GB2310284A

GB2310284A - A piezo-electric effect vibrating gyrometric device

Info

Publication number: GB2310284A
Application number: GB9602876A
Authority: GB
Inventors: Pierre Leger; Alain Jeanroy
Original assignee: Sagem SA
Current assignee: Sagem SA
Priority date: 1994-08-11
Filing date: 1996-02-13
Publication date: 1997-08-20
Anticipated expiration: 2016-02-13
Also published as: FR2723635B1; DE19605648B4; DE19605648A1; GB9602876D0; FR2723635A1; GB2310284B

Abstract

The appts. includes a disc (12) constructed from piezoelectric material and is seated on a base of low expansion which bears thirty-two electrodes on its surface. Exciting electrodes (Em+,Em-), bearing potentials of opposite polarity, are designed to produce a motor field and 4R type vibrations in the disc. Electrodes (Dm+,Dm-) detect the motor field but not any vibrations due to Coriolis acceleration and they are used for regulation. Electrodes (Dc+,Dc-) detect the Coriolis field which is compensated by further electrodes (Ec+,Ec-). The exciting and compensating electrodes are interleaved with the detecting electrodes but separated from them by guarding electrodes (26) at earth potential. The excitation electrodes provide radial nodes of the fourth order and are dependent on a predetermined source. The source induces radial nodes having a frequency less than the nodes at the tangential node.

Description

2310284 A PIEZO-ELECTRIC EFFECT VIBRATING GYROMETRIC DEVICE

The invention relates to gyrometric devices, i.e. devices for measuring a speed or angle of rotation around a sensitive axis. More particularly it relates to gyrometric devices of the kind comprising a vibrating detecting means in the form of a substantially flat, circular plate consisting at least mainly of piezo-electric material, the major surfaces of which bear electrodes for exciting the plate to resonance and distributed regularly around the axis of the plate, and electrodes for detecting the vibration of the plate and situated in the same plane as the excitation electrodes and disposed so as to detect the stresses caused by the driving field due to the excitation electrodes and by the Coriolis field when the plate is rotating around the sensitive axis consisting of the axis of the disc.

The detecting means in a device of this kind is a resonator whose vibration field vibrates as a result of rotation around the axis, at a speed which depends on the geometry of the resonator and is usually different from the speed applied to the resonator.

Depending on the construction of an electronic excitation and detection circuit connected to the electrodes, the device can be either a gyrometer or a gyroscope. In the first case the excitation electrodes must maintain the initial vibration field in a fixed direction relative to the resonator. A speed of rotation results in a shift in the vibration, which can be considered due to superposition on the driving field of a vibration field having the same nature but offset at an angle. In the second case, for operation as a gyroscope, the supply circuit is designed so that the driving field follows the rotation of the vibration relative to the resonator without interfering with it.

Numerous gyrometers of the previously-defined kind are already known. Basically they use one or other of two types of vibration, one of which may be called the radial mode or with radial lobes, whereas the other has tangential lobes. In each kind there are successive orders of vibration each identified by a serial number which represents the number of spatial periods of vibrations over 3600.

A gyrometer using second-order vibrations with radial lobes has inter alia already been proposed. A gyrometer of this kind is described in the article "The theory of a piezo-electric disk gyroscope" by J. S. Burdess et al, IEEE Transactions on Aerospace and Electronic Systems, Vol. AES-22, No. 4, July 1986, page 410.

Fig. 1, greatly exaggerated for clarity, shows the deformations undergone by the plate of such a gyrometer during vibration in resonance, in the absence of a speed of vibration. The detecting means is a plate 12 having a circular cross-section and of piezo-electric material connected at its centre to the structure whose speed of rotation is to be measured. The rear surface of the plate 12 is covered with a thin layer of conductive material which can be continuous, for ease of manufacture. Excitation electrodes EM are disposed along an axis M and supplied by a signal processed by an amplifying and oscillating circuit 13 on the basis of a signal taken from the detection electrodes Du aligned along the axis Y at right angles to X. The measurement is made by sampling a signal on the detection electrodes Dr aligned along an axis A at 450 from axes X and Y. A differential amplifier 14 supplies the output signal Sr.

