilating and air conditioning blower motors, windshield wiper motors and fuel pumps are examples of d.c. operated, variable speed motors that are used in a motor vehicle. The heating, venting and air conditioning blower motor is one example of an "open loop" control system. The operator sets the switch to cause the motor to operate at one speed. This speed is maintained until the operator resets the switch to another setting. An example of a "closed loop" or feedback control system is the electronic fuel pump controller which uses a pressure sensor to automatically maintain fuel system pressure regardless of fuel flow. Traditional methods used to control heating, venting and air conditioning fan control includes switching various resistance values into the power supply line coupled to the motor. An alternate prior art technique is to utilize a tapped motor winding which allows switching of the motor winding to control motor speed. With increasing use of permanent magnet motors, however, the resistive switching technique has become the method of choice for use with a d.c. motor. - Certain shortcomings of this prior art technique has been observed. The resistors have typically been cooled by air flow within a discharge duct of the motor. Occasionally, a fan motor seizes during prolonged idle periods and the current within the motor becomes very high. Since the resistors have no cooling air flow after the motor seizes, the temperature of the resistors can increase and could start a fire.

An additional shortcoming with the prior art resistive switching technique was the limited range of fan speed. The system architect designed the range of settings as well as the discrete values of those settings. There is no ability of the motorist to use settings between these discrete ranges. While adequate in perfor¬ mance, these prior art fan control systems have drawbacks which are addressed b the present invention.

Fuel injection systems used on gasoline engines use an electronic fuel pum which runs at a constant speed that produces a volume of fuel flow adequate for full throttle operation. At low and moderate highway speeds, a considerable amount of fuel is returned to the fuel tank via a return line. During the round trip, the fuel picks up heat from the engine which also warms the fuel tank. Warm fuel is less dense than cold fuel and thus contains fewer molecules of hydrocarbon per pound of fuel. Engine performance degrades and the noise of the pump operating at maximum speed can be annoying to vehicle passengers.

The present invention addresses these perceived problems by controlled ener¬ gization of a d.c. activated fuel pump motor.

Disclosure of the Invention The present invention concerns a pulse width modulation control system for activating a control device. The pulse width modulation scheme allows a con¬ tinuously variable range of control of power delivered to a device and is accom¬ panied by certain other advantages and features described more completely below. Generation of the PWM signal may be simple or complex, depending on linearity required, manual or automatic (feedback) operation and, of course, cost.

The simplest form of PWM generator is a simple astable multi-vibrator with variable time constants. For manual control of a HVAC blower, for example, this could work well. Likewise, a fixed frequency oscillator driving a one-shot multi¬ vibrator is effective.

Several more circuits using diode-gated charge/discharge paths with Logic gates (inverters), OP-amps, ramp generator/comparator schemes, or even micro- processor-generated waveforms are feasible. Examples of many variations are included herein, as are controller circuits showing some peripheral features.

A circuit constructed in accordance with the invention activates a control device such as a direct current motor with an adjustable signal to control the speed of the direct current motor. The circuit includes an input for setting motor speed and a pulse width modulation circuit for generating a repetitive series of pulses corresponding to a signal from the input means. The pulse width duty cycle is adjusted to vary the speed of the direct current motor. A filter may optionally be coupled to the pulse width modulation circuit to smooth the repetitive series of pulses to provide a motor drive signal. The input means allows the speed of the motor to be set over a continuous range of values. The duty cycle of the pulse width modulation circuit responds to this change in the input signal and the output to the motor is varied to adjust motor speed. One application of the present invention is for use with a heating, venting and air conditioning blower motor. Other applications include control of windshield wiper motors, doors, windows, vents, power seats, air conditioning, and sunroof motors in both non-automotive and automotive applications. Other power control applications include control of heated seats, heated steering wheel, rear window defroster, and heated windshields.

An alternate use of the invention is in a closed loop application wherein a sensor generates the control input to the circuit and is utilized to activate a control motor.

