DE3249801C2 - Inductive measurement transducer - Google Patents

Abstract

Published without abstract.

Description

The invention relates to an inductive measuring transducer according to the preamble of patent claim 1.

A known linear position detector of this type (Handbook for electrical measurement of mechanical quantities, 1967, p. 188/189) is designed as a differential transformer sensor, in which there is only a single primary coil, to which two secondary coils are assigned. Although the primary coil can be divided into two winding halves, which are arranged one behind the other in the longitudinal direction, even in these cases both winding halves, which are connected in series, are excited by a single (common) alternating current signal. Two winding halves, which are excited by a common alternating current signal, form only a single coil. The secondary coils are connected in series, either in such a way that their voltages add up or in such a way that their voltages are in opposite phase and therefore subtract. According to the linear displacement of the core, the reluctances of the magnetic circuits of the two secondary coils change differently, causing the amplitude values (peak values) of the AC signals generated by the current in the primary coil in the secondary coils to be different. When the two secondary coils are operated in antiphase, the combined output signal of these two coils corresponds to the difference in the reluctances. The combined output signal of the secondary coils is converted by rectification into a DC signal whose amplitude corresponds to the linear position of the core.

The measuring principle of the known differential transformers is based on the fact that the inductive transfer behavior from the primary coil to each of the secondary coils changes depending on the position of the core. First, an alternating voltage signal is generated, which then requires rectification in order to generate a measuring voltage.

The disadvantage of the known differential transformers is essentially that the amplitude of the output signal is evaluated to determine the position of the core. Such amplitude evaluations are particularly susceptible to interference, since the signal amplitude depends on the coil resistance, among other things. The coil resistances, however, are dependent on the temperature, the length of the line paths, etc. In addition, the amplitude-dependent evaluation is particularly susceptible to interference from interfering external signals.

The invention is based on the object of creating an inductive measuring transducer of the type specified in the preamble of patent claim 1, which is capable of determining the linear position with high accuracy, independently of resistance changes and other influences which act on the amplitude of the output signal.

This object is achieved according to the invention with the features specified in the characterizing part of patent claim 1.

The position detector according to the invention is constructed in such a way that primary coils and secondary coils, as used in the known position detector with a differential transformer, are arranged in such a way that a plurality of primary coils are arranged along the linear movement path of a core. The respective primary coils are individually excited by reference alternating signals which are phase-shifted with respect to one another, so that an output signal is produced in a secondary coil which is phase-shifted in accordance with the linear position of the core. In this way, the linear position of the core is determined by measuring the phase difference between the output signal of the secondary coil and one of the reference signals. Since the linear position is determined by measuring a phase difference, the measurement is not influenced by amplitude changes which arise as a result of disturbances. In this way, a very precise position measurement is carried out. The resolution for detecting the linear position can be very easily improved by increasing the resolution for detecting the phase difference by suitable circuit measures in the phase difference detector. For example, the frequency of a pulse that is used to count the phase angle can be increased. The position detector according to the invention is therefore suitable for applications that require high measurement accuracy.

Preferably, the signal generator has a clock pulse generator which supplies clock pulse signals to a frequency reducer, the output of the frequency reducer is connected to a waveform circuit which generates the reference alternating signals from the output signal pulses of the frequency reducer, and the phase difference detector contains a counter which is clocked by the clock pulse signals of the clock pulse generator and is controlled by a control circuit such that, starting from the point in time at which one of the reference alternating signals passes through a predetermined phase value, it counts the clock pulses until the output signal of the secondary coil has reached the same phase value, and outputs the counter reading then taken as a measure of the angle of rotation position.

The reference alternating signals for the primary coils are formed from the clock pulses of the clock pulse generator after frequency reduction. The same clock pulse signals are used to determine the phase difference of the output signal of the secondary coil in relation to a reference alternating signal, in that the counter counts the number of clock pulses in the time period that arises from the time shift of the respective signals due to the phase difference. The reference alternating signal is therefore coupled to the clock pulses and precisely synchronized. This enables the phase difference and thus the angle of rotation of the rotor to be determined with high accuracy, which essentially only depends on the pulse frequency of the clock pulses. By using the same clock pulses for both the reference alternating signal generation and for determining the phase difference, no additional pulse generator for determining the phase difference or a synchronization device is required.