1 A gyrometer of this kind has disadvantages. A first disadvantage is that the second-order vibration in radial mode tends to cause considerable displacement near the centre. Fixing the plate modifies the vibration of the disc and reduces its sensitivity. Any lack of homogeneity causes defects in symmetry. Furthermore the gyrometer is less sensitive, since the field of stresses created in piezo-electric material produces very little current irrespective of the geometry given to the electrodes.

Some of these disadvantages are avoided in a gyrometer with two tangential lobes described in the document FR-A-2 684 729 (patent application FR 91 15096 by the applicants). The electrodes and the excitation circuit of this gyrometer are constructed so as to bring about second-order oscillation in tangential mode. However, this mode likewise results in considerable displacement near the centre, so that the central fixing appreciably interferes with operation.

The displacement brought about near the centre of the disc decreases as the order increases and it has been found that the displacements become practically negligible from order 4. Vibration above mode 4 would require a very complex network of electrodes, resulting in a high resonance frequency and low sensitivity.

The invention accordingly proposes a gyrometric device of the kind defined hereinbefore, characterised in that the electrodes for exciting the plate to resonance have a geometry such as to bring about fourth- order vibration and are connected to a supply circuit adapted to bring about resonance with radial lobes by limiting the frequency of the circuit to a limiting value below the resonance frequency of the plate in fourth-order tangential mode.

The advantage of the device according to the invention will be clear from a consideration of the values of the ratio between the maximum displacemen at the surface of the disc during resonance and the average displacement over an internal circumference secured in conventional manner. It has been found, in the case of discs intended for miniature gyrometric devices, that this ratio is below 2 for second-order modes and about 18 for fourth-order modes. Use of the fourth-order radial mode, called 4R hereinafter for simplicity, results in a resonance frequency which remains relatively near the fundamental frequency of the disc during contraction and expansion, and much lower than the resonance frequency of mode 4T.

The invention will be more clearly understood from the following description of embodiments of the invention given by way of nonlimitative example, and from the comparison made therein with gyrometers of known kind.

The description refers to the accompanying drawings in which:

Fig. 1, already mentioned, is a block diagram showing a possible arrangement of excitation and detection electrodes on the vibrating plate of a gyrometer according to the prior art, operating in mode 2R;

Fig. 2 gives the ratio of the frequencies of the various vibration modes to the frequency fn of the fundamental radial mode (disc is deformed while remaining circular).

Fig. 3 shows a possible distribution of excitation and detection electrodes on the plate of piezoelectric material of a gyrometer according to the invention; Fig. 4 is a diagram showing the directions and relative amplitude of the displacement in a free disc resonating at mode 4R, in the absence of rotation and having a central opening; Fig. 5 shows a possible method of assembling the resonator; Fig. 6 shows a possible construction of an excitation and detection circuit for association with the arrangement of electrodes in Fig. 3, and Fig. 7 shows an electronic circuit usable in conjunction with the electrodes in Fig. 3 for operation as a gyroscope.

Before describing the gyrometric device according to the invention, it may be useful to recall the basic operation of a resonator comprising a disc-shaped plate vibrated in a mode sensitive to a speed of rotation around the axis of the disc.

The plate can be considered as constituting a double resonator vibrating in its plane under the action of a driving field with sinusoidal variation (generated by the electrodes Ev in the case of Fig. 1) and also, in the event of rotation around the sensitive axis, vibrating as a result of coupling due to the Coriolis acceleration. Any vibration mode is sensitive to a speed of rotation if it has a cyclic geometry with alternating axes of symmetry and antisymmetry. In the case of Fig. 1, the axes A are axes of anti-symmetry for the driving field and axes of symmetry for the field of vibrations produced by the Coriolis acceleration.

