FR2724463A1 - METHOD FOR MEASURING AN ACCELERATION BY A MOVABLE ARMOR SENSOR AND ACCELERATION SENSOR FOR IMPLEMENTING THE METHOD - Google Patents

Abstract

Method for measuring an acceleration, and corresponding acceleration sensor, in which an excitation signal (1M) is applied to a movable armature (2) of an acceleration sensor in order to identify a rest position of the armature (2), the moving armature (2) is subjected to the acceleration to be measured, a position deviation signal (EC) of the armature (2) generated by the signal d is detected (H2) 'excitation (lM) and depending on the acceleration to be measured, to provide a control signal (ME) regulating the intensity and direction of an electric force which is applied to the mobile armature (2) in order to return it to its rest position and the value of the acceleration is then determined from the control signal (ME), a process in which the excitation signal (lM) is applied in the form of a modulation of the control signal. control (ME) and the control signal (ME) is continuously applied and generated by integration (20) of the position deviation signal (EC).

Description

t Method of measuring an acceleration by a moving arm sensor

  and acceleration sensor for implementing the method.

  The measurement of the acceleration undergone by a sensor is used to determine the stresses or the displacement undergone by a mechanical part integral with the sensor. Thus, in an automobile, a shock sensor controls, in the event of a sudden shock, the inflation of a shock absorbing air bag deploying in front of the driver. Sensors are known comprising a movable electrically conductive armature, of determined mass, which is resiliently mounted on a support and moves, under the action of acceleration, towards one or the other of two conductive armatures of electricity surrounding it, rigidly fixed to the support. The fixed armatures, or electrodes, are respectively brought to two opposite potentials and each form, with the mobile armature, two capacitors. In the absence of acceleration, the two armatures, or electrodes, fixed have, on the movable armature occupying a position equidistant from the two armatures

  fixed, electrical effects that cancel each other out.

  When the mobile armature approaches one of the fixed electrodes, under the effect of an acceleration, the value of the capacitance between the mobile armature and the nearest fixed electrode increases, conversely the value of the capacity between the movable armature and the farthest fixed electrode decreases. This variation in differential capacity can be read by a circuit for measuring

  capacity connected to fixed electrodes.

  3) The measurement voltage of this differential capacity is a position signal which makes it possible to control a feedback loop providing a control signal determining the average voltages of the two fixed electrodes in order to restore, by the electrostatic forces applied to the movable armature , a mechanical and therefore also an electrical balance. The variation of the signal

2 2724463

  control, from its value for the rest position of the movable armature,

  then represents the acceleration to be measured.

  It is known to measure variations in capacity by measuring charge transport resulting from the application of voltage steps on the electrodes.

  with devices called switched capacitor circuits.

  Thus, it is known to modulate by the control signal, individually and in opposite directions, the voltage pulse widths, or their rate of appearance, of two pulse trains applied respectively to the two

electrodes, or armatures, fixed.

  The amplitude of each of these pulses is seen by the capacity measurement circuit as a step producing charge transport regardless of the duration of these pulses (there is sampling of voltage levels). The capacitance measurement voltage thus obtained determines the duration of the pulses with reference to a setpoint. This modulation of

  pulse width results in the appearance of an average counter voltage

  reaction on the electrodes. The (fixed) level of the pulses is used to measure the position (variable capacity) and the duration (at fixed level) of the pulses is used to

position control.

  However, these pulses generate a significant electrical noise disturbing the measurement, since they must reach a large amplitude to have the necessary electrostatic efficiency. In addition, this efficiency is

  limited by the fact that it is necessary to plan guard times for opening-

  closing of electronic doors, and especially for the measurement sampling, in the sequencing of pulses. The sampling frequency of the measurement is therefore necessarily the same as that of

  enslavement, therefore limited by the response of the electromechanical system.

  The present invention aims to improve the efficiency of the control voltages applied in feedback on the two fixed plates and to limit electrical noise.

3 2724463

  To this end, the invention relates to a method of measuring an acceleration, in which an excitation signal is applied to a movable frame of an acceleration sensor to identify a rest position of the frame, subjecting the movable armature to the acceleration to be measured, detecting a signal of position difference of the armature generated by the excitation signal and dependent on the acceleration to be measured, to provide a control signal adjusting the intensity and direction of an electric force which is applied to the movable armature in order to bring it back to its rest position and the value of the acceleration is then determined from the control signal, characterized by the fact that the excitation signal is applied in the form of a modulation of the control signal and the control signal is applied continuously and generated by integration of the

position deviation signal.

