The invention relates to a method for operation of a vibration gyro, and a sensor arrangement having a vibration gyro, which represents a resonator and is part of at least one control loop which excites the vibration gyro by supplying an exciter signal at its natural frequency, in which case an output signal can be tapped off from the vibration gyro, from which output signal the exciter signal is derived by filtering and amplification.
By way of example, EP 0 461 761 B1 discloses rotation rate sensors in which a vibration gyro is excited on two axes which are aligned radially with respect to a major axis, for which purpose a primary and a secondary control loop are provided, with appropriate transducers on the vibration gyro. When rotation rate sensors such as these are used to stabilize the vehicle movement in vehicles, the rotation rate sensor must operate immediately after the vehicle has been started up. However, this is delayed by the transient response of the control loop.
The object of the present invention is therefore to speed up the transient response of the control loop. This object is achieved according to the invention in that once a sensor arrangement with the vibration gyro has been switched on, an initial value of the natural frequency is calculated from a previously measured value, stored in a memory, of the natural frequency and from parameters which have resulted in a change in the natural frequency since the measurement, and in that the exciter signal is supplied to the vibration gyro with the calculated initial value of the frequency.
The method according to the invention makes it possible to approximate the initial value sufficiently closely to the actual natural frequency that an extremely short time is required for stabilization. In the method according to the invention, the initial value is preferably calculated from the stored value, its temperature dependency and the temperature measured on switch on. If the temperature of the vibration gyro is not constant during the measurement of the value to be stored, with this measurement normally using an adjustment process during the manufacturing process, the value of the temperature during the adjustment process can also be stored. In many cases, the temperature dependency of the natural frequency is constant, so that there is no need to specifically determine and store that during the adjustment process, and, instead, this can be written as a constant to the memory.
In order to speed the stabilization process further, one development of the method according to the invention provides that a phase detector in the control loop has the output signal applied to it in an initial phase, and in that phase control is switched on with a delay. In this case, provision is preferably made for gain control to be switched on once the phase control has been switched on.
This development can be improved by the exciter signal being supplied to the vibration gyro with the maximum permissible amplitude in an initial phase.
Another development of the invention is used to identify malfunctions or failures as quickly as possible. This is achieved
This development preferably provides that a correction temperature, which takes account of the change in the thermal conditions since the adjustment of the temperature sensor, is included in the calculation.
One advantageous refinement of the development consists in that the correction temperature is calculated from the measured power loss of an integrated circuit which contains the temperature sensor, from the power loss measured during adjustment and stored in the memory, and from the thermal resistance of the integrated circuit with respect to the environment.
The sensor arrangement according to the invention has a temperature sensor and a microcontroller, in any case, so that, according to this development, no additional hardware complexity is required to carry out the method, but only an implementation of the measures according to the invention as a program.
In the sensor arrangement according to the invention, means are provided to rapidly stabilize the control loop in that once the sensor arrangement has been switched on, an initial value of the natural frequency is calculated from a previously measured value, stored in a memory, of the natural frequency and from parameters which have resulted in a change in the natural frequency since the measurement, and in that the exciter signal is supplied to the vibration gyro with the calculated initial value of the frequency.
One advantageous refinement of this arrangement preferably consists in that the means comprise a microcontroller with a non-volatile memory and a frequency synthesizer.
In order to monitor the sensor arrangement according to the invention, means can be provided
The measures described in the other dependent claims allow further advantageous developments and improvements of the sensor arrangement according to the invention.
The invention allows numerous embodiments. One of these is illustrated schematically in the drawing and will be described in the following text. The drawing shows a block diagram of a sensor arrangement with a vibration gyro, with the elements which are used to carry out the method according to the invention.
The sensor arrangement as well as parts of it are admittedly shown in the form of block diagrams. However, this does not mean that the sensor arrangement according to the invention is restricted to an implementation based on individual circuits corresponding to these blocks. In fact the sensor arrangement according to the invention can be implemented particularly advantageously using large-scale-integrated circuits. Microprocessors may be used for this. purpose which, when suitably programmed, carry out the processing steps illustrated in the block diagrams.
The sensor arrangement has a vibration gyro 1 with two inputs 2, 3 for a primary exciter signal PD and a secondary exciter signal SD. The excitation is provided by suitable transducers, for example electromagnetic transducers. The vibration gyro also has two outputs 4, 5 for a primary output signal PO and a secondary output signal SO. These signals reflect the respective vibration at physically different points on the gyro. Gyros such as these are known, for example, from EP 0 307 321 A1 and are based on the Coriolis force effect.
The vibration gyro 1 represents a high Q-factor filter, with the path between the input 2 and the output 4 being part of a primary control loop 6, and that the path between the input 3 and the output 5 being part of a secondary control loop, which is not illustrated, since there is no need to explain it in order to understand the invention. The primary control loop 6 is used to excite oscillations at the resonant frequency of the vibration gyro, for example at 14 kHz. The excitation is in this case produced on one axis of the vibration gyro, with respect to which the oscillation direction used for the secondary control loop is offset through 90°. The signal SO is split in the secondary control loop, which is not illustrated, into two components, one of which can be tapped off, after suitable processing, as a signal which is proportional to the rotation rate.