This closed loop system has particular use in controlling the speed of a fuel pump motor. The speed of the motor is controlled to produce a required volume of fuel at a constant pressure. As the motorist opens the throttle, the pressure sensed by a sensor begins to drop. The control system responds to the reduced pressure by causing the fuel pump motor speed to increase and thus maintain the pressure. The constant pressure is necessary to keep the fuel flowing properly through the fuel injectors. Since the pump motor runs slower most of the time, it is quieter. At full throttle, when the motor is running at full speed, its noise is masked by the engine and wind noise. Various other objects, advantages and features of the invention will become better understood from a detailed description of two alternate embodiments of the invention which are described in conjunction with the accompanying drawings.

Brief Description of the Drawings Figure 1 is a schematic drawing showing a d.c. motor coupled to an energization circuit having a pulse width modulation speed control component;

Figure 2 shows a pulse width modulation control scheme for use with the Figure 1 circuit;

Figures 3A-3C are detailed schematics showing one embodiment of a motor speed control circuit;

Figure 4 is an elevation view of a module used for mounting the Figure 1 circuit;

Figure 5 is an end elevation view of the Figure 4 module; Figure 6 is a rear elevation view of the Figure 4 module; Figures 7A-7C are schematics of an alternate motor speed control circuit used for generating pulse width modulation energization signals for a direct current motor;

Figures 8-10 are views of a pressure transducer for monitoring pressure and activating a fuel pump; and Figures 11-17 are schematics of alternate pulse width modulation gener¬ ation techniques.

Best Mode for Practicing the Invention

Turning now to the drawings, Figure 1 depicts an open loop motor control circuit 10 coupled to a direct current motor 12. The circuit 10 both energizes and controls the speed of the motor 12. A battery input 14 energizes a motor speed control circuit 16 having an output 18 coupled to the motor 12. The speed control implemented by the Figure 1 circuit 10 is considered "open loop" since the motorist or operator sets the desired condition or speed and this setting is maintained until the motorist manually adjusts the setting. A variable resistor or rheostat 20 coupled to ground through a fixed resistor 22 is energized by the battery whenever an ignition switch 24 is closed. A control input

26 to the motor speed control circuit 16 is grounded when the ignition switch 24 is open. When the ignition switch 24 is closed, the input 26 rises to an adjustable level to control motor speed.

The speed control circuit 16 includes a pulse width modulation generation circuit for delivering a pulse width modulation controlled d.c. signal at the output

18 to the motor 12. As seen in Figure 1, one side of the motor 12 is grounded.

Other applications of the circuit 10 use feedback or "closed loop" control to adjust motor speed. One such application is use of the speed control circuit in conjunction with a fuel pump motor wherein, rather than having a control signal at the input 26 adjusted by the^' motorist, the signal at the input 26 is derived from a pressure sensor which automatically updates a pressure input signal to the control circuit.

The output 18 from the speed control circuit 16 in the open loop control carries a filtered d.c. signal whose amplitude is controlled to adjust the motor speed. In the closed loop application, the output 18 transmits a pulsed signal having a frequency such that the motor 12 responds to the average voltage related to the duty cycle of the pulsed signal.

Figure 2 illustrates one technique for producing a pulse width modulated signal 30 which provides an output which can be used to energize the motor 12. As seen in Figure 2, the pulse width modulated signal 30 has an adjustable or controllable duty cycle, which can be filtered by appropriate filtering circuitry to provide a d.c. signal directly related to the duty cycle of the signal 30.

The pulse width modulation of the signal is adjusted by use of a comparator circuit within the speed control circuit 16. The comparator circuit compares a reference signal 32 which changes as the input 26 to the speed control is adjusted

(as an example, to adjust the speed of a blower motor) or varies as a sensed condition changes (as in a use of the invention wherein the input 26 changes with sensed pressure). A saw tooth wave form 34 also depicted in Figure 2 increases and decreases at a constant rate and has a constant amplitude. The reference input 32 and saw tooth wave form 34 form inputs to a comparator amplifier which compares the amplitude of the two signals and generates an output based upon 5 the comparison. This output conforms generally to the wave form 30 and forms a pulse width modulated signal for controlling motor speed.