  Thus, a continuous control is applied, which therefore does not generate electrical noise, and the excitation signal can be fixed at a relatively low level, not very disturbing, since it is used only for measurement, and not for electric force control. The excitation signal pursues the electrical equilibrium of the movable armature, each instant excitation signal measuring the residual instantaneous imbalance, i.e. the remaining error on the control signal, and tending gradually, by

  integration, to cancel this error.

  We can then inject the excitation signal at a frequency much higher than the natural frequency of the sensor, represented by the control signal, that is to say that we can have only a very weak signal d '' excitement, little disturbing, since, to control the integrating means, the number

  high of instant signals of excitement compensates for their weakness.

  Advantageously, the position deviation signal is measured by a measurement of

  the electric charge induced by the excitation signal on the movable armature.

  Preferably, the control signal is modulated by two trains of pulses of the same amplitudes and opposite polarities, which are applied respectively to two fixed armatures framing the movable armature and the position deviation signal is integrated. in synchronism with the modulation pulses. Advantageously. the signals required for the measurement do not limit

$ 4 2,724,463

  not the dynamics of the command, so we can consider that there is

  separation between measurement and control.

  The invention also relates to an acceleration sensor for implementing the method of claim 1, comprising, between two fixed frames, a movable frame intended to be moved from a rest position under the action of acceleration to be measured and an excitation signal, characterized in that it includes integrating means connected to the movable armature for integrating a position difference signal of the armature generated by the excitation signal and dependent acceleration to be measured, and adding means connected at

  entry to integrating means and, at exit, to fixed armatures.

  The invention will be better understood using the following description of the form

  preferred embodiment of the acceleration sensor implementing the method of the invention, with reference to the accompanying drawing, in which: - Figure 1 is an electrical diagram of the sensor, - Figures 2 to 4 illustrate the evolution during of the time L of sensor control signals - Figures 5 and 6 illustrate the time response of the sensor during a measurement and

  - Figures 7 to 9 illustrate this same response during a test.

  The sensor comprises a housing 1 containing a movable armature 2 in the form of a flexible blade, electrically conductive, elastically connected to the housing 1, but electrically isolated from the latter, and occupying a rest position equidistant from two fixed plane armatures 3 and 4, also electrically conductive, which are arranged parallel to the frame 2 and surround it. The armature 2 is electrically connected to the inverting input of an operational amplifier 10, which is also connected to its output by a capacitor 11 in parallel with an electronic switch 12 controlled by a signal

2724463

  H1 clock. A non-inverting input of the amplifier 10 is connected to ground. Amplifier 10 is thus mounted as an electrical charge amplifier

  to measure the electrical charge of the armature 2.

  The output of the amplifier 10 provides, through an electronic switch 13, a position deviation signal EC, explained below, to a terminal of a

  capacitor 15 connected to another ground return switch 14.

  Switch 13 is controlled by a clock signal H2, while

  the switch 14 is controlled by the clock signal Hi.

  The other terminal of the capacitor 15 supplies, through an electronic switch 16 controlled by the clock H1, the inverting input of an amplifier 20 and an electronic switch 17, controlled by the clock signal H2, for recall to the mass. A capacitor 21 connects the input

  reverser at the output of amplifier 20, which is thus mounted as an integrator.

  The output of amplifier 20 provides a signal ME for acceleration measurement and drives, through a resistor 31 connected in parallel to a series RC circuit, the inverting input of an amplifier 30 also receiving, through a resistor 32 , IM pulses of low amplitude, much lower than the volt,

  from a pulse generator 6.

  A non-inverting input of the amplifier 30 is connected to ground.

  Amplifier 30 has an inverted or supplemented output and a

  direct output connected to its negative input through a resistor 33.

  The inverting inputs of two amplifiers 40 and 50 are respectively connected to the midpoint of two resistive dividing bridges 41-42 and 43-44 respectively supplied, at one end, by the two outputs of 3 (amplifier 30 and connected, by their other end, at the output of the associated amplifier, 50. The fixed plates 3 and 4 are respectively connected to the outputs of amplifiers 40 and 50. The positive inputs of amplifiers 40 and 50 are biased by a DC voltage source 9. This DC voltage source 9 serves to prepolarize the fixed electrodes 3 and 4 to the same voltage value developing two equal and opposite forces on the movable electrode 2 to which will be added the differential voltages (forces)

6 2724463

  enslavement. Two pulse generators, 7 and 8, provide

  the clock signals H1 and H2 respectively.

