WO2017027792A1

WO2017027792A1 - Method and apparatus for solenoid position measurement and control

Info

Publication number: WO2017027792A1
Application number: PCT/US2016/046737
Authority: WO
Inventors: Jaret LEDAIN; David Lynch; Jeff Tyler; David J. Trapasso
Original assignee: G.W. Lisk Company, Inc.
Priority date: 2015-08-13
Filing date: 2016-08-12
Publication date: 2017-02-16

Abstract

A method of control using the measurement of the inductance of the solenoid coil to predict the physical position of the armature within the stator coil. As the resistance of the windings will vary not only from unit to unit, but also with the temperature of the coil, it is preferable to continuously measure the resistance and compensate the calculations for the change in current. The use of solenoid position sensing through closed loop control of the sensing of the inductance and temperature produces highly accurate and repeatable control of the valve assembly.

Description

METHOD AND APPARATUS FOR SOLENOID POSITION

MEASUREMENT AND CONTROL

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION The invention pertains to the field of electrical solenoids. More particularly, the invention pertains to methods of determining and controlling the position of a variable- position or variable-force solenoid.

DESCRIPTION OF RELATED ART

Presently marketed solenoids use linear variable differential transformers (LVDT) or Hall-effect type position feedback sensors to provide position sensing. Mechanical or optical limit switches are also commonly used, which provide a binary indication of when the solenoid has reached a determined position, but cannot provide continuously variable feedback of solenoid position.

Current vendors of solenoids with position feedback include Hydraforce and Sun Hydraulics.

Bergstrom, US Patent 6,300,733, "System to Determine Solenoid Position and Flux Without Drift", shows a system for measuring and controlling solenoid armature position. The system determines inductive voltage in the drive winding of the solenoid, integrates that voltage to obtain flux, and uses the current/flux ratio to measure armature position. To overcome integration drift, the current/flux position measure is compared to an independent position measure, this comparison leading to a drift correction. This US patent is related to Bergstrom, PCT Published Application WO 2001/063626, "An Improved System to Determine Solenoid Position and Flux Without Drift". The Bergstrom application uses a separate sensing coil, not the solenoid working coil, to establish position. SUMMARY OF THE INVENTION

Using the teaching of this disclosure, hardware and software is provided both to drive and to enable the linear position of a solenoid to be determined in such a way that a closed loop servo control system can be implemented. The method of control involves the measurement of the inductance of the solenoid coil, and uses this information to predict the physical position of the armature within the stator coil. As the resistance of the windings will vary not only from unit to unit, but also with the temperature of the coil, it is preferable to continuously measure the resistance and compensate the calculations for the change in current. BRIEF DESCRIPTION OF THE DRAWING

Fig. 1 shows a block diagram of a solenoid position measurement and control system.

Fig. 2 shows a block diagram of an analog solenoid driver sensing circuit of an

embodiment of the invention.

Fig. 3 shows a block diagram of an analog solenoid driver sensing circuit of an

embodiment of the invention.

Fig. 4 shows a block diagram of a digital solenoid driver sensing circuit of an alternate embodiment of the invention.

Fig. 5 shows a flowchart of a method of setting up the method.

Figs. 6A-6B show a flowchart of the measurement and control method. Fig. 7 shows an example of a scope trace of a coil's current waveform displayed on an oscilloscope.

DETAILED DESCRIPTION OF THE INVENTION

Referring to the block diagram of figure 1 , a variable force solenoid 1 has a coil 2 which acts on an armature 3. By varying the electrical signal fed to the coil 2, the force exerted on the armature 3 against the force of a bias such as a spring 6 will vary, so that the armature 3 moves continuously between positions, for example those shown as positions 4 and 5 in figure 1.

The method of control would involve measuring the inductance of the solenoid coil 2 and using this information to predict the physical position 4-5 of the armature 3 within the solenoid coil 2. As the resistance of the windings will vary not only from unit to unit, but also with the temperature of the coil, it will be necessary to continuously measure the resistance and compensate the inductance calculations for the change in resistance.

