Liquid level detection device

Abstract

A liquid level detection device includes a float that ascends and descends following a liquid level, a magnet attached to the float, a guide that guides the float to ascend and descend, magnetic sensors attached to the guide and sensing a magnetic flux density varying in response to a position assumed by the magnet as the magnet ascends/descends and outputting an electrical signal corresponding to the magnetic flux density, and a detection circuit that detects a position of the float based on electrical signals respectively output from the magnetic sensors. The detection circuit detects the position of the float based on electrical signals output from two adjacent ones of the magnetic sensors.

Claims

1. A liquid level detection device comprising: a float that is rotatable, and ascends and descends following a liquid level; a magnet attached to the float; a guide that guides the float to ascend and descend; a plurality of magnetic sensors attached to the guide to sense a magnetic flux density varying in response to a position of the magnet as the magnet ascends/descends and output an electrical signal corresponding to the magnetic flux density; and a detection circuit that detects a position of the float based on electrical signals respectively output from the plurality of magnetic sensors; the detection circuit detecting the position of the float based on electrical signals output from two adjacent ones of the plurality of magnetic sensors; and the detection circuit extracting electrical signals output from the two adjacent magnetic sensors based on a comparison with only an intermediate voltage among electrical signals respectively output from all of the plurality of magnetic sensors.

2. The liquid level detection device according to claim 1, wherein the detection circuit calculates angular information with the two extracted electrical signals with one electrical signal represented as a sine wave and the other electrical signal represented as a cosine wave, and detects the position of the float based on the calculated angular information.

3. The liquid level detection device according to claim 2, wherein at least one of the plurality of magnetic sensors outputs an output signal corresponding to a variation of a magnetic flux density of an external magnetic field, and senses a polarity of a magnetic field applied to the magnetic sensor.

4. The liquid level detection device according to claim 1, wherein at least one of the plurality of magnetic sensors outputs an output signal corresponding to a variation of a magnetic flux density of an external magnetic field, and senses a polarity of a magnetic field applied to the magnetic sensor.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a diagram for illustrating an external configuration of a liquid level detection device according to a first preferred embodiment of the present invention.

(2) FIG. 2 is a diagram for illustrating a plurality of magnetic sensors 5 attached to a guide 10 according to the first preferred embodiment of the present invention.

(3) FIG. 3 is a circuit configuration diagram of liquid level detection device 1 according to the first preferred embodiment of the present invention.

(4) FIG. 4 is a diagram for illustrating a pattern of a magnetoresistive element of magnetic sensor 5 according to the first preferred embodiment of the present invention.

(5) FIGS. 5A and 5B are diagrams for illustrating a principle of detection via magnetic sensor 5 according to the first preferred embodiment of the present invention.

(6) FIG. 6 is a diagram for illustrating an arrangement of magnet 2 attached to a float 20 according to the first preferred embodiment of the present invention.

(7) FIG. 7 is a diagram for illustrating a layout of magnets 2A to 2D attached to float 20 and magnetic sensors 5A to 5C according to the first preferred embodiment of the present invention.

(8) FIGS. 8A-8G are diagrams for illustrating a relationship of float 20 with magnetic sensors according to the first preferred embodiment of the present invention when float 20 positionally varies as it ascends/descends.

(9) FIG. 9 illustrates waveforms of signals output from a plurality of magnetic sensors as float 20 ascends/descends in the first preferred embodiment of the present invention.

(10) FIG. 10 is an enlarged image of a prescribed region of FIG. 9.

(11) FIG. 11 is a diagram for schematically illustrating a relationship between magnetic sensor 5 and a magnetic vector P according to the first preferred embodiment of the present invention.

(12) FIGS. 12A and 12B are diagrams for illustrating accuracy of angular information according to the first preferred embodiment of the present invention.

(13) FIG. 13 is a flowchart for illustrating a detection method via liquid level detection device 1 according to the first preferred embodiment of the present invention.

(14) FIG. 14 is a diagram for illustrating a layout of magnets 2E and 2F attached to float 20 and magnetic sensors 5A, 5B, 5C according to the first preferred embodiment of the present invention.

(15) FIG. 15 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and magnetic sensors 5A, 5B, 5C according to a second preferred embodiment of the present invention.

(16) FIG. 16 illustrates waveforms of signals output from a plurality of magnetic sensors as float 20 ascends/descends in the second preferred embodiment of the present invention.

(17) FIG. 17 is an enlarged image of a prescribed region of FIG. 16.

(18) FIG. 18 is a diagram for schematically illustrating a relationship between magnetic sensor 5 and magnetic vector P according to the second preferred embodiment of the present invention.

(19) FIGS. 19A and 19B are diagrams for illustrating accuracy of angular information according to the second preferred embodiment of the present invention.

(20) FIG. 20 is a diagram for illustrating a layout of magnets 2I-2T attached to float 20 and magnetic sensors 5A, 5B, 5C according to a third preferred embodiment of the present invention.

(21) FIG. 21 illustrates waveforms of signals output from magnetic sensors 5 as float 20 ascends/descends in the third preferred embodiment of the present invention.

(22) FIG. 22 is an enlarged image of a prescribed region of FIG. 21.

(23) FIGS. 23A-23C are diagrams for schematically illustrating a relationship between magnetic sensor 5 and magnetic vector P according to the third preferred embodiment of the present invention.

(24) FIG. 24 is a diagram for illustrating a method of extracting waveforms of two signals output in each of regions T1 to T3 from waveforms of signals output from a plurality of magnetic sensors 5A to 5C according to the third preferred embodiment of the present invention.

(25) FIGS. 25A-25F are diagrams for illustrating accuracy of angular information according to the third preferred embodiment of the present invention.

(26) FIG. 26 is a flowchart for illustrating a detection method via liquid level detection device 1 according to the third preferred embodiment of the present invention.

(27) FIG. 27 is a diagram for illustrating a layout of magnets 2U and 2V attached to float 20 and magnetic sensors 5A, 5B, 5C according to a first exemplary variation of the third preferred embodiment of the present invention.

(28) FIG. 28 is a diagram for illustrating a layout of magnets 2I-2P attached to float 20 and magnetic sensors 5A, 5B, 5C according to the third preferred embodiment of the present invention.

(29) FIG. 29 illustrates waveforms of signals output from magnetic sensors 5 as float 20 ascends/descends in a second exemplary variation of the third preferred embodiment of the present invention.

(30) FIG. 30 is an enlarged image of a prescribed region of FIG. 29.

(31) FIGS. 31A-31C are diagrams for illustrating an angle of a magnetic vector received by magnetic sensor 5 according to the second exemplary variation of the third preferred embodiment of the present invention.

(32) FIG. 32 is a diagram for illustrating a method of extracting waveforms of two signals output in each of regions T1 to T3 from waveforms of signals output from a plurality of magnetic sensors 5A to 5C according to the second exemplary variation of the third preferred embodiment of the present invention.

(33) FIGS. 33A-33F are diagrams for illustrating accuracy of angular information according to the second exemplary variation of the third preferred embodiment of the present invention.

(34) FIG. 34 is a diagram for illustrating a layout of magnets 2W and 2X attached to float 20 and magnetic sensors 5A, 5B, 5C according to a third exemplary variation of the third preferred embodiment of the present invention.

(35) FIG. 35 is a diagram for illustrating a pattern of a magnetoresistive element of a magnetic sensor according to a fourth preferred embodiment of the present invention.

(36) FIGS. 36A and 36B are diagrams for illustrating a case where the position of float 20 is displaced with respect to guide 10 according to a fifth preferred embodiment of the present invention.

(37) FIG. 37 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and magnetic sensors according to the fifth preferred embodiment of the present invention.

(38) FIGS. 38A and 38B are diagrams for illustrating variation of a bias magnetic field vector according to the fifth preferred embodiment of the present invention.

(39) FIGS. 39A and 39B are diagrams for illustrating accuracy of angular information according to the fifth preferred embodiment of the present invention.

(40) FIG. 40 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and a plurality of magnetic sensors 5Qs according to an exemplary variation of the fifth preferred embodiment of the present invention.

(41) FIG. 41 illustrates a magnetic sensor 5R based on another preferred embodiment of the present invention.

(42) FIG. 42 is a diagram for illustrating a bias magnetic field vector for a magnetic sensor 5P.

(43) FIG. 43 illustrates a circuit configuration of a magnetic sensor 5R.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(44) Preferred embodiments of the present invention will be described hereinafter in detail with reference to the drawings. Note that in the figures, identical or corresponding components are identically denoted, and accordingly, will not be described redundantly.

First Preferred Embodiment

(45) FIG. 1 is a diagram for illustrating an external configuration of a liquid level detection device according to a first preferred embodiment of the present invention.

(46) With reference to FIG. 1, liquid level detection device 1 includes a float 20 that ascends and descends following a liquid level, a guide (a guide) 10, and a detection circuit 50.

(47) Detection circuit 50 detects the position of float 20 based on output signals detected from a plurality of magnetic sensors (AMR (Anisotropic Magneto Resistance) elements) attached to guide 10.

(48) FIG. 2 is a diagram for illustrating a plurality of magnetic sensors 5 attached to guide 10 according to the first preferred embodiment.

(49) Referring to FIG. 2, the plurality of magnetic sensors are disposed at predetermined intervals in an ascending/descending direction.

(50) Float 20 is provided with a magnet 2. Specifically, magnets 2A, 2B, 2C and 2D are attached as two magnet units. Magnets 2A and 2B define a magnet unit. Magnets 2C and 2D define a magnet unit.

(51) The plurality of magnetic sensors 5 sense a magnetic flux density according to an ascending/descending movement of magnet 2 attached to float 20, and output an electrical signal corresponding to the magnetic flux density. While in this example, a configuration of 4-pin magnetic sensor 5 will be described as one example, the number of pins is not limited thereto, and those skilled in the art would be able to redesign it as appropriate.

(52) FIG. 3 is a circuit configuration diagram of liquid level detection device 1 according to the first preferred embodiment. Referring to FIG. 3, liquid level detection device 1 according to the first preferred embodiment includes the plurality of magnetic sensors (AMR elements) 5 and detection circuit 50. In this example, a case where n magnetic sensors are provided is shown.

(53) Detection circuit 50 includes an A/D circuit 60 which is an analog/digital conversion circuit, a P/S conversion circuit 30 which is a parallel/serial conversion circuit, and an MPU (Micro-processing unit) 40 which performs a computing process.

(54) A/D circuit 60 is connected to the plurality of (or n) magnetic sensors 5 and converts a received analog signal into a digital signal.