In the following, embodiments of the invention are explained in more detail with reference to the drawings.

It shows

Fig. 1 is a block diagram of the reference AC signal generator and a phase difference detector included in the position detector,

Fig. 2 is a timing diagram of the phase difference measurement process in the phase difference detector,

Fig. 3 is a block diagram of a modified embodiment of the AC reference signal generator and the phase difference detector of Fig. 1,

Fig. 4 is a block diagram of another embodiment of the phase difference detector according to Fig. 1, in which the phase difference is determined as an analog value,

Fig. 5 is a timing diagram of waveforms of the output signals appearing at various points in the circuit of Fig. 4,

Fig. 6 is a block diagram of an embodiment in which a phase shifter circuit is provided for adjusting the origin between the output of the secondary coil and the phase difference detector in the detector head and

Fig. 7 is a waveform diagram of an output signal of the detector head generated when the core of the detector head in Fig. 1 is moved at a certain speed.

In the detector head of Fig. 1, an axially movable core 3 is inserted into the hollow middle part of a coil body 2. Two primary coils 4A and 4B are wound onto the coil body 2 at axially offset locations. A secondary coil 5 is wound onto a suitable location on the coil body 2 (e.g. in the illustrated embodiment in the middle between the primary coils). The two primary coils 4A and 4B are excited separately by alternating current signals that are phase-shifted relative to one another (e.g. by a sine signal a sin small omega t and a cosine signal a cos small omega t). A voltage is generated in the secondary coil 5, which is made up of a component induced by the primary coil 4A and a component induced by the primary coil 4B.

Since the primary coils 4A and 4B are excited by the phase-shifted alternating voltage signals a sin small omega t and a cos small omega t, a signal Y is generated in the secondary coil, which results from the phase shift or phase modulation of the alternating voltage signal sin small omega t depending on the position l of the core 3. The principle of generating this signal is explained below. If the position of the core 3 is denoted by l and the coupling coefficient of the primary coils with the secondary coil is denoted by x, the result is

x = l / k (1)

where k is a constant depending on the number of turns of the coils, the permeability of the core, etc. Using the coupling coefficient x, the output signal Y of the secondary coil 5 can be expressed by the following equation:

Y = a (1-x)sin small omega t - a (1 + x) cos small omega t (2)

In the above equation it is assumed that a position in which the inductance between the primary coil

4A and the secondary coil 5 is in equilibrium with the inductance between the primary coil 4B and the secondary coil 5, the zero position of the core 3 (ie l = 0), and that the mutual inductance in this position is 1. If one inserts a (1-x) = A and a (1 + x) = B in equation (2), this results in

The output signal Y of the secondary coil 5 is therefore an alternating voltage signal which is phase-shifted by large Phi compared to the reference signal A sin small Omega t. The phase difference large Phi is, as equation (5) shows, a function of x and x is in turn a function of the position l of the core 3. By measuring the phase difference large Phi between the output signal Y of the secondary coil 5 and the reference signal sin small Omega t (or cos small Omega t), the position l of the core 3 can be obtained on the basis of the phase difference large Phi. In other words, when equation (5) is solved for x, x can be expressed as a function f (large Phi) by using the phase difference large Phi as a variable, i.e. the position l can be expressed using the function f (large Phi) and on the basis of equation (1) as

l = k x f(large Phi) (7)