Any change in resistance will change the apparent inductance. As the temperature of the device increases, the resistance will also increase. In order to maintain a position, a given of force is required which in turn requires a given level of current. Therefore, an increase in temperature and resistance will require an increase in duty cycle to obtain the same level of current and maintain a given position. The duty cycle's on and off times enter into the calculation of inductance. Therefore, the change in resistance will change the apparent inductance of the device. If the temperature is sensed externally, then a reverse calculation will be necessary, that is the calculation of the resistance from the temperature. The temperature itself is used for diagnostic purposes such as abnormal operation to generate a fault etc...

The preferred method of control is by Pulse Width Modulation of the power 9 supplied to the coil 2, as this technique offers the maximum efficiency of switching using semiconductors. It also has peripheral benefits in this particular application, allowing the pulse width to be asynchronously modulated, giving a 'jitter' effect which prevents static friction (stiction) from occurring in the armature 3. The basic principle of using inductive positional detection is a known effect, however in order to achieve accurate information, a considerable amount of data processing will be required to be performed by the on-board microcontroller. The application of closed loop control of the PWM signaling to the working coil can yield the higher accuracies needed for better system control of the engines they are implemented on.

The control system will be capable of operating from a range of input supply voltages 7 typically in the range of 9VDC to 48VDC +/- 10%. It would be desirable to make the control unit 'auto- sensing' in that it would operate with any coil in the range specified. However, after analysis of the coil parameters it may be found necessary to set up in the software the actual coil operating voltage. If necessary this could be achieved either by a communication link, discrete inputs or by some other processor input

No second coil is used by the system to read induction and coil temperature, as was common in the prior art. Instead, all sensing is done through electronic board interaction based on measurements of the solenoid working coil.

Positional accuracy is assumed to be maintainable in the 0% - 80% control range. Although desirable to extend control beyond this range, actual positional accuracy will be dependent on the characteristics of each coil. The characteristics of the coil are dependent on application conditions such as applied load, spring rate, and other characteristics.

The theory of operation assumes minimal impact to the positional determination from the parasitic capacitance and resistance on the closed loop controls. The initial positions of the solenoid armature, from the measurements made, will determine whether the parasitic capacitance and resistance values are needed for better position accuracy. The impacts of these parasitics are understood and can be used for each integrated solenoid coil assembly to gain higher accuracy of the armature positions. A calibration procedure can be performed during manufacturing to improve positional accuracy.

Applications include On-Highway and Off-Highway Engine control Valves and Electro-hydraulic, Electro-pneumatic control valves, actuator drives or any other solenoid- operated device.

Figures 2 and 3 are the combined diagrams of a sensing circuit of one embodiment of the invention, using analog sensing of the working coil inductance without the sensing of temperature through the working coil.

Figure 4 contains a diagram of a sensing circuit of an embodiment of the invention, that utilizes the measuring the response of the digital PWM waveform applied to the working coil to determine the inductance and the temperature. Solenoid Control Apparatus Fig. 1 shows a block diagram of a solenoid position measurement and control system.

The solenoid position measurement and control apparatus of the present disclosure will preferably have the following characteristics:

Input/Output Interface:

The apparatus of the invention will preferably have an input/output (I/O) interface 8 for receiving commands from an external controller, for example, an Engine Control Unit (ECU) in the case of solenoid used in internal combustion engines, and for exchanging data with the controller. The I/O interface may also be used to implement inputs from sensors or outputs to other systems. Preferably, this will be in the form of a bus interface 10 for input and output, for example using the CANBus, Lin Bus, I2C or via some other means of communication. Alternatively, a dedicated I O connection to the controller could be provided, using an appropriate interface and protocol as required by the controller or specific reference signals from the ECU, depending on overall system design. This could include, but is not limited to I/O such as discrete PWM, analog command and feedback signals.

Outputs:

(a) Coil Driver: The apparatus will have a driving output 9 for driving the solenoid coil 2. The driving output will preferably be Pulse Width Modulated (PWM), preferably in at least the range 10Hz to 500Hz, with an output current as appropriate to the solenoid.