(55) P/S conversion circuit 30 receives digital signals input from A/D circuit 60 in parallel in synchronization with a clock CLK received from MPU 40, converts the received digital signals serially to a signal, and outputs it to MPU 40.

(56) MPU 40 computes signals output from the plurality of (or n) magnetic sensors 5 and received from P/S conversion circuit 30 and detects the position of float 20.

(57) While in this example MPU 40 is described such that it is configured to receive for a signal output from A/D circuit 60 an output of P/S conversion circuit 30 synchronized with clock CLK, it is not limited to this configuration and may also be changed to a configuration to receive a digital signal from A/D circuit 60 via a multiplexer.

(58) FIG. 4 is a diagram for illustrating a pattern of a magnetoresistive element of magnetic sensor 5 according to the first preferred embodiment.

(59) Referring to FIG. 4, magnetic sensor 5 has a bridge structure including four magnetoresistive elements MR1 to MR4 (collectively referred to as a magnetoresistive element MR).

(60) When a magnetic field is applied to magnetic sensor 5, magnetoresistive elements MR1 to MR4 have their resistance values varied, and in response thereto, magnetic sensor 5 outputs signals V+ and V. Magnetic sensor 5 outputs a difference V between signals V+ and V.

(61) Magnetoresistive element MR of magnetic sensor 5 is an anisotropic magnetoresistive element and has a folded pattern structure.

(62) A resistance value of magnetoresistive element MR when a magnetic field is applied thereto has a characteristic which is minimized when a magnetic field perpendicular (i.e., 90 degrees) to the element's longitudinal direction (or a direction of a current) is applied and which is maximized when a magnetic field parallel (i.e., 0 degree) to the element's longitudinal direction is applied.

(63) Furthermore, magnetic sensor 5 is provided with bias magnets 3A and 3B. Bias magnets 3A and 3B are disposed such that a bias magnetic field is applied to magnetoresistive elements MR1 to MR4 in a direction from an upper left side to a lower right side.

(64) Although magnetoresistive element MR of magnetic sensor 5 of this example is described as a folded pattern structure by way of example, it is not limited to the folded shape in particular, and a person skilled in the art would be able to redesign the pattern structure as appropriate to allow magnetic sensor 5 to have enhanced detection characteristics. Furthermore, while bias magnets 3A and 3B are arranged (or oriented) in this example such that a bias magnetic field vector with an angle of 45 degrees is applied in a direction from an upper left side to a lower right side by way of example, this arrangement or angle can also be redesigned by a skilled artisan as appropriate to allow magnetic sensor 5 to have enhanced detection characteristics.

(65) Furthermore, while in this example a configuration will be described in which a bias magnetic field vector is applied based on two bias magnets 3A, 3B, it is also possible to dispose one bias magnet 3A instead of two bias magnets at a center portion of magnetoresistive elements MR1 to MR4 with an inclination of 45 degrees to apply a similar bias magnetic field vector. By this configuration, the number of bias magnets and hence the cost of magnetic sensor 5 is able to be reduced. Alternatively, a bias magnet may be disposed on a substrate provided with magnetoresistive elements MR1 to MR4, or a bias magnet may be disposed on a back surface of the substrate.

(66) FIGS. 5A and 5B are diagrams for illustrating a principle of detection via magnetic sensor 5 according to the first preferred embodiment.

(67) FIG. 5A is a diagram for illustrating a bias magnetic field vector varying with an external magnetic field.

(68) As shown in FIG. 5A, the bias magnetic field vector of magnetic sensor 5 has a direction varying with an external magnetic field with respect to an ascending/descending direction. In this example, a bias magnetic field vector V0 with no external magnetic field applied is indicated by a solid line. The bias magnets are set to attain a magnetic field strength allowing magnetic sensor 5 to reach a saturated sensitivity region.

(69) Bias magnetic field vector V0 varies to a bias magnetic field vector V1 according to an external magnetic field (applied in a right to left direction).

(70) In contrast, bias magnetic field vector V0 varies to a bias magnetic field vector V2 according to an external magnetic field (applied in a left to right direction).

(71) The bias magnetic field vector varies as the external magnetic field's magnetic flux density varies. Magnetic sensor 5 detects variation of the bias magnetic field vector and outputs a signal (a potential difference V) corresponding to the result of the detection.

(72) FIG. 5B shows a characteristic of variation of a signal output from magnetic sensor 5 according to variation in magnetic flux density of an external magnetic field.

(73) As shown in FIG. 5B, a predetermined magnetic flux density ST is applied based on a bias magnetic field according to bias magnets 3A and 3B. The output in this case is preferably set to an intermediate value, and potential difference V varies as a magnetic field applied to magnetic sensor 5 varies in direction.

(74) Potential difference V shifts toward V1 in response to variation in magnetic flux density of an external magnetic field in a right to left direction.

(75) In contrast, potential difference V shifts toward V2 in response to variation in magnetic flux density of an external magnetic field in a left to right direction.

(76) According to an increase/decrease of potential difference V from the intermediate value, a polarity of the magnetic field applied to magnetic sensor 5 (in which direction the magnetic field is applied) is able to be sensed. Further, saturated magnetic field strength is also able to be increased by changing bias magnets 3A, 3B in magnetic force strength.

(77) As will be described hereinafter, the position of float 20 is able to be detected based on a signal waveform (potential difference V) corresponding variation in magnetic flux density of the external magnetic field.

(78) FIG. 6 is a diagram for illustrating an arrangement of magnet 2 attached to float 20 according to the first preferred embodiment.

(79) FIG. 6 shows a case in which float 20 is viewed from above. Furthermore, magnets 2A and 2B defining a magnet unit are provided to face each other with a guide interposed therebetween. In this example, magnets 2A and 2B are disposed opposite to each other with their respective N poles facing each other. Magnets 2A and 2B may be disposed opposite to each other with their respective S poles facing each other.

(80) This arrangement allows the magnetic force to have a direction along the guide and cancels a magnetic force component having a direction perpendicular to the direction along the guide. Furthermore, even when float 20 rotates, the magnetic force's direction, the magnetic flux density and the like do not have substantial variation, and magnetic sensor 5 accurately measures an amount of displacement of the magnetic flux density.

(81) FIG. 7 is a diagram for illustrating a layout of magnets 2A to 2D attached to float 20 and magnetic sensors 5A to 5C according to the first preferred embodiment.

(82) As shown in FIG. 7, magnets 2A, 2B define a set to define a magnet unit. Magnets 2C, 2D define a set to define a magnet unit.

(83) Magnets 2A and 2B defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2C and 2D defining a magnet unit are disposed to have their respective S poles facing each other. Adjacent magnet units are disposed such that their magnets' magnetic poles are different.

(84) In this example, a distance between magnet 2A and magnet 2B is preferably set to a spacing of twice a distance a, and magnetic sensors 5 are disposed to pass through a center thereof. An interval (or center-to-center distance) between the adjacent magnet units is also set to twice distance a. Magnetic sensors 5 are also disposed at intervals of a distance a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(85) While in this example a case where three magnetic sensors 5A to 5C are disposed to detect the position of float 20 will be described, a case where more magnetic sensors are disposed is also similarly discussed.

(86) In this example, as a position of float 20, for example, a center of magnet 2A and magnet 2C (or magnet 2B and magnet 2D) in the ascending/descending direction is set as a reference position (or a center point). In this case, a case where magnetic sensor 5B is positioned at the reference position (or center point) is shown.

(87) FIGS. 8A-8G are diagrams for illustrating a relationship of float 20 with magnetic sensors according to the first preferred embodiment when float 20 positionally varies as it ascends/descends.

(88) In this example, a case where float 20 varies in a right to left direction (an ascending direction by way of example) will be described.

(89) In FIG. 8A, a case where float 20 ascends and approaches magnetic sensor 5A (i.e., a state S0) is shown.

(90) Magnetic sensor 5A is affected by a magnetic field (or a line of magnetic force) generated by magnets 2A and 2B of float 20. Specifically, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies, potential difference V decreases. The other magnetic sensors 5B and 5C are also affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(91) In FIG. 8B, a case where float 20 further ascends from FIG. 8A by distance a (i.e., a state S1) is shown.

(92) Magnetic sensor 5A is in a state where it is located on a center line between magnet 2A and magnet 2B. In this example, this state is set as an initial state.

(93) Magnetic sensor 5B is affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B. Accordingly, bias magnetic field vector V0 of magnetic sensor 5B varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies, potential difference V decreases. Magnetic sensor 5C is also affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(94) In FIG. 8C, a case where float 20 further ascends from FIG. 8B by distance a (i.e., a state S2) is shown.

(95) A case is shown in which magnetic sensor 5A is exposed to a magnetic field maximally in the ascending/descending direction by a magnetic field generated by magnets 2A, 2B, 2C, 2D. Specifically, magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force from magnet 2A to magnet 2C (or a line of magnetic force from magnet 2B to magnet 2D). Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases (or is maximized).

(96) Magnetic sensor 5B is in a state where it is located on the center line between magnet 2A and magnet 2B. Thus, it is in the initial state.

(97) Magnetic sensor 5C is affected by a magnetic field generated by magnets 2A and 2B. Specifically, magnetic sensor 5C is affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(98) In FIG. 8D, a case where float 20 further ascends from FIG. 8C by distance a (i.e., a state S3) is shown.

(99) Magnetic sensor 5A is in a state where it is located on the center line between magnet 2C and magnet 2D. Thus, it is in the initial state.

(100) A case is shown in which magnetic sensor 5B is exposed to a magnetic field maximally in the ascending/descending direction by a magnetic field generated by magnets 2A, 2B, 2C, 2D. Specifically, magnetic sensor 5B is affected by a left to right magnetic field as a line of magnetic force from magnet 2A to magnet 2C (or a line of magnetic force from magnet 2B to magnet 2D). Accordingly, bias magnetic field vector V0 of magnetic sensor 5B varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases (or is maximized).

(101) Magnetic sensor 5C is in a state where it is located on the center line between magnet 2A and magnet 2B. Thus, it is in the initial state.

(102) In FIG. 8E, a case where float 20 further ascends from FIG. 8D by distance a (i.e., a state S4) is shown.

(103) A case is shown in which magnetic sensor 5A is exposed to a magnetic field in the ascending/descending direction by a magnetic field generated by magnets 2C, 2D. Specifically, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2C and 2D. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(104) Magnetic sensor 5B is in a state where it is located on the center line between magnet 2C and magnet 2D. Thus, it is in the initial state.