Open Loop Speed Control A detailed schematic of one embodiment of the speed control circuit 16 is ^:'lϋ depicted in Figures 3A-3C. An output 18 (Figure 3C) from the speed control circuit is coupled across a d.c. motor 12. In this embodiment of the invention, the motor 12 would typically comprise a motor used in an open looped control such as a fan blower motor. As seen in Figure 3 A, the circuit is energized by a d.c. signal such as a switched d.c. signal applied to the circuit through a motor vehicle

15 ignition switch. This d.c. signal is filtered by two filter capacitors 110, 112 and is regulated by a VOM 114 which breaks down in the event the battery voltage exceeds a predetermined value. This regulated d.c. signal passes through a diode 116 which prevents damage to the speed control circuit 16 in the event the connection from the battery is reversed. The signal also passes through a small

20 resistor 118 used to sense current and inhibit operation of the circuit in the event current to the motor 12 exceeds a certain value.

The input from the ignition is also coupled to a junction 120 coupled to a circuit portion depicted in Figure 3B. Turning to Figure 3B, this circuit has a saw tooth wave generator which includes two operational amplifiers 122, 124 that

25 produce a saw tooth wave form at an output 126. This wave form is coupled to the noninverting input of a third operational amplifier 130 configured as a com¬ parator amplifier. The inverting input to this comparator amplifier 130 is coupled to a control input 26. In the disclosed embodiment, the signal at the input 26 is generated external to the speed control circuit 16 and is based upon a motorist

30 setting of the desired blower motor speed. Turning briefly to Figure 2, the output at the junction 126 corresponds generally to the saw tooth wave form 34 and the input signal at the input 26 corresponds to the reference input 32. The reference input is continuously variable over a range of settings controlled by the motorist.

An output from the comparator 130 corresponds to the pulse width modulating signal 30 of Figure 2. This signal is used to turn on and off a switching transistor 140 (Figure 3A) which in turn controls the state of a gate drive circuit

142 of an N-channel field effect transistor 150.

One potential problem with using a pulse width modulation activation scheme is the potential for electromagnetic interference due to the steep edges of the switching wave form 30. Harmonics of the fundamental frequency extend up into the radiowave spectrum and can produce unwanted noise in the vehicle radio or interfere with other circuit components used to control vehicle operation. The operating frequency of the pulse width modulation also becomes important if is too low. Too low a frequency will require the use of oversize conductors and capacitors for filtering and thereby increases the cost of the circuit. The frequency of the speed control circuit pulse width modulation is chosen to be as low as possible consistent with economic choice of filtering components and in the disclosed embodiment is approximately 22K hertz. Filtering of the output signal from the speed control circuit eliminates EMI problems and still produces good power control of the motor. The use of an N-channel field effect transistor 150 for switching at the pulse width modulation frequency introduces complexity to the circuit 16 but is implemented due to the cost difference between N-channel switching transistors and equivalent P-channel transistors. The complexity is encountered since the gate electrode 151 of the transistor 150 must be raised 7-10 volts above the battery voltage to achieve saturated turn-on of the transistor 150. This requires the use of a boost voltage which must be generated to provide a gate voltage greater than the power supply or battery voltage.

The Figure 3A circuit includes a voltage doubling circuit 160 constructed from two diodes 161, 162 and two capacitors 163, 164. A constant frequency 50% duty cycle signal output from the operational amplifier 124 (Figure 3B) switches on and off two transistors 170, 172 which control the voltage at a junction 174 coupled to the voltage doubler circuit 160. Alternate turn-on of the transistors 170, 172 charges the capacitors 163, 164 and provides an output signal which is essentially twice the peak-to-peak value of the constant duty cycle output from the amplifier 124. This doubling of the battery signal forms a pulsed d.c. signal at an output junction 180 from the voltage doubler 160 of a magnitude sufficient to activate the FET gate.