More generally, whatever the mode, the field of Coriolis vibrations has the same distribution as the driving field but with an angular shift denoted by a in Fig. 1, which changes over the axes of symmetry and antisymmetry. The shift is equal to n/2n, where n is the order of the mode.

At any point M on the surface of the disc, the Coriolis acceleration yc can be written:

yc(M,t)=2Q(t) A V(M,t) (1) where Q is the speed of rotation and V is the speed due to the driving field.

The acceleration is therefore perpendicular to the direction of displacement of the point M. The efficiency with which the field of Coriolis vibrations is excited by this acceleration is at a maximum when the acceleration and the Coriolis field are co-linear. It is therefore desirable that the driving field of vibrations and the field of Coriolis vibrations should be perpendicular at all points on the surface of the disc.

This condition exists for the radial modes, which gives them greater sensitivity than the tangential modes, where the condition of orthogonality is not always respected, particularly in the zones where the displacements are purely tangential and do not make any contribution to the sensitivity.

The sensitivity to speed of a given mode can be determined by calculating the following integral over the entire surface of the resonator:

f-VOriving(M) / VCoriolis(M)dS fV2.9 S Drivin (M) ds (2) As this calculation shows, the sensitivity of the radial mode 4R to the speed of rotation is ten times as great as that of the tangential mode 4T.

The overall sensitivity of the resonator depends not only on the sensitivity to the speed of rotation but also on the piezo-electric sensitivity, i.e. the current supplied by the field of stresses produced by the Coriolis vibration.

Calculation shows that this sensitivity is very low for the radial mode 2R and remains low for the tangential mode 2T. On the other hand the sensitivity obtained for the fourth order, both in tangential mode and in radial mode, is sufficient for satisfactory operation of the electronic circuit.

Overall, the 4R mode gives higher sensitivity than the 4T mode and above all, as will be shown hereinafter, enables the disc to be fixed centrally without interference with the vibrations. For this reason, this mode was adopted. It was found that use of a higher-order mode was not advantageous. The resolution frequency increases, as shown in Fig. 2, so that the scale factor (inversely proportional to the square of the frequency) decreases. The electronic circuits have increased complexity. The number of electrodes becomes very large, which complicates the wiring.

Fig. 3 shows a possible geometry of the electrodes used for operating a gyrometric device in 4R mode. This geometry uses alternatively positive and negative electrodes, both for excitation and for detection, so as to have maximum symmetry. The result is a total of 32 electrodes. In Fig. 3, as in Fig. 1, the following notation is used:

Em+ and Er: excitation electrodes of the driving field. The electrodes EMare supplied in phase opposition to the electrodes EV+.

Dv+ and Du-: electrodes for detecting the driving field. These electrodes are insensitive to the field of Coriolis vibrations, and enable the driving field to be controlled at a constant amplitude;

Dc+ and Dr: electrodes for detecting the Coriolis f ield; Ec+ and Er: electrodes for compensating the Coriolis f ield.

Other arrangements are possible, more particularly the electrodes for compensating the Coriolis field can be omitted, at least during operation as a gyrometer. The electrodes for detecting the driving field can also be omitted if operation with variable excitation is acceptable.

The detection electrodes Du and Dr are distributed at the periphery of the disc where the displacement brought about by the driving field and the Coriolis field is the greatest, as shown in Fig. 4. Electrodes Dp and Du are prolonged towards the centre by tracks 20 ending in studs 22 for soldering wires which are connected to the measuring circuit and can extend inside the inner periphery 24 of the disc 12. Electrodes R, and DC all have the same surface area. Each pair of successive electrodes having the same function and the same polarity are disposed at 900 to one another.

The excitation electrodes EC and Ev are triangular in shape with a central aperture for the tracks of the detection electrode. They are all identical. Each driving or compensating electrode is aligned with a detection electrode corresponding to the same field and having the same polarity.