(105) A case is shown in which magnetic sensor 5C is exposed to a magnetic field maximally in the ascending/descending direction by a magnetic field generated by magnets 2A, 2B, 2C, 2D. Specifically, magnetic sensor 5C is affected by a left to right magnetic field as a line of magnetic force from magnet 2A to magnet 2C (or a line of magnetic force from magnet 2B to magnet 2D). Accordingly, bias magnetic field vector V0 of magnetic sensor 5C varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases (or is maximized).

(106) In FIG. 8F, a case where float 20 further ascends from FIG. 8E by distance a (i.e., a state S5) is shown.

(107) Magnetic sensor 5A is affected by a magnetic field generated by magnets 2C and 2D. Specifically, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2C and 2D. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A slightly varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(108) A case is shown in which magnetic sensor 5B is exposed to a magnetic field in the ascending/descending direction by a magnetic field generated by magnets 2C and 2D. Specifically, magnetic sensor 5B is affected by a right to left magnetic field as a line of magnetic force of magnets 2C and 2D. Accordingly, bias magnetic field vector V0 of magnetic sensor 5B varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(109) Magnetic sensor 5C is in a state where it is located on the center line between magnet 2C and magnet 2D. Thus, it is in the initial state.

(110) In FIG. 8G, a case where float 20 further ascends from FIG. 8F by distance a (i.e., a state S6) is shown.

(111) Magnetic sensors 5A and 5B are slightly affected by a magnetic field generated by magnets 2C and 2D. Specifically, magnetic sensors 5A and 5B are affected by a right to left magnetic field as a line of magnetic force of magnets 2C and 2D. Accordingly, bias magnetic field vector V0 of each of magnetic sensors 5A and 5B slightly varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(112) A case is shown in which magnetic sensor 5C is exposed to a magnetic field in the ascending/descending direction by a magnetic field generated by magnets 2C and 2D. Specifically, magnetic sensor 5C is affected by a right to left magnetic field as a line of magnetic force of magnets 2C and 2D. Accordingly, bias magnetic field vector V0 of magnetic sensor 5C varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(113) FIG. 9 illustrates waveforms of signals output from a plurality of magnetic sensors as float 20 ascends/descends in the first preferred embodiment.

(114) As shown in FIG. 9, a positional relationship of states S0 to S6 and a relationship of the signals output is shown. For example, when focusing on magnetic sensor 5A, magnetic sensor 5A outputs a signal corresponding to a magnetic flux density of an external magnetic field received by magnetic sensor 5A.

(115) In state S0, a case is shown in which in accordance with an external magnetic field received by magnetic sensor 5A the bias magnetic field vector varies and is decreased as an output signal (or potential difference V).

(116) In state S1, magnetic sensor 5A is located on the center line between magnets 2A and 2B, or in the initial state, and in this example, magnetic sensor 5A in the initial state outputs a signal (potential difference V) with a voltage having an intermediate value (i.e., an intermediate voltage).

(117) In state S2, a case is shown where magnetic sensor 5A outputs a maximized signal when the bias magnetic field vector of magnetic sensor 5A varies toward the bias magnetic field vector V2 side.

(118) In state S3, a case is shown where magnetic sensor 5A is located on the center line between magnets 2C and 2D, or in the initial state, and outputs the intermediate voltage.

(119) In state S4, a case is shown in which in accordance with an external magnetic field received by magnetic sensor 5A the bias magnetic field vector varies and the output signal (or potential difference V) is decreased.

(120) In state S4 et seq., a case is shown where the signal output from magnetic sensor 5A varies based on an external magnetic field varying with distance.

(121) Furthermore, for magnetic sensor 5B, a waveform obtained by shifting the output signal of magnetic sensor 5A by distance a (or by a phase of 90 degrees) is shown. For magnetic sensor 5C, a waveform obtained by shifting the output signal of magnetic sensor 5B by distance a (or by a phase of 90 degrees) is shown.

(122) FIG. 10 is an enlarged image of a prescribed region of FIG. 9.

(123) FIG. 10 represents waveforms of signals output from a plurality of magnetic sensors 5A to 5C in a hatched region of FIG. 9 indicating the prescribed region.

(124) The waveforms of the signals output from magnetic sensors 5A and 5B can be modeled (or approximated) into a horizontal component (in the ascending/descending direction) of a magnetic vector P of an external magnetic field varying along a circle, as will be described hereinafter, when the intermediate voltage serves as a reference.

(125) Specifically, signal waveforms having phases offset by 90 degrees can be detected as electrical signals output from two adjacent magnetic sensors.

(126) In this example, as the signals have phases offset by 90 degrees, one output signal (electrical signal) can be represented as a sine wave (sin ) and the other output signal (electrical signal) as a cosine wave (cos ). Based on the two output signals (electrical signals), an angle of magnetic vector P of the external magnetic field is calculated.

(127) In the present preferred embodiment, electrical signals output from two adjacent ones of a plurality of magnetic sensors are detected and therefrom an angle of a magnetic vector of an external magnetic field is calculated, and the position of the float is detected based on the angle of the magnetic vector as calculated.

(128) FIG. 11 is a diagram for schematically illustrating a relationship between magnetic sensor 5 and magnetic vector P according to the first preferred embodiment.

(129) FIG. 11 represents a magnetic vector applied to magnetic sensors 5A and 5B in the ascending/descending direction of float when state S2 shifts to state S3. Note that the ascending/descending direction is a direction along the x axis. Magnetic vector P indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2A and the S pole of magnet 2C for example.

(130) Although a line of magnetic force of a magnetic field generated by the N pole of magnet 2B and the S pole of magnet 2D is omitted for the sake of simplicity, a component of magnetic vector P perpendicular to the ascending/descending direction is canceled by a magnetic vector of the line of magnetic force of the magnetic field generated by the N pole of magnet 2B and the S pole of magnet 2D. Accordingly, magnetic sensors 5A, 5B are exposed to an external magnetic field composed only of a component in the ascending/descending direction. As has been previously discussed, the bias magnetic field vector in each magnetic sensor 5 varies according to this external magnetic field.

(131) As one example, as a magnitude of a magnetic vector representing an external magnetic field and a magnetic flux density (an AMR output) are correlated, an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P cos and an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P sin . Based on the two output signals (electrical signals), angle of magnetic vector P is calculated.

(132) Specifically, tan (P sin /P cos ) is calculated based on two output signals (electrical signals), and angular information is calculated by calculating arctan .

(133) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(134) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(135) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 positionally varies by distance a.

(136) For example, as a position of float 20, a center of magnets 2A to 2C in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in state S2 in FIG. 8C is the same position as the position of magnetic sensor 5A.

(137) In this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a/2.

(138) While in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship from magnetic vector 5A, a positional relationship from magnetic vector 5B can also be determined. Furthermore, as a matter of course, a similar method can also be followed to use electrical signals of magnetic sensors 5B, 5C to calculate angular information of a magnetic vector and determine a positional relationship from magnetic vector 5B. A different method is also similarly discussed.

(139) FIGS. 12A and 12B are diagrams for illustrating accuracy of angular information according to the first preferred embodiment.

(140) FIG. 12A compares arctan with a reference value when angle is changed from 0 to 90 degrees with one output signal (an electrical signal) set to P cos and the other output signal (an electrical signal) set to P sin .

(141) As a result of a simulation, there is no substantial difference from the reference value.

(142) Furthermore, for angular accuracy, as shown in FIG. 12B, a case is shown in which there is only a deviation of 2 degrees from the reference value, and the position of float 20 can be detected with high accuracy.

(143) FIG. 13 is a flowchart for illustrating a detection method via liquid level detection device 1 according to the first preferred embodiment.

(144) As shown in FIG. 13, two adjacent signals both exceeding the intermediate voltage are extracted (step SP2). In this example, the intermediate voltage is preferably set to a voltage of an output signal for the initial state as one example. Specifically, as has been described with reference to FIGS. 8A-8G, for example, the state is a state in which magnetic sensor 5A is located on the center line between magnets 2A and 2B, and by previously measuring voltage, the intermediate voltage can be set. There are various methods for setting the intermediate voltage and setting the intermediate voltage is not limited to the method described above, and the intermediate voltage may be set to a value intermediate between a peak value's maximum value and minimum value for example.

(145) Then, two electrical signals in a region surrounded by a dotted line as described with reference to FIG. 9 are extracted. Subsequently, a magnetic vector's angle is calculated based on the extracted two signals (step SP4). More specifically, of the two electrical signals, one output signal (an electrical signal) is preferably set to P cos and the other output signal (an electrical signal) is preferably set to P sin , and based on the two output signals (electrical signals), the magnetic vector's angle is calculated. And tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(146) Subsequently, the position of float 20 is calculated based on the magnetic vector's angle (step SP6). Based on the calculated angular information , the reference position (or center point) of float 20 is calculated from the positions of the magnetic sensors. For example, as has been described above, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a/2.

(147) Subsequently, the process ends (END).

(148) Liquid level detection device 1 according to the first preferred embodiment allows two electrical signals to be used to detect the position of float 20 with high accuracy. According to this method, it is not necessary to provide a switching circuit or the like for switching a signal, so that a circuit configuration is able to be simplified and miniaturization is able to be achieved.

(149) Further, while there is a possibility that, following variation in the ambient temperature, the magnets and the magnetic sensors may have their characteristics varied and an output signal may vary, the angle is calculated with two output signals' tan (P sin /P cos ) calculated and accordingly, an amount of variation following the ambient temperature is canceled, and an error due to an effect of the ambient temperature is thus able to be reduced and highly accurate positional detection is able to be achieved.

(150) While in this example a case where a distance between magnet 2A and magnet 2C preferably is preferably set to an interval of twice distance a has been described, those skilled in the art would be able to adjust magnet 2A or the like in the direction of its thickness (i.e., the N pole and the S pole in width) and thus change a design as appropriate to allow magnetic sensor 5 to have enhanced detection characteristics.

(151) FIG. 14 is a diagram for illustrating a layout of magnets 2E and 2F attached to float 20 and magnetic sensors 5A, 5B, 5C according to the first preferred embodiment.

(152) As shown in FIG. 14, magnets 2E and 2F define a set to define a magnet unit.

(153) The magnet unit defined by magnets 2E and 2F is divided into two regions and arranged such that S poles or N poles face each other in each region. Divisional, adjacent regions are arranged such that the magnets have magnetic poles arranged differently.