The switching transistor or control transistor 140 which is driven by the output of the comparator (Figure 3B) turns on and off two transistors 182, 184 which drive the gate voltage of the FET 150. As the FET 150 turns on and off in response to the output from the comparator, a^" signal corresponding to the battery voltage is coupled across two junctions 186, 188. This pulse width modulated output is filtered and coupled to a positive side of the motor. The resultant output is a d.c. signal having a value corresponding to the duty cycle at the output from the comparator 130. As the control input 26 is adjusted up and down, the duty cycle and corresponding voltage across the motor is adjusted. The current passing through the resistor 118 in Figure 3A is monitored and used ro shut down the circuit in the event an overcurrent condition is sensed. The current passing through the resistor 118 provides a voltage drop in series with the switching field effect transistor 150. Should the voltage across the resistor 118 exceed a threshold, the emitter-base voltage of a transistor 190 increases to a point where the transistor turns on. This in turn raises the voltage at the base input to the switching transistor 140 and turns that transistor on. This effectively shuts down the output from the circuit and keeps the output de-activated so long as a capacitor 192 connected across the base-emitter junction of the transistor 190 remains charged. Once this capacitor 192 discharges, the circuit will begin operation again unless the overcurrent condition is again sensed.

The circuit 16 dissipates heat through a heat sink coupled to a module that houses the circuit 16. A negative temperature coefficient device 200 (Figure 3B) coupled to the heat sink changes resistance in response to changes in the temper¬ ature of the heat sink. If the heat sink temperature rises above a threshold, the voltage at a noninverting to an operational amplifier 202 causes the output of the operational amplifier to go low, pulling the inverting input to the comparator amplifier 130 low. As seen in Figure 2, when the control input signal 32 to the operational amplifier 130 is pulled low, the saw tooth wave form output at the junction 126 remains higher than the reference input and the pulse width modulating signal 30 goes low and stays low so long as the reference input is low. So long as the temperature remains above a threshold value, the motor is de-activated.

The circuit of Figure 3B also includes a low-voltage and high-voltage shut¬ down circuit for de-activating the motor in the event the battery voltage either exceeds or is below certain threshold values.

A low-voltage shut-down circuit 210 includes a zener diode 212 and switching transistor 214. The collector of the transistor 214 is coupled, to the noninverting input of the comparator 202 through a resistor 216. The transistor 214 is normally conductive. In the event the voltage at the junction 120 falls below the breakdown voltage of the zener 212, the transistor 214 will turn off and an output from the comparator 202 will go low inhibiting further circuit operation. An overvoltage sensing circuit 220 includes two diodes 222, 224 which couple the noninverting input to the comparator 202 to ground. With the diodes 222, 224 forward biased, the noninverting input is limited to about 1.4 volts. As the battery voltage at the junction 120 increases, the voltage at the inverting input (-) increases until the comparator output goes low and deactivates the circuit. At low power settings, the d.c. motor 12 may experience problems in starting due to its inertia and bearing friction. To overcome this problem, the Figure 3B circuit may be modified to include a "hard start" feature accomplished by applying full power to start the motor armature rotating followed by a drop in power to the level set at the control input 26. This is optionally accomplished by a brief application of battery voltage to the input 26.

A modification to Figure 3B would require a parallel combination of a resistor and a capacitor between the control input 26 and the switched battery voltage. As the capacitor charges at a rate dictated by the RC time constant of the parallel combination, the signal at the control input returns to the user adjusted control input signal. Closed Loop Speed Control Figures 7A-7C depict an alternate embodiment of a speed control circuit wherein a sensor input is used to control the operational speed of the direct current motor. This is to be contrasted with the Figures 3A-3C embodiment in which the operator or motorists adjust the motor speed by means of a rheostat setting. An input to the Figure 7 circuit is generated by a silicon pressure sensor 250 which continuously monitors fuel line pressure. An output from the sensor 250 is used to adjust a pulse width modulation duty cycle which controls the speed of a d.c. motor used in the motor vehicle fuel -pump. The pulse width modulation frequency produced by the Figure 7 circuit is large enough that the motor cannot respond to individual pulses but only to an average voltage that is developed. Thus, output filtering of the motor drive signal is not used in this embodiment.