(b) Position Output: The controller will also have a position output for providing position information, preferably in the form of a variable voltage 0 to 5V +/- 10% or a variable current 4-20mA +/- 10%, although it is not intended to limit the scope of the invention to those values or technologies. The position output could also be in the form of a digital output, serial or parallel, or the position output could be sent on an I/O interface, as required by the application. An internal closed loop control can be implemented into the design for higher resolution control. (c) Status Indicator: Preferably, there may also be an LED output (or on-board LED 11) to indicate operational conditions such as a good/fault condition, flashing a fault code.

Inputs: The apparatus will preferably have the following inputs. These inputs may be direct connections to the sensors, or may be in the form of reading sensor data from the I/O bus 10.

(a) Temperature input 12: A thermistor 13 or similar device may be used for measuring solenoid coil temperature, or the temperature can be determined using the solenoid working coil, for example by measuring the resistance of the coil. The temperature range to be sensed could differ by the proposed application, for example, commercial grade 0°C to 85°C, industrial grade -40°C to 100°C or military grade -55°C to a minimum of 125°C. The temperature of the solenoid may not be in or near the above values but can be thermally limited or isolated from the apparatus drivers to allow for lower operating temperatures of the apparatus.

(b) Supply Sensing: An input for sensing the voltage and/or current from the power supply hardware 7.

(c) Pressure sensors: If desired, the apparatus may have inputs 14 for sensors 15 measuring other parameters, such as pressure or change in pressure (delta P). (d) Media temperature sensor; if desired, the apparatus may have inputs 16 for sensor 17 measuring temperature of a working media.

(e) Other inputs as required: inputs (not shown) can be provided for other types of measurement sensors to provide feedback for the closed loop control.

Additional Features: The apparatus may contain on-board non- volatile memory for storing information on the operation of the solenoid, for example calibration coefficients, a record of accumulated hours of operation, a record of maximum temperature profile, or a record of maximum coil voltage and current profile. Schematics of Examples of Analog and Digital Embodiments

Figs. 2 and 3 show a block diagram of a sensing circuit of an embodiment of the invention of a superimposed signal to sense the inductance. In addition, a measurement of the coil temperature by measuring the working coil resistance of the solenoid can be added to the design. The superimposed analog signal on the digital PWM signal to the working coil will have a frequency shift enabling the sensing of the inductance. This requires sensing the frequency shift of the superimposed signal to determine the inductance and thereby the position of the armature of the solenoid.

Fig. 4 shows a block diagram of a solenoid driver circuit of an embodiment of the invention using a digital approach of measuring inductance and temperature via the working coil of the solenoid.

The power transistor controls the level of current through the solenoid's coil by application of Pulse Width Modulation (PWM). The solenoid is in series with a current sense resistor, so that the level of current present in the coil is also passing sense resistor (typically a low resistance as to minimize the insertion loss). A differential amplifier is used to increase the level of the voltage measured across current sense resistor, thus providing an analog signal that's proportional to the current flowing through the coil. Another differential amplifier is used to measure the voltage across the coil. The differential property is require to bring the higher level output voltages to voltage levels that the ADC unit can read, which are typically from zero to less than 5 volts.

When the power transistor turns off, the polarity of the voltage across the coil reverses. It is necessary to bias (Vbias) the Voltage Sense Amplifier so that negative voltages can be sensed correctly. As a large voltage will normally be present across the coil as it is turned off by the transistor, it is necessary to control the voltage by clamping across it with a flyback control diode. Since a negative voltage is across the coil, this diode will then become forward biased during turn-off and cause conduction back to the other end of the coil. This diode then protects the circuit components that would be damaged by high voltages. Since the diode is connected to the sense resistor rather than directly across the coil, the diode also acts to slow the collapse of the coil's field, this acting as an integrator, which allows the average current through the coil to be more easily determined. The level of the flyback voltage level can be adjusted if necessary by adding other components such as resistors, capacitors and/or Zener diodes.

