GB2435112A

GB2435112A - Control System for a Piezo-electric Actuator

Info

Publication number: GB2435112A
Application number: GB0702192A
Authority: GB
Inventors: Paul Weaver; Ferdi Graser
Original assignee: PBT IP Ltd
Current assignee: PBT IP Ltd
Priority date: 2006-02-03
Filing date: 2007-02-05
Publication date: 2007-08-15
Also published as: GB0702192D0; WO2007088390A1; GB0602255D0

Abstract

A method and control system for reducing power consumption due to charge leakage during operation of a piezo-ceramic actuator 1. The system compensates for ceramic hysteresis without the use of a temperature sensor by supplying a charge to the actuator, measuring the voltage across the same by a voltage measuring unit and then varying the supply of charge accordingly. The actuator charge may be measured by timing the discharge through a current source. The charge may be adjusted by changing the duration of the charge pulses or the time period between them. The effects of temperature and humidity on the actuator's insulation can therefore be overcome.

Description

Control System for Piezo-electric Actuator

Field of the 1nventio

The present invention relates to piezo-electric actuators and more particularly to a control system for such an actuator.

Background of the Invention

A typical simple drive circuit is shown schematically in Figure 1.

A control A is used to charge the piezo to the required voltage. This is typically 200-250V, but any voltage is possible depending on the actuator construction. The actuator voltage may be obtained from a low voltage source such as a battery or microcontroller output, by means of a switched inductor or other means of achieving the required voltage level. It is assumed in the following that the charge circuit includes components to prevent reverse charge leakage through the charging circuit. Control B is used to discharge the actuator when its operation is no longer required. This can be implemented electronically by means of a MOSFET or similar.

Piezo-ceramic actuators behave electrically approximately as a capacitor. A mechanical deflection is produced by delivering charge to the actuator. For example, a typical actuator may have a capacitance of 7OnF at room temperature, with a working voltage in the range 200-250V.This equates to a stored charge of approximately I 5iC. As with a capacitor, these devices have a finite insulation resistance which leads to charge leakage. To maintain the actuator with a constant deflection, it is necessary to provide additional charge to compensate for this leakage. In the simple circuit described above, the initial charge is provided by a period of continuous charging., This is followed by brief pulses at a fixed frequency to compensate for charge leakage whilst maintaining low power consumption.

This is shown schematically in Figure 2.

The duty cycle of the pulses is fixed, and designed to accommodate a reasonable variation in insulation resistance without excessive power consumption.

It is known that when an actuator is exposed to long periods of high humidity the insulation resistance decreases. In the fixed duty cycle scheme described above, if the insulation resistance drops below a certain level, the actuator will lose deflection and operation of the device will be impaired. This could be remedied by increasing the pulse duty cycle, but this would result in higher power consumption even in low leakage conditions.

In general, many devices will not be exposed to permanently high humidity. in outdoor locations the humidity and temperature varies during the day, and daily averages vary throughout the year. A system that delivers power only when required would provide improved reliability when the humidity is high, and lower power consumption in low humidity conditions.

Additionally, variations in temperature cause changes in the actuator properties including temperature dependency of the following properties * Hysteresis. This causes a loss of deflection at low temperature for unipolar drive.

* Capacitance -increases with increasing temperature.

* Coercive field -decreases with increasing temperature.

The simplest way of controlling the actuator is by means of a unipolar drive system i.e. positive only voltage (aligned with the poling direction) applied to the ceramic. However, actuator performance can be considerably improved particularly at low temperature by applying a negative voltage (P.M. Weaver, A. Ashwell, Y. Zheng, S.C. Powell, "Extended temperature range piezo actuator system with very large movement" SPIE Smart Structure and Materials conference: Active Materials: Behaviour and Mechanics, San Diego, March 2003). Normally this technique requires a temperature sensor to permit application of the optimum reverse voltage without exceeding the coercive field (this would re-polarise the ceramic in the opposite sense leading to a loss in performance). it is also possible to measure the capacitance of the ceramic, and use this to infer the temperature. However, the actuator does not represent a precision capacitance and this technique will suffer from productions spread in performance, poor correlation between capacitance and temperature and non-linear time dependent capacitance.