  The shapes of the clock signals H1 and H2 are respectively represented, as a function of time t, in FIGS. 2 and 4, while the pulses IM of the

  generator 6 appear in FIG. 3 as a function of the same time t.

  In this example, there is provided a self-test circuit consisting of a generator 34 of test pulses ST, in series with a resistive potentiometer 35, connected

  between the inverting input of amplifier 30 and ground.

  The operation of the sensor will now be explained.

  The IM pulses from the generator 6, which, by the amplifier 30, are split into two opposite pulses reaching the fixed armatures 3 and 4 respectively, each induce, on the mobile armature 2, an electric charge which, for reasons of symmetry , is equal and of opposite sign to that induced by the other fixed armature 4 and 3. The movable armature 2 is therefore not loaded

  steps and amplifiers 10 and 20 remain at rest.

  During the application to the sensor of an acceleration directed vertically in FIG. 1, the movable armature 2 approaches one of the fixed armatures 3 and 4 and presents with it an increased capacity. The IM pulse received by this nearest fixed electrode 3 or 4 then induces an increased number of electrical charges on the movable armature 2, while the opposite effect, of reduction of the

  number of induced charges, occurs for the other fixed armature 4, 3.

  The difference in charges induced on the movable armature 2 provides the signal

  position difference EC appearing at the output of amplifier 10.

  The switch-capacitor assembly 13-17 samples the position deviation signal EC and transmits it to the integrating amplifier 20, which outputs

  ME measurement of acceleration, up to several volts.

  The measurement signal ME is then split by the amplifier 30, the two measurement signals split, that is to say reversed one with respect to the other by

7 2724463

  on either side of a rest value, being respectively applied to the fixed armatures 3 and 4 in order to generate two electrostatic forces, the

  resulting tends to recall the movable frame 2 to its rest position, median.

  During the transient state to return to the rest position, the measurement ME supplied by the amplifier 20 is false. The IM pulses from the generator 6 induce, as explained, the position deviation signal EC, here substantially proportional to the position deviation, which continues to be integrated by the amplifier 20 including the measurement output ME, and regulation, thus evolves, until equilibrium, in a direction tending to cancel the signal of position difference EC. FIG. 6 illustrates the evolution, as a function of time t, of the measurement signal ME, in response to an acceleration AC represented in FIG. 5, and shows, in the form of steps of decreasing amplitudes, the successive integrations of the

  position deviation signal EC in the measurement signal ME.

  The regulation thus takes place in a closed loop and is carried out by means of the "continuous" voltage which constitutes the measurement signal ME and which is scalable as a function of the pulses IM, or stimuli, of the generator 6 which measure the servo error and constitute a modulation of the ME measurement signal

  applied to fixed reinforcement 3 and 4.

  In short, in this method, the excitation signal IM is applied to the mobile armature 2 to identify the rest position of the armature 2, the mobile armature 2 is subjected to the acceleration to be measured, the position deviation signal EC of the armature 2 generated by the excitation signal IM and dependent on the acceleration to be measured, to supply the control signal ME regulating the intensity and the direction of an electric force that l 'is applied to the movable armature 2 in order to return it to its rest position and the value of the acceleration is then determined from the control signal ME, the excitation signal IM being applied in the form of a modulation of the control signal ME, which is continuously applied and generated by integration of the

position deviation signal EC.

  Figures 2-4 illustrate the sequencing of the operation of the switches.

8 2724463

  In general, all the switches are blocked off when the clock signal H1 or H2 which controls them is in the blocked state, here

corresponding to a low state.

  IM pulses from generator 6 have a leading edge, here rising, which follows the falling edge of signal H1, while their falling edge is

  produced at the same time as the rising front of Hi.

  The H2 signal has positive pulses strictly "inside" the IM pulses, that is to say that the H2 signal has a time window of

  reading of the effect of IM pulses. The sequencing is as follows.

  The signal Hi1 being initially in the high state, the switches 14 and 16 are closed, that is to say passers-by, and connect the capacitor 15 to ground and to

amplifier 20.