The ADC unit reads both of the analog voltages that represent the voltage across the coil, and the current flowing through the coil and convert these voltages into digital counts.

The output voltage of the DAC unit is set by the processor to adjust the full scale range of the ADC unit to better maximize the resolution of the ADC unit at lower power levels.

The processor's software, through a control algorithm, adjusts its PWM output logic signal to the drive transistor to obtain the desired level of solenoid operation.

The processor I/O has auxiliary connections for external sensors, the ability to accept a position command in various forms such as analog, PWM and discrete signals as well as the output of any required status signals such as actual position, fault codes etc. Many systems also require a communication link to other system processors. This communications can take many different forms such as CAN and Lin Bus, I2C, SPI etc...

In some situations, the variable force solenoid is controlling the position of a valve, such as a flow control or regulating valve and a change in flow relative to the change in position of the valve relative to a seat is most sensitive nearest the closed position as the valve approaches its seat. This can result in a significant change in flow for a small change in the valve's position. Therefore accuracy in the measurement of the position of the valve in this situation is very important. It should be noted that it is at this point of operation that the coil's currents are also at their lowest levels. In order to obtain finer position resolution, finer sample resolution of the current is also required.

The digital hardware used to sense the current level is an Analog to Digital Converter (ADC) unit or device as shown in Figure 4. The ADC device has a maximum bit resolution, preferably of 12 bits or more.

Using a typical ADC device as an example, this represents a maximum resolution of one part in 2¹² = 1 part in 4096. The unit must be scaled so that it is capable of reading the maximum current level that the device is being operated at. For example, if the maximum current level of the device is 4 amps, then the 12 bit device could resolve 4 amps / 4096 = 0.92 mA.

ADC units use a reference voltage that sets the full scale range of the ADC device. If a variable reference voltage is applied to the ADC unit to set the full range scale based upon the operating point, then increased resolution may be obtained as the current level decreases.

For example, if the ADC unit uses a 4.096 Volt reference, then the ADC unit is able to resolve 1 mV of change. If the reference voltage is set to 1.024 Volts, then the ADC unit can resolve 0.25 mV of change and so on.

This may be implemented by connecting a Digital to Analog Converter (DAC) unit (shown in Figure 4) to the reference voltage of the ADC unit. A DAC unit does the opposite of an ADC unit, in that the processor can set an output voltage from a digital count. In this fashion the processor can automatically set the DAC output to use nearly the full scale of the ADC unit. In practice, using as much of the maximum range of the ADC unit as possible results the best possible measurement resolution.

The processor shown in Figure 4 is typically programmed to increase the output of the DAC unit as the ADC unit reading approaches a settable maximum reading limit, for example 95% of full scale, and decreases the DAC output as the ADC unit drops below some settable lower limit such as 90% of full scale. This keeps the ADC unit operating near its full scale range at all times. To determine the magnitude of the reading of an ADC unit that uses variable reference voltages, a mathematical scaling calculation of using both the ADC unit's output and the known applied ADC reference voltage is used.

For example if the current level is at 480 mA, and the conditioning circuit is scaled at 1 mV per mA, then the conditioning circuitry would output 480 mV to the ADC's input. If a reference voltage of 512 mV is applied from the DAC unit to the ADC unit then a 12 bit ADC unit would output:

480 mV * 4095 Counts Full Scale / 512 mV full scale = 3839 counts (from the hardware). The processor, through software, applies the known reference voltage to number of counts obtained from the ADC unit to scale the result back into mA:

Round (3839 counts * 512mV full scale / 4095 counts full scale ) =480 mV

This 480mV reading represents 480 mA current.

As can be shown, one count of change in output of the ADC at this level of operation represents 0.125 mA of current change. If the ADC unit was simply left scaled to the maximum current level of about 4 amps, then one count of change results in about 1 mA of resolution. In this example, this about an 8 to 1 improvement in the sample resolution, and a corresponding improvement in the position resolution.