The present invention proposes to use the discharge of the actuator during the transition from ON to OFF state to gauge the reverse charge that needs to be applied. Because an intermediate temperature measurement is not required, errors due to variations in the actuator capacitance and the correlation between capacitance and temperature are eliminated. This is achieved with very simple low cost circuit components, many of which are already implemented for the power scaling regime described above DE1995862 discloses a means of controlled charging of an actuator from an initial uncharged state. In this document, control is achieved by adjusting inductor current by controlling the timing of an inductor switch. This document however concentrates on optimization of actuator dynamics and does not disclose the voltage measurement process.

US publication 2004169436 discloses that voltage feedback is achieved by use of a buffer circuit with relatively high impedance and does not concern itself with the voltage measuring process.

GB 2334164 also deals with control of initial charging of an actuator but does not deal with measuring the voltage, the method of maintaining a constant voltage or with reverse charging to compensate actuator hysteresis.

US 5130598 uses current feedback to approximate a constant current source, and resistive dividers to provide feedback. However, this document is not concerned with compensation for leakage currents or the application of a negative voltage for temperature compensation.

Summary of the invention

Accordingly the present invention discloses a method and control system for compensating for the effects of humidity and temperature during operation of a piezo-ceramic actuator with low cost and low power consumption. The system comprises the steps of providing an initial charge pulse from a charge circuit to the actuator; measuring the obtained voltage across the actuator without consuming excessive power in the measuring circuit; calculating the charge required to be supplied from the charge circuit to the actuator to reach the voltage measured previously; adjusting the charge supplied to the actuator according to the charge calculated;providing charge pulse to the actuator based on the adjusted charge; repeating steps the previous steps until end of operation of the actuator; discharging the actuator and applying a negative pulse at the end of operation which compensates for temperature dependent ceramic hysteresis.

Brief Description of the Drawings

Figure 1 defines a simple piezo actuator drive circuit existing in the art.

Figure 2 shows a simple timing diagram for an actuator Figure 3 is a diagram of a voltage measuring unit as per the present invention Figure 4 is a diagram of a sampling switch for use in the present invention Figure 5 is a diagram of the current source for the present invention Figure 6 is a graphical representation of the transfer function in respect of a mE sample capacitor Figure 7 is a graphical representation of voltage regulation across a 0.1 nF sample capacitor and having 250 Hz variable duty cycle.

Figure 8 is a graphical representation of current draw across a 0.lnF sample capacitor, having 250 Hz variable duty cycle and with no sleep' condition Figure 9 is a schematic diagram of the power scaling system according to the present invention Figure 10 is a schematic diagram of the reverse charge circuit according to the present invention

Detailed Description of the Invention

in order that the present invention be more readily understood, an embodiment thereof will now be described by way of examples with reference to Figs 3-10 of the accompanying drawings.

The preferred embodiment of the present invention proposes a system to deliver the above benefits in a robust and cost effective mariner. It works by measuring the voltage on the actuator and adjusting the power delivery to maintain this voltage within acceptable limits. It has the following features: * Very low power monitoring circuit does not add significantly to power consumption.

* Low impedance voltage measurement for rapid measurement and high noise immunity.

* Maintains satisfactory voltage stability without calibration or the use of exotic components.

* Maintains satisfactory voltage stability with variations in temperature and actuator capacitance.

* Pulse duration output for simple interface with microprocessor.

* Variable initial charge dose accommodates variations in actuator capacitance * Control of steady state position of the actuator.

* Accommodation of changes in low level leakage currents whilst maintaining low power consumption.

This achieved by the addition of a capacitor and a single transistor current source as shown in Figure 3. In the case of a mosfet switch for B, this will also require an additional high end driver transistor.

A control system for a piezoelectric actuator includes a charge control circuit to maintain actuator deflection and extend actuator stroke in the event of changes in temperature and/or prolonged exposure to conditions of high humidity. The circuit uses a sampling technique to measure the actuator voltage, and adjusts the charge delivered to the actuator to maintain this voltage within set limits. This ensures reliable mechanism operation in the event of high leakage current through the actuator e.g. caused by exposure to high levels of humidity. It also results in considerable power saving when the leakage current is low. Therefore the measurement of charge on a sample capacitor compensates the effect of humidity. This provides a low power voltage-sampling regime. This low power measurement is capable of being readily measured by a digital system e.g. a microcontroller, and may be used to adjust the amount of power delivered to the actuator by said control system.

The system is also able to take account of changes in actuator capacitance due to temperature variation so that energy is not wasted during the initial charge-up period.