(154) Furthermore, in this example, a distance between magnet 2E and magnet 2F is preferably set to a spacing of twice distance a. Furthermore, magnetic sensors 5 are disposed to pass through a center thereof. Furthermore, a distance of the two divisional regions of the N pole and the S pole is preferably set to an interval of twice distance a, and magnetic sensors 5 are also disposed at intervals of distance a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(155) Magnets 2E and 2F generate a magnetic field (a line of magnetic force) which is basically the same as that generated by the layout of FIG. 7, and signals output therefor have waveforms similar to those described with reference to FIG. 9. Accordingly, the position of float 20 can be detected with high accuracy according to a method similar to that described above.

(156) By this configuration, the number of magnets to be disposed is able to be reduced and the layout of the magnets is also able to be easily done.

(157) While in this example, a configuration has been described in which magnets are disposed opposite to each other so that magnetic sensor 5 is able to measure a magnetic field with high accuracy even when float 20 rotates, in a case where float 20 does not rotate and only moves along the guide in the ascending/descending direction, it is also possible to adopt a configuration without opposite magnets (i.e., with a magnet only on one side). The following configuration is also similarly discussed. In that case, it is also possible to provide a configuration in which a bias magnet provided in a magnetic sensor is not provided.

Second Preferred Embodiment

(158) While in the first preferred embodiment, a configuration of liquid level detection device 1 that detects the position of float 20 preferably by using a plurality of magnet units has been described, in the present preferred embodiment, a liquid level detection device that detects the position of float 20 preferably by using a single magnet unit will be described.

(159) FIG. 15 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and magnetic sensors 5A, 5B, 5C according to the second preferred embodiment.

(160) As shown in FIG. 15, magnets 2G and 2H define a set to define a magnet unit.

(161) Magnets 2G and 2H defining a magnet unit are disposed to have their respective N poles facing each other. While in this example an example will be described in which the N poles face each other, it is also possible to adopt a configuration in which the S poles face each other.

(162) Furthermore, in this example, a distance between magnets 2G and 2H is preferably set to a spacing of twice a distance a, and magnetic sensors 5 are disposed to pass through a center thereof. Magnetic sensors 5 are also disposed at intervals of distance a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(163) While in this example a case where three magnetic sensors 5A to 5C are disposed to detect the position of float 20 will be described, a case where more magnetic sensors are disposed is also similarly discussed.

(164) In this example, as a position of float 20, for example, a center of magnet 2G (or magnet 2H) in the ascending/descending direction is set as a reference position (or a center point). In this case, a case where magnetic sensor 5B is positioned at the reference position (or center point) is shown.

(165) FIG. 16 illustrates waveforms of signals output from a plurality of magnetic sensors as float 20 ascends/descends in the second preferred embodiment.

(166) As shown in FIG. 16, similarly as has been described for waveforms of signals output as shown in FIG. 9, as float 20 approaches magnetic sensor 5A, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2G and 2H. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases. The other magnetic sensors 5B and 5C are also affected by a right to left magnetic field as a line of magnetic force of magnets 2A and 2B, and accordingly, as the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(167) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2G and 2H (i.e., a state S7). In this example, this state is set as an initial state. And a voltage of an output signal in this state is set as an intermediate voltage.

(168) Magnetic sensor 5B is affected by a right to left magnetic field as a line of magnetic force of magnets 2G and 2H. Accordingly, bias magnetic field vector V0 of magnetic sensor 5B varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases. Magnetic sensor 5C is also slightly affected by a right to left magnetic field as a line of magnetic force of magnets 2G and 2H. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(169) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2G and 2H (a state S8). Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(170) Magnetic sensor 5B is in a state where it is located on the center line between magnets 2G and 2H. Thus, it is in the initial state.

(171) Magnetic sensor 5C is affected by a right to left magnetic field as a line of magnetic force of magnets 2G and 2H. Accordingly, bias magnetic field vector V0 of magnetic sensor 5C varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(172) Thus, magnetic sensors 5B and 5C output signals having waveforms, respectively, offset from a waveform of an output signal of magnetic sensor 5A by distance a (a phase of 90 degrees) and further by distance a (a phase of 90 degrees), respectively.

(173) FIG. 17 is an enlarged image of a prescribed region of FIG. 16.

(174) FIG. 17 represents waveforms of signals output from a plurality of magnetic sensors 5A and 5B in a hatched region of FIG. 16 indicating the prescribed region.

(175) The waveforms of the signals output from magnetic sensors 5A and 5B can be modeled (or approximated) into a horizontal component (in the ascending/descending direction) of magnetic vector P of an external magnetic field varying along a circle, as will be described hereinafter, when the intermediate voltage serves as a reference.

(176) Specifically, signal waveforms having phases offset by 90 degrees can be detected as electrical signals output from two adjacent magnetic sensors.

(177) In this example, as the signals have phases offset by 90 degrees, one output signal (electrical signal) can be represented as a sine wave (sin ) and the other output signal (electrical signal) as a cosine wave (cos ). Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(178) FIG. 18 is a diagram for schematically illustrating a relationship between magnetic sensor 5 and magnetic vector P according to the second preferred embodiment.

(179) FIG. 18 represents a magnetic vector applied to magnetic sensors 5A and 5B in the ascending/descending direction of float when state S7 shifts to state S8. Note that the ascending/descending direction is a direction along the x axis. Magnetic vector P indicates a direction of a line of magnetic force of a magnetic field generated by the N pole and S pole of magnet 2G for example.

(180) Although a line of magnetic force of a magnetic field generated by the N pole and S pole of magnet 2H is omitted for the sake of simplicity, a component of magnetic vector P perpendicular to the ascending/descending direction is canceled by a magnetic vector of the line of magnetic force of the magnetic field generated by the N pole and S pole of magnet 2H. Accordingly, magnetic sensors 5A, 5B are exposed to an external magnetic field composed only of a component in the ascending/descending direction. As has been previously discussed, the bias magnetic field vector in each magnetic sensor 5 varies according to this external magnetic field.

(181) As one example, as a magnitude of a magnetic vector representing an external magnetic field and a magnetic flux density (an AMR output) are correlated, an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P sin and an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P cos . Based on the two output signals (electrical signals), angle of magnetic vector P is calculated.

(182) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(183) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(184) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(185) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 positionally varies by distance a.

(186) For example, as a position of float 20, a center of magnet 2G or 2H in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 15 is the position of magnetic sensor 5B. Furthermore, magnetic sensor 5A outputs a signal having the intermediate voltage (or state S7 in FIG. 16 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5A. Furthermore, magnetic sensor 5A outputs a maximized signal (or state S8 in FIG. 16 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5B.

(187) In this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a/2.

(188) While in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship from magnetic vector 5A, a positional relationship from magnetic vector 5B can also be determined. Furthermore, as a matter of course, a similar method can also be followed to use electrical signals of magnetic sensors 5B, 5C to calculate angular information of a magnetic vector and determine a positional relationship from magnetic vector 5C. A different method is also similarly discussed.

(189) FIGS. 19A and 19B are diagrams for illustrating accuracy of angular information according to the second preferred embodiment.

(190) FIG. 19A compares arctan with a reference value when angle is changed from 0 to 90 degrees with one output signal (an electrical signal) set to P cos and the other output signal (an electrical signal) set to P sin .

(191) As a result of a simulation, there is no substantial difference from the reference value.

(192) Furthermore, for angular accuracy, as shown in FIG. 19B, a case is shown in which there is only a deviation of approximately 2 degrees from the reference value, and the position of float 20 can be detected with high accuracy.

(193) Liquid level detection device 1 according to the second preferred embodiment allows a single magnet unit to be used to detect the position of float 20, and hence allows a simplified circuit configuration and hence miniaturization.

Third Preferred Embodiment

(194) While in the first preferred embodiment a case where an interval between magnetic sensors 5 preferably is preferably set to half an interval between magnet units, or distance a, has been described, it is also possible to change this distance.

(195) More specifically, a case where an interval of magnet units is preferably set to a distance 2a and an interval of the magnetic sensors 5 is preferably set to a distance 3a will be described.

(196) FIG. 20 is a diagram for illustrating a layout of magnets 2I-2T attached to float 20 and magnetic sensors 5A, 5B, 5C according to a third preferred embodiment.

(197) As shown in FIG. 20, magnets 2I and 2J define a set to define a magnet unit. Magnets 2K, 2L define a set to define a magnet unit. Magnets 2M, 2N define a set to define a magnet unit. Magnets 2O, 2P define a set to define a magnet unit. Magnets 2Q, 2R define a set to define a magnet unit. Magnets 2S, 2T define a set to define a magnet unit.

(198) Magnets 2I, 2J defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2K, 2L defining a magnet unit are disposed to have their respective S poles facing each other. Magnets 2M, 2N defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2O, 2P defining a magnet unit are disposed to have their respective S poles facing each other. Magnets 2Q, 2R defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2S, 2T defining a magnet unit are disposed to have their respective S poles facing each other. Adjacent magnet units are disposed such that their magnets' magnetic poles are different.

(199) Furthermore, in this example, a distance between magnets 2I and 2J is preferably set to a spacing of twice distance a, and magnetic sensors 5 are disposed to pass through a center thereof. Furthermore, an interval (or center-to-center distance) between adjacent magnet units is also set to twice distance a. Magnetic sensors 5 are disposed at intervals of distance 3a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(200) In this example, as a position of float 20, for example, a center of magnet 2I to 2S (or 2J to 2T) in the ascending/descending direction is set as a reference position (or a center point). In this case, a case where magnetic sensor 5B is positioned at the reference position (or center point) is shown.

(201) FIG. 21 illustrates waveforms of signals output from magnetic sensors 5 as float 20 ascends/descends in the third preferred embodiment.

(202) As shown in FIG. 21, similarly as has been described for waveforms of signals output as shown in FIG. 9, as float 20 approaches magnetic sensor 5, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2I and 2J. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(203) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2I and 2J. In this example, this state is set as an initial state.

(204) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2I and 2K. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(205) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2K and 2L. Thus, it is in the initial state.

(206) Float 20 further ascends, and magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2K and 2M. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(207) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2M and 2N. Thus, it is in the initial state. In this example, as one example, a voltage of an output signal in this state is set as an intermediate voltage.

(208) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2M and 2O. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(209) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2O and 2P. Thus, it is in the initial state.

(210) Float 20 further ascends, and magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2O and 2Q. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(211) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2Q and 2R. Thus, it is in the initial state.

(212) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2Q and 2S. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(213) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2S and 2T. Thus, it is in the initial state.