The sensor 250 is a silicon strain gauge which is internally linked to a diaphragm. Flexure of the diaphragm due to pressure changes results in an imbalance of otherwise equal arms of a resistive bridge. A power supply 260 sends current to the bridge and a resultant output voltage is fed to a differential amplifier constructed from three operational amplifiers 262, 264, 266.

Two outputs 270, 272 from the sensor 250 are coupled to the noninverting inputs of two amplifiers 262, 264. The voltage across these two outputs 270, 272 is caused by flexing of the diaphragm and unequal balance of resistive arms of a within the sensor 250. The difference in magnitude between the signals at the outputs 270, 272 is output at a junction 274 from the amplifier 266.

An array 280 of resistors is coupled to the sensor 250. The specific value of these resistors is dependent upon the sensor and is finalized during fabrication of the control system. The sensor 250 is a resistive bridge powered by the system power supply 260. To reduce the voltage across the sensor 250 to a recommended level, a dropping resistance is used and this resistor is split into two equal values such that the zero output quiescent voltage is one half the power supply value. Specific values for resistors Rl, R2, R3, R4, R5 (Figure 7B) are determined by the manufacturer of the sensor 250 and are supplied by the manufacturer. For each of these resistances, a second resistor is shown in the Figure 7B depiction. These shunt resistance values are also provided by the manufacturer to achieve the desired specific resistance value. The preferred sensor 250 is a Nova PS series ^•^ sensor for measuring pressures up to 30 pounds per square inch.

The output at the junction 274 is a signal whose voltage changes with the pressure sensed by the sensor 250. As seen in Figure 7A, this signal is input to a filter constructed using an operational amplifier 282 which shapes the response to eliminate effects of hydraulic and mechanical resonances in the fuel line leading to the fuel injectors. An output from the filter 282 passes to a unity gain amplifier 284 which isolates the filter circuit from an operational amplifier configured as an integrator 286. An output 288 from the integrator 286 is fed to a comparator constructed from an operational amplifier 290. A noninverting input to the comparator 290 is coupled to a double-slope linear ramp generator 292. The frequency of this generator is determined by an RC time constant of a resistor 294 and a capacitor 296. The capacitor 296 integrates the output of a threshold detector 298 having a reference input coupled to a voltage divider 299. The output of the threshold detector 298 is a square wave pulse train of constant frequency which, when integrated, produces a dual-slope ramp signal. The ramp signal is fed to the comparator 290 and used to compare the output 288 corresponding to the sensed pressure. When the ramp signal is more positive than the sensor voltage, the com¬ parator output goes high. If the pressure increases as sensed by the pressure sensor 250, the signal applied to the inverting input of the comparator 290 decreases, increasing the portion of the wave form from the ramp generator 292 which is more positive than the signal from the sensor. This causes the output from the comparator 290 to stay high longer, causing a switching transistor 300 to conduct for longer periods.

An output field effect transistor 310 is configured as a source follower. This means that the source goes to full battery voltage when the transistor 300 turns on. In the configuration depicted in Figure 7A, the gate voltage must be between 7 to 15 volts more positive than the battery voltage. In the Figure 7A embodiment, this is accomplished through means of a half wave voltage doubling circuit 320. The square wave output of the operational amplifier 298 is fed to a switching transistor 322 to turn this transistor on and off. When the transistor 322 turns on, a capacitor 324 charges through a diode 326. When the transistor 322 turns off, a pull-up resistor 328 pulls the bottom of the capacitor 324 up to the battery voltage causing the top of the capacitor 324 to rise above battery voltage. This causes current to flow through a diode 330 into a capacitor 332. The signal on the capacitor 332 is applied to the gate of the transistor 310 by three resistors 334, 336, 338. When the switching transistor 300 turns on, the voltage on the gate is pulled to ground causing conduction of the transistor 310 to stop. Conduction resumes again when the switching transistor 300 turns off allowing the gate voltage on the FET 310 to again rise to the level of the capacitor 332.