Initial Set-up of System

Figure 5 shows a flowchart of the approach described herein for determination of the valve stroke involves metering of inductance, capacitance, and resistance to:

Step 51. Discover the variable characteristics of a coil intended to be controlled for

absolute position control. The variable characteristics may be derived from a starting look up table which may be a table of inductance readings, correlating inductance to absolute position. Other means of establishing the variable characteristics may be used such as polynomials, interpolation and algorithms.

Step 52. Choose an appropriate PWM frequency for a specific coil to maximize accuracy (see step 51 above).

Step 53. Measure characteristics of the coil to generate calibration coefficients to generate the position. Generation of the calibration coefficients may be determined by adjusting the look-up table for specific measured characteristics of the coil to set the initial accuracy and then adjusting look-up table value(s) based on current coil temperature, since the inductance will change relative to temperature. Temperature can be directly sensed by a sensor such as a thermistor, or can be indirectly determined through sensing coil resistance. Other coefficients may also be used. Step 54. Storing the calibration coefficients in a repository. The repository may be stored in disk and/or solid-state memory in an external control system, and/or in a readonly-memory (ROM, EPROM, EEPROM, etc.) associated with the solenoid and/or in the microprocessor control memory of the control circuitry of the solenoid. The calibration coefficients may be look-up table value(s).

Method of Position Control

Figure 6 shows a flowchart of a method of position control. Upon receipt of a command to move the armature of the solenoid to a desired position, a closed, firmware based control loop would be utilized to: Step 61 : Determine a startup PWM duty cycle for setting the desired position based on the look-up table.

Step 62: Apply drive waveform to coil.

Step 63: Measure coil waveform attributes. The attributes may be characteristics of the PWM voltage and current waveform. These characteristics could be the rise and fall of the waveform, or a frequency of the waveform applied in step 62. An example of waveform is shown in Figure 7.

Referring to Figure 7, an oscilloscope trace of a coil's current waveform is shown on a first channel and a second channel illustrates where the sample timing of the voltage and current readings occur within the system. The first waveform trace on the first channel is indicated by reference number 100. The second waveform trace on the second channel is indicated by reference number 101. Vertical lines at VI, V2 show where both voltage and current samples are obtained. Points PI, P2, are where the PWM transitions from Off- to-On (PI) and from On-to-Off (P2). Vertical line VIA and point PIA show the first point in the repeat of the cycle. The values at the points of transition times (PI, P2) may be used to calculate inductance within the time settings (On time/Off time).

Step 64: If an external temperature sensor, such as thermistor is applied, the externally measured temperature is measured (Step 70) and the method continues to step 65.

Step 64: If an external temperature sensor is not applied, the method continues to step 65. Step 65 : Calculate resistance of the coil.

Step 66: Calculate inductance based on temperature (either derived from the resistance or measured by the external temperature sensor) and resistance of the coil. Inductance may be calculated using the following equation: L = /(V( , I( , R( ) (1.1)

Where:

V(t) is the voltage across the coil as a function of time

I(t) is the current through the coil as function of time

R(t) is the resistance of the coil as a function of time

t is the time

Step 67: Determining the armature position and change in position from the inductance and working coil temperature measured in steps 64 through 66 relative to the calibration coefficients determined in step 53 of Figure 5. The parameters may be stored in non- volatile memory), such as a record of accumulated hours of operation and a record of profiles of temperature, current, voltages, and hours.

Step 68: Compare the external desired position command to the determined position in step 67 to determine position error.

If the actual position is not within the anticipated range, a raised condition occurs and a fault is recorded and stored in a repository and systems such as fail safe operation and limp home mode operation are set (step 72). The method then continues to Step 73.

Step 71 : If the actual position is within the anticipated range, increase or decrease duty cycle (current) at a controlled rate so that inductance measured = inductance set, which should directly correlate to absolute position (step 74). This will allow overcoming variable loads on the actuator and allow limits to be set in the system. The method then continues to Step 73.

Step 73 : Set PWM command and return to step 62. The method of Figures 5-6 may be implemented using the sensing circuits of Figures 2-4 as discussed above.