Because the system is constantly monitoring actuator performance the system can issue a warning or request for maintenance if leakage levels are too high. Additionally, the system measures the positive charge stored on the actuator in the ON position and uses this to gauge a negative charge applied to reach the OFF position. This is achieved without the use of an independent temperature sensor, or intermediate temperature measurement.

Operation is summarised as follows: I. Initial charge pulse sufficient for low temperature actuator.

2. Perform voltage measurement 3. Calculate additional charge required to reach target voltage 4. Apply second charge pulse to reach target voltage 5. Perform voltage measurement 6. Supply default duty cycle and frequency pulses 7. Perform voltage measurement 8. Adjust duty cycle and frequency to compensate for maintain required voltage 9. Repeat from 7 until end of operation 10. Discharge actuator measuring the charge 11. Calculate a negative charge based on the discharge measurement 12. Apply the calculated negative charge pulse The charge circuit is the same as described above, and a long initial charge pulse is provided to the actuator as described previously. A deficiency in the simple charge circuit is that the initial charge burst must be long enough to charge a high capacitance actuator at high temperature. At low temperature the actuator capacitance is low, and the excess charge is wasted in the voltage limiting circuitry. The new circuit can compensate for capacitance variation by measuring the voltage after a fixed charge pulse designed to charge a low capacitance (cold) actuator. A second long pulse is then applied to provide a much more accurate charge dose for the actuator capacitance.

To measure the actuator voltage a small amount of the actuator charge is sampled by connecting the sample capacitor, Cl, to the piezo-ceramic actuator (the current source is OFF). The value of Cl is chosen to be small enough that only a very small amount of charge is drawn from the actuator, and large enough to provide reasonable measurement accuracy.

The charge consumed by a lOOpF sample capacitor is just 2OnC at 200V which reduces that actuator voltage by about 2V (this is more than expected due to the non-linearity of the ceramic). A typical circuit for connecting the sample capacitor from a microcontroller output is shown in Figure 4.

In this circuit no power is drawn from the actuator until control B goes high. Power is only drawn while the charge is being transferred -typically under 30jis. Ri needs to be a high value to limit the power drawn by the switch -values of Ri 1OM, R2390k, R3470k are typical. Current draw is approximately 20pA. if the switch is operated for I OOp.s then it consumes just 2nC -a small fraction of the charge transferred. The total charge lost from the actuator by the sampling procedure is therefore approximately 22nC. If a sample is taken once every 0.5s, then this equates to an average current of 44nA. The same loss would be caused by an insulation resistance of about 5Gohms, so the losses due to the sampling regime should be negligible.

When the sample has been taken the switch is turned OFF and the current source is turned ON. A simple transistor implementation of the current source is shown in Figure 5.

When the transistor is turned ON an approximately constant current flows from the sample capacitor. The current is set by R4. 0.lmA is easily achieved using convenient component values and this discharges 200V from lOOpF in 200Rs. While the current flows a constant voltage appears at D. This 200 jis pulse can be timed accurately by a microprocessor. The measurement is taken during discharge of the sample capacitor. The discharge time is directly proportional to the original voltage on the actuator. The whole measuring process therefore takes much less than 500p.s. Apart from the sample capacitor charge, the power consumed by the current source is negligible. If the microcontroller needs to be awake during the timing operation, then this will consume typically 2mA. If a 500!.ts measurement is performed every 0.5s, then the average current consumption is just 2pA which is insignificant.

This simple circuit suffers from some limitations -mainly the temperature stability of the current, especially at low supply voltage. For a I.8V reference voltage, the current could change by 8% with a 50C change in temperature. For a temperature drop this will result in a slight increase in power consumption, but for temperature increase the voltage regulation will drop. Consultation with a textbook on electronic design will yield numerous methods for improvement all of which (at varying levels of cost and complexity) can be applied to this system. Some simple methods include using a forward biased diode in the base circuit, selection of a zener Z 1 with an appropriate temperature coefficient, or use of an op-amp to stabilise the emitter voltage.

The relationship between sample capacitor voltage and pulse time for the example system is shown in Figure 6. As expected a linear response is achieved.

The voltage measurement is used to adjust the power level supplied to the actuator. If the voltage drops below a threshold (say 21 OV) the power is increased. If the voltage increases above a threshold (say 240V), the power is decreased. The power is easily modulated by the microcontroller either by changing the duration of the charge pulse or the time between charge pulses. The latter may be preferable because it will scale the frequency with increasing leakage current. This will help stabilise actuator deflection. Regulation in the example system is shown in Figure 7.