  When the signal H1 goes down, the switches 12, 14 and 16 open and then the IM pulse is emitted and generates the position deviation signal EC at the output of the amplifier 10. The signal H2 controls the closing, during the reading or sampling window, of the switches 13 and 17 and the capacitor charges at the value of the position difference signal EC. The signal H1 then rises and the pulse IM drops. Switch 12 eliminates, between two

  measurements, the charges induced on the movable armature 2.

  A sampling and maintenance assembly is thus carried out.

  The integration of the position deviation signal EC is thus carried out, here, in

  synchronism with the modulation pulses.

  In this example, two fixed armatures have been provided which induce electric charges on the movable armature. Generally, at least one fixed armature is required, exerting on the movable armature a return force opposite to that due to acceleration, which is proportional to the mass of the armature.

  mobile. The restoring force can also, for example, be magnetic.

9 2724463

  The self-test circuit 34-35 has an effect of the same type as that of the generator 6, that is to say that the application of a voltage pulse by the generator 34 creates an electrostatic force, but which, then , unbalances the electrode, or seismic mass, 2. The position deviation signal EC, integrated by the amplifier 20, has the effect of shifting the output voltage ME by an amount proportional to

  the amplitude of the ST test pulses, in the ratio of resistors 31 and 35.

  FIGS. 7 to 9 respectively illustrate the temporal evolution of the test ST pulses, the position modulation P of the mobile electrode 2 which results therefrom and

  the corresponding modulation of the measurement signal ME.

  It is thus possible to perform the self-test in real time, without deactivating the functions

  of the sensor, and in particular by maintaining closed loop operation.

  In addition, the self-test is bidirectional and the controlled movement of the movable armature 2 is adjustable by action on the potentiometer 35 and can reach full scale. In addition, the self-test creates a mechanical excitation of the seismic mass 2, equivalent to an acceleration, without it being necessary, when one wants, in production, to perform a sort of running-in of the sensor,

  mount the sensor on a vibrating pot.

2724463

Claims (8)

  1. A method of measuring an acceleration, in which an excitation signal (IM) is applied to a movable armature (2) of an acceleration sensor to locate a rest position of the armature (2 ), the movable armature (2) is subjected to the acceleration to be measured, a position deviation signal (EC) of the armature (2) generated by the excitation signal (IM) is detected (H2) and depending on the acceleration to be measured, to provide a control signal (ME) regulating the intensity and the direction of an electric force which is applied to the movable armature (2) in order to bring it back to its position of rest and the acceleration value is then determined from the control signal (ME), characterized in that the excitation signal (IM) is applied in the form of a modulation of the control signal ( ME) and the control signal (ME) is applied continuously and generated by integration (20) of the position deviation signal (EC).   2. Method according to claim 1, in which the position deviation signal (EC) is measured by a measurement of the electric charge induced by the signal.   excitation (IM) on the movable frame (2).   3. Method according to one of claims 1 and 2, wherein the module is modulated   control signal (ME) by two pulse trains of the same amplitudes and opposite polarities, which are respectively applied to two armatures   fixed (3,4) framing the movable frame (2).   4. Method according to claim 3, in which the integration of the position deviation signal (EC) is carried out in synchronism (H1) with the pulses of modulation (IM).   5. Acceleration sensor for implementing the method of claim 1, comprising, between two fixed frames (3,4), a movable frame (2) intended to be moved from a rest position under the action of the acceleration to be measured and of an excitation signal (IM), characterized in that it comprises integrating means (20) connected to the movable armature (2) for integrating a position deviation signal (EC) of the armature (2) generated by the excitation signal (IM) and dependent on the acceleration to be measured, and there 2724463 adding means (30) connected as input to the integrating means (20) and, in outlet, with fixed armatures (3,4).   6. The sensor as claimed in claim 5, in which an amplifier (10) for measuring electrical charges is connected at the input to the mobile armature (2) and, in   outlet, to the integrating means (20).   7. Sensor according to one of claims 5 and 6, wherein means   sampling and holding (13-17) are arranged to take from the mobile armature (2), in synchronism with a rate of appearance of the excitation signal (IM), the position deviation signal ( EC) and to present it to   the input of the integrating means (20).   8. Sensor according to one of claims 5 to 7, in which there are provided   self-test means (34, 35) arranged to, via the fixed armatures (3, 4), transmit a test signal (ST) to the mobile armature (2)   applying a determined acceleration.