The majority of the signal processing in both control loops is carried out digitally. The clock signals required for signal processing are produced in a crystal-controlled digital frequency synthesizer 10, whose clock frequency in the illustrated example is 14.5 MHz. The method according to the invention is carried out essentially by using the primary control loop, as a consequence of which FIG. 1 illustrates one exemplary embodiment of the primary control loop.
The control loop has an amplifier 11 for the output signal PO, to which an anti-aliasing filter 12 and an analog/digital converter 13 are connected. Multipliers 14, 15, to which carriers Ti1 and Tq1 are supplied, are used for splitting into an in-phase component and a quadrature component. The two components then respectively pass through a (sinx/x) filter 16, 17 and a low-pass filter 18, 19. The filtered real part is supplied to a PID regulator 20 which controls the digital frequency synthesizer, thus closing a phase control circuit which results in the phase angles of the carriers Ti1 and Tq1 being correct. Furthermore, a carrier Tq2 is produced and is modulated in a circuit 22 with the output signal from a further PID regulator 21, which receives the low-pass-filtered imaginary part. The output signal from the circuit 22 is supplied to the input 2 of the vibration gyro 1 as the exciter signal PD. Depending on the preconditions in detail, other regulators, for example PI regulators, may also be provided instead of the PID regulators.
A microcontroller 27 is provided in order to carry out the method according to the invention, controls the individual steps of the method according to the invention and has access to a non-volatile memory 28 which is in the form of EEPROM. Furthermore, a temperature sensor, which is provided in any case for many circuit arrangements and comprises an actual sensor 29 and an analog/digital converter 30, is used for the method according to the invention. A bus system 31 connects the components that have been mentioned to one another and to the digital frequency synthesizer 10 as well as to the circuit 22.
The initial value is calculated in a simple form using the equation F0Ta=(VTA−VRT)/TCv×TCf0+FORT. In this case, F0Ta is the natural frequency of the vibration gyro at the temperature measured by means of the temperature sensor 29, that is to say the initial value, and F0RT is the natural frequency measured at room temperature during the adjustment process, and stored in the memory 28. TCv is the temperature coefficient of the temperature sensor 29, while TCf0 represents the temperature coefficient of the natural frequency of the vibration gyro. VTA and VRT are the two output voltages from the temperature sensor 29 at the temperature on switch on, and at the temperature during adjustment, as stored in the memory 28.
Once the initial value has been calculated, the microcontroller 27 sets the frequency synthesizer 10 to produce an exciter signal with the initial value as the frequency. At this stage, the control loop has not yet being closed, for example by interrupting the clocks Ti1 and Tq1. As soon as the vibration gyro has received the exciter signal, the control loop is closed, to be precise first of all the phase control followed by the gain control.
For monitoring purposes, these variables are read from the memory 28 from time to time during operation and are compared with the respective current natural frequency, taking account of the current temperature (output voltage VTA) measured by the sensor 29. The comparison process is based, for example, on the following equations:
T
adelta1=(F0TA−F0RT)/TCF0.
T
adelta2=(VTA−VRT)/TCv
In this case, Tadelta2 is the temperature change determined by the temperature sensor, Tadelta1 is the temperature change determined from the change in the frequency, TCv is the temperature coefficient, stored in the memory, of the temperature sensor 29, F0TA is the current frequency, FORT is the frequency stored in the memory, and TCF0 is the temperature coefficient, likewise stored in the memory, of the natural frequency of the vibration gyro.
The current value of the natural frequency can be obtained from the respective setting of the divider for the digital frequency synthesizer 10 and its clock frequency. However, it is also possible to calculate the current value using a frequency measurement device which comprises a further amplifier 24, a Schmitt trigger 25 and a counter 26.
Ideally, Tadelta1 and Tadelta2 are the same; if any difference assumes values which exceed a predetermined level, the presence of one of a plurality of possible faults can be deduced and, for example, a fault can be signaled in the form of illumination of a warning lamp, or can be stored in the memory in order to be available for subsequent diagnostic purposes.
In order to take account of any power loss that differs from that during the adjustment process in the circuit which contains at least the primary control loop 6, the current drawn I of the circuit is measured by means of a measurement resistor 32 whose resistance is R. The operating voltage U for the circuit is supplied to a connection 33, and is distributed via a circuit point 34 to the various components. The voltage drop Ui across the measurement resistor 32 is amplified by a factor v in an amplifier 35 and is supplied via a multiplexer 36 to the analog/digital converter 30. The microcontroller 27 then calculates the power loss using the equation P=U*I=U*U1/R*v). During the adjustment process, the power loss PRT and the associated ambient temperature TRT are stored in the memory. The power loss calculated during operation is referred to in the following text as PA.
This can be used to calculate a correction temperature, specifically in the form Tcor=TRT+(PA−PRT)*RTH, where RTH represents the thermal resistance between the circuit and the environment. The equation mentioned above is therefore supplemented to calculate Tadelta2:
T
adelta2cor
=T
adelta2
−T
cor
As already mentioned above, a fault signal can be emitted in the event of inequality, specifically when:
T
adelta1≠Tadelta2cor.
Number | Date | Country | Kind |
---|---|---|---|
10 2005 043 560.2 | Sep 2005 | DE | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP2006/065815 | 8/30/2006 | WO | 00 | 3/12/2008 |