As the transistor 310 turns on and off, the motor winding is energized with a pulsed signal of controlled duty cycle at a rate of 2 kilohertz. At this rate the motor responds as though a d.c. signal having an average voltage related to the switching duty cycle is presented across the motor windings. When motor current is interrupted, energy stored in the winding inductance is returned by current in a diode 350. This avoids generation of high voltage spikes. A zener diode 352 provides transient protection by turning on the FET 310 whenever battery voltage exceeds a safe value. The transient energy is thereby dissipated in the motor rather than damaging the transistor 310. A diode 354 prevents the gate boost voltage from being clamped to the battery when the zener 352 is forward biased. An additional zener 356 prevents the gate voltage from rising to damaging levels.

Note that for resistors R1-R5, a second identifier is present ("S" suffix). In order to exactly match the required value, the sensor manufacturer selects a specific resistor. It may be necessary in production to shunt (hence the "S") one resistor with another to obtain exactly the value needed. To provide for this on the printed circuit board layout, we show a second resistor in parallel with the first.

We have described two systems: examples of both manual ("open loop") and automatic ("closed loop") control systems. The PWM waveform generator is a triangle-wave circuit with a comparator to produce the pulse width in direct proportion to the DC voltage supplied by the sensor circuit (pump control) or potentiometer (Blower Control). The following describes alternative circuit designs for pulse width modulation.

PWM Waveform Generation Techniques Direct PWM waveform generation techniques encompass circuits such as the two-transistor astable multi-vibrator, various R-C oscillator circuits using diode- gated charge/discharge "R" values, oscillator one-shot combinations, or even microprocessors. Examples of various forms of these circuits are provided.

Figures 11-13 are variations of a similar theme; the "R" of an R-C oscillator circuit is divided into a charge and a discharge path through diode gating. By varying the resistance values, the duty cycle will vary. The repetition rate is substantially constant if a potentiometer 360 is used as the variable element as shown. Limit resistors may be included as desired to set minimum and maximum duty cycle. In Figure 12, the resistors R R₂ set a minimum and maximum duty cycle and may be omitted.

A method employing a microprocessor can directly generate PWM wave¬ forms with excellent linearity and resolution. A typical 8-bit microprocessor can easily produce a resolution of 255 discrete ON/OFF ratios. Basically, the device counts clock pulses and changes state of the output at "X" counts. A version of this technique is used in certain microprocessors having a "timer" output. This

PWM pulse train is produced independently of whatever else the microprocessor is doing once it has been programmed.

Figures 14 and 15 are an alternate variation of the R-C timer method. A free-running oscillator is used to trigger a one-shot multi-vibrator. Repetition rate is fixed by the oscillator and ON time is set by the one-shot. The resultant pulse train exhibits fixed-frequency, variable duty cycle output. If the "R" is replaced with a voltage-controlled current source, this system can be used in closed loop or other voltage-controlled applications.

Indirect means such as those disclosed in Figures 3 and 7 utilize a com- parato to sense coincidence between a reference voltage and the instantaneous value of a time-variable ramp voltage. The time interval between intersecting points on the rising and falling ramps determines the proportionate ON-time, for example. This system is capable of 100 percent to 0 percent duty cycles; if the ramp voltage never exceeds the reference voltage, the output is constant. Similar¬ ly, if the ramp always exceeds the reference voltage, the output is constant at the opposite level. The ramp waveform may be a dual ramp with similar or dissimilar rise/fall times, or even a saw-tooth with a virtually-instantaneous fall time. The waveform ramp may be linear or logarithmic (following an R-C charge/discharge curve), the primary consideration is means of detecting the ramp/reference voltage coin¬ cidence as the voltage relationship between the two changes, thus producing a variable pulse width output to control the load driver.

Figure 16 depicts the ramp generator/comparator approach which results in a means of voltage controlling the PWM waveform. A further variation uses an integrator which results in an extremely linear waveform, hence a linear vol¬ tage/pulse width transfer function for more precision applications. This modif- ication is shown in Figure 17.