In the embodiment given by way of example in Fig. 3, each detection electrode and its track are surrounded by a guard electrode 26 which is earthed and appreciably reduces the stray capacitative coupling between excitation electrodes and detection electrodes. In the case illustrated, the guard electrode is a spider-web structure having radial metal-coated tracks distributed at regular angular intervals, each placed between two successive measuring or excitation electrodes, a central circular track 28 which can easily be earthed by a wire extending through the central hole in the plate, and arms 30 surrounding the tracks 20.

In this embodiment, the surface of the disc 12 on the other side from the surface bearing the electrodes shown in Fig. 3 will usually be completely covered by an earthed metal covering.

The plate can be assembled as shown in Fig. 5. The plate 12 is fixed to a base 30 of material having a low coefficient of expansion and a central tubular prolongation which fits into the central hole of the plate 12 and is stuck to the plate. A cap 32 bearing an annular printed circuit 34 is secured to the base. Electrodes having the same function and the same polarity can be placed in parallel to form groups of four, and connected to a single connecting wire.

As Fig. 4 shows, displacements near the inner periphery 24 are extremely small, so that this method of fixing does not interfere with the measurement.

The circuits associated with the vibrating detecting means in Fig. 3 can be of various kinds, more particularly one of those described in the prior patent applications by the applicants. Fig. 6 shows another possible construction of the excitation circuit and the measuring circuit.

The excitation circuit shown in Fig. 6 operates as a closed loop in order to give the excitation vibration a constant amplitude and a pulsation equal to that of the proper mode of vibration. It comprises an amplitudemodulated oscillator 40. The output signals f rom the electrodes De and Iare added in absolute value in the summing integrator 42, which receives them via amplifiers 44. The circuit also comprises a multiplier 46 for controlling the output voltage of the oscillator 40. A multiplier input receives the output signal from a phase-locking loop 48 for selecting the vibration mode 4R from among other modes having neighbouring frequencies and the same phase relations, inter alia the mode 4T. To this end, the phase locking loop 46 has at least one high-frequency stop which prevents oscillation at the frequency of mode 4T. It also advantageously comprises a low-frequency stop which is used only during starting-up. Accordingly, when voltage is applied, oscillation builds up at a frequency between the two stops and the control gain has the maximum amplitude. The detected signal supplied by the summing integrator 42 is compared in phase with the signal from the phase-locking loop which controls the oscillator 40 until the signals are put in phase. proper mode 4R.

The device thus oscillates on the The second input of the multiplier 46 receives an amplitude control signal obtained by comparing the rectified output voltage of the integrator 42 with the reference signal. The circuit formed by the rectifier 50 and the comparator 52 constitutes an amplitude control circuit.

The measuring circuit comprises amplifiers 54 which receive the output signals f rom the electrodes Dp and drive an absolute-value integrator 56. This integrator is followed by two synchronous demodulators 57 which receive reference signals from the oscillator circuit and bring about demodulation in phase and in quadrature. The reference signals comprise a cosine signal supplied by a converter circuit 61 followed by a 900 phaseshifter and a sine signal supplied by a circuit 63, which can be connected in cascade to the circuit 61, followed by a threshold circuit which does not give a phase shift. The demodulated signals enter correcting networks 58 and 60 which fix the overall pass band. The outputs of the networks are remodulated at 62 and 64. They are then applied to an amplifier 66 which supplies the electrodes Er+ and E,-. The phase-remodulated signal which appears at the output of one of the networks 58 and 60 represents the rotation speed 0 of the resonator around its axis.

To bring about operation of the gyroscope with unlimited deflection of the device, the excitation and measuring circuits can have various constructions, inter alia that shown in Fig. 7.

In all cases, a rotation 0 of the plate casing around its sensitive axis (the axis of the plate) brings about a rotation 0 of the field of vibrations relative to the resonator. The measuring circuit must measure sin 0 and cos 0 on the basis of the amplitudes of the vibrations of the input resonator associated with the excitation circuit and of the output resonator, these vibrations being electrically orthogonal.

The relation between 0 and 0 depends on the geometry of the resonator.

The excitation circuit has the same function as in gyrometer mode, i.e. to compensate losses. It must perform this function irrespective of the direction of the field of vibrations relative to the resonator.