(214) Float 20 further ascends, and magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2S and 2T. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(215) Magnetic sensors 5B and 5C output signals having waveforms, respectively, similar to the waveform of the signal output from magnetic sensor 5A, offset therefrom by distance 3a (a phase of 270 degrees) and further by distance 3a (a phase of 270 degrees), respectively.

(216) FIG. 22 is an enlarged image of a prescribed region of FIG. 21.

(217) FIG. 22 represents waveforms of signals output from a plurality of magnetic sensors 5A and 5B in a hatched region of FIG. 21 indicating the prescribed region.

(218) The waveforms of the signals output from magnetic sensors 5A and 5B can be modeled (or approximated) into a horizontal component (in the ascending/descending direction) of magnetic vector P of an external magnetic field varying along a circle, as will be described hereinafter, when the intermediate voltage serves as a reference.

(219) Specifically, signal waveforms having phases offset by 270 degrees can be detected as electrical signals output from two adjacent magnetic sensors.

(220) In this example, the 270 degrees can be divided by 90 degrees into three to obtain divisional regions T1-T3, and in the regions one of two output signals (an electrical signal) can be represented as a sine wave (sin ) and the other output signal (an electrical signal) as a cosine wave (cos ). Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(221) FIGS. 23A-23C are diagrams for schematically illustrating a relationship between magnetic sensor 5 and magnetic vector P according to the third preferred embodiment.

(222) FIGS. 23A-23C represent a magnetic vector applied to magnetic sensors 5A and 5B in the ascending/descending direction of float 20 when state S9 shifts to state S10. Note that the ascending/descending direction is a direction along the x axis. Magnetic vector P indicates directions of lines of magnetic force received by magnetic sensors 5A and 5B exposed to magnetic fields generated by the N pole of magnet 2M and the S pole of magnet 2K, the N pole of magnet 2M and the S pole of magnet 2O, and the N pole of magnet 2Q and the S pole of magnet 2O for example.

(223) Although the lines of magnetic force of the magnetic fields generated by facing magnets 2L, 2N, 2P, 2R are omitted for the sake of simplicity, a component of magnetic vector P perpendicular to the ascending/descending direction is canceled by the magnetic vectors of the lines of magnetic force of the magnetic fields generated by the N and S poles of magnets 2L, 2N, 2P, 2R. Accordingly, magnetic sensors 5A, 5B are exposed to an external magnetic field composed only of a component in the ascending/descending direction. As has been previously discussed, the bias magnetic field vector in each magnetic sensor 5 varies according to this external magnetic field.

(224) FIG. 23A represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T1 obtained by dividing states S9-S10 into three. Note that the ascending/descending direction is a direction along the x axis.

(225) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2O, for example.

(226) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2K, for example.

(227) As one example, as a magnitude of a magnetic vector representing an external magnetic field and a magnetic flux density (an AMR output) are correlated, an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P cos and an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P sin . Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(228) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(229) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(230) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(231) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 positionally varies by distance a.

(232) For example, as a position of float 20, a center of magnets 2I to 2S in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 20 is the position of magnetic sensor 5B. Furthermore, magnetic sensor 5A outputs a maximized signal (or state S9 in FIG. 21 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5A. Furthermore, magnetic sensor 5A outputs a signal having the intermediate voltage (or state S10 in FIG. 21 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5B.

(233) In the detection region of region T1 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a/2.

(234) FIG. 23B represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T2 obtained by dividing states S9-S10 into three. Note that the ascending/descending direction is a direction along the x axis.

(235) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2Q and the S pole of magnet 2O, for example.

(236) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2K, for example.

(237) As one example, as a magnitude of a magnetic vector representing an external magnetic field and a magnetic flux density (an AMR output) are correlated, an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P sin . Furthermore, an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P cos . Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(238) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(239) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(240) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(241) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 also positionally varies by distance a.

(242) For example, as a position of float 20, a center of magnets 2I to 2S in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 20 is the position of magnetic sensor 5B.

(243) In the detection region of region T2 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a+a/2.

(244) FIG. 23C represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T3 obtained by dividing states S9-S10 into three. Note that the ascending/descending direction is a direction along the x axis.

(245) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2Q and the S pole of magnet 2O, for example.

(246) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2O, for example.

(247) As one example, as a magnitude of a magnetic vector representing an external magnetic field and a magnetic flux density (an AMR output) are correlated, an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P cos . Furthermore, an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P sin . Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(248) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(249) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(250) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(251) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 also positionally varies by distance a.

(252) For example, as a position of float 20, a center of magnets 2I to 2S in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 20 is the position of magnetic sensor 5B.

(253) In the detection region of region T3 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of 2a+a/2.

(254) FIG. 24 is a diagram for illustrating a method of extracting waveforms of two signals output in each of regions T1 to T3 from waveforms of signals output from a plurality of magnetic sensors 5A to 5C according to the third preferred embodiment.

(255) Referring to FIG. 24, herein, a plurality of threshold values TH0 to TH2 are set. Threshold value TH0 is preferably set to the intermediate voltage as one example. Threshold value TH2 is preferably set to an intermediate peak value as observed from an output signal waveform having fallen to a minimum value as one example. Threshold value TH1 is preferably set to an intermediate value between threshold value TH0 and threshold value TH2 as one example. It should be noted that setting threshold values TH0 to TH2 as above is merely an example and a different method may be used to set a threshold value.

(256) In the third preferred embodiment, divisional regions T1 to T3 are provided based on a relationship between threshold values TH0 to TH2 and output signal waveforms, and two output signal waveforms are extracted.

(257) For region T1, when magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) outputs a signal having a waveform having a value below threshold value TH1 and magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) outputs a signal having a waveform having a value below threshold value TH0 or when magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) outputs a signal having a waveform having a value below threshold value TH2 and magnetic sensor 5A outputs a signal having a waveform having a value exceeding threshold value TH0, the output of magnetic sensor 5A is assumed to be P cos and the output of magnetic sensor 5B is preferably set to P sin .

(258) For region T2, when magnetic sensor 5A and magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) output signals having waveforms having values below threshold value TH0 and magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) outputs a signal having a waveform having a value exceeding threshold value TH0, the output of magnetic sensor 5A is assumed to be P sin and the output of magnetic sensor 5B is preferably set to P cos .

(259) For region T3, when magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) and magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) output signals having waveforms having values exceeding threshold value TH0 and magnetic sensor 5A outputs a signal having a waveform having a value below threshold value TH0, the output of magnetic sensor 5A is assumed to be P cos and the output of magnetic sensor 5B is preferably set to P sin .

(260) According to the above method, angle of a magnetic vector is calculated based on the two output signals (electrical signals).

(261) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(262) Note that while in this example, divisional regions T1 to T3 are provided based on a relationship between threshold values TH0 to TH2 and output signal waveforms and two output signal waveforms are extracted, this method is not exclusive and a different method may be used to extract two output signal waveforms.

(263) FIGS. 25A-25F are diagrams for illustrating accuracy of angular information according to the third preferred embodiment. FIGS. 25A, 25C and 25E compare arctan with a reference value when angle is changed from 0 to 90 degrees with one output signal (an electrical signal) set to P cos and the other output signal (an electrical signal) set to P sin .

(264) As a result of a simulation, there is no substantial difference from the reference value.

(265) Furthermore, for angular accuracy, as shown in FIGS. 25B, 25D and 25F, a case is shown in which there is only a deviation of approximately 2 degrees in each region, and the position of float 20 can be detected with high accuracy.

(266) FIG. 26 is a flowchart for illustrating a detection method via liquid level detection device 1 according to the third preferred embodiment.

(267) As shown in FIG. 26, two signals are extracted based on a prescribed signal relationship combination (step SP2#). Specifically, according to the method described with reference to FIGS. 23A-23C, based on a combination of threshold values TH0 to TH2 and each output signal waveform, divisional regions T1 to T3 are provided, and two output signal waveforms are extracted in each region.

(268) The intermediate voltage, or threshold value TH0, is preferably set to a voltage of a signal output in a state where magnetic sensor 5A is located on the center line between magnet 2M and magnet 2N, for example. Threshold value TH2 can be set to an intermediate peak value as observed from an output signal waveform having fallen to a minimum value as one example. Threshold value TH1 can be set to an intermediate value between threshold value TH0 and threshold value TH2 as one example.

(269) Subsequently, a magnetic vector's angle is calculated based on the extracted two signals (step SP4). More specifically, of the two electrical signals, one output signal (an electrical signal) is preferably set to a sin wave (sin ) and the other output signal (an electrical signal) is preferably set to a cosine wave (cos ), and based on the two output signals (electrical signals), the magnetic vector's angle is calculated. Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(270) Subsequently, the position of float 20 is calculated based on the magnetic vector's angle (step SP6). Based on the calculated angular information , the reference position (or center point) of float 20 is calculated from the positions of the magnetic sensors.

(271) Subsequently, the process ends (END).

(272) Liquid level detection device 1 according to the third preferred embodiment allows two electrical signals to be used to detect the position of float 20 with high accuracy. Furthermore, as two magnetic sensors can be used to detect information for 270 degrees (or distance 3a), the number of magnetic sensors is able to be further reduced to achieve miniaturization. Furthermore, a distance between facing magnets is also able to be reduced and further miniaturization is also able to be achieved.

(273) Further, while there is a possibility that, following variation in the ambient temperature, the magnets and the magnetic sensors may have their characteristics varied and an output signal may vary, the angle is calculated with two output signals' tan (P sin /P cos ) calculated and accordingly, an amount of variation is canceled, and an error due to an effect of the ambient temperature is thus able to be reduced and highly accurate detection is also able to be achieved.

(274) FIG. 27 is a diagram for illustrating a layout of magnets 2U and 2V attached to float 20 and magnetic sensors 5A, 5B, 5C according to a first exemplary variation of the third preferred embodiment.

(275) As shown in FIG. 27, magnets 2U and 2V define a set to define a magnet unit.

(276) The magnet unit defined by magnets 2U and 2V is divided into six regions and arranged such that S poles or N poles face each other in each region. Divisional, adjacent regions are arranged such that the magnets have magnetic poles arranged differently.

(277) Furthermore, in this example, a distance between magnet 2U and magnet 2V is preferably set to a spacing of twice distance a. Furthermore, magnetic sensors 5 are disposed to pass through a center thereof.