Circuit Mounting Figures 4-6 depict a housing and support for the circuit of Figures 3A-3C. This circuit is mounted in close proximity to an existing d.c. motor used for circulating air through a conduit. A circuit module 400 housing circuitry depicted in Figures 3A-3C is mounted to a printed circuit board 401 which extends perpen¬ dicularly away from a plate 402 attached by connectors to the conduit. As in the prior art, the electronic circuitry of the Figures 3A-3C circuit is cooled by air flow in the conduit. A foam rubber gasket 404 insures an airtight seal between the plate 402 and the rest of the conduit. Attached to the plate is an electrical connector 410 having four pins 412, 414, 416, 418 for routing signals to and from the printed circuit board 401. These four contacts provide a switched battery connection, a ground connection, a speed control connection to the fan motor, and a speed control .input signal connection. These connections are identified in Figure 1.

Attached beneath the printed circuit board 400 is a heat sink 420 that is connected to the mounting plate 402 and dissipates heat generated during oper- ation of the Figure 3 circuit. The heat sink 420 includes a plurality of fins 422 which transmit heat from the circuit to air flowing past the circuit in the air flow conduit.

Turning now to Figures 8-10, these figures disclose a module 430 for housing the circuitry depicted in Figures 7A-7C coupled to an insert 432 placed within a fuel delivery line feeding from the gas tank to the vehicle engine. The insert 432 has an input 434 and an output 436 comprising couplings that allow easy insertion of the insert 432 into the fuel delivery line. As seen in Figure 9, the module supports a transducer such as the pressure sensor 250 in a position where the pressure of fuel flowing through the insert 432 can be monitored- on a real time basis. As the pressure is monitored, an output from the circuitry of Figure 7A-7C provides a 2 kilohertz pulsed signal for activating the fuel pump to achieve a constant pressure in the line leading to the fuel injectors regardless of the speed demands placed upon the vehicle by the motorist. The present invention has been described with a degree of particularity. It is the intent, however, that the invention include all modifications from the disclosed design falling within the spirit or scope of the appended claims.

Claims

1. A circuit for actuating a component with an adjustable signal to control the power delivered to that component comprising; a) input means for setting desired powered level; b) pulse means for generating a repetitive series of pulses corresponding to a signal from the input means wherein a pulse width duty cycle and repetition rate is adjusted to vary the power delivered to the component; and c) output means coupled to an output from the pulse means to provide a drive signal.

2. The circuit of Claim 1 additionally comprising means for deactivating the component if the components' current exceeds a threshold current.

3. The circuit of Claim 1 additionally comprising means for deactivating the component if a sensed temperature exceeds a threshold temperature.

4. The circuit of Claim 1 additionally comprising a voltage source monitoring circuit for deactivating the component if tn voltage source for driving the component exceeds a threshold voltage.

5. The circuit of Claim 4 where the voltage source monitoring circuit additionally comprises means for deactivating the component if the voltage source provides less than a minimum component drive voltage.

6. The circuit of Claim 1 additionally comprising a start-up means for adjusting the repetitive series of pulses when the component is first activated.

7. The circuit of Claim 1 additionally comprising means to filter high- current surges during pulse width modulation to allow for reduced EMI and RFI emissions and/or mechanical vibration.

8. The circuit of Claim 7 where the filter means comprises an analog filter between the PWM drive and the power drive.

9. The circuit of Claim 8 where the analog filter is supplied in part by the reactance of the powered device.

10. A circuit for activating an electric motor with an adjustable signal to control the speed of the electric motor comprising: a) input means for setting motor speed; b) pulse means for generating a repetitive series of pulses corresponding to a signal from the input means wherein a pulse width duty cycle repetition rate of fixed width pulses or the duty cycle and repetition rate is adjusted to vary the speed of the motor; and c) output means coupled to an output from the pulse means to provide a motor drive signal.