The power scaling function is shown in Figure 8. As the load resistance decreases, the power drawn is scaled smoothly to accommodate the increased load. The high resistance current limit is a function of the static current draw of the circuit and can be reduced to very low levels by selection of low power components and management of microprocessor power.

The power scaling system is shown in Figure 9.

When the operation is complete, the actuator is discharged by turning on both the switch B and the current source C. The discharge time for the actuator is measured at point D. This is a measure of the positive charge stored on the actuator.

To operate with negative voltage a slightly more elaborate system is required as shown in Figure 10.

The actuator is preferably configured in an H bridge to permit positive and negative charge from a single charge circuit. H bridge designs using transistors for this type of application are known. For positive charge BI and B4 are turned on with B2 off. B3 and the current source are used to control the voltage on the actuator in the same manner as described above.

To discharge the actuator B! and B2 are turned OFF, and B4 ON (or diode connected). The dischargc is timed through B3 and the current source as described above.

To charge negative, B 1 and B4 are OFF, while B2 is ON. B3 and the current source are turned ON to provide an accurate quantity of negative charge on the actuator. The amount of negative charge is controlled simply by timing of the charging pulse. The duration of the negative charge pulse is calculated from the measured duration of the positive discharge. A long positive discharge is obtained at high temperature, and a correspondingly short negative pulse is applied to prevent re-poling the ceramic. At low temperature, a much shorter positive charge is measured, and a correspondingly long reverse charge is applied. Thus the need for a temperature sensor or an intermediate temperature measurement is removed.

Claims

CLAIMS: 1. A method for compensating for the effects of humidity and

temperature variation during operation of a piezo-ceramic actuator comprising the steps of: a) providing an initial charge pulse from a charge circuit to the actuator; b) measuring the obtained voltage across the actuator in a voltage measurement unit; c) calculating the charge required to be supplied from the charge circuit to the actuator to reach the voltage measured in step (b); d) adjusting the charge supplied to the actuator according to the charge calculated in step (c); e) providing charge pulse to the actuator based on the adjusted charge; f) repeating steps (b) to (e) until end of operation of the actuator; and g) measuring the charge stored on the actuator during discharging of the actuator.

h) using the measured stored charge to calculate a suitable negative charge to remove hysteresis in the ceramic i) applying the calculated negative charge pulse to the actuator
2. The method as claimed in claim I wherein in step (b), said measurement is obtained by timing the discharge of a sampling capacitor through a current source.

3. The method as claimed in claim I wherein, for calculating the charge to be supplied in step (c), the charge is increased if the voltage measured in the voltage measurement unit drops below a predetermined threshold and is decreased if said voltage increases above the predetermined threshold.

4. The method as claimed in claim 1 wherein in step (d) the charge is adjusted by changing the duration of the charge pulses supplied to the actuator.

5. The method as claimed in claim I wherein in step (d) the charge is adjusted by changing the time period between consecutive charge pulses.

6. The method as claimed in claim 1 wherein the charge circuit is adapted to supply positive andlor negative charge to the actuator.

7. The method as claimed in claims 1 to 6 wherein in step (g) the actuator charge is measured by timing the discharge through a current source.

8. The method as claimed in claim I to 7 wherein the same current source as in claim 2 is used.

9. The method as claimed in claims 1 to 8 wherein step (h) uses a correlation between the charge measured in claim 7 and the negative charge that can be applied without exceeding the coercive field strength of the ceramic to calculate a negative charge pulse without the use of a temperature sensor or intermediate temperature measurement 10. The method as claimed in claims 1 to 9 wherein in step (i) applies the negative charge pulse calculated in Claim 9.

11. The method as claimed in claim 10 wherein the negative charge pulse is applied by means of a timed current pulse 12. A control system for a piezo-ceramic actuator for preventing excess charge leakage in the actuator by the method claimed in any one of claims I to 8, said system comprising a charge circuit (A) for supplying charge pulses to the actuator and a voltage measurement unit for measuring voltage obtained across the actuator, said charge circuit (A) being adapted to vary the charge pulses supplied to the actuator on the basis of the voltage measured by the voltage measurement unit.

13. The control system as claimed in claim 12 wherein said charge circuit (A) is implemented by a transistor current source.

14. The control system as claimed in claim 12 wherein said voltage measuring unit is implemented by a capacitor (C 1).

15. The control system as claimed in claim 14 wherein said capacitor (Cl) has a small capacitance value.