(278) Furthermore, a distance of two divisional regions of the N pole and the S pole is preferably set to an interval of twice distance a, and magnetic sensors 5 are disposed at intervals of distance 3a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(279) Magnets 2U and 2V generate a magnetic field (a line of magnetic force) which is basically the same as that generated by the layout of FIG. 20, and signals output therefor have waveforms similar to those described with reference to FIG. 21. Accordingly, the position of float 20 is able to be detected with high accuracy according to a method similar to that described above.

(280) By this configuration, the number of magnets is able to be reduced and the layout of the magnets is also able to be easily accomplished.

(281) While in the third preferred embodiment a float composed of six magnet units has been described, the number of magnet units is also able to be reduced.

(282) FIG. 28 is a diagram for illustrating a layout of magnets 2I-2P attached to float 20 and magnetic sensors 5A, 5B, 5C according to the third preferred embodiment.

(283) As shown in FIG. 28, magnets 2I and 2J define a set to define a magnet unit. Magnets 2K, 2L define a set to define a magnet unit. Magnets 2M, 2N define a set to define a magnet unit. Magnets 2O, 2P define a set to define a magnet unit.

(284) Magnets 2I, 2J defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2K, 2L defining a magnet unit are disposed to have their respective S poles facing each other. Magnets 2M, 2N defining a magnet unit are disposed to have their respective N poles facing each other. Magnets 2O, 2P defining a magnet unit are disposed to have their respective S poles facing each other. Adjacent magnet units are disposed such that their magnets' magnetic poles are different.

(285) Furthermore, in this example, a distance between magnets 2I and 2J is preferably set to a spacing of twice distance a, and magnetic sensors 5 are disposed to pass through a center thereof. Furthermore, an interval (or center-to-center distance) between adjacent magnet units is also set to twice distance a. Magnetic sensors 5 are disposed at intervals of distance 3a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(286) In this example, as a position of float 20, for example, a center of magnet 2I to 2O (or 2J to 2P) in the ascending/descending direction is set as a reference position (or a center point). In this case, a case where magnetic sensor 5B is positioned at the reference position (or center point) is shown.

(287) FIG. 29 illustrates waveforms of signals output from magnetic sensors 5 as float 20 ascends/descends in the second exemplary variation of the third preferred embodiment.

(288) As shown in FIG. 29, similarly as has been described for waveforms of signals output as shown in FIG. 9, as float 20 approaches magnetic sensor 5, magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2I and 2J. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(289) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2I and 2J. In this example, this state is set as an initial state.

(290) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2I and 2K. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(291) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2K and 2L. Thus, it is in the initial state.

(292) Float 20 further ascends, and magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2K and 2M. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(293) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2M and 2N. Thus, it is in the initial state. In this example, a voltage of an output signal in this state is set as the intermediate voltage.

(294) Float 20 further ascends, and magnetic sensor 5A is affected by a left to right magnetic field as a line of magnetic force of magnets 2M and 2O. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V2 side. As the bias magnetic field vector varies toward the bias magnetic field vector V2 side, potential difference V increases.

(295) Float 20 further ascends, and magnetic sensor 5A assumes a state where it is located on the center line between magnets 2O and 2P. Thus, it is in the initial state.

(296) Float 20 further ascends, and magnetic sensor 5A is affected by a right to left magnetic field as a line of magnetic force of magnets 2O and 2P. Accordingly, bias magnetic field vector V0 of magnetic sensor 5A varies toward the bias magnetic field vector V1 side. As the bias magnetic field vector varies toward the bias magnetic field vector V1 side, potential difference V decreases.

(297) Magnetic sensors 5B and 5C output signals having waveforms, respectively, similar to the waveform of the signal output from magnetic sensor 5A, offset therefrom by distance 3a (a phase of 270 degrees) and further by distance 3a (a phase of 270 degrees), respectively.

(298) FIG. 30 is an enlarged image of a prescribed region of FIG. 29.

(299) FIG. 30 represents waveforms of signals output from a plurality of magnetic sensors 5A and 5B in a hatched region of FIG. 29 indicating the prescribed region.

(300) The waveforms of the signals output from magnetic sensors 5A and 5B can be modeled (or approximated) into a horizontal component (in the ascending/descending direction) of magnetic vector P of an external magnetic field varying along a circle, as will be described hereinafter, when the intermediate voltage serves as a reference.

(301) Specifically, signal waveforms having phases offset by 270 degrees can be detected as electrical signals output from two adjacent magnetic sensors.

(302) In this example, the 270 degrees can be divided by 90 degrees into three to obtain divisional regions T1-T3, and in the regions one of two output signals (an electrical signal) is set as a sine wave (sin ) and the other output signal (an electrical signal) as a cosine wave (cos ), and based on the two output signals (electrical signals), angle of a magnetic vector is calculated.

(303) FIGS. 31A-31C are diagrams for illustrating an angle of a magnetic vector received by magnetic sensor 5 according to the second exemplary variation of the third preferred embodiment.

(304) FIGS. 31A-31C represent a magnetic vector applied to magnetic sensors 5A and 5B in the ascending/descending direction of the float when state S11 shifts to state S12. Note that the ascending/descending direction is a direction along the x axis. Magnetic vector P indicates directions of lines of magnetic force of magnetic fields generated the N pole of magnet 2I and the S pole of magnet 2K, the N pole of magnet 2M and the S pole of magnet 2K, and the N pole of magnet 2M and the S pole of magnet 2O for example.

(305) Although lines of magnetic force of magnetic fields generated by facing magnets 2J, 2L, 2N, 2P are omitted for the sake of simplicity, a component of magnetic vector P perpendicular to the ascending/descending direction is canceled by magnetic vectors of the lines of magnetic force of magnetic fields generated by the N and S poles of magnets 2J, 2L, 2N, 2P. Accordingly, magnetic sensors 5A, 5B are exposed to an external magnetic field composed only of a component in the ascending/descending direction. As has been previously discussed, the bias magnetic field vector in each magnetic sensor 5 varies according to this external magnetic field.

(306) FIG. 31A represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T1 obtained by dividing states S11-S12 into three. Note that the ascending/descending direction is a direction along the x axis.

(307) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2K, for example.

(308) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2I and the S pole of magnet 2K, for example.

(309) As one example, as a magnitude of a magnetic vector and a magnetic flux density (an AMR output) are correlated, according to angle an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P cos . Furthermore, an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P sin . Based on the two output signals (electrical signals), angle of a magnetic vector is calculated.

(310) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(311) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(312) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(313) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 also positionally varies by distance a.

(314) For example, as a position of float 20, a center of magnets 2I to 2O in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 28 is the position of magnetic sensor 5B. Furthermore, magnetic sensor 5A outputs a minimized signal (or state S11 in FIG. 29 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5A. Furthermore, magnetic sensor 5A outputs a signal having the intermediate voltage (or state S12 in FIG. 29 is assumed) when the reference position (or center point) of float 20 is located at the position of magnetic sensor 5B.

(315) In the detection region of region T1 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a/2.

(316) FIG. 31B represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T2 obtained by dividing states S11-S12 into three. Note that the ascending/descending direction is a direction along the x axis.

(317) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2O, for example.

(318) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2I and the S pole of magnet 2K, for example.

(319) As one example, as a magnitude of a magnetic vector and a magnetic flux density (an AMR output) are correlated, according to angle an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P sin . Furthermore, an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P cos . Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(320) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(321) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(322) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(323) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 also positionally varies by distance a.

(324) For example, as a position of float 20, a center of magnets 2I to 2O in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 28 is the position of magnetic sensor 5B.

(325) In the detection region of region T2 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of a+a/2.

(326) FIG. 31C represents magnetic vectors input to magnetic sensors 5A and 5B, respectively, with respect to the ascending/descending direction of float 20 in region T3 obtained by dividing states S11-S12 into three. Note that the ascending/descending direction is a direction along the x axis.

(327) Magnetic vector P affecting magnetic sensor 5A indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2O, for example.

(328) Magnetic vector P affecting magnetic sensor 5B indicates a direction of a line of magnetic force of a magnetic field generated by the N pole of magnet 2M and the S pole of magnet 2K, for example.

(329) As one example, as a magnitude of a magnetic vector and a magnetic flux density (an AMR output) are correlated, according to angle an output signal detected at magnetic sensor 5A with respect to the ascending/descending direction can be represented as P cos . Furthermore, an output signal detected at magnetic sensor 5B with respect to the ascending/descending direction can be represented as P sin . Based on the two output signals (electrical signals), angle of the magnetic vector is calculated.

(330) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(331) Note that a sine wave P sin and a cosine wave P cos have an amplitude value P, which is cancelled by calculating tan .

(332) The above process is a process performed in detection circuit 50. Specifically, the above calculation process is performed in MPU 40.

(333) In response to the magnetic vector's angular information varying from 0 degree to 90 degrees, float 20 also positionally varies by distance a.

(334) For example, as a position of float 20, a center of magnets 2I to 2O in the ascending/descending direction is set as a reference position (or a center point) as one example. In that case, a reference position (or center point) of float 20 shown in FIG. 28 is the position of magnetic sensor 5B.

(335) In the detection region of region T3 in this example, electrical signals of magnetic sensors 5A, 5B are used to calculate angular information of a magnetic vector and determine a positional relationship therefor. For example, when angular information is calculated as 45 degrees, it can be detected that the float's reference position (or center point) is located at a position moved from the position of magnetic sensor 5A toward magnetic sensor 5B by a distance of 2a+a/2.

(336) FIG. 32 is a diagram for illustrating a method of extracting waveforms of two signals output in each of regions T1 to T3 from waveforms of signals output from a plurality of magnetic sensors 5A to 5C according to the second exemplary variation of the third preferred embodiment.

(337) Referring to FIG. 32, herein, a plurality of threshold values TH0 to TH2 are set. Threshold value TH0 is preferably set to the intermediate voltage as one example. Threshold value TH2 is preferably set to an intermediate peak value as observed from an output signal waveform having fallen to a minimum value as one example. Threshold value TH1 is preferably set to an intermediate value between threshold value TH0 and threshold value TH2 as one example. It should be noted that setting threshold values TH0 to TH2 as above is merely an example and a different method may be used to set a threshold value.

(338) In the second exemplary variation of the third preferred embodiment, divisional regions T1 to T3 are provided based on a relationship between threshold values TH0 to TH2 and output signal waveforms, and two output signal waveforms are extracted.

(339) For region T1, when magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) outputs a signal having a waveform having a value exceeding threshold value TH0 and magnetic sensor 5A and magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) output signals having waveforms having values below threshold value TH0, the output of magnetic sensor 5A is assumed to be P cos and the output of magnetic sensor 5B is preferably set to P sin .