Accordingly, it is to be understood that the embodiments of the invention herein described are merely illustrative of the application of the principles of the invention. Reference herein to details of the illustrated embodiments is not intended to limit the scope of the claims, which themselves recite those features regarded as essential to the invention.

Claims

What is claimed is:

1. A method of controlling the position of an armature of a solenoid having a coil with a set inductance and a temperature, comprising the steps of: a) determining a start up PWM duty cycle for setting a desired position based a

determined PWM duty cycle to the initial duty cycle; b) applying a PWM signal having voltage and current waveforms to the solenoid at the determined PWM duty cycle; c) measuring coil waveform attributes; d) determining a resistance of the coil; e) calculating inductance of the coil based on coil waveform attributes and resistance of the coil; f) determining the position and change in position of the armature from the inductance of the coil and the resistance of the coil; g) adjusting the determined PWM duty cycle, so that the measured inductance equals the set inductance; h) repeating the method from step (b).

2. The method of claim 1, further comprising the step, after step (g), of storing coil

waveform attributes in non- volatile memory.

3. The method of claim 2, in which the coil waveform attributes are selected from the group consisting of accumulated values of operation, resistance, current, voltages, and time.

4. The method of claim 1, further comprising the step, after step (a) of applying ramp-up or special signaling to enter position quickly and/or overcome initial friction and momentum utilizing closed- loop control.

5. The method of claim 1, wherein the starting PWM duty cycle is based on a look up table.

6. The method of claim 5, in which the look-up table is a table of values, correlating inductance to absolute position.

7. The method of claim 1, wherein the starting PWM duty cycle is determined by a

method comprising the steps of:

(a) discovering variable characteristics of the coil;

(b) choosing an appropriate PWM frequency for the coil to maximize accuracy;

(c) adjusting the look-up table for measured characteristics of the coil, thereby setting an initial accuracy;

(d) generating calibration coefficients based on the measured characteristics of the coil to generate a position of the armature;

(e) storing the calibration coefficients in a repository.

8. The method of claim 1, in which the coil waveform attributes measured in step (c) is a rise and fall of the PWM waveform based on current and voltage measurements of the coil at transitions of the PWM waveform of off to on and on to off.

9. The method of claim 1, in which the coil waveform attributes measured in step (c) is a change in frequency of the PWM signal that is applied in step (b).

10. The method of claim 1, in which the coil temperature is measured in step (d) from a coil temperature sensor.

11. The method of claim 1 , further comprising the step of modifying the determined PWM duty cycle based on inputs from sensors measuring conditions external to the solenoid.

12. The method of claim 11, in which at least one of the sensors is a working fluid

temperature sensor and at least one of the conditions external to the solenoid is a temperature of a working fluid controlled by a valve connected to the armature of the solenoid.

13. The method of claim 11, in which at least one of the sensors is a working fluid

pressure sensor and at least one of the conditions external to the solenoid is a pressure of a working fluid controlled by a valve connected to the armature of the solenoid.

14. The method of claim 1, further comprising the steps of: comparing an actual position of the armature to a nominal range for system operation; and raising a condition if the actual position is out of the nominal range.

15. The method of claim 14, in which determining the nominal range is accomplished by: a look-up table is used to determine the nominal range.

16. The method of claim 1, in which the determined PWM duty cycle is adjusted at a controlled and rate.

17. The method of claim 1, further comprising introducing a dither at determined times to the PWM duty cycle to overcome stiction.

18. The method of claim 1, wherein the inductance is further calculating using calibration coefficients.

19. The method of claim 1, in which step (c) of measuring coil waveform attributes

comprises measuring coil current using an analog-to-digital converter having a reference voltage input for setting a full range scale.

20. The method of claim 19, in which the step further comprises applying a selected

voltage applied to the reference voltage input of the analog-to-digital converter.

21. The method of claim 20, in which the voltage applied to the reference voltage input is selected to scale the resolution of the analog-to-digital converter to a maximum current level applied to the coil.

22. The method of claim 1 , wherein determining a start up PWM duty cycle is based on derived variable characteristics consisting of a look up table, polynomials, interpolation and algorithms.