11. The circuit of Claim 10 additionally comprising means for deac¬ tivating the motor if motor current exceeds a threshold current.

12. The circuit of Claim 10 additionally comprising means for deac¬ tivating the motor if a sensed temperature exceeds a threshold temperature.

13. The circuit of Claim 10 additionally comprising a voltage source monitoring circuit for deactivating the motor if the voltage source for driving the motor exceeds a threshold voltage.

14. The circuit of Claim 13 where the voltage source monitoring circuit additionally comprises means for deactivating the motor if the voltage source provides less than a minimum motor drive voltage.

15. The circuit of Claim 10 additionally comprising a start-up means for adjusting the repetitive series of pulses when the motor is first activated.

16. The circuit of Claim 10 where the motor is an AC motor.

17. The circuit of Claim 10 where the motor is a direct current motor.

18. The circuit of Claim 17 where the direct current motor actuates a fan.

19. The circuit of Claim 18 where the fan is mounted to circulate air in a motor vehicle.

20. The circuit of Claim 18 where the fan is used as a heating, venting and air conditioning blower.

21. The circuit of Claim 17 where the direct current motor actuates a door.

22. The circuit of Claim 21 where the door is a motor vehicle door.

23. The circuit of Claim 17 where the direct current motor actuates a window.

24. The circuit of Claim 23 where the window is mounted to a motor vehicle.

25. The circuit of Claim 17 where a direct current motor actuates a vent.

26. The circuit of Claim 25 where the vent is in a motor vehicle.

27. The circuit of Claim 17 where a direct current motor actuates a windshield wiper.

28. The circuit of Claim 27 where the windshield wiper is in a motor vehicle.

29. The circuit of 17 where a direct current motor actuates a power seat.

30. The circuit of Claim 29 where the power seat is in a motor vehicle.

31. A circuit for actuating a component with a "closed loop" or feedback control system which uses a sensor to automatically control the power delivered to that component comprising: a) a sensor in a "closed loop" or feedback configuration for setting desired power level; b) - pulse means for generating a repetitive series of pulses corresponding to a signal from the input means wherein a pulse width duty cycle repetition rate of fixed width pulses, or the duty cycle and repetition rate is adjusted to vary the power delivered to the component; and c) output means coupled to an output from the pulse means to provide a drive signal.

32. . The circuit of Claim 31 where the component comprises a direct current motor.

33. The circuit of Claim 32 where the direct current motor is used to actuate a door where position and/or force and/or load feedback is used to sense an obstruction that inhibits closing the door.

34. The circuit of Claim 33 where the door is in a motor vehicle ap¬ plication.

35. The circuit of Claim 32 where the direct current motor is used to actuate a window where position and/or force and/or load feedback is used to sense an obstruction in the path of window movement.

36. The circuit of Claim 35 where the window is in a motor vehicle application.

37. The circuit of Claim 32 where the direct current motor is used to actuate a vent where position and/or force and/or load feedback is used to sense an obstruction.

38. The circuit of Claim 37 where the vent is in a motor vehicle ap¬ plication.

39. The circuit of Claim 32 where the direct current motor is used to actuate a power seat where position and/or force and/or load feedback is used to sense an obstruction.

40. The circuit of Claim 37 where the power seat is in a motor vehicle application.

41. The circuit of Claim 31 where the power device consists of an AC motor.

42. The circuit of Claim 31 where the component is a motor used in a compressor based air conditioning system.

43. The circuit of Claim 42 where the air conditioning system is in a motor vehicle application.

44. A circuit for actuating a fuel pump motor with a "closed loop" or feedback control system which uses a pressure sensor to automatically maintain a fuel system pressure regardless of fuel flow comprising: a) a pressure sensor in a "closed loop" or feedback configuration for monitoring fuel pressure; b) pulse means for generating a repetitive series of pulses corresponding to a signal from the input means wherein a pulse width duty cycle repetition rate of fixed pulses, or the duty cycle and repetition rate is adjusted to vary the speed of the motor; and c) output means coupled to the output from the pulse means to provide a fuel pump motor drive signal.