An example of a circuit for determining the position l of the core 3 by measuring the phase difference large Phi between the output signal Y of the detector head 1 and the reference signal a sin small Omega t is shown in Fig. 1. In Fig. 2, the reference signal generator 13 generates an alternating voltage signal which is fed to the primary coils 4A and 4B of the detector head 1. The phase difference detector 14 serves to measure the phase difference large Phi. In the example according to Fig. 1, the circuit 14 measures the phase difference large Phi digitally. An oscillator 15 generates a high-frequency pulse clock CP. The frequency divider circuit 16 divides the frequency of the pulse clock CP by M and generates pulses Pb with a division ratio of 50% and inverted pulses Pa. M is an integer. More specifically, the frequency divider circuit comprising a 2/M frequency divider 17 and a ½ frequency divider circuit 18 generates at the 2/M frequency divider 17 the pulse clock Pc which corresponds to the pulse clock CP after frequency division by a factor of 2/M, and this pulse clock Pc is divided by two by the flip-flop circuit 18. As a result, the flip-flop circuit 18 outputs rectangular pulses with a pulse division ratio of 50% and the frequency of the pulse clock CP divided by M at the output Q and rectangular pulses in the form of the inverted pulse clock Pb at the inverted output Q [with tilde]. The pulse clocks Pb and Pa which are 180° out of phase with each other are supplied to the ½ frequency divider flip-flops 19 and 20, respectively, which produce pulse trains with frequencies ½ Pb and ½ Pa. Note that the pulse trains with frequencies ½ Pb and ½ Pa respectively output from the flip-flop circuits 19 and 20 have a 2M-th of the frequency of the pulse clock CP and are 90° out of phase with each other. The pulse trains in frequencies ½ Pb and ½ Pa are respectively supplied to low-pass filters 21 and 22 to obtain the fundamental wave components. It is assumed that the low-pass filter 21 outputs a cosine wave signal cos small omega t. In this case, the low-pass filter 22 necessarily produces a sine wave signal sin small omega t. The signal cos small omega t output by the low-pass filter 21 is amplified by the amplifier 23 to produce a signal a cos small omega t which is fed to the primary coil 4B. The signal sin small omega t output by the low-pass filter 22 is amplified by the amplifier 24 and fed to the primary coil 4A as a signal a sin small omega t.

As explained above, the alternating voltage signal Y = K sin (small omega t - large phi) is obtained at the secondary coil 5, which is phase-shifted by a phase angle large phi corresponding to the position l of the core 3 compared to the signal K sin small omega t. The output signal Y is fed to a polarity discriminator 26 via an amplifier 25. One of the reference signals a sin small omega t is fed to a further polarity discriminator 27 via the amplifier 24. The polarity discriminators 26 and 27, which consist of comparators, output a "1" signal when the amplitude of the input signal

(K sin (small Omega t - large Phi), a are small Omega t)

of positive polarity and a "0" signal if this amplitude is of negative polarity.

The output signals of the polarity discriminators 26 and 27 are respectively fed to rising edge detection circuits 28 and 29, which are monostable multivibrators that output a short pulse when the input signal goes to "1".

Therefore, when the phase angle (small omega t - large phi) of the output signal Y of the detector head 1 is 0°, the rising edge detector 28 outputs a rising edge detection pulse Ts as shown in Fig. 2. On the other hand, when the phase angle small omega T of the reference signal a sin small omega t is 0°, the rising edge detector 29 outputs a rising edge detection pulse T0. The signal Y = K sin (small omega t - large phi) lags the reference signal a sin small omega t by a phase angle large phi corresponding to the position I. Therefore, the rising edge detection pulse Ts follows the rising edge detection pulse T0 with a delay corresponding to the phase difference large phi.

The value of the phase difference large Phi can be obtained by counting the time interval between the rising edge detection pulses T0 and Ts using a counter 30 to which the pulse clock CP of the oscillator 15 is fed. The counter 30 is fed as a reset pulse the pulse T0 which corresponds to the phase angle 0 of the reference signal a sin small Omega t. The counter 30 is therefore reset each time the reference signal a sin small Omega t assumes the phase 0.

The output signal of the counter 30 is fed to a buffer register 31 which receives at its sampling input a pulse Ts representing the phase angle small omega t large phi = 0 of the signal K sin (small omega t - large phi). The count of the counter 30 is transferred to the buffer register 31 each time the pulse Ts is generated, so that the buffer register 31 receives the count D[low]large phi corresponding to the phase difference large phi.

Although it is possible to take the count D[deep]Phi of the buffer register 31 directly as the position value I, this count value D[deep]Phi can also be converted into the position value I using a suitable function generator. The relationship between the phase shift large Phi and the position I is calculated, for example, as follows:

1. If the position I = - k, the coupling coefficient x = -1. Therefore,

large Phi = cos[to the power of]-1 1 = 0.

2. If the position I = 0, the coupling coefficient x = 0. Therefore,

3. If the position I = k, the coupling coefficient x = 1. Therefore

From the above calculation example, it can be seen that the relationship between the phase shift large Phi and the position I is almost linear when the position I of the core 3 is in the range of -k <= I <= k. Since the detector head 1 is intended to indicate the position very finely, there is no difficulty in restricting the displacement range of the core 3 to the order of magnitude of -k <= 1<= k. Accordingly, the value D[deep]Phi, which indicates the phase shift large Phi, can be used directly as the value indicating the position I of the core 3. In this case, the zero position (I = 0), i.e. the neutral position, should not be taken to be the origin of the core 3, but a position in which the phase shift large Phi is "0", i.e. in the above calculation example, the position I = k. In the case that the phase shift value D[deep]large Phi is converted to determine the position I by a function generator, the function generator can, for example, consist of a read-only memory in which the functions of equations (5) and (1) are stored.

The generator circuit 3 for generating the AC voltage reference signal and the phase difference detector 14 are not limited to the embodiments of Fig. 1, but can also be designed in a different way.

Fig. 3 shows an example in which the 2/M frequency divider 17 of Fig. 1 is omitted, while the reference signal generator 13A and the phase difference detector 14A together contain the counter reading 30 which counts modulo 2/M. In Fig. 3, circuits which perform similar functions to those of Fig. 1 are provided with the same reference numerals. The bit whose weight corresponds to a quarter of that of the most significant bit, namely the output signal of the 2/M frequency divider stage, is fed to the flip-flop 18 as a pulse Pc. On the basis of these pulses Pc, the sine signal a sin small omega t and the cosine signal a cos small omega t are generated by the circuits 18 to 24 as in the embodiment of Fig. 1. The output signal Y = K sin (small Omega t - large Phi) of the detector head 1 is processed by the circuits 25, 26 and 28 as in the case of Fig. 1 and as a result the pulse Ts corresponding to the output signal Y in phase 0 is fed to the sampling control input of the register 31. The count of the counter 30 is fed to the data input of the register 21. The digital value D[low]large Phi corresponding to the phase difference large Phi is stored in the register 31 in this way as in the embodiment according to Fig. 1.

While the value D[deep]large Phi is determined in digital form by the phase difference detectors 14 and 14A according to the embodiments of Fig. 1 and 3, this determination can also be carried out in analogue technology according to Fig. 4. According to Fig. 4, the reference signal a sin small Omega t and the output signal Y of the detector head 1 are fed to a phase difference detector 14B. If the output signal Y has the waveform shown in Fig. 5a, the polarity discriminator 50 outputs a "1" signal in response to a positive polarity and a "0" signal in response to a negative polarity, as shown in Fig. 5b. The rising edge detector 51 outputs the short pulses shown in Fig. 5c at those times at which the output signal b of the polarity discriminator has a rising edge. The reference signal a sin small omega t shown in Fig. 5d is converted into a square wave by the polarity discriminator 52 and then supplied to the ½ frequency divider 53 to produce the output signal f which goes to "1" for one period of the reference signal a sin small omega t and to "0" for the next period. The output signal f of the ½ frequency divider 53 is supplied to an integration circuit 54 to produce an analog voltage g corresponding to the length of time elapsed from the rising point or falling point of the output signal f of the frequency divider circuit as shown in Fig. 5g. The output signal g of the integration circuit 54 is supplied to a sample and hold circuit 55 to be sampled at the instants when the phase angle of the signal Y = 0°. The sampling control input of the sample and hold circuit 55 receives the output signal c of the rise detection circuit 51 via a gate 56 which is opened so that the sample and hold circuit 55 can receive the output pulse c of the rise detection circuit 51 when the output signal f of the ½ frequency divider 53 is "1", but blocks the pulse c when the output signal f is "0". The gate 56 serves to suppress the sampling of the negative branch of the output signal g which the integration circuit 54 generates when the output signal f of the 1/2 frequency divider circuit 53 is "0" according to Fig. 5g. Accordingly, a sampling pulse h is fed to the circuit 55 via the gate 56 when the output signal g of the integration circuit 54 has a positive slope, as shown in Fig. 5h. The sample and hold circuit 55 thus performs the sampling every other cycle to produce a DC voltage V[low] large Phi which is analogous to the phase difference between the reference signal a sin small Omega t and the output signal Y of the detector head.

When attaching the detector head 1 to an object, it is difficult to align the origin of the detector head 1 exactly with that of the object, so that assembly errors can occur. In order to correct such errors in the mechanical zero point adjustment, a phase shift circuit 12 is preferably provided between the detector head 1 and the phase difference detector 14. The phase shift circuit 12 outputs

a signal that is generated by shifting the input signal Y by an adjustable phase angle and whose amount of phase shift can be set on an actuator. The zero point adjustment is carried out with this phase shifter holder 12 in the following way: First, the position of the object to be detected is set as the origin and the detector head 1 is mounted on the object. If the origin of the detector head 1 matches the origin of the object, the output signal Y of the detector head 1 indicates the position for the origin (i.e. the phase shift large Phi is 0). If there is an assembly error, the output signal Y shows the phase difference large Phi corresponding to the error. In this case, the actuator 11 is manually adjusted to set the amount of phase shift in the phase shift circuit 12 so that the phase shift circuit 12 outputs a signal Y' which indicates the position for the origin or zero point (i.e., the phase difference large Phi of this signal Y' with respect to the reference signal a sin small Omega t 0).

In other words, phase adjustment is performed in the phase shift circuit 12 so that the phase difference caused by the assembly error is eliminated. For example, the amount of phase shift can be changed by adjusting a variable resistor in the phase shift circuit 12 by means of the actuator 11, thereby regulating the time constant of the phase shift circuit 12.

When the core 3 moves, the speed and acceleration of the movement of the core 3 can be obtained on the basis of the output signal Y of the detector head 1. When the core 3 is displaced as a function of time t, the phase difference large Phi of the output signal Y with respect to the reference signal a sin small Omega t also varies with time. The phase difference large Phi in equation (6) can therefore be expressed as a function large Phi (t) of time according to the following equation (8):

(8th)

If the angular velocity of the magnitude of the phase difference large Phi (t) is denoted by small Omega[deep]M, then

(9)

Since the integral of the angular velocity small Omega[low]M corresponds to the phase difference large Phi (t), equation (8) can be rewritten as

(10),

where large Phi[low]0 denotes the initial phase.

On the other hand, if the angular acceleration of the magnitude of the phase difference large Phi (t) is denoted by small Alpha [low]M, then

(11)

and thus

(12)

Equation (8) can therefore be rewritten as

(13)

An example of the output signal Y of the detector head 1 when the phase difference large Phi (t) is generated at the angular velocity small Omega [low]M is shown in Fig. 7. In the embodiment according to Fig. 7, the initial phase large Phi [low]0 is considered to be 0 for better explanation. In Fig. 7, the reference signal a sin small Omega t is also shown. The reference symbol t[low]0 denotes a period of the reference signal and t[low]s denotes a period of the output signal Y of the detector head. As can be seen from Fig. 7 and equation (10), the frequency of the output signal Y deviates from the reference frequency small Omega and the amount of the deviation corresponds to the angular frequency small Omega[low]M. If the frequency of the output signal Y of the detector head is denoted by small Omega[low]s, equation (10) can be written as

(14)

This means that the frequency small Omega[low]s = small Omega - small Omega[low]M and that the angular velocity small Omega[low]M can be written as

(15)

The angular velocity small Omega[deep]M is thus obtained by calculating one cycle t[deep]s of the output signal Y of the detector head and by obtaining the operation result of t[deep]s and the reference cycle t[deep]0.

In detail, t[low]s is obtained by counting one cycle of the output signal Y with the clock pulse CP. If the count value corresponding to t[low]s is n[low]s and one period of the pulse contact CP is T(sec), then

(16)

The one period t[low]0 of the reference signal a sin small Omega t is known in advance and equation (15) can be written as follows:

(17)

Here, n[deep]0 denotes the count value of the pulse clock CP corresponding to one cycle t[deep]0. Since 2 small Pi, T and n[deep]0 are constants, the angular velocity small Omega[deep]M can be obtained by counting one cycle of the output signal Y of the detector head and by calculating according to equation (17) on the basis of this count value n[deep]s.

The following relationship exists between the angular acceleration small Alpha[deep]M and the angular velocity small Omega[deep]M:

(18)

where large delta small omega[deep]M is the amount of change in the angular velocity small omega[deep]M during the time interval large delta t. If the angular velocity at time t[deep]1 is small omega[deep]M 1 and the angular velocity at time t[deep]s, which is later than t[deep]1 by the time period t[deep]2, is small omega[deep]M 2, then:

Since according to equation (16) t[deep]s = n[deep]sx T, equation (18) can be rewritten as:

(19)

The angular velocity small Alpha[low]M can therefore be calculated by measuring the angular velocity small Omega[low]M for each individual period t[low]s of the output signal Y and then determining the difference between the angular velocity small Omega[low]M 2 and the angular velocity small Omega[low]M 1 and then dividing this difference by the product of the count value n[low]s and the period T of the pulse clock CP.

A circuit for counting a period t[low]s of the output signal Y of the detector head 1 and for realizing equations (17) and (19) is not shown, but such a circuit can be easily realized according to the above explanation.

Claims (7)

1. Inductive measuring transducer for detecting linear positions with at least two primary coils, at least one secondary coil and with a core that can be moved relative to the coils, wherein a) the second primary coil is supplied with a reference alternating current signal from a reference signal generator which has a phase difference to the alternating current signal of the first primary coil, and b) an output alternating current signal is generated in the secondary coil, c) and wherein the phase difference between the reference and the output alternating current signal serves as a measure of the position of the core, characterized in that d) the phase difference detector (14, 14a, 14b) measures the phase difference between one of the reference signals at the primary coil (4a) and the output AC signal of the secondary coil (5). 2. Inductive measuring transducer according to claim 1, characterized in that the signal generator (13, 13A) has a clock pulse generator (15) which supplies clock pulse signals (CP) to a frequency reducer (16, 17, 30), that the output of the frequency reducer (16, 17, 30) is connected to a waveform circuit (18 to 24) which generates the reference alternating signals from the output signal pulses of the frequency reducer (16, 17, 30), and that the phase difference detector (14, 14A) contains a counter (30) which is clocked by the clock pulse signals (CP) of the clock pulse generator (15) and is controlled by a control circuit (26 to 29) in such a way that, starting from the point in time at which one of the reference alternating signals passes through a predetermined phase value, it counts the clock pulses until the output signal (Y) of the secondary coil reaches the same phase value and outputs the counter reading as a measure of the angle of rotation position. 3. Inductive measuring transducer according to claim 2, characterized in that the control circuit (26 to 29) contains a first amplitude discriminator (26, 28) for the output signal (Y) of the secondary coil. 4. Inductive measuring transducer according to one of claims 2 or 3, characterized in that the control circuit (26 to 29) contains a second amplitude discriminator (27, 29) for a reference alternating signal. 5. Inductive measuring transducer according to claim 3 or 4, characterized in that the first and/or the second amplitude discriminator is a zero-crossing detector. 6. Inductive measuring transducer according to one of claims 2 to 5, characterized in that the counter (30) is connected to a holding circuit (31) which takes over the counter reading when the output signal (Y) of the secondary voltage has reached the predetermined phase value. 7. Inductive measuring transducer according to claim 6, characterized in that the counter (30) is connected downstream of the signal generator (13A) as a frequency reducer and gives a signal pulse to the waveform circuit (18 to 24) after a predetermined number of clock pulse signals (CP) and that the holding circuit (31) takes over the count value of the counter when the predetermined phase value of the output signal (Y) of the secondary coil is reached.