(340) For region T2, when magnetic sensor 5A and magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) output signals having waveforms having values exceeding threshold value TH0 and magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) outputs a signal having a waveform having a value below threshold value TH0, the output of magnetic sensor 5A is assumed to be P sin and the output of magnetic sensor 5B is preferably set to P cos .

(341) For region T3, when magnetic sensor 5C (i.e., a magnetic sensor next to magnetic sensor 5A but one) outputs a signal having a waveform having a value below threshold value TH1 and magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) outputs a signal having a waveform having a value below threshold value TH0, or when magnetic sensor 5B (i.e., a magnetic sensor next to magnetic sensor 5A) outputs a signal having a waveform having a value below threshold value TH2 and magnetic sensor 5A outputs a signal having a waveform having a value exceeding threshold value TH0, the output of magnetic sensor 5A is assumed to be P cos and the output of magnetic sensor 5B is preferably set to P sin .

(342) According to the above method, angle of a magnetic vector is calculated based on the two output signals (electrical signals).

(343) Specifically, tan is calculated based on the two output signals (electrical signals), and angular information is calculated by calculating arctan .

(344) Note that while in this example, divisional regions T1 to T3 are provided based on a relationship between threshold values TH0 to TH2 and output signal waveforms and two output signal waveforms are extracted, this method is not exclusive and a different method may be used to extract two output signal waveforms.

(345) FIGS. 33A-33F are diagrams for illustrating accuracy of angular information according to the second exemplary variation of the third preferred embodiment.

(346) FIGS. 33A, 33C and 33E compare arctan with a reference value when angle is changed from 0 to 90 degrees with one output signal (an electrical signal) set to cos and the other output signal (an electrical signal) set to sin .

(347) Furthermore, for angular accuracy, as shown in FIGS. 33B, 33D and 33F, a case is shown in which there is only some extent of deviation, and highly accurate detection is also able to be achieved.

(348) Liquid level detection device 1 according to the second exemplary variation of the third preferred embodiment allows two electrical signals to be used to detect the position of float 20 with high accuracy. Furthermore, as two magnetic sensors are able to be used to detect information for 270 degrees (or distance 3a), the number of magnetic sensors is able to be further reduced to achieve miniaturization.

(349) Further, while there is a possibility that, following variation in the ambient temperature, the magnets and the magnetic sensors may have their characteristics varied and an output signal may vary, the angle is calculated with two output signals' tan (P sin /P cos ) calculated and accordingly, an amount of variation is canceled, and an error due to an effect of the ambient temperature is thus able to be reduced and highly accurate detection is also able to be achieved.

(350) FIG. 34 is a diagram for illustrating a layout of magnets 2W and 2X attached to float 20 and magnetic sensors 5A, 5B, 5C according to a third exemplary variation of the third preferred embodiment.

(351) As shown in FIG. 34, magnets 2W and 2X define a set to define a magnet unit.

(352) The magnet unit defined by magnets 2W and 2X is divided into four regions and arranged such that S poles or N poles face each other in each region. Divisional, adjacent regions are arranged such that the magnets have magnetic poles arranged differently.

(353) Furthermore, in this example, a distance between magnets 2W and 2X is preferably set to a spacing of twice distance a. Furthermore, magnetic sensors 5 are disposed to pass through a center thereof. Furthermore, a distance of two divisional regions of the N pole and the S pole is preferably set to an interval of twice distance a, and magnetic sensors 5 are disposed at intervals of distance 3a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(354) Magnets 2W and 2X generate a magnetic field (a line of magnetic force) which is basically the same as that generated by the layout of FIG. 28 and signals output therefor have waveforms similar to those described with reference to FIG. 28, and the position of float 20 can be detected according to a similar method.

(355) By this configuration, the number of magnets is able to be reduced and the layout of the magnets is also able to be easily accomplished.

Fourth Preferred Embodiment

(356) FIG. 35 is a diagram for illustrating a pattern of a magnetoresistive element of a magnetic sensor 5# according to a fourth preferred embodiment of the present invention.

(357) Referring to FIG. 35, in this example, magnetic sensor 5 has a bridge structure including four magnetoresistive elements MR1# to MR4#.

(358) The four magnetoresistive elements MR1# to MR4# are symmetrically arranged with respect to a center line. A barber pole electrode structure is provided such that magnetoresistive elements 1# and MR3# have a magnetoresistive effect characteristic allowing them to both have a resistance value increasing as magnetic fields opposite in direction increase and magnetoresistive elements 2# and 4# have a magnetoresistive effect characteristic allowing them to both have a resistance value decreasing as magnetic fields opposite in direction increase. By this configuration, an output characteristic which accompanies variation in magnetic flux density described with reference to FIGS. 5A and 5B is provided. Thus, a liquid level detection device including magnetic sensor 5# of this configuration instead of magnetic sensor 5 is also able to detect the position of float 20 in a method similar to that described above.

Fifth Preferred Embodiment

(359) In a fifth preferred embodiment of the present invention, a configuration capable of highly accurate liquid level detection even when the position of float 20 is displaced with respect to guide 10 will be described.

(360) FIGS. 36A and 36B are diagrams for illustrating a case where the position of float 20 is displaced with respect to guide 10.

(361) FIG. 36A shows float 20 as observed from above. Furthermore, magnets 2G and 2H defining a magnet unit are provided to face each other with the guide interposed therebetween.

(362) In this example, while a case where the center axis of a guide 10 and the center of float 20 are matched with each other is shown, a case where a relative positional relationship between magnetic sensor 5 and float 20 is changed as float 20 is positionally displaced will be described. A case where float 20 has magnet 2H closer to magnetic sensor 5 than float 20 has magnet 2G will be described.

(363) FIG. 36B shows a layout of magnets 2G and 2H attached to float 20 and magnetic sensors 5A to 5C.

(364) As shown in FIG. 36B, when magnetic sensors 5A to 5C approach magnet 2H, they are affected not only by an external magnetic field (a line of magnetic force) of a horizontal component but also by an external magnetic field (a line of magnetic force) of a perpendicular component. This results in a bias magnetic field vector with a rotational angle varied and an output signal in response thereto varies. As the output signal varies, there is a possibility or less accurate liquid level detection.

(365) FIG. 37 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and magnetic sensors according to the fifth preferred embodiment.

(366) As shown in FIG. 37, magnets 2G and 2H define a set to define a magnet unit.

(367) Magnets 2G and 2H defining a magnet unit are disposed to have their respective N poles facing each other. Float 20 has a configuration similar to that described with reference to FIG. 15.

(368) As has been described with reference to FIG. 15, a distance between magnets 2G and 2H is preferably set to a spacing of twice a distance a, and magnetic sensors 5 are disposed to pass through a center thereof. Magnetic sensors 5 are also disposed at intervals of distance a. Magnetic sensors 5 are attached to the guide in the ascending/descending direction.

(369) Note that the bias magnetic field vector of magnetic sensor 5 is provided to have a direction symmetrically with reference to a horizontal direction as compared with an adjacent magnetic sensor. In this example, magnetic sensors 5PA to 5PC are provided as one example. The bias magnetic field vectors respectively of magnetic sensors 5PA to 5PC are arranged to have a direction in a horizontal direction perpendicular to the ascending/descending direction of float 20.

(370) An arrangement is such that a bias magnetic field vector applied to magnetoresistive elements MR1 to MR4 has a direction in the horizontal direction perpendicular to the ascending/descending direction of the float 20. In this respect, magnetic sensors 5PA to 5PC may include magnetoresistive element MR having a configuration similar to that of FIG. 4, and this arrangement or angle is able to be appropriately redesigned by a skilled artisan to allow magnetic sensors 5PA to 5PC to have enhanced detection characteristics.

(371) The other magnetic sensors attached to the guide in the ascending/descending direction are also disposed in the same manner.

(372) FIGS. 38A and 38B are diagrams for illustrating variation of a bias magnetic field vector of a magnetic sensor.

(373) FIG. 38A shows variation of a bias magnetic field vector of magnetic sensors 5A to 5C.

(374) As has been described with reference to FIGS. 5A and 5B, when an external magnetic field in the ascending/descending direction is applied to magnetic sensors 5A to 5C, bias magnetic field vector V0 varies to bias magnetic field vector V1 according to the external magnetic field (applied in a right to left direction). In contrast, bias magnetic field vector V0 varies to bias magnetic field vector V2 according to the external magnetic field (applied in a left to right direction).

(375) In contrast, when float 20 has magnet 2H approaching magnetic sensors 5A to 5C, the sensors are affected by an external magnetic field of a horizontal component (in the ascending/descending direction) as well as an external magnetic field of a perpendicular component (in the horizontal direction) (lines of magnetic force).

(376) Specifically, when magnetic sensors 5A to 5C are not displaced, external magnetic fields of perpendicular components (in the horizontal direction) from magnets 2G and 2H cancel each other and the magnetic sensors will not be affected thereby (or by a line of magnetic force), whereas when the magnetic sensors are displaced, external magnetic fields of perpendicular components (in the horizontal direction) from magnets 2G and 2H do not cancel each other and the magnetic sensors will be affected thereby. For example, magnetic sensors 5A to 5C of FIGS. 36A and 36B are exposed to an external magnetic field of a perpendicular component (in the horizontal direction) from magnet 2H to magnet 2G.

(377) More specifically, in the state shown in FIGS. 36A and 36B, magnetic sensor 5A varies to a bias magnetic field vector V2#. Furthermore, magnetic sensor 5B varies to a bias magnetic field vector V0#. Furthermore, magnetic sensor 5C varies to a bias magnetic field vector V1#.

(378) Herein, when bias magnetic field vector V0# of magnetic sensor 5B in FIGS. 36A and 36B is set as a reference, bias magnetic field vector V1# of magnetic sensor 5C rotates from the state of bias magnetic field vector V0# clockwise by an angle . Furthermore, when bias magnetic field vector V0# of magnetic sensor 5B is set as a reference, bias magnetic field vector V2# of magnetic sensor 5C rotates from the state of bias magnetic field vector V0# counterclockwise by an angle .

(379) Bias magnetic field vector V0# serving as a reference deviates from bias magnetic field vector V0 and the rotation angle also varies, and accordingly, magnetic sensors 5A to 5C output signals having different values in amplitude and therefore there is a possibility of angular detection with an error.

(380) FIG. 38B shows variation of bias magnetic field vectors of magnetic sensors 5PA to 5PC.

(381) The bias magnetic field vectors respectively of magnetic sensors 5PA to 5PC are arranged to have a direction in a horizontal direction perpendicular to the ascending/descending direction of float 20.

(382) When an external magnetic field in the ascending/descending direction is applied to magnetic sensors 5PA to 5PC, a bias magnetic field vector V3 varies to a bias magnetic field vector V4 according to the external magnetic field (applied in a right to left direction). In contrast, bias magnetic field vector V3 varies to a bias magnetic field vector V5 according to the external magnetic field (applied in a left to right direction).

(383) In contrast, when float 20 has magnet 2H approaching magnetic sensors 5PA to 5PC, the sensors are affected by an external magnetic field of a horizontal component (in the ascending/descending direction) as well as an external magnetic field of a perpendicular component (in the horizontal direction) (lines of magnetic force).

(384) Specifically, when magnetic sensors 5PA to 5PC are not displaced, external magnetic fields of perpendicular components (in the horizontal direction) from magnets 2G and 2H cancel each other and the magnetic sensors will not be affected thereby (or by a line of magnetic force), whereas when the magnetic sensors are displaced, external magnetic fields of perpendicular components (in the horizontal direction) from magnets 2G and 2H do not cancel each other and the magnetic sensors will be affected thereby. For example, magnetic sensors 5PA to 5PC of FIG. 37 are exposed to an external magnetic field of a perpendicular component (in the horizontal direction) from magnet 2H to magnet 2G.

(385) More specifically, in the state shown in FIG. 37, magnetic sensor 5PA varies to a bias magnetic field vector V5#.

(386) Furthermore, as bias magnetic field vector V3 and an external magnetic field (a line of magnetic force) of a perpendicular component (in the horizontal direction) are the same in direction, bias magnetic field vector V3 of magnetic sensor 5PB maintains a state in the horizontal direction perpendicular to the ascending/descending direction. Furthermore, magnetic sensor 5PC varies to a bias magnetic field vector V4#.

(387) Herein, when bias magnetic field vector V3 of magnetic sensor 5PB in FIG. 37 is set as a reference, bias magnetic field vector V4# of magnetic sensor 5PC rotates from the state of bias magnetic field vector V3 clockwise by an angle . Furthermore, when bias magnetic field vector V3 of magnetic sensor 5PB is set as a reference, bias magnetic field vector V5# of magnetic sensor 5PA rotates from the state of bias magnetic field vector V3 counterclockwise by angle .

(388) Bias magnetic field vector V3 serving as a reference is fixed and the rotation angle also does not vary, and accordingly, magnetic sensors 5PA to 5PC output signals having equal values in amplitude and an error in angular detection is thus able to be suppressed.

(389) Specifically, as has been described above, in this example also, as has been described with reference to FIG. 15, of two electrical signals, one output signal (an electrical signal) is preferably set to P cos and the other output signal (an electrical signal) is preferably set to P sin , and based on the two output signals (electrical signals), tan (P sin /P cos ) is calculated, and pseudo angular information is calculated by calculating arctan .

(390) In calculating angular information , when float 20 has magnet 2H approaching magnetic sensors 5PA to 5PC, then in the example of FIG. 37 an amplitude value will vary from P to an amplitude value P3 based on angle by which the bias magnetic field vector rotates, however, as two output signals (electrical signals) have their amplitude values equally varied, they are canceled in calculating angular information . Thus, even when float 20 has magnet 2H approaching magnetic sensors 5PA to 5PC, the position of float 20 is able to be detected with high accuracy.

(391) In contrast, in the example of FIG. 36B, the amplitude values of the two output signals (electrical signals) vary from P to amplitude values P1 and P2 based on rotation angles and , respectively, by which the bias magnetic field vectors rotate, and thus cannot cancel each other, and accordingly, the ratio varies, resulting in a possibility of angular detection with an error.

(392) FIGS. 39A and 39B are diagrams for illustrating accuracy of angular information according to the fifth preferred embodiment.

(393) FIG. 39A compares arctan with a reference value when angle is changed from 0 to 90 degrees with one output signal (an electrical signal) set to P cos and the other output signal (an electrical signal) set to P sin .

(394) While, as has been described for the configuration of the first preferred embodiment with reference to FIG. 36B, a case is shown in which when float 20 is displaced, accuracy is decreased, the configuration of FIG. 37 according to the fifth preferred embodiment is free of reduction in accuracy and allows the position of float 20 to be detected with high accuracy.

(395) Furthermore, for angular accuracy, as shown in FIG. 39B, a case is shown in which there is only a deviation of 5 degrees from the reference value, and the position of float 20 is able to be detected with high accuracy.

(396) When float 20 is positionally displaced, adjacent magnetic sensors is each affected by an external magnetic field of a horizontal component (in the ascending/descending direction) as well as an external magnetic field of a perpendicular component (in the horizontal direction) (lines of magnetic force).

(397) For example, when float 20 is positionally displaced, magnetic sensor 5C shown in FIG. 36B is exposed to a first external magnetic field composed of an external magnetic field of a horizontal component (in the ascending/descending direction) and an external magnetic field of a perpendicular component (in the horizontal direction) (lines of magnetic force) composited together. Magnetic sensor 5A is exposed to a second external magnetic field composed of an external magnetic field of a horizontal component (in the ascending/descending direction) and an external magnetic field of a perpendicular component (in the horizontal direction) (lines of magnetic forces) composited together. The first external magnetic field and the second external magnetic field are symmetrical with respect to the horizontal direction.

(398) When float 20 is positionally displaced, and bias magnetic field vectors are arranged in a fixed direction in an inclined state, as seen in magnetic sensors 5A to 5C, the first and second external magnetic fields are applied to the bias magnetic field vectors at different incident angles. Accordingly, the aforementioned rotation angles and from a bias magnetic field vector serving as a reference are also different.

(399) In the configuration according to the fifth preferred embodiment, the bias magnetic field vectors of adjacent magnetic sensors are preferably set to be in a horizontal direction perpendicular to the ascending/descending direction of float 20. If float 20 is positionally displaced, the first and second external magnetic fields are applied to the bias magnetic field vectors at equal incident angles. Accordingly, the aforementioned rotation angle from a bias magnetic field vector serving as a reference does not vary. As a rotation angle from a bias magnetic field vector serving as a reference does not vary and accordingly, magnetic sensors 5PA to 5PC output signals having equal values in amplitude, and positional detection with high accuracy is thus able to be achieved.

(400) FIG. 40 is a diagram for illustrating a layout of magnets 2G and 2H attached to float 20 and a plurality of magnetic sensors 5QA to 5QC according to an exemplary variation of the fifth preferred embodiment.

(401) As shown in FIG. 40, magnets 2G and 2H define a set to define a magnet unit.

(402) Magnets 2G and 2H defining a magnet unit are disposed to have their respective N poles facing each other. While in this example an example will be described in which the N poles face each other, it is also possible to adopt a configuration in which the S poles face each other.

(403) Furthermore, in this example, a distance between magnets 2G and 2H is preferably set to a spacing of twice distance a, and magnetic sensors 5QA to 5QC are disposed to pass through a center thereof. Magnetic sensors 5QA to 5QC are also disposed at intervals of distance a. Magnetic sensors 5QA to 5QC are attached to the guide in the ascending/descending direction.

(404) In this example, magnetic sensor 5QB is provided at a center, and the bias magnetic field vectors of magnetic sensors 5QA and 5QB are provided symmetrically in direction with respect to the horizontal direction. Furthermore, the bias magnetic field vectors of magnetic sensors 5QC and 5QB are provided symmetrically in direction with respect to the horizontal direction.

(405) This arrangement allows the bias magnetic field vectors to be exposed to the first and second external magnetic fields at equal incident angles. This allows a rotation angle from a bias magnetic field vector serving as a reference to be set to be invariable, and magnetic sensors 5QA to 5QC output signals having equal values in amplitude, and by subjecting the output signals to a corrective operation of a prescribed value, an error in angular detection can be suppressed and positional detection with high accuracy can be performed. The prescribed value can be set to a maximum value of variation in amplitude divided by 2, and a skilled artisan would be able to redesign it as appropriate.

Other Preferred Embodiments

(406) FIG. 41 illustrates a magnetic sensor 5R based on another preferred embodiment.

(407) Referring to FIG. 41, for magnetic sensor 5R, a substrate 6 on which a magnetoresistive element is provided, and a bias magnet 4 are shown. It is assumed that bias magnet 4 has an S pole facing substrate 6 on which the magnetoresistive element is provided. While in this example a case is described in which bias magnet 4 has an S pole facing substrate 6 on which the magnetoresistive element is provided, a case in which bias magnet 4 has an N pole facing substrate 6 on which the magnetoresistive element is provided is also similarly discussed.

(408) FIG. 42 is a diagram for illustrating a bias magnetic field vector for magnetic sensor 5P.

(409) As shown in FIG. 42, four magnetoresistive elements MR1 to MR4 are provided on substrate 6 axisymmetrically with respect to a direction in which substrate 6 ascends/descends. Bias magnet 4 is provided on the four magnetoresistive elements MR1 to MR4. As one example, a case is shown in which bias magnet 4 having a cylindrical shape is disposed. It should be noted that the bias magnet is not thus limited in shape and a square-shaped bias magnet may be disposed. Further, while a configuration in which bias magnet 4 is provided on magnetoresistive elements MR1 to MR4 will be described, a configuration may be adopted in which bias magnet 4 is provided on the opposite side with substrate 6 interposed.

(410) In this example, a case is shown where the magnetoresistive elements are disposed in the vicinity of an end portion of bias magnet 4. This arrangement allows a bias magnetic field to be applied to the magnetoresistive elements at a position where the bias magnetic field has a large strength.

(411) As one example, a case is shown in which a bias magnetic field vector VB0 is applied to magnetoresistive elements MR1 and MR2 as a bias magnetic field in one direction inward from the center of bias magnet 4.

(412) Furthermore, a case is shown in which a bias magnetic field vector VA0 is applied to magnetoresistive elements MR3 and MR4 as a bias magnetic field in the other direction inward from the center of bias magnet 4. As a bias magnetic field applied with reference to the center line of bias magnet 4 are directed differently, bias magnetic field vector VA0 and bias magnetic field vector VB0 are opposite in direction.

(413) While magnetoresistive element MR of magnetic sensor 5R of this example is described as a folded pattern structure by way of example, it is not limited to the folded shape in particular, and the pattern structure is able to be redesigned as appropriate