For operation as a gyroscope, it is necessary to have detection electrodes corresponding to the input and output modes of operation as a gyrometer (electrodes Dr and Du in Fig. 3). These electrodes, which can be considered as corresponding to the axes X and Y respectively, are denoted by references DX and DY in Fig. 7. The vibration is reconstituted by the sum of the integral of one detection channel and the other detection channel. The signal obtained represents the vibration in amplitude and in phase.

The amplitude control circuit has a construction comparable with that shown in Fig. 6. It also comprises a rectifier 70 and a reference value comparator 72 driving a control multiplier 74.

The measuring circuit also comprises a phase-locking loop 76 controlled by the signal representing the vibration, with a phase reference given by an angular coding circuit 68 which uses the vibration amplitudes to calculate an evaluation 0 of the rotation 0. The angular coding circuit 78 demodulates the detected sine and cosine signals relative to a reference coding coming from the output 80 of the phase-locking loop 76.

The signal obtained, after amplitude control at 74, is distributed among the excitation electrodes Ex and Ey (corresponding to electrodes EC and Em in Fi xg. 3) by two multipliers 84 and 86 actuated by cos 0 and sin i.e. the trigonometric functions of the evaluation 0 of the rotation 0 of the vibration.

The output of the angular coding circuit 78 represents an evaluation 0 of the orientation of the vibration relative to the resonator. The angle 0 through which the resonator has turned is obtained by dividing this angle by a known driving coefficient which can be determined experimentally and has good stability.

Claims

C L A I M S

1. A gyrometric device comprising a vibrating detecting means in the form of a substantially flat, circular plate consisting at least mainly of piezoelectric material, the major surfaces of which bear electrodes for exciting the plate to resonance and distributed regularly around the axis of the plate, and electrodes for detecting the vibration of the plate and situated in the same plane as the excitation electrodes and disposed so as to detect the stresses caused by the driving field due to the excitation electrodes and by the Coriolis field when the plate is rotating around the sensitive axis consisting of the axis of the disc, characterised in that the electrodes for exciting the plate to resonance have a geometry such as to bring about fourth-order vibration with radial lobes and are connected to a supply circuit adapted to bring about resonance with radial lobes by limiting the frequency of the circuit to a limiting value below the resonance frequency of the plate in fourth-order tangential mode.

2. A device according to claim 1, characterised in that the detection electrodes are distributed around the periphery of the disc and are prolonged towards the centre by tracks (20) ending in studs (22) for welding wires connected to the measuring circuit.

3. A device according to claim 2, characterised in that the excitation electrodes (EC, E,4) are triangular, with a central gap for the tracks of the detection electrodes, each excitation electrode, whether driving or for compensating the Coriolis field, being aligned with a detection electrode -is- corresponding to the same field and having the same polarity.

4. A device according to claim 2 or 3, characterised in that each detection electrode and its track are surrounded by an earthed guard electrode.

5. A device according to any of the preceding claims, adapted for operation as a gyrometer, characterised in that the supply circuit comprises an amplitude-modulated oscillator (40) controlled by the output signals of the electrodes for detecting the driving field via a phaselocking loop having low and high frequency stops.

6. A device according to any of the preceding claims, adapted to operate as a gyroscope, characterised in that:

it comprises a measuring circuit having an angular coder (78) connected to two sets of detection electrodes (DX, Dy) corresponding to electrically orthogonal vibrations and supplying a phase reference to a phase-locking loop (76) which in turn gives a reference coding to the coder, and the excitation circuit comprises means for summing the integral of the signal from one set with the signal coming from the other set, a comparator (72) having a reference value, a control multiplier (74) receiving the output signal from the loop (80) and two output multipliers (84, 86) receiving the output signal from the multiplier and controlled by the trigonometrical functions of the evaluation supplied by the angular coding circuit (78).

7. A gyrometric device substantially as hereinbefore described with reference to and as shown in Figs. 2 to 7 of the accompanying drawings.