Magnetic detecting circuit having magnetoresistance effective elements oriented in at least two different directions

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a magnetic detection device which converts a magnetic state to an electric signal utilizing a magnetoresistance effective element (MRE), whose electric resistance is determined by the magnetic state, for speed detection by magnetic changes etc. and to a physical quantity detection device utilizing the same.
Especially, this invention relates to a magnetic detection device for detecting the amount of magnetic change with a high accuracy and sensitivity utilizing a means which converts the change of the magnetism to a digital pulse signal.
2. Description of the Related Art
Generally speaking, a position sensor Dr a rotational speed sensor needs to have a high level of detection accuracy, a wide range of ambient temperature of use, and simple construction. Use is now made of sensors working by detecting magnetism. The reason is that an MRE has a high sensitivity to magnetism, is relatively stable against temperature changes, and can be easily produced. In addition, in an electronic circuit for converting a magnetic state into an electric signal, the magnetic state directly determines the resistance, so it is important that the design work of the detecting on circuit also be easy. These characteristics make it effective to mount both the MRE and detection circuit on one chip.
In a conventional sensor or circuit, use is made of a resistance bridge circuit, widely used for detecting a resistance value, and a voltage amplifying circuit, which can detect and amplify an output voltage.
The output voltage of the resistance bridge circuit is stable against temperature changes. The voltage amplifying circuit is an analog circuit, however, so when the temperature changes, the characteristics of the transistor in the voltage amplifying circuit changes and the resistance of the resistor shifts, whereby an offset voltage is generated. Therefore, the range of ambient temperature in which the voltage amplifying circuit can correctly work is relatively narrow.
This problem can be alleviated to a certain extent by utilizing a temperature compensating circuit, but the construction of the circuit becomes complicated and the production efficiency is reduced. Also, when the temperature increases to a much higher level, the temperature compensating circuit itself cannot work correctly any more. Therefore, the circuit is not suitable for a circuit used, for example, for detection of rotational speed in automobiles, when the range of ambient temperature of use is wide.
Note that when this kind of magnetic detection device is used as a sensor for detecting the rotational speed of a wheel of an automobile, especially used for an antilock brake system (ABS), a wide range of temperatures under which the device can be guaranteed to normally operate is required, because the sensor is usually mounted near a brake, which is usually heated to a high temperature.
In the future, sensors will have to be built on one chip and the range of ambient temperature of use of a detection circuit will become wider, so a detection circuit which can operate stably in a wide range of ambient temperatures of use will become necessary.
Note also that the mounting position of the magnetic sensor greatly affects the output signal and the adjustment thereof is very difficult.
SUMMARY OF THE INVENTION
The present invention was created to overcome the problems mentioned above. The primary object of the present invention is to provide a magnetic detection device which can stably work in a much wider range of ambient temperatures and is suitable for being made by an integrated circuit.
Another object of the present invention is to improve the sensitivity of the magnetic detection device.
A still other object of the present invention is to provide a magnetic detection device which can detect movement of an object to be detected in a good condition regardless of the distance between the device and the object.
A still other object of the present invention is to provide various physical quantity detection circuits.
According to the present invention, there is provided a magnetic detection device which comprises at least one oscillating circuit having a magnetoresistance effective element (MRE) and a comparator for comparing an oscillating frequency of the oscillating circuit with another oscillating frequency to detect magnetic changes. According to the present invention, there is further provided a physical quantity detection device comprising a detecting device for detecting a magnetic change comprising at least one oscillating circuit having an MRE and a comparator for comparing an oscillating frequency of the oscillating circuit with another oscillating frequency and outputting a pulse signal indicating a magnetic change, and a converter for converting the output pulse signal output from the comparator to a pulse signal indicating a change of the physical quantity. Moreover according to the present invention, there is provided another type of a magnetic detection circuit and a physical quantity detection circuit utilizing the same, which comprises;
a magnetic field generating means for generating a bias magnetic field toward an object to be detected having magnetic materials therein and MREs provided on a predetermined plane arrange in the bias magnetic field, a resistance value thereof being changed in response to a change of conditions of the bias magnetic field due to a movement of the object to be detected,
the change of conditions of the bias magnetic field being detected by the change of the resistance value of said MRES,
the predetermined plane formed by a direction of the bias magnetic field generated from the magnetic field generating means and a moving direction of the object to be detected,
the MREs arranged on the plane so that the change of the resistance value displays at least one of a monotonous increment and monotonous reduction due to said change of the bias magnetic field.
Further according to the present invention, there is provided a different type of a magnetic detection circuit utilizing a pulse phase difference which comprises;
a detecting circuit for detecting pulse phase difference, which comprises
a delayed pulse signal output means comprising a signal delay circuit which outputs a plurality of pulse signals when a first pulse signal is input thereto, each pulse signal being delayed from the previous pulse signal by a predetermined different delay time,
a pulse signal selecting means which receives as input second pulse signal delayed from the first pulse signal by any desired time and a plurality of pulse signals output from the delayed pulse signal output means and which selects from the plurality of pulse signals, a certain pulse signal having a specific condition with respect to a timing at which the second pulse signal is input, and
a detecting means for detecting a phase difference between the first and the second pulse signals utilizing the pulse signal selected by the pulse signal selecting means.

BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a cross-sectional view of a construction of a magnetic detection device of a first aspect of the present invention;
FIG. 2 is a cross-sectional view of a construction of a molded chip;
FIGS. 3(a) to 3(d) are cross-sectional views showing a method of producing a thin portion in an insulating substrate;
FIG. 4 is a block diagram of the detection circuit of the first aspect of the present invention,
FIGS. 5 and 6 are timing charts each indicating a timing condition of a block output signal in the circuit;
FIG. 7(a) shows an embodiment of an oscillating circuit used in the present invention, and FIG. 7(b) shows an operational waveform obtained in the circuit shown in FIG. 7(a);
FIG. 8 shows one embodiment of a timing comparator;
FIG. 9 shows one embodiment of a hysteresis control circuit;
FIG. 10 is a timing chart showing an operation of the hysteresis control circuit:
FIG. 11 is a block diagram of another detection circuit of the first aspect of the present invention;
FIG. 12(a) shows one embodiment of an oscillating circuit used in the circuit of FIG. 11, and FIG. 12(b) shows an operational waveform obtained in the circuit shown in FIG. 12(a);
FIGS. 13(a) and 13(b) show results of experiments on the sensitivity of an oscillating circuit;
FIG. 14 shows another magnetic detection device of the first aspect of the present invention;
FIGS. 15(a), 15(b), and 15(c) show another embodiment oscillating circuit used in the circuit of FIG. 11;
FIGS. 16(a) to 16(f) are illustrating figures explaining a method making a thin portion in an insulating substrate;
FIGS. 17(a) to 17(c) show a circuit construction in which an MRE and a conductive layer are formed on the insulating substrate;
FIG. 18 is a graph of a charge/discharge operation at point H of the oscillating circuit shown in FIG. 7(a),
FIG. 19 is a graph of a charge/discharge operation at point D of the oscillating circuit shown in FIG. 12(a),
FIG. 20 is a graph of the relationship between the sensitivity and the resistance ratio R.sub.2 /R.sub.3,
FIGS. 21(a) and 21(b) are a front view and a perspective view, respectively, indicating an arrangement among a magnet, an insulating substrate on which MREs are mounted, and an object to be detected in a conventional device;
FIG. 22(a) is a perspective view of a magnetic field angle in a plane perpendicular to the insulating substrate on which MREs are formed and parallel to the longitudinal direction of the MRE;
FIG. 22(b) is a perspective view of a magnetic field angle in a plane perpendicular to the insulating substrate on which MREs are formed and perpendicular to the longitudinal direction of the MRE;
FIG. 22(c) shows the change in resistance of the MREs with respect to the magnetic field angle in FIG. 22(a) and 22(b);
FIG. 23 shows a conventional circuit construction use in a magnetic detection device utilizing an MRE;
FIG. 24(a) shows the change in resistance of the MREs and an output of the magnetic detection device with respect to the magnetic field angle in the conventional device shown in FIGS. 21(a) and 21(b) when the air gap is large;
FIG. 24(b) shows the change in resistance of the MREs and an output of the magnetic detection device with respect to the magnetic field angle in the conventional device shown in FIGS. 21(a) and 21(b) when the air gap is small;
FIG. 25 is a cross-sectional view of the magnetic detection device of the third aspect of the present invention,
FIG. 26 shows the mutual relationship of arrangements among a magnet, an insulating substrate on which MREs are mounted, and an object to be detected in the third aspect of the present invention,
FIG. 27(a) is a schematic view of a magnetic field angle in an insulating substrate on which an MRE is formed;
FIG. 27(b) is a graph of the change in resistance of the MRE with respect to the magnetic field angle in FIG. 27(a);
FIG. 28(a) is a graph of the change in resistance of the MRE and an output of the magnetic detection device with respect to the magnetic field angle with a large air gap;
FIG. 28(b) is a graph of the change in resistance of the MRE and an output of the magnetic detection device with respect to the magnetic field angle with a small air gap;
FIGS. 29(a), 29(b), and 29(c) show patterns of arrangement of MREs on the insulating substrate of the magnetic detection device shown in FIG. 25;
FIGS. 30(a), 30(b), and 30(c) show comb-type patterns of arrangement of MREs on the insulating substrate of the magnetic detection device;
FIG. 31 shows the mutual relationship of arrangements among a magnet, an insulating substrate on which MREs are mounted, and an object to be detected in the device shown in FIG. 26;
FIG. 32 shows a circuit construction of a pulse phase difference detection circuit of a fourth aspect of the present invention;
FIG. 33 is a time chart illustrating the operation of the pulse phase difference detection circuit shown in FIG. 32;
FIG. 34 is a block diagram of a magnetic detection device used for a physical quantity detection device of a fifth aspect of the present invention;
FIG. 35 is a timing chart illustrating an operation of the magnetic detection device shown in FIG. 34;
FIGS. 36(a), 36(b), and 36(c) are graphs illustrating the theory behind the magnetic detection circuit shown in FIG. 34;
FIG. 37 shows a construction of a comparison circuit used in the magnetic detection circuit shown in FIG. 34;
FIG. 38 shows a construction in which a Hall device is used instead of the MRE in the magnetic detection circuit shown in FIG. 34;
FIG. 39 shows of another embodiment of the signal delay circuit of the present invention;
FIG. 40 shows another embodiment of the oscillating circuit used for the magnetic detecting device of the present invention;
FIG. 41 shows operational waveforms of the circuit shown in FIG. 40;
FIG. 42 is a chart indicating the transistor size of each inverter in FIG. 40;
FIG. 43 is a chart indicating the transistor size of each inverter in a conventional circuit;
FIG. 44 shows waveforms indicating currents flowing in the inverters having the transistor sizes indicated in FIG. 43;
FIG. 45 shows waveforms indicating currents flowing in the inverters having the transistor sizes indicated in FIG. 42;
FIG. 46(a) is a cross-sectional view of an integrated circuit used for the magnetic detection device of the present invention; and
FIG. 46(b) is a plane view of the construction in the case shown in FIG. 46(a).

DESCRIPTION OF THE PREFERRED EMBODIMENTS
Embodiments of the present invention will be explained with reference to attached drawings hereunder.
FIG. 1 shows the construction of a magnetic detection device of a first aspect of the present invention. In the magnetic detection device, a magnetic sensor is arranged between a bias magnet and a gear made of a magnetic material. It is suitably used for detecting a rotational speed of a wheel of an automobile with ABS.
The magnetic detection device of the first aspect basically includes a molded chip 1, an MRE molded therein with resin, an inner case 2 made of a metallic material, an external case 3, a bias magnet 4, an output pin 5, and a connector 6. The bias magnet 4 is inserted into a grooved portion 7 provided in the connector 6, and the inner case 2 is fixed to the connector 6 with a suitable adhesive, not shown in FIG. 1. The molded chip 1 is fixed to the portion just on the upper surface of the bias magnet 4 with a suitable adhesive, not shown in FIG. 1, and is also connected to the output pin 5 which is integrally mounted within the connector 6 by welding or the like. Finally, the outer case 3 is press fit to the connector 6, and the end 8 of the case 3 is calked.
In FIG. 1, a rotating gear 9 made of a magnetic material is provided as an example of an object to be detected.
Note that when the magnetic detection device is used for detecting the rotational speed of a wheel, the shaft axis of the wheel penetrates into the center of the rotating gear 9.
The construction of the molded chip 1 is indicated in FIG. 2. As shown in FIG. 2, an MRE 1b made of Ni--Co or Ni--Fe is formed in a U-configuration by a photoetc.hing method or the like on a surface of a glass plate or an insulating substrate 1a made of high resistance Si or made of Si with an oxide film formed on a surface thereof. One end of a conductive layer 1c is arranged to come into contact with the MRE 1b. The insulating substrate la provided with the MRE 1b and the conductive layer 1c is mounted on a lead frame 1d by a die bonding paste 1e, while another end of the conductive layer 1c is connected to the lead frame 1d with a bonding wire 1f. After that, the whole device is filled with a molding resin utilizing the injection molding method.
When the insulating substrate 1a is Si, a thin portion li may be formed in the substrate 1a other than at the portion on which the MRE 1b is intended to be arranged, utilizing an etc.hing method, providing a tapered portion between them. The bonding wire 1f as mentioned above is connected to this thin portion 1i of the substrate.
When a substrate made of Si having an insulating film on the surface thereof is used as the insulating substrate 1a, not only the oscillating circuit explained later but also other circuits may be simultaneously formed inside or on the surface of the silicon substrate 1a.
FIGS. 3(a) to 3(d) show a method to form the thin portion 1i on the insulating substrate made of Si.
In this embodiment, the crystal plane (100) is used as the surface direction of the silicon substrate.
In FIG. 3(a), first, an oxide film 1j is formed on a portion of the surface of the Si substrate 1a as shown by hatching. The cross-section thereof is shown in FIG. 3(b). Anisotropic etc.hing using an alkaline substance is applied to the substrate so that a thin portion 1i is formed leaving the thick portion 1h unchanged as shown in FIG. 3(c) and a tapered portion 1k having an angle of about 54 degrees is formed between them. The cross-section thereof is shown in FIG. 3(d). After that, an oxide film is formed on the surface thereof to provide the insulating substrate used in this invention.
FIG. 4 is a block diagram of a detection circuit for the main element of this device. This circuit outputs pulses such as a high level and a low level indicating changes in the magnetic field affecting the MRE. This system can be used as a detection circuit for a rotational speed sensor, for example.
FIGS. 5 and 6 are timing charts indicating timing conditions of each signal output from each block in the block diagram of FIG. 4 and show the operating theory of this embodiment.
The MRE's 10 and 20 shown in FIG. 4, which have the same pattern and the same size, are formed on the surface of the insulating substrate 1a as shown in FIG. 2 and mounted on one chip in such a way that the longitudinal directions of one pattern are arranged perpendicular (90.degree.) to those of other longitudinal directions of another pattern so that the resistance of each MRE changes in the opposite direction from each other with respect to the effect of the same magnetic field.
Oscillating circuits 30 and 40 have oscillating frequencies defined by the resistances of the MREs 10 and 20 and have output clock pulses B.sub.C and A.sub.C provide frequencies equal to those frequencies. The clock pulses B.sub.C and A.sub.C are input to counters 50 and respectively. The counters output divided pulses A.sub.0 and B.sub.0 which are obtained by dividing the clock pulses B.sub.C and A.sub.C by one-eighth, for example, respectively.
A timing comparator 70 detects the interrelation between timings of rising edges of the pulse A.sub.0 and the pulse B.sub.0 utilizing the pulse A.sub.0 and the pulse B.sub.0 output from the counter 50 as input data. The timing comparator 70 outputs a high level signal (H) when the timing of the rising edge of the pulse B.sub.0 is later than that of the pulse A.sub.0 and outputs a low level signal (L) when the timing of the rising edge of the pulse B.sub.0 is earlier than that of the pulse A.sub.0. Accordingly, one pulse can be obtained in one cycle of change of a magnetic field or magnetism, and the frequency of the output pulse coincides with the frequency of the magnetic change. Therefore, when a constant relationship exists between the rotating frequency and the frequency of the magnetic change, the frequency of the output pulse C.sub.0 indicates the rotating frequency.
FIG. 5 shows the timing relationship among the clock pulses B.sub.C and A.sub.C, the divided pulses A.sub.0 and B.sub.0 output from the counter 50 and 60, respectively, and the output pulse C.sub.0 output from the timing comparator 70 when the frequency of the clock pulses A.sub.C is higher than that of the clock B.sub.C.
The clock pulses B.sub.C and A.sub.C are divided, for example, by one-eighth in the counters 50 and 60, respectively. Therefore, a phase difference between the clock pulses B.sub.C and A.sub.C caused by a frequency difference T.sub.C can be obtained as the phase difference T.sub.D between the divided pulses A.sub.0 and B.sub.0 output from the counters 50 and 60, respectively.
Note that T.sub.D can be obtained as a phase difference eight times the frequency difference T.sub.C, so the pulse phase difference can be detected stably.
The timing comparator 70 outputs signal C.sub.0 corresponding to the output signal level (H) of the divided pulses A.sub.0 at the time when the level of the divided pulses B.sub.0 increases. Accordingly, the output signal C.sub.0 is the high level when the level of the divided pulses B.sub.0 increases.
FIG. 6 shows the timing relationship among the clock pulses B.sub.C and A.sub.C, the divided pulses A.sub.0 and B.sub.0 output from the counters 50 and 60, respectively, and the output pulse C.sub.0 output from the timing comparator 70 when the frequency of the clock pulses B.sub.C is higher than that of the clock A.sub.C.
As explained above, the timing comparator 70 outputs signal C.sub.0 corresponding to the output signal level (L) of the divided pulses A.sub.0 at the time when the level of the divided pulses B.sub.0 increases. Accordingly, the output signal C.sub.0 is the low level when the level of the divided pulses B.sub.0 increases.
FIG. 7(a) shows one example of a circuit construction of the oscillating circuit 30 and 40 and further explains the application of the known oscillating circuit to this embodiment. FIG. 7(b) shows operational waveforms obtained at the points K, H, I, and J indicated in FIG. 7(a). Serially arranged inverters 31, 32, and 33 are provided, and an input of the inverter 31 is connected to an output of the inverter 33 through an MRE 10 or 20 (having resistances of R.sub.0). Further, an input of the inverter 31 is connected to an output of the inverter 32 through a capacitor 34.
According to this circuit, the oscillating frequency f.sub.0 is defined by the following equation; ##EQU1##
When the threshold value V.sub.TH of the inverters 31, 32, and 33 is V.sub.TH =1/2 VDD , the frequency f.sub.0 is as follows: ##EQU2##
Note that the frequency f.sub.0 above is determined by the charge/discharge time a time constant defined by the resistance R.sub.0 of the MRES 10 and 20 and the capacitance of the capacitor 34.
FIG. 8 shows one example of a circuit construction which can be used in the timing comparator 70. In that, the output A.sub.0 of the divided pulses from the counter 60 is input to a data input terminal 72 of a D-type flip-flop 71 triggered by a rising edge of an input pulse, while an output pulse B.sub.OH from a hysteresis control circuit 74 is input to the clock input terminal 73 of the D-type flip-flop 71.
The D-type flip-flop 71 outputs a signal corresponding to the output signal A.sub.0 of the divided pulses input to the input terminal 72 of the flip-flop 71 when the output B.sub.OH of the hysteresis control circuit 74 is increased. Therefore, the outputs C.sub.0 of the timing comparator as shown in FIGS. 5 and 6 correspond to the pulse signal output from the output terminal 75 of the D-type flip-flop 71 shown in FIG. 8.
FIG. 9 shows one example of a circuit construction of the hysteresis control circuit 74. This hysteresis control circuit includes a NAND gate circuit 74b to which are input an output pulse B.sub.OD from a pulse delay circuit 74a, which receives the output of the divided pulses B.sub.0, and an output C.sub.0 from the D-type flip-flop, a NAND gate circuit 74d to which are input an output of an inverter 74c inverting the pulse signal C.sub.0 and the output of the divided pulses B.sub.0, and a NAND gate 74e to which are input the outputs of the NAND gates 74b and 74d and which outputs signal B.sub.OH used as a clock input of the D-type flip-flop 71.
FIG. 10 is a timing chart indicating operations of the hysteresis control circuit 74. As shown, the pulse delay circuit 74a outputs pulse B.sub.OD which is delayed from the output pulse B.sub.0 by the delay time T.sub.HD. When the output C.sub.0 of the timing comparator is low, the output pulse B.sub.0 is selected, whereby the pulse B.sub.0 becomes the output B.sub.OH of the hysteresis control circuit, while when the output C.sub.0 is high, the output pulse B.sub.OD is selected, whereby the pulse B.sub.OD becomes the output B.sub.OH of the hysteresis control circuit.
Note that, in this embodiment, the oscillating frequencies of the oscillating circuits 30 and 40 are determined by the resistances R.sub.0 of the MREs 10 and 20, the capacitance C.sub.0 of the capacitor, the characteristics of transistors in the inverters 31 to 33, and the like. These elements can be provided on one chip, whereby a pair of oscillating circuits having the same oscillating frequencies can be produced. Accordingly, when no magnetism affects the MRE 10 and 20 in the oscillating circuits 30 and 40, the oscillating frequencies of the two oscillating circuits 30 and 40 in one pair are almost the same.
Even if the temperature changes, as long as the whole chip is subjected to the same temperature, each of the three factors determining the oscillating frequencies as explained above equally changes so while the absolute values of the oscillating frequencies vary, the oscillating frequencies of the two oscillating circuits 30 and 40 in one pair are always equal. Therefore, the influence on the oscillating frequencies by temperature changes can be eliminated by detecting the ratio between the oscillating frequencies of the oscillating circuits 30 and 40 in one pair.
When magnetism affects a pair of the oscillating circuits 30 and 40, the resistances of the MREs 10 and 20 vary in accordance with changes in the magnetism and thereby the ratio of the oscillating frequencies will be changed. Note that this change of the frequency ratio is caused only by the magnetic condition.
In the present invention, when the source voltage changes, the effect on the oscillating frequencies can be compensated for by taking the ratio in the same manner as in the case of a temperature change.
As explained above, the operation of detecting magnetic conditions can be stably carried out under a widened range of ambient temperature of usage by detecting with the timing comparator 70 the change in the ratio between two oscillating frequencies generated from a pair of oscillating circuits 30 and 40, oscillating frequency of which is determined in accordance with the magnetic condition.
The detection circuit of the present invention is basically constructed with digital circuits and therefore problems which used to occur in analog circuits, for example, changes in the characteristics of transistors, changes of resistance values of resistors, and changes of the capacitance value of capacitors, do not occur. Thus, the detection device of the present invention can correctly operate at a high temperature at which no detection device consisting of analog circuits can correctly operate. Further, the detection device of the present invention can be easily formed by an integrated circuit.
Generally speaking, when a conductive layer 1c is connected to a lead frame 1d with a bonding wire 1f, the bonding wire if should be bonded with a predetermined curvature in order to keep a certain tensile strength. However, in this embodiment, since the bonding wire 1f is connected to the surface of the thin portion 1i of the insulating substrate 1a, the gap between the MRE 1b (10, 20) and the molding resin 1g can be reduced, and the distance between the MRE 1b (10, 20) and the rotating gear 9 can also be reduced as small as possible, thus improving the sensitivity.
Another detection circuit of the first aspect of the present invention by which the sensitivity of the detection device is improved will be explained hereunder. FIG. 11 is a block diagram of a detection circuit. In the first example, the MRE was used in both the oscillating circuits 30 and 40, in this example, an MRE 80 is used only in the oscillating circuit 30. In the oscillating circuit 40, an ordinary reference resistor 90 is used, and the resistance does not change with magnetic changes.
FIG. 12(a) shows the specific constructions of the oscillating circuits 30 and 40, important elements in this example, and FIG. 12(b) shows operational waveforms obtained at the points B, D, E, F, and A shown in FIG. 12(a).
In FIGS. 12(a), 31(41), 32(42), and 33(43) denote inverters serially connected in the same manner as shown in FIG. 7.
In that circuit, an input of the inverter 31(41) is connected to an output of the inverter 33(43) through a resistor R.sub.1X (R.sub.1) and the input of the inverter 31(41) is connected to an output of the inverter 31(41) through a serially connected capacitor 35(45) and resistor R.sub.2X (R.sub.2), while a node point formed between the capacitor 35(45) and the resistor R.sub.2X (R.sub.2)is connected to an output of the inverter 32(42) through a resistor R.sub.3X (R.sub.3). As explained later, at least one of the resistors R.sub.1X, R.sub.2X, and R.sub.3X in the oscillating circuit 30 is replaced by the MRE 80.
The oscillating frequency f.sub.1 of the oscillating circuits 30 and 40 is represented by the following equation; ##EQU3##
When a threshold value V.sub.TH of the inverters 31(41), 32(42), and 33(43) is V.sub.TH =1/2 V.sub.DD, the frequency f.sub.1 is represented as follows; ##EQU4##
The potential at point A is determined by dividing the voltage between the points E and F utilizing the resistors (R.sub.2(X), R.sub.2(X)).
Here, a voltage having a high level is represented by V.sub.AH, and a voltage having a low level is represented by V.sub.AL. The difference therebetween is represented by V.sub.AS. When a voltage at the point D (an input voltage to the inverter 31(41) reaches a threshold value V.sub.TH of the inverter 31(41), the voltage of each point E, F, and B is reversed in turn and the voltage at the point D is shifted to both the plus and minus directions with an amplitude of V.sub.AS with respect to a center level of V.sub.TH. Then, when the voltage at the point D again approaches the threshold value V.sub.TH due to the charge/discharge operation of a capacitor C.sub.1 and a resister R.sub.1(X), the same operation as mentioned above is repeated.
Note that when the differential voltage V.sub.AS is increased, the frequency f.sub.1 falls, and vice versa.
In the first example, the oscillating frequency f.sub.0 was determined by the charge/discharge time caused by the capacitor C.sub.0 and resistor R.sub.0. In this example, the voltage at the point A can be controlled by the resistors R.sub.2(X) and R.sub.2(X), increasing the flexibility of control of the frequency.
Next, the difference between the sensitivity of the oscillating circuit used in the first example shown in FIG. 7 and that of the second example shown in FIG. 12 will be explained utilizing experimental data.
The situation mentioned above will be explained with reference to FIG. 19. In the oscillating circuit as shown in FIG. 7(a), the resistance of the resistor R.sub.0 is changed so as to modulate the frequency.
Note that when the resistance of the resistor R.sub.0 is changed to R.sub.0 +.DELTA.R.sub.0, the slope indicating the charge/discharge and which determines the frequency thereof, is shown by the characteristic curve L in FIG. 18(a). Conversely, when the resistance is changed to R.sub.0 -.DELTA.R.sub.0, the slope is shown by the characteristic curve M. Thus, modulation of the frequency .DELTA.T.sub.0 will occur due to the resistance being changed from R.sub.0 +.DELTA.R.sub.0 to R.sub.0 -.DELTA.R.sub.0. In this situation, the voltage at point H at the time=0 is increased to V.sub.TH +V.sub.DD and is kept constant.
On the other hand, in the oscillating circuit shown in FIG. 12(a), change of the resistance of the resistor R.sub.1 as well as the resistors R.sub.2 and R.sub.3 is used as a frequency modulating means. The resistors R.sub.1, R.sub.2, and R.sub.3 are formed by MREs. They are arranged so that the direction of change of resistance are R.sub.1 +.DELTA.R.sub.1, R.sub.2 +.DELTA.R.sub.2, and R.sub.3 -.DELTA.R.sub.3, respectively, under a predetermined magnetic field. Namely, when the resistors R.sub.1 and R.sub.2 are arranged in an in-phase relationship with each other and the resistors R.sub.3 and R.sub.1 are arranged in an opposite-phase relationship,
(1) When R.sub.1 .fwdarw.R.sub.1 +.DELTA.R.sub.1, R.sub.2 .fwdarw.R.sub.2 +.DELTA.R.sub.2, and R.sub.3 .fwdarw.R.sub.3 -.DELTA.R.sub.3
The charge/discharge slope is identical to that in the oscillating circuit shown in FIG. 7(a). The voltage at time t=0 is shown by the following equation as shown in FIG. 19(b): ##EQU5##
Note that when the time constant of the charge/discharge operation is elongated due to the resistance of the resistor R.sub.1 increasing to R.sub.1 +.DELTA.R.sub.1, the voltage (voltage for starting charge/discharge operation) at time=0 is increased.
(2) When R.sub.1 .fwdarw.R.sub.1 -.DELTA.R.sub.1, R.sub.2 .fwdarw.R.sub.2 -.DELTA.R.sub.2, and R.sub.3 .fwdarw.R.sub.3 +.DELTA.R.sub.3
The charge/discharge slope is identical to that in the oscillating circuit shown in FIG. 7(a). The voltage for starting the charge/discharge operation at time t=0 is shown by the following equation: ##EQU6##
Note that when the time constant of the charge/discharge operation is shortened due to the resistance of the resistor decreasing to R.sub.1 -.DELTA.R.sub.1, the voltage at time=0 is decreased.
As explained above, the phase of the change of resistance can be arbitrarily set as above by adding the resistors R.sub.2 and R.sub.3 formed as MRES to the oscillating circuit shown in FIG. 7(a). The voltage for starting the charge/discharge operation can also be modulated as one of the means to modulate of frequency other that a conventional modulating method utilizing adjustment of a time constant in a charge/discharge operation. Accordingly, in this example, a large modulation .DELTA.T.sub.1 of the period can be obtained and thus the frequency thereof can be changed considerably.
Next, the difference between sensitivities of the oscillating circuit shown in FIG. 7(a) and FIG. 12(a), will be explained. Simultaneously, a method for selection of the resistance values of the resistors in the oscillating circuit shown in FIG. 12(a) for improving the sensitivity thereof will be explained.
Note that in this example, the sensitivity is represented by the following equation; ##EQU7##
First the sensitivity of the oscillating circuit shown in FIG. 7 is calculated as follows: ##EQU8##
Therefore, when the change in resistance is 1%, the change of the frequency is 1% (absolute value).
On the other hand, the sensitivity of the oscillating circuit shown in FIG. 12(a) is shown in Table 1. In this table, the resistance is assumed to be varied as follows:
R1.fwdarw.increase by 1%
R2.fwdarw.increase by 1%
R3.fwdarw.decrease by 1%
Further, the ratio R.sub.2 /R.sub.3 is used as a parameter, and a capacitance C.sub.1 is set at 220 PF.
TABLE 1__________________________________________________________________________No change in resistance 1% change in resistance Sensitivity R.sub.2 /R.sub.3Parameter R.sub.1(.OMEGA.) R.sub.2(.OMEGA.) R.sub.3(.OMEGA.) f.sub.1(Hz) R.sub.1x(.OMEGA.) R.sub.2x(.OMEGA.) R.sub.3x(.OMEGA.) f.sub.1x(Hz) ##STR1##__________________________________________________________________________1.5 10K 1.5K 1K 642K 10.1K 1.515K 1.01K 675K 4.82.0 .uparw. 2.0K .uparw. 432K .uparw. 2.02K .uparw. 445K 3.03.0 .uparw. 3.0K .uparw. 321K .uparw. 3.03K .uparw. 328K 2.04.0 .uparw. 4.0K .uparw. 283K .uparw. 4.04K .uparw. 288K 1.75.0 .uparw. 5.0K .uparw. 264K .uparw. 5.05K .uparw. 268K 1.5__________________________________________________________________________
From Table 1,the curve shown in FIG. 20 is obtained.
As clear from the graph, when the resistance ratio R.sub.2 /R.sub.3 is increased, the sensitivity is reduced and approaches that of the oscillating shown in FIG. 7(a). While, when the resistance ratio R.sub.2 /R.sub.3 approaches 1, the sensitivity becomes infinity. In this situation, the oscillating frequency is also infinity, so the oscillation per se becomes unstable.
Therefore, the most suitable resistance values of the resistors R.sub.1, R.sub.2, and R.sub.3 are determined by the following;
(1) The resistance values of the resistors R.sub.1, R.sub.2 and R.sub.3 should be sufficiently larger than the internal resistance of the inverter.
(2) When the resistance values of the resistors R.sub.1, R.sub.2, and R.sub.3 are small, the amount of current consumed will increase.
(3) When the resistance values of the resistors R.sub.1, R.sub.2 and R.sub.3 are large, the impedance of the sensing portion will increase, whereby the sensitivity will be weakened against noise.
From these, the resistance values of the resistors R.sub.1, R.sub.2, and R.sub.3 are suitably set at 1 k.OMEGA. to 10 k.OMEGA..
Further,
(1) When the ratio R.sub.2 /R.sub.3 approaches 1, the sensitivity will be improved, but the oscillation thereof will be unstable.
(2) When the ratio R.sub.2 /R.sub.3 exceeds 10, the sensitivity is not so different from that of the oscillating circuit shown in FIG. 7(a).
For these reasons, the ratio R.sub.2 /R.sub.3 is suitably set at 1.5 to 4.
Experimental data of the sensitivity characteristics in the oscillating circuits shown in FIG. 7(a) and FIG. 12(a) will be disclosed hereunder.
(1) Oscillating Circuit Shown in FIG. 7(a)
When devices TC40H004 (produced by Toshiba Co. Ltd.) are used as the inverters, and the resistance R.sub.0, the capacitance C.sub.0, and source voltage V.sub.DD are set at 10 k.OMEGA., 220 pF, and 5 V, respectively, the sensitivity is 0.82 at room temperature (this means a change of frequency of 0.82% with respect to a 1% change of the resistance value).
(2) Oscillating Circuit Shown in FIG. 12
The sensitivity is examined in the following four cases in the case of devices TC40H004 (produced by Toshiba Co. Ltd.) used as the inverters, and the resistances R.sub.1X and R.sub.3X, the capacitance C.sub.1, and source voltage V.sub.DD set at 10 k.OMEGA. and 2 or 1 k.OMEGA., 220 pF, and 5 V, respectively, utilizing a parameter of R.sub.2(X) /R.sub.2(X) and changing the value R.sub.2(X) at room temperature.
(1) The value R.sub.1X is varied from 0 to 2%,
(2) The value R.sub.2X is varied from 0 to 2%,
(3) The values of both R.sub.1X and R.sub.2X are simultaneously varied from 0 to 2% by the same ratio,
(4) The absolute values of both R.sub.2X and R.sub.3X are simultaneously varied from 0 to 2% by the same ratio, but one positive increment and the other negative.
The results obtained by these experiments are shown in FIGS. 13(a) and 13(b), respectively.
From these results, it will be understood that the sensitivity of the oscillating circuit shown in FIG. 12 can be increased more than 2.5 times that of the oscillating circuit shown in FIG. 7 with a combination of changes of the resistances. Various modifications of the above examples are possible as shown by (9a) to (5) below:
(1) While FIG. 1 show a gear 9 which is not magnetized, if a gear 9a as shown in FIG. 14 is magnetized with alternately arranged N poles and S poles, the bias magnet 4 shown in FIG. 1 can be omitted.
(2) The construction of the oscillating circuit in the second example may be formed as shown in any one of FIGS. 15(a), 15(b), and 15(c). When the construction shown in FIG. 15(a) is used, the positions to which the resistors R.sub.2X and R.sub.3X are connected may be changed to the outputs of the inverters 32 and 133, respectively. When the construction shown in FIG. 15(b) is used, the number of the inverters may be varied to another odd number exceeding 3. Further, when the construction shown in FIG. 15(c) is used, the number of the inverters connected between the resistors R.sub.2X and R.sub.3X may be any odd number exceeding 1.
In other words when an oscillating circuit having a high sensitivity is formed by adding the resistors R.sub.2X and R.sub.3X to the circuit construction as shown in FIG. 7(a), use may be made of MRES for resistors R.sub.1X to R.sub.3X so that voltage may be set at the point A between the resistors R.sub.2X and R.sub.3X and so that the voltage at the point A can be varied when the oscillating circuit subjected to magnetism.
In this case, when the ratio of the resistors R.sub.2X and R.sub.3X (R.sub.2X /R.sub.3X), is less than 1, the oscillating circuit stops the oscillation because that ratio falls outside of condition in which the circuit can operate stably. Thus, the ratio is set at exceeding 1. Further, when at one or more of the resistors R.sub.1X, R.sub.2X, and R.sub.3X is formed by an MRE, the circuit can be used as a frequency modulator, but when the sensitivity is required to be much higher, at ,least two of the resistors R.sub.1X, R.sub.2X, and R.sub.3X must be formed by an MRE.
(3) In the second example, the cases were shown in which the resistors R.sub.1X and R.sub.2X or the resistors R.sub.2X and R.sub.3X were replaced by MREs and the resistance changed. If the resistances are changed simultaneously by the same ratio, the design work for these oscillating circuits can be simplified.
(4) In the above examples, two oscillating circuits were used, but three or more oscillating circuits can be used and the outputs thereof compared with each other.
(5) In the first example, the sensitivity of the device was improved by connecting the bonding wire if to the thin portion 1i of the insulating substrate 1a as shown in FIG. 2. This may be produced by the method shown in FIG. 16. First, as shown in FIG. 16(a), an oxide film 1m (indicated by hatching) is formed on the surface of a silicon (Si) substrate 11, the cross-sectional views of which taken from a line A--A and a line B--B in FIG. 16(a), shown in FIGS. 16(b) and 16(c), respectively.
After that, anisotropic etc.hing using an alkaline substance is applied and the configuration of the substrate is changed to that as shown in FIG. 16(d), whereby the thin portion In and tapered portions 1o and 1p are formed. Cross-sectional views taken from a line A--A and a line B--B in FIG. 16(d) are given in FIGS. 16(e) and 16(f), respectively.
FIGS. 17(a) to 17(c) show a circuit construction in which an MRE 1q and an conductive layer 1r are formed on the insulating substrate thus produced above. In that, the MRE 1q is formed on the surface of the thick portion 1s of the insulating substrate 1a with a zig-zag configuration as shown in FIG. 17(a). The conductive layer 1r is formed to cover the thick portion 1s, the tapered portion 1p, and the thin portion 1n of the insulating substrate 1a. Finally, a bonding pad 1t is provided on the thin portion 1n. Cross-sections along a line A--A and a line B--B in FIG. 17(a) are shown in FIGS. 17(b) and 17(c), respectively.
In FIG. 17, the bonding pad it provided on the surface of the thin portion 1n is surrounded by the thick portions in three directions, so the strength of the thin portion will be increased at the bonding operation. Further, the bonding pad It may be mounted on the surface of the insulating substrate without applying the etc.hing operation to the substrate to make the tapered portions.
As explained above, in accordance with the first invention of this application, a magnetoresistance device which can stably work in widened range of a ambient temperature in which the device is used and is suitable for making an integrating circuit, can be provided and further, in accordance with the second invention, the sensitivity of the magnetoresistance device can be improved.
According to the first aspect of the present invention, when a magnetism does not affect the magnetoresistance effective device (MRE) provided in an oscillator circuit, the oscillating frequency of both first and second oscillator circuits are mostly coincided with each other.
And when a temperature is varied, the absolute value of each oscillating frequency is changed, although the condition is not changed as long as the whole circuit is suffered from uniform temperature.
Therefore, the influence caused by a temperature can be eliminated by comparing and processing the oscillating frequencies of both oscillator circuits.
When a magnetism affects the oscillator circuit, a resistance value of the MRE is varied in accordance with a condition of a magnetism whereby a difference is created between both oscillating frequencies which is equal to each other causing to change a ratio of both oscillating frequencies.
Note that this variation of the ratio of both oscillating frequencies is created depending only upon a condition of a magnetism.
On the other hand, to a variation of a voltage source, such variation can be compensated by utilizing the same ratio of both oscillating frequencies as used in a case of the temperature variation.
Accordingly, a detecting operation for a magnetism condition can be carried out stably under wide range of ambient temperature in which the detector is used.
In accordance with another embodiment of the present invention as above, since a potential of said node portion defined between said second and third resistors is changed when the MRE detects a magnetism, a oscillating frequency can be controlled not only by a charge/discharge operation caused by the first resistor and the condenser but also by the potential of the node point and thus a variation of the frequency can be amplified with respect to a variation of a resistance value leading to improve a detecting sensitivity thereof.
A magnetoresistance effective element (MRE) used in this embodiment, is an element, a resistance value thereof can be varied in response to a variation of a magnetism and any kind of elements having such a function as mentioned above can be used, for example a Hall device, a ferromagnetic magnetoresistance element (MRE), and an element including a capacitor instead of a resistor or the like.
As explained above, when the magnetoresistance effective element is used in a magnetism detecting device utilizing a bias magnet, a relationship between the amount of oscillation of a magnetic field and a position on which the magnetoresistance effective element is arranged is very important and the sensitivity thereof depends upon the relationship and therefore the setting operation of the magnetoresistance effective element in this device and what kind of magnetoresistance effective element is used, are key factors to obtain an accurate output signal indicating an amount of the variation of the magnetism.
Therefore, the next embodiment of the present invention relates to a magnetism detecting device which can be used in a physical quantity detecting device as one embodiment later explained and which can be replaced with the magnetism detecting circuit as shown in previous embodiments.
FIG. 21(a) shows another example of a conventional magnetic detection device which detects a rotational speed of a rotating body with a rotating gear 204. The gear 204 is made of magnetic material, and a magnet 203 for generating a bias magnetic field is provided. When the gear is rotated, the lines of magnetic force of the bias magnetic field are periodically changed by projecting portions and concave portions to vary with a sine wave form in an A--A' direction in accordance with the relative position of the gearteeth, as shown in FIG. 21(a). This change has been detected heretofore by a plurality of MREs 201a, 201b, 201c, 201d, etc. as shown in FIG. 21(b) arranged on an insulating substrate 202 which is provided in a plane perpendicular to a direction of the bias magnetic field.
The magnetic detection device explained above detects the change of intensity of the magnetic field generated inside the plane of the MREs, caused by the change of the angle of the magnetic field, as a change of resistance. The intensity of the magnetic field generated in the plane of the MREs is usually not sufficient to saturate charges in the resistance variation of the MREs and thus the sensitivity and the accuracy of the device depends significantly upon the intensity and angle of the bias magnetic field generated by the bias magnet.
The theory of the change of the resistance of the MREs will be explained with reference to FIGS. 22(a) to 22(c). These Figures show the change of resistance of the MREs 201 when a direction of the magnetic field is changed in a plane perpendicular to a surface of the insulating substrate 202 on which the MREs 201 are provided and either parallel to or perpendicular to the direction of a current I flowing in the MREs (direction indicated by arrow) under an intensity of a magnetic field saturating the MREs 201.
An shown in FIG. 22(a), when the angle .theta. of a magnetic field B is changed in a plane perpendicular to the surface of the insulating substrate 202 and parallel to the current direction, the change of the resistance, i.e., the change of the resistance in the direction parallel to the current direction, is shown by a curve R.sub.p in FIG. 22(c), while when the angle .theta. thereof is changed in a plane perpendicular to the surface of the insulating substrate 202 and perpendicular to the current direction, the change of the resistance, i.e., the change of the resistance in the direction perpendicular to the current direction, is shown by a curve R.sub.R in FIG. 22(c).
Accordingly, as shown in FIG. 21(a), when the MRE is arranged on a center axis of the bias magnet, the angle .theta. of the magnetic field applied to the MREs is finely changed in accordance with the movement of an object to be detected.
As in FIG. 22(c), the resistance characteristics of R.sub.p and R.sub.R steeply change in a vicinity of 90.degree.. Therefore, the change of the resistance of the MREs can be detected with high sensitivity in response to a change of the magnetic field angle .theta..
Movement of an object to be detected (for example rotation of a gear 204) is generally detected by a change in resistance of the MREs by an analog type detecting system, in which an MRE resistance bridge is formed and an output voltage of the resistance bridge is detected by a comparator, as shown in FIG. 23. In FIG. 23, MREs 201a, to 201d correspond to the MREs 201a to 201d shown in FIG. 21(b).
FIG. 24 shows a relationship between oscillation of the magnetic field as shown in FIG. 21(b) (change of the magnetic field angle .theta.) and the change of resistance of the MREs as well as an output of the circuit shown in FIG. 23 at that time, i.e., the output of the magnetic detection device. FIG. 24(a) shows the same relationship as mentioned above when an air gap between the object to be detected and the MREs is large, while FIG. 24(b) shows the same relationship when the air gap is small.
Note that the MREs 201a and 201b and MREs 201c and 201d change in resistance in accordance with a change of the magnetic field angle .theta., as shown in FIGS. 24(a) and 24(b). In order to obtain normal outputs 0 and 1, the pattern of the MREs is generally arranged to be displaced from the center of the magnet by a slight offset so that the magnetic field is biased at a cross point A of a curve of the resistance change R.sub.p of the MREs in the parallel direction and that of the resistance change R.sub.R of the MREs in the perpendicular direction.
At this time, the magnetic field angle changes around this cross point A and when the air gap is large as shown in FIG. 24(a), the magnetic field angle is small, so another cross point B is not reached as shown in FIG. 24(a). Accordingly, normal signals of 0 and 1 are output from the detection circuit as shown in FIG. 23, whereby outputs having the same frequency as that of the movement of the object to be detected can be obtained, enabling detection of the movement of the object with a high accuracy.
However, when the air gap is small, the change of the magnetic field angle is large, therefore a cross point D other than the cross point C which is usually used can be reached as shown in FIG. 24(b). In this situation, an output signal 0, 1 other than the normal signal 0, is erroneously output from the detection circuit shown in FIG. 23. Thus, an output having a frequency double the frequency of the movement of the object to be detected is generated, preventing accurate detection.
That is, a problem arises in that erroneous outputs can be included in the output from the magnetic detection device depending on the air gap formed between the MREs and the object to be detected.
To overcome this, the magnetic detection device may be mounted at a position with an air gap resulting in a small oscillation of the magnetic field angle, however, the cross point as mentioned above exists at a nearly 90.degree. magnetic field angle, and the characteristics of the change of resistance of the MREs are symmetric at the angle of 90.degree.. Therefore, it is very difficult to avoid erroneous detection with extremely precise positioning.
According to another aspect of the present invention there is provided. A magnetic detection circuit which comprises;
a magnetic field generating means for generating a bias magnetic field toward an object to be detected having magnetic materials therein and MREs provided on a predetermined plane arranged in the bias magnetic field, a resistance value thereof being changed in response to a change of conditions of the bias magnetic field due to a movement of the object to be detected,
the change of conditions of said bias magnetic field being detected by the change of the resistance value of the MREs,
the predetermined plane formed by direction of the bias magnetic field generated from the magnetic field generating means and a moving direction of said object to be detected,
the MREs arranged on the plane so that the change of the resistance value displays at least one of a monotonic increment or a monotonic reduction due to the change of the bias magnetic field. The MREs will only change in resistance by a single monotonic increment or a monotonic reduction due to the change of the bias magnetic field, due to the arrangement. The resistances will never exceed the maximum point or minimum point of the curve shown in FIG. 24(b). Therefore, the change of resistance of the MREs due to the change of the bias magnetic field does not include a change of resistance of the MREs not caused by a change of a condition of the bias magnetic field. The change of resistance therefore, becomes a clean sine wave corresponding to the change of the condition of the bias magnetic field. The meaning of the word monotonic as used herein connotes a signal which is either continually increasing or continually decreasing until reaching a maximum or minimum for any given change in resistance.
FIG. 25 is a cross-sectional view of magnetic detection device according to this aspect. In the figure, an inner plate 213 made of a low thermal expansion metal (for example KOVAR, stainless steel 430, or the like) is fixed to a housing 210a made of a non-magnetic metal material by soldering, welding, press-fitting, or the like. The inner plate 213 is provided with an output pin 220 sealed with a sealing glass 250. A varistor 221 is provided on the inner plate 213 and adhered to the plate 213 with glass having a low melting point or a suitable adhesive.
An insulating substrate 202 on which the MREs are formed, is mounted on a substrate holder 210 with an adhesive. The substrate holder 210 is fixed to the plate 213 so that the surface of the substrate on which the MREs are formed is at a right angle with the surface of the plate 213. The output pin 220 and the MRE are electrically connected with bonding wires 280 through the varistor 221. An outer case 223 made of a non-magnetic material is press-fitted to housing 210a. The end most portion of a peripheral portion thereof 212a is hermetically connected with the housing 210a by laser welding or the like. The outer case 223 is previously fixed to a bracket 260 by soldering or the like to solidly fix the magnetic detection device. A pin 270 for positioning the device is provided. A magnet 230 for applying a bias magnetic field to the MREs, is mounted on the surface of the inner plate 213 so that the magnetizing direction is perpendicular to the surface of the plate 213 (magnetic direction shown in FIG. 26) In this situation, the direction of the magnetic field applied to the plate is parallel to the surface on which the MREs are formed. Ferrite beads 240 are provided for preventing erroneous operations caused by electromagnetic noise. The magnetic detection device is completed by filling a molding resin 290 into the device and curing the resin with heat after the magnet 230 and the ferrite beads 240 are fixed.
FIG. 26, shows one pattern of arrangement of MREs 201e, 201f, 201g, and 201h formed on the insulating substrate 202. As shown in FIG. 26, a group of MREs 201e and 201f and another group of the MREs 201g and 201h are arranged so that the longitudinal direction of the MREs 201e and 201f is inclined with respect to the direction of the bias magnetic field generated by the bias magnet by about 45.degree. while the longitudinal direction of the MREs 1g and 1h is inclined with respect to the same by about 45.degree. in the opposite direction.
Preferably the MREs used in this invention are ferromagnetic magnetoresistance elements made of an alloy, the main substance of which is Ni, for example, Ni--Co alloy or Ni-Fe alloy. The film is preferably in a thin film like layer of these alloys formed on the insulating substrate 202 by the vapor deposition method or the like and thereafter etc.hing the film by a photo lithographic method to form a predetermined pattern.
The insulating substrate 202 on which the MREs are provided is arranged on a plane parallel to the direction of the bias magnetic field generated from the bias magnet 3 and parallel to the moving direction of the object to be detected.
The operation of this embodiment will be explained hereunder. As shown in FIG. 27(a), when a saturated magnetic field B is changed by an angle .theta. in the plane 224, corresponding to the surface of the insulating substrate 202 on which the MREs are formed, the change of the resistance of the MRE 201 is as shown in FIG. 27(b).
Note, that in FIG. 26, when the projecting portions or concave portions of the gear 24, i.e., the object to be detected, reach the center of the bias magnetic field, the MREs 201e and 201f are biased with a bias angle of 45.degree. in that plane while the MREs 201g and 201h are biased with a bias angle of 135.degree..
It is apparent from FIG. 27(b), that the resistance at the magnetic field angle of 45.degree. is identical with that at the magnetic field angle of 135.degree.. On the other hand, at the magnetic field angle of 45.degree., when the magnetic field angle is increased, the resistance is decreased. Conversely, at the magnetic field angle of 135.degree., when the magnetic field angle is increased, the resistance is increased. Therefore, in a coordinate system as shown in FIG. 26, when the magnetic field angle .theta. is 90.degree., the MREs 201e and 201f are biased with a bias angle of 45.degree. while the MREs 201g and 201h are biased with a bias angle of 135.degree..
Characteristic curves of this magnetic detection device are shown in FIGS. 28(a) and 28(b) in which the abscissa represents the magnetic field angle .theta. and the ordinate represent the change of resistance of the MREs 201e to 201h.
In the case of a long air gap between the MREs and the object to be detected, when the magnetic field angle is changed due to movement of the object to be detected, the resistance of one group of the MREs 201e and 201f and another group of the MREs 201g and 201h is increased and decreased, respectively, with respect to a crossing point E. Accordingly, when the MREs 201e to 201h are wired to each other to form a bridge circuit to make a detection circuit as shown in FIG. 26, the detection circuit outputs correct signals 0, 1. In FIG. 26, the MREs 201f to 201h are electrically connected so as to correspond to the MREs 201a to 201d of FIG. 21(d), respectively.
On the other hand, in the case of a small air gap the change of the magnetic field angle .theta. is enlarged, so the resistances of the MREs 201e to 201h are changed in the same manner as explained above as shown in FIG. 28(b).
As shown in FIG. 24(b), in a conventional magnetic detection device, when the air gap is small, the change of the magnetic field angle .theta. is enlarged, thereby erroneous operations frequently occur due to use of both cross points, but as is apparent from FIGS. 27(b), 28(a), and 28(b), in the invention, the magnetic field will arrive at another cross point only when the change of the magnetic field angle .theta. exceeds .+-.90.degree.. Generally speaking, the change of the magnetic field angle is defined by the pitch of the gear and the length of the air gap or the like, but it is at most about .+-.20.degree.. Therefore, erroneous detection, which used to occur in a conventional device when the air gap was small, can be prevented.
A second pattern of arrangement of the magnetic detection device of the aspect will be explained hereunder.
In the first example as mentioned above, the group of MREs 201g and 201h and the group of MREs 201e and 201f were arranged as in FIG. 29(a).
In this arrangement, the MREs 201e to 201h are arranged on the insulating substrate 202 as shown in FIG. 29(b). That is, one group of the MREs 201e and 201f are arranged on the substrate inclined with respect to an axis of a magnetic field by about 45.degree. while another group of the MAEs 201g and 201h are arranged on the same substrate inclined to the same axis by about 45.degree. in an opposite direction. Note that in the first pattern of arrangement shown in FIG. 29(a), the same magnetic field might not exactly be applied to both the group of the MREs 201e and 201f and the group of the MREs 201g and 201h and therefore a slight phase difference might conceivably occur between them, but in the second pattern of arrangement shown in FIG. 29(b), no phase difference will occur between them because the same magnetic field is applied exactly to all of the MREs 201e to 201h.
A third pattern of arrangement will be explained hereunder with reference to FIG. 29(c). The MREs 201e to 201h are formed in a two-layer construction in order to minimize the area occupied. That is, one group of the MREs 201e and 201f are formed on a surface of the insulating substrate, and another group of the MREs 201g and 201h are formed on top of a separate insulating film (not shown) provided therebetween.
In the above patterns of arrangement, the MREs had a strip-like configuration, but they may be formed in a comb configuration.
In a fourth pattern of arrangement as shown in FIG. 30(a), the MREs 1i comprise a plurality of alternating long strip portions 1il and short strip portion 1is joined end to end. The long strip portions 1il are inclined with respect to a magnetic field B by about 45.degree., while the short strip portions 1is are at right angles to the long strip portions 1il. In this pattern of arrangement, the total resistance value R1i of the MREs with respect to the magnetic field B shown by the continuous line is represented by the following equation;
R1i=Rl+Rs
wherein, Rl denotes the total resistance of the long strip portions 1il and Rs denotes the total resistance of the short strip portions 1is.
In this situation, when the axis of the magnetic field B shown by the continuous line is changed to a magnetic field B' shown by a dotted line due to a movement of an object to be detected the total resistance value R1i' of the MREs is represented by the following equation; ##EQU9## wherein, both .DELTA.Rl and .DELTA.Rs are positive values and denote amounts of change of the total resistance Rl in the long strip portions 1il and the total resistance Rs in the short strip portions 1is caused by the change of the magnetic field, respectively.
Note that since the long strip portions 1il and the short strip portions 1is are arranged at right angles with each other, the changes of the resistances of the long strip portions 1il and the short strip portions 1is are opposite to each other, i.e., the change of the resistance of one is positive, while that of the other is negative. Therefore, the amounts of change with respect to the total resistance may slightly offset each other but this does not adversely affect the accuracy of the output in the circuit shown in FIG. 23.
FIG. 30(b) shows a fifth pattern of arrangement of the MREs. The MREs 1j comprises alternating, end-to-end long strip portions 1jl inclined with respect to the magnetic field B by about 45.degree. and short strip portions 1js perpendicular to the direction of the magnetic field B. The total resistance R1j of the MREs 1j is represented by the following equation, with respect to the magnetic field B;
R1j=Rl+Rs
wherein, Rl denotes the total resistance of the long strip portions 1jl and Rs denotes the total resistance of the short strip portions 1js.
On the other hand, with the magnetic field B', the total resistance R1 of the long strip portions 1jl increases by .DELTA.Rl, but the total resistance of the short strip portions 1js hardly changes. The reason is that since the short strip portions are almost all perpendicular to both the magnetic fields B and B', the change of the resistance is saturated, as shown in FIG. 27(b). Namely, the total resistance R1j' with respect to the magnetic field B' is represented by the following equation;
R1j'=Rl+.DELTA.Rl+Rs
Comparing this pattern with the pattern shown in FIG. 30(a), the amount of the change in the pattern of FIG. 30(a) is (.DELTA.Rl-.DELTA.Rs) while the amount in the pattern of FIG. 30(b) is .DELTA.Rl. Thus, the amount of change in the resistance with respect to the total resistance is increased to improve the detecting sensitivity.
FIG. 30(c) shows a sixth pattern of arrangement of the MRES. The MREs 1k comprise alternating, end-to-end long strip portions 1kl inclined with respect to the magnetic field B by about 45.degree. and short strip portions 1ks parallel to the direction of the magnetic field B. The change of the resistance of the short strip portions 1ks is saturated in the case of the magnetic field B, the same as in FIG. 30(b), and does not change, so the same effect as in the fifth pattern can be obtained.
FIG. 31 shows a specific pattern of arrangement of the MREs in the sixth pattern.
In the pattern examples of the present invention as mentioned above, at least one pair of MREs are provided and a main portion of a first MRE of the pair is inclined in a first direction at a predetermined angle with respect to the magnetic field; and a main portion of a second MRE of the pair, is inclined in a second direction, different than the first direction, at an angle having the same absolute value as the angle at which the first MRE is inclined, with respect to the same magnetic field as mentioned above, and in that both of the first and second MREs, forming the pair thereof, are symmetrically arranged with respect to an axis of the magnetic field caused by the bias magnet and a substance to be detected and the pair of MREs are inclined by about 45.degree., but the angle is not restricted to 45.degree., it may be any angle as long as change of the resistance of the MREs displays only a monotonic increment and/or a monotonic reduction with respect to a change of the angle of the bias magnetic field. If such a condition is realized, the change of the angle of the magnetic field never includes two cross points. Note that when the MREs are inclined at about 45.degree., however, the resistance changes the most steeply with respect to the change of the angle of the magnetic field, so the detecting sensitivity is improved.
Also, in these examples, the insulating substrate 202 may be made a silicon substrate the MREs provided on the surface of an insulating film formed on the silicon substrate, and a detecting circuit, for example, a voltage comparator as shown in FIG. 23, formed on the silicon substrate, thereby forming the element on one chip.
Further, while one or more pairs of MREs were provided to make the magnetic detection device, combination of MREs and fixed resistors may be used.
In the previous embodiments, the magnetic detection device and the physical quantity detecting device mainly included oscillating circuits having MREs and a comparison means for comparing the oscillating frequencies output from the oscillating circuits by taking a ratio of two oscillating frequencies. However, in these devices, these oscillating circuits are required to generate output signals indicating a change of resistance with the signals crossing to form cross points. Therefore the construction and positioning of the oscillating circuits are important. Some restrictions exist on the oscillating circuits, and, further it is difficult for the oscillating circuit to be properly positioned.
On the other hand, when analog data is handle by a digital circuit, the physical resolution is determined by the number of bit used in the digital circuit, and the chronological resolution is determined by the clock frequency of the digital circuit. Therefore, a phase difference detecting circuit for detecting, for example, two pulse signals having an arbitrary phase relationship is constructed with the digital circuit. When the phase difference, however, is smaller than the chronological resolution determined by the clock frequency, it is difficult for such a phase difference to be detected. Note that in the present silicon MOS transistor technology, the maximum clock frequency is around 50 MHz and the chronological resolution is limited to around 20 ns. Therefore, in the magnetic detection devices of the previous aspect of the invention, the resolution is restricted and it is difficult to detect a very fine change which should be detected with a resolution of less than 20 ns.
Accordingly, to the fourth aspect of the present invention there is provided a magnetic detection device in which the magnetic change can be detected utilizing a pulse phase difference.
That is the fourth aspect provided a magnetic detection device which further comprises; a delayed pulse signal output means comprising a delay signal circuit which outputs a plurality of pulse signals when a fist pulse signal is input thereto, each pulse signal being delayed from the previous pulse signal by a predetermined different delay time,
a pulse signal selecting means which receives as input a second pulse signal, delayed from the first pulse signal by any desired time, and a plurality of pulse signals output from said delayed pulse signal output means, and which selects from the plurality of pulse signals, a certain pulse signal having a specific condition with respect to a timing at which the second pulse signal is input, and,
a detecting means for detecting a phase difference between the first and the second pulse signals utilizing the pulse signal selected by the pulse signal selecting means.
In the pulse phase difference detecting circuit, When a first pulse signal is input to the delayed pulse signal output means, a plurality of pulse signals are output from the signal delay circuit. Each pulse signal is delayed from the first pulse signal by a predetermined different delay time. The delay time caused by the signal delay circuit can be set extremely shorter than the chronological resolution of a conventional digital circuit. Accordingly, the phase difference between the first pulse signal and the second pulse signal can be detected with a high accuracy by selecting a specific pulse signal out of the plurality of pulse signals, i.e., the first pulse signals, by a pulse signal selecting means and having a specific condition with respect to a timing at which the second pulse signal is input thereto.
Further, in the pulse phase difference detecting circuit as mentioned above, a pulse signal with an oscillating frequency changed in accordance with a detected value of a physical quantity, is output from the first oscillating circuit. The phase difference between this pulse signal and the pulse signal output from the second oscillating circuit is detected and further the condition of change of this phase difference is detected. By detecting the condition of change of the phase difference, the accuracy of the detection is not affected by the absolute oscillating frequencies of the first and second oscillating circuits. Therefore, even if the oscillating frequency changes due to changes of the temperature characteristic and device characteristic, or due to a time elapsed and a variation of the voltage source or the like, the condition of change of the physical quantity can be detected with a high accuracy without regard to any influences as mentioned above.
Specific examples will be explained with reference to the attached drawings hereunder. FIG. 32 shows the construction of a pulse phase difference detecting circuit. In FIG. 32, a delay pulse signal generating circuit 310 is provided which comprises a gate delay circuit serving as a signal delay circuit consisting of a plurality of inverters 341 to 34l. The delay pulse signal generating circuit 310 generates a plurality of output pulse signals P.sub.1 to P.sub.n when a level of the pulse signal PA is turned to 1, i.e., turned ON. Each pulse signal is output with a suitable equal delay time T.sub.D from the time when the previous pulse signal is output, as shown in FIG. 33. In this gate delay circuit 310, the pulse signal PA is input to an inverter 341 and an output of the inverter 34l is connected to an input of an inverter 342. Further, an output of the inverter 342 is connected to an input of an inverter 343 and simultaneously output therefrom as an output pulse signal P.sub.1. Outputs and inputs of the following inverters 343 to 34l are connected in the same manner as explained above, and outputs of the inverters having an even number are output therefrom as output pulse signals P.sub.1 to P.sub.n.
In the delay pulse signal generating circuit 310 above, the outputs of the even number inverters 342 to 34l, for example, the inverters 342, 344, 346, and 348, respectively, are used as a group of the output pulse signals P.sub.1 to P.sub.n. It is apparent that the outputs of odd numbered inverters 341 to 341-l, i.e., the transistors 341, 343, 347, and 341-l, can be added to the above-mentioned group of output pulse signals P.sub.1 to P.sub.n, if necessary.
The group of the pulse signals P.sub.1 to P.sub.n are output utilizing the delay time caused only by the inverters 341 to 34l, although it is possible to adjusting the delay time by changing the number of steps of the inverters 341 to 34l or by adding a certain capacitance in a wire connecting the inverters 341 to 34l.
Further, the delay circuit 310 may be constructed utilizing resistors 700 to 70P and condensers 800 to 80P as shown in FIG. 39 without using the gate circuit consisting of the inverters.
In FIG. 32, a pulse signal selecting means 320 is provided as a synchronized pulse detecting circuit. Each of the group of the pulse signals P.sub.1 to P.sub.n, output from the delayed pulse signal output means 310, and a pulse signal PB which has an arbitrary phase relationship with the pulse signal PA are input to the pulse signal selecting means 320. The pulse signal selecting means 320 detects one output pulse signal out of the group of the output pulse signals P.sub.1 to P.sub.n, which has the phase nearest to that of the pulse signal PB.
Namely, the group of the output pulse signals P.sub.1 to P.sub.n are input to this synchronized pulse detecting circuit 320 as data. Also, the pulse signal PB which has an arbitrary phase relationship with the pulse signal PA is input thereto, while this synchronized pulse detecting circuit 320 detects and outputs the one output pulse signal out of the group of the output pulse signals P.sub.1 to P.sub.n which has the phase nearest to that of the pulse signal PB.
The synchronized pulse detecting circuit 320 comprises D-flip-flops (D-FF) 351 to 35n. To one of the D-FFs, each one of the group of the output pulse signals P.sub.1 to P.sub.n is input. The pulse signal PB is also input to each one of the D-FFs as a clock. The synchronized pulse detecting circuit 320 has an AND gate 61 to which an output Q of the D-FF 351 and an inverted output Q of the D-FF 352 are input and outputs a synchronized pulse P01. Other AND gates 62 to 6n, to which outputs Q and the inverted outputs Q of the D-FFs 352 to 35n are input, output a synchronized pulse P02 to P0m in the same manner explained above.
The operation of the pulse phase difference detecting circuit as constructed above will be explained hereunder with respect to FIG. 33. Note that, the number of the signals of the group of the output pulse signals P.sub.1 to P.sub.n and the synchronized pulse P01 to P0m are set at 4 and 3 respectively, although any number of signals which is suitable for a circuit applied may be used since there is theoretically no limitation on the number of signals.
In FIG. 33, when the level of the pulse signal PA is changed from 0 to 1, i.e., turned ON, at a time T.sub.0, the level of the output signal of the inverter 342 (that is the output pulse signal P1) increases from 0 to 1, at a time T.sub.1 delayed by a gate delay time T.sub.D from the time T.sub.0, which is a time for the signal to transmit through the two inverters 341 and 342.
In the same manner, each time T.sub.2 to T.sub.4 when the level of each of the output pulse signals P.sub.2 to P.sub.4 increases, is delayed in accordance with the number of the inverters 343 to 348 through which the signal is transmitted.
In this example, when a circuit size of the inverters 341 to 348 is equal the gate delay time T.sub.D of the inverters 341 to 348 can be made equal to each other.
Thus, the time T.sub.n can be represented as follows;
T.sub.n =n.times.T.sub.D
The gate delay time T.sub.D can be set at around 2 ns at a minimum by a conventional silicon MOS producing technology.
On the other hand, a clock frequency of a conventional digital circuit is around 50 MHz maximum and therefore the chronological resolution in the conventional technology is limited to around 20 ns.
Therefore, in accordance with this example of the present aspect of the invention, a resolution of a high accuracy of 10 times that in the prior art is obtained.
As explained above, each signal of the group of the output pulse signals P.sub.1 to P.sub.n, increased in turn by a time interval of about 2 ns, is input to the synchronized pulse detecting circuit 320. At a time T.sub.B (wherein T.sub.2 <T.sub.B <T.sub.3), when a level of the pulse signal PB is changed from 0 to 1, each of the D-FFs 351 to 354 is triggered by the pulse signal PB. At the same time, a signal corresponding to the level of the output pulse signals P.sub.1 to P.sub.4, given to the D-FFs at the time T.sub.B, is output from both the output Q and inverted output Q of each of the D-FFs 351 to 354. For example, at the time T.sub.B, the level of the output pulse signals P.sub.1 and P.sub.2 is 1, while the level of the output pulse signals P3 and P.sub.4 is 0. Therefore, signal having a level of 1 is output from the output Q of the D-FF 351, while signal having a level of 0 is output from the inverted output Q of the D-FF 352. Thus, the output level of the AND gate 361 is 0.
On the other hand, the level of the signal output from the output Q of the D-FF 352 is 1, and that of the signal output from the inverted output Q of the D-FF 353 is 1. Thus, the output level of the AND gate 362, i.e., synchronized pulse signal P02, is 1.
Regarding the AND gates 363 and 364, the levels of the outputs of the D-FFs 352 to 354 are 0, since the signal level output from the D-FFs 352 to 354 and input thereto is 0.
In accordance with the synchronized pulse detecting circuit 320, the level of the synchronized pulse signal P02, which corresponds to the output pulse signal P.sub.2, selected from the group of the output pulse signals P.sub.1 to P.sub.n and increased just before the time T.sub.B at which the output pulse signal PB is input when selected, can be different from the level of other synchronized pulse signals P01 and P03. Therefore, the phase difference between the two pulse signals PA and PB can be detected with accuracy for detection utilizing the gate delay time T.sub.D as a minimum resolution.
In the above-mentioned example, the number of the pulse signals to be compared for detecting a phase difference is set at 2, but when the number of the pulse signals PB is increased, the detecting operation can be carried out by adding another synchronized pulse detecting circuit thereto. Accordingly, the number of the pulse signals to be compared is not restricted.
Also, one may not only detect the pulse signals increased to the level 1 Just before the pulse signal PB is input, but may also detect the pulse signals increased to the level 1, Just after the pulse signal PB is input. Further, the detecting operation may be carried out on an nth pulse signal counted from and before the time when the pulse signal PB is input, wherein n is a predetermined number.
The pulse phase difference detecting circuit can also be used for a physical quantity detecting device.
According to a fifth aspect of the present invention, is provided a physical quantity detection device, which comprises a first oscillating circuit generating a pulse signal having an oscillating frequency which changes in response to a detected value of a physical quantity to be detected,
a second oscillating circuit generating a pulse signal having an oscillating frequency to be compared with the oscillating frequency of the first oscillating circuit,
a pulse phase difference detecting circuit for detecting a phase difference between the pulse signals output from the first oscillating circuit and the second oscillating circuit,
a changing condition detecting means for detecting the changing condition of the phase difference based upon the phase difference detected by the pulse phase difference detecting circuit, and
a detecting means for detecting a changing condition of the physical quantity depending upon the changing condition of the phase difference detected by the changing condition detecting means.
FIG. 34 is a block diagram of a magnetic detection circuit for detecting a physical quantity. In FIG. 34, MREs 410 and 420 are provided. The resistances thereof change in accordance with a change of magnetism. When the two MREs 410 and 420 are formed in the same configuration and arranged so as to be displaced with angle of 90.degree., the resistances thereof can be changed in an opposite direction with respect to magnetic changes so the sensitivity to magnetic changes can be improved.
In this example, oscillating circuits 430 and 440 are also provided. The oscillating frequency can be changed in accordance with the change of resistance of the MREs 410 and 420. The oscillating circuits 430 and 440 are further provided with wave shaping circuits and output pulse signals CKB and CKA having the same frequency as an oscillating frequency determined by the resistances of the MREs 410 and 420, respectively, as shown in FIGS. 34 and 35.
Conventional counters 450 and 460 are provided. The counters count the pulse signals CKB and CKA output from the oscillating circuits 430 and 440 and output count signals C0B to C3B and C0A to C3A, respectively, as shown in FIGS. 34 and 35.
Decoders 451 and 461 are provided. The decoder 451 outputs reset signals RSTB and RSTA to the counters 450 and 460, respectively, when the count of the counter 450 reaches a predetermined value (for example, 9), as shown in FIGS. 34 and 35. The decoder 451 outputs a synchronizing signal SYNC to oscillating circuit 440 to synchronize a starting time for the next counting operation. The decoder 451 further outputs a dividing pulse signal PB1 when the count of the counter 450 reaches 9. In the same manner, the decoder 461 outputs a dividing pulse signal PA1 when the count of the counter 460 reaches 9.
Note that in this example, since the dividing pulse signals PB1 and PA1 are output from the decoders 451 and 461, respectively, when the same number of the pulse signals CKB and CKA are counted by the counters 450 and 460, respectively, the difference between the rising times of the dividing pulse signals PB1 and PA1 corresponds to a time in which all of the phase difference of the pulse signals are accumulated. Therefore, the detecting sensitivity to magnetic change is improved, and thus change of the oscillating frequency due to a slight magnetic change can be detected.
Reference numeral 470 denotes a pulse phase difference detecting circuit as mentioned above, which turns the level of the synchronized pulse signal corresponding to the phase difference between the two pulse signals PB1 and PA1 selected from the synchronized pulse P01 to P0m, to 1, turns the level of other synchronized pulse signals to 0, and outputs both levels.
Reference numeral 480 denotes a decoding circuit, which outputs decoded binary digital signals D0 to Di-1 (wherein i denotes a positive integer) corresponding to the synchronized pulse signals selected from the synchronized pulses P01 to P0m output from the pulse phase difference detecting circuit 470, the level being 1.
In this example, the decoding circuit 480 comprises a ROM utilizing the group of the synchronized pulse signals P01 to P0m as address signals. In each address, predetermined binary digital data is stored which indicates the phase difference represented by the synchronized pulse signals P01 to P0m.
Reference numeral 490 denotes a comparison circuit which processes the difference between the binary digital data stored in the decoding circuit 480 in a previous step and the binary digital data output at this step and thereafter outputs a pulse signal P.sub.OUT with a level 1 or 0 which is reversed at the maximum value and the minimum value of this difference, respectively. The comparison circuit 490 is provided with memory circuits 491 for storing binary digital data signals D-10 to D-1i-1 output from the decoding circuit 480. The data is stored therein in a previous step, as shown in FIG. 37. The comparison circuit 490 is further provided with a digital subtracter 492, for processing a difference between the binary digital data D-10 to D-1i-1 stored in the memory circuit 491 at a previous step and the binary digital data D-0 to D i-1 output from the decoding circuit 480 at this time.
A flip-flop 493 is provided which receives as input coded bit signals indicating if the resultant processed data in the digital subtracter 492 is positive or negative and which output a pulse signal P.sub.OUT in response to the input data. Accordingly, the pulse signal P.sub.OUT is a signal with a level reversed from 1 to 0 or vice versa when the binary digital data signals D-0 to D i-1, which indicate a phase difference between the divided pulse signals PB1 and PA1, are reversed from an increment to a decrement or from a decrement to an increment.
In this processing circuit, a so-called hysteresis circuit may be provided in order to avoid erroneous output data caused by noise.
FIG. 36(A) shows a configuration relating to a change of oscillating frequencies f.sub.a and f.sub.b output from the oscillating circuit 430 and 440 as shown in FIG. 34.
FIG. 36(B) shows a condition in which how a difference .DELTA.f between the oscillating frequencies f.sub.a and f.sub.b as shown in FIG. 36(A) varies in accordance with an elapsed time. The oscillating frequency difference .DELTA.f has a frequency corresponding to a magnetic change. Therefore, as shown in FIG. 36(C), a pulse signal P'.sub.OUT with a level reversed from 0 to 1 or 1 to 0 in response to the time t.sub.max when the signal indicating the oscillating frequency difference .DELTA.f comes to the maximum value and to the time t.sub.min when the signal indicating the oscillating frequency difference .DELTA.f comes to the minimum value, is a signal having a frequency corresponding to the magnetic change.
On the other hand, the phase difference between two divided pulse signals PB1 and PA1 corresponds to the difference of frequencies of the pulse signals CKB and CKA output from two oscillating circuits 430 and 440. (Note that when the difference of the oscillating frequencies is increased, the phase difference is also increased, while when the difference of the oscillating frequencies is decreased, the phase difference is also decreased.) Therefore, the pulse signal P.sub.OUT output from the flip-flop 493 is also a signal having a frequency corresponding to the magnetic change and thus the magnetic change can be detected by detecting the pulse signal P.sub.OUT.
In this method, a pulse signal P.sub.OUT having a frequency corresponding to the magnetic change formed based upon the change of the difference between the oscillating frequencies f.sub.a and f.sub.b. Thus, the values of the oscillating frequencies f.sub.a and f.sub.b per se do not adversely affect the detecting accuracy. Namely, when the oscillating frequencies f.sub.a and f.sub.b change due to the difference of characteristics of devices in the production process or the change of characteristics of devices caused by time, the detecting accuracy is not deteriorated.
In the above-mentioned example, the magnetic change is detected utilizing the MREs 410 and 420, but any detecting device can be used instead so long as the oscillating frequencies of the oscillating circuits 430 and 440 can change in response to magnetic change. One example of this will be explained hereunder with reference to FIG. 38, in which a Hall device 200 is used as a detecting means.
In FIG. 38, one of the output signals V.sub.1 of the Hall device 500 is connected to a gate of an MOS transistor 510, while another output signal V.sub.2 is connected to a gate of an MOS transistor 520. The voltages of the output signals V.sub.1 and V.sub.2 of the Hall device 500 change in opposite directions with respect to magnetic change. Thus, the ON resistances of the MOS transistors 510 and 520 also change in opposite directions. Therefore, the oscillating frequencies of the oscillating circuits 430 and 440 can be changed in opposite directions.
In the magnetic detection circuit mentioned above, the pulse signals CKB and CKA output from the oscillating circuit 430 and 440 are divided by the counters 450 and 460, respectively, to produce the divided pulse signals PB1 and PA1 formed by accumulating the phase differences of the pulse signals. Thus, the detecting accuracy of the device with respect to magnetic change can be improved by detecting the phase difference between the divided pulse signals PB1 and PA1.
In this method, to further increase the detecting sensitivity to magnetic change, a value of a dividing ratio of the pulse signals PB1 and PA1 to be divided may be further increased. However, when the dividing ratio is simply increased, while the detecting sensitivity can be improved, conversely a problem arises in that the detecting period (sampling period T.sub.s) is elongated. To overcome this, other examples of the magnetic detection circuit will be explained hereunder.
A magnetic detection circuit for this end may comprise a plurality of counters corresponding to the counter 460 as shown in FIG. 34 and dividing a pulse signal CKA output from an oscillating circuit corresponding to the oscillating circuit 440 as shown in FIG. 34, whereby starting times for counting pulses of the counters are delayed in turn by a sampling period T.sub.s.
Note that in this circuit, the same number of synchronized pulse signals as the number of counters may be output in one sampling period T.sub.s from a counter corresponding to the counter 450 counting a clock signal CKB output from another oscillating circuit 430 as shown in FIG. 34. These synchronized pulse signals are input to a synchronized pulse detecting circuit similar to the circuit 320 shown in FIG. 32, except that a plurality of lines each comprising a plurality of D-FFs and AND gates with a input terminals for each line and thus synchronized pulse signals as above, is input to each input terminal, which corresponds to the input terminal to which the pulse signal PB is input.
From each circuit, a plurality of pulse signals each newly starts to oscillate depending upon the time when each of the synchronized pulse signals is input thereto. This new pulse can be used to define a pulse PA1 as shown in FIG. 34, with which a pulse phase difference is detected with respect to another pulse PB1 as shown in FIG. 34. Accordingly, in this circuit, the pulse signal CKA a shown in FIG. 34 can be divided into a divided period having the same number of sampling periods T.sub.s as the number of the counters. Therefore, the detecting sensitivity can be increased three times compared with the pulse phase difference detecting circuit as shown in FIG. 32.
As explained above, in accordance with the fourth aspect of the invention, a plurality of gate delay circuits are used to generate a plurality of pulse signals having slightly different delay times, and a pulse signal having a delay time closest to the phase difference between the first and the second pulse signals can be selected among the plurality of the pulse signals. Since the delay time defined by the gate delay circuit can be set extremely short, the phase difference between the first and the second pulse signals can be detected with a high accuracy.
On the other hand, in accordance with a fifth aspect of the present invention, the change of a phase difference between pulse signals of an oscillating circuit with an oscillating frequency, which changes in response to a detected physical value, and other pulse signals is detected, so even when the oscillating frequency changes due to changes of temperature characteristic, time elapse, characteristic of devices, source voltage, or the like, the condition of change can be detected with a high accuracy.
Further, even if the gate delay time T.sub.b is increased due to a temperature increment and the detecting accuracy is reduced, within a reasonable range, the devices can correctly work at high temperatures (exceeding 200.degree. C.). The temperature range in which the detection devices can correctly work exceeds that of a conventional detection device utilizing analog circuits. Turning to still another aspect of the invention, in the oscillating circuits used in previous aspects, a plurality of inverters are used with the resistors, including MREs and a capacitor. These oscillating circuits, however, require large electric power due to the current flowing in the inverters. It has been considered to reduce the amplitude of the working voltage in the oscillating circuits in order to reduce the amount of the power consumption. When the amplitude of the working voltage is reduced, however, the reliability of the circuit is reduced with respect to changes of the voltage, noise, or changes of temperature.
Therefore, oscillating circuit comprising a plurality of inverters should reduce the power consumption while keeping the amplitude of the working voltage as is.
According to a sixth aspect of the present invention, there is provided a magnetic detection device, wherein the oscillating circuit includes a plurality of serially arranged inverters, a resistor, one end connected to an output terminal of a predetermined inverter among the plurality of inverters and another end connected to an input terminal of a first inverter among the plurality of inverters, and a capacitor, one end connected to an output terminal of an inverter output opposite to the output of the predetermined inverter, and another end connected the input terminal of the first inverter, the oscillating circuit working as a resistance-capacitance type oscillating circuit, an oscillating frequency thereof being defined by a time constant determined by the resistor and the capacitor, each of the plurality of inverters comprising FETs, a size of the FET of the first inverter being smaller than that of other inverters.
In FIG. 40, inverters Iv1, Iv2, and Iv3 are serially connected to form a ring oscillating circuit. The inverter Iv1 is a complementary MOS circuit and thus has a P-channel type field effect transistor (FET) P1 and N-channel type PET N1. To a drain terminal of the P-channel type FET P1, a direct voltage V.sub.DD supplied from a direct current source through a switch, not shown, is supplied. A source terminal of the N-channel FET N1 is grounded. On the other hand, as shown in FIG. 41, the inverter Iv1 has a threshold voltage V.sub.TH lower than the direct voltage V.sub.DD. When a voltage V.sub.A applied to an input terminal A of this circuit is higher than the threshold voltage V.sub.TH, the N channel. FET N1 is made conductive, whereby an output voltage V.sub.B generated at an output terminal B is turned to 0. At this time, the P-channel FET P.sub.1 is in a non-conductive condition. When the input voltage V.sub.A is lower than the threshold voltage V.sub.TH, the P-channel FET P.sub.1 is made conductive, whereby the output voltage V.sub.B is the direct voltage V.sub.DD. At this time, the N-channel FET N1 is in a non-conductive condition.
As shown in FIG. 42, the inverters Iv2 and Iv3 have the same construction as the inverter Iv1, except the sizes (represented by a ratio of channel width W/channel length L) of the transistors used are larger. A capacitor C.sub.0 is connected between an output terminal D of the inverter Iv2 and an input terminal A of the inverter Iv1, while a resistor R.sub.0 is connected between an output terminal E of the inverter Iv3 and the input terminal A of the inverter Iv1. Thus, the oscillating frequency of this oscillating circuit is determined by a time constant defined by the capacitor C.sub.0 and resistor R.sub.0.
The operation of this oscillating circuit will be explained hereunder. In FIG. 41, at a time when power has just been applied to this circuit, the charge of the capacitor is zero. Thus, an input voltage V.sub.A of the input terminal A of the inverter Iv1 is 0 volt (GND level). Therefore, the output voltage V.sub.B of the inverter Iv1, the output voltage V.sub.D of the inverter Iv2, and the output voltage V.sub.E of the inverter Iv3 are a source voltage V.sub.DD, 0 volt, and a source voltage V.sub.DD, respectively.
Since the output voltage V.sub.E of the inverter Iv3 is a source voltage V.sub.DD and the output voltage V.sub.D of the inverter Iv2 is 0 volt, the capacitor C.sub.0 is charged through the resistor R.sub.0. By this charging operation, the voltage of the input terminal A of the inverter V.sub.A is increased in accordance with the time constant defined by the capacitor C.sub.0 and the resistor R.sub.0.
When the input voltage V.sub.A exceeds the threshold voltage V.sub.TH, each output voltage V.sub.B, V.sub.D, and V.sub.E of each inverter Iv1, Iv2, and Iv3 respectively is reversed. Therefore, the output voltage V.sub.B, the output voltage V.sub.D, and the output voltage V.sub.E are zero, the source voltage V.sub.DD, and 0 volt, respectively. At this time, when the output voltage V.sub.D of the inverter Iv2 is the source voltage V.sub.DD, the input voltage V.sub.A of the input terminal A of the inverter Iv1 is the source voltage V.sub.DD +the threshold voltage V.sub.TH. Since the output voltage V.sub.E of the inverter Iv3 is 0 volt, the charge stored in the capacitor C.sub.0 is discharged through the resistor R.sub.0 in this time. Due to this discharge operation, the input voltage V.sub.A of the input terminal A of the inverter Iv1 is decreased in accordance with the time constant defined by the capacitor C.sub.0 and the resistor R.sub.0.
When the input voltage V.sub.A falls below the threshold voltage V.sub.TH, the output voltages V.sub.B, V.sub.D, and V.sub.E of the inverters Iv1, Iv2, and Iv3, respectively, are again reversed, so the input voltage of the inverter Iv1 is the threshold voltage V.sub.TH and the output voltage V.sub.D of the inverter Iv2 is the source voltage V.sub.DD. Therefore, the charging operation is again started in the capacitor C.sub.0.
The oscillating operation continues due to the repetition of the operation as explained above.
At this time, when the threshold voltage V.sub.TH is set at 1/2 of the source voltage V.sub.DD, the frequency F is represented by the following equation: ##EQU10##
In the oscillating circuit as shown in FIG. 40, when the oscillation as mentioned above is continued, the voltage input to the inverters Iv1, Iv2, and Iv3 is changed from the source voltage V.sub.DD to 0 volt and vice versa. At this time, when the voltage input to the inverters Iv1, Iv2, and Iv3 becomes close to the threshold voltage V.sub.TH, a through current transiently flows from the P-channel FET to the N-channel FET. The current demand in the oscillating circuit as shown in FIG. 40 includes this through current and a charge/discharge current caused by the capacitor C.sub.0 and resistor R.sub.0. The charge/discharge current is a key factor for determining the oscillating frequency, however, thus the charge/discharge current cannot be reduced. Accordingly, to reduce the current demand in this oscillating circuit, it is necessary to reduce the through current. This through current I.sub.D, in a saturated condition, is represented by the following equation: ##EQU11## In a non-saturated condition, it is represented by the following equation: ##EQU12## wherein, W denotes a channel width of a gate;
L denotes a channel length of the gate;
.mu. denotes the surface mobility of a carrier;
Cox denotes a gate capacity per unit area;
V.sub.G denotes a gate voltage;
V.sub.D denotes a drain voltage;
V.sub.T denotes a threshold voltage; and
I.sub.D denotes a drain current.
From these equations, the through current I.sub.D is found to be proportional to the transistor size W/L. This transistor size W/L is a parameter for determining the current carrying capacitance. The narrower the channel width W and the longer the channel length, the more difficult for the current to flow therein. It is considered that the resistance value thereof increases equivalently.
As shown in FIG. 43, when the channel width W and the channel length L of the P channel FET in the inverters Iv1, Iv2, and Iv3 are set at 1250 .mu.m and 5 .mu.m, respectively; the channel width W and the channel length L of the N channel FET in the inverters Iv1, Iv2, and Iv3 are set at 500 .mu.m and 5 .mu.m respectively; and the transistor size W/L of the inverters Iv1, Iv2, and Iv3 is set at the same value as in the conventional inverter, the through current flowing in the inverters Iv1, Iv2, and Iv3 is as shown in FIG. 44.
It is apparent from FIG. 44 that when the inverters Iv1, Iv2, and Iv3 having the same transistor size W/L are used, the through current of the current Il' which flows through the inverter Iv1 becomes the largest. The reason is that a voltage varied in accordance with a time constant defined by the capacitor C.sub.0 and the resistor R.sub.0 is applied to the input terminal A of the inverter Iv1, as shown in FIG. 41. Note that the ratio of change of the voltage changed in accordance with a time constant defined by the capacitor C.sub.0 and the resistor R.sub.0 is generally small, and thus a period during which the input voltage V.sub.A of the inverter Iv1 is closed to the threshold voltage V.sub.TH is longer than that of the input voltage V.sub.B and V.sub.D of other inverters Iv2 and Iv3. Therefore, the through current of the current I1' which flows through the inverter Iv1 becomes larger than that of each current I2' and I3' which flow through other inverters Iv2 and Iv3, respectively.
The transistor size W/L of the inverter Iv1 is set at a value one-fifth of the transistor size W/L of the inverters Iv2 and Iv3, in order to reduce the through current of the current I1' which flows through the inverter Iv1 as shown in FIG. 42. The waveforms of the through currents of the currents I1', I2', and I3' which flow through each inverter Iv1, Iv2, and Iv3 respectively are shown in FIG. 45. It is apparent from FIG. 45 that the through current of the current I1' which flows through the inverter Iv1 can be sufficiently reduced.
Note that the P-channel FET and N-channel FET of the inverter Iv2 should have a driving performance because it has the capacitor C.sub.0 charged and discharged and thus it is difficult to reduce the transistor size W/L. On the other hand, since both the P-channel FET and N-channel FET of the Iv3 are connected to the resistor R.sub.0, it is also difficult to reduce the transistor size W/L. Conversely, since both the P-channel FET P.sub.1 and N-channel FET N1 of the Iv1 need only have a driving performance for charging and discharging a parasitic capacitance generated in a gate of the inverter Iv2, the transistor size W/L can be reduced.
Note also that, when the number of stages of the inverters is increased, the through current can be further reduced.
In these aspects mentioned above, the magnetic detection device may be formed with a plurality of oscillating circuits, each having MREs and a logic processing circuit integrated into one chip. Such a magnetic detection device is generally required to have an anti-noise characteristic and an improved surge characteristic.
According to a seventh aspect of the present invention, there is provided an integrated circuit device used for detecting magnetism, which comprises;
an integrated circuit, a metallic casing containing said integrated circuit; a varistor having electrodes formed on opposite surfaces thereof, one surface thereof being connected to an inside surface of the casing in an electrically insulated condition; a plurality of lead wires for transmitting electrical signals to an external circuit from the metallic casing; and a conductive means for electrically connecting the integrated circuit to the plurality of lead wires through the electrodes formed on each side of the varistor.
FIG. 46(a) is a cross-sectional view of the integrated circuit device, and FIG. 46(b) is a plane view of the construction in a case 906. Shown is a one chip MRE sensor with two lines utilizing a source line in which signals overlap. In that, the IC chip 901 is a silicon chip including a logic processing means, for example, a one chip MRE sensor. A varistor 902 is a non-linear resistance element made mainly of ZnO and having a resistance which can steeply increase or decrease in accordance with an increment of an applied voltage. On both surfaces of the varistor 902, electrode 907s are formed. One of the electrodes is fixedly mounted on a surface of a case 906 made of a metallic material through a melted insulating glass 908. On the other hand, the IC chip 901 is fixedly mounted on the opposite surface of the varistor 902 with a melted insulating glass 908 interposed. One electrode 907 formed on one of the surfaces thereof is printed so that another end of the electrode is extended to the opposite surface of the varistor through a side surface thereof.
In this construction, a lead wire 903, introducing a signal, for example, an electrical source signal, thereto, is connected on the electrode 907 formed on the surface of the varistor 902 through a bonding wire 904. Therefore, the signal introduced thereto is applied to the IC chip 901. A hermetical glass 905 is used for fixing the lead wire 903 to the case 906 insulatingly. A junction coat 910 is used to protect the IC chip 901 and the bonding wire 904.

Number	Date	Country
1-10235	Jan 1989	JPX
1-12526	Jan 1989	JPX
1-96873	Apr 1989	JPX
1-264319	Oct 1989	JPX
1-324290	Dec 1989	JPX
1-336089	Dec 1989	JPX

Number	Name	Date
3808524	Tarassoff et al.	Apr 1974
4104619	Hesler	Aug 1978
4150573	Iinuma et al.	Apr 1979
4182986	Parker	Jan 1980
4296377	Ohkubo	Oct 1981
4308501	Tuccinardi et al.	Dec 1981
4316158	Akita et al.	Feb 1982
4319188	Ito et al.	Mar 1982
4321539	Trump	Mar 1982
4341998	Castera et al.	Jul 1982
4361805	Narimatsu et al.	Nov 1982
4492922	Ohkubo	Jan 1985
4506217	Rothley et al.	Mar 1985
4533872	Boord et al.	Aug 1985
4649342	Nakamura	Mar 1987
4754221	Ao et al.	Jun 1988
4816748	Tazawa	Mar 1989
4875008	Lorenzen	Oct 1989

Number	Date	Country
0281733	Sep 1988	EPX
3621710	Jan 1988	DEX
0068266	Jun 1979	JPX
57-142336	Sep 1982	JPX
58-61583	Apr 1983	JPX
59-37611	Sep 1984	JPX
60-104925	Jun 1985	JPX
60-173434	Sep 1985	JPX
54-138457	Oct 1989	JPX
WO8807206	Sep 1988	WOX
805926	Dec 1958	GBX
1397246	Jun 1975	GBX

Magnetic detecting circuit having magnetoresistance effective elements oriented in at least two different directions

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Priority Claims (6)

Parent Case Info

US Referenced Citations (18)

Foreign Referenced Citations (12)

Non-Patent Literature Citations (6)

Divisions (1)

Continuations (1)

Entry
V. P. Chirkov, "Magnetic Induction-To-Frequency Transducer", Measurement Techniques vol. 12, No. 12, Dec. 1978, pp. 1728-1729.
"Signal Injector", Elektor, vol. 4, Nos. 7/8, Jul./Aug. 1978, p. 27.
W. Kwiatkowski, et al., "The permalloy magnetoresistive sensors-properties and applications", Journal of Physics E: Scientific Instr., vol. 19, No. 7, Jul. 1986, 508-515.
Sensor Device Hand Book (Feb. 25, 1985) pp. 54-59 and translation.
Macario, R. C. V.; Discriminative Metal Detector, Wireless World, Jul. 1978, vol. 84, No. 1511.
Agrawal et al.; Transistor probe simplifies solid-state gaussmeter, Electronics, Oct. 9, 1980, vol. 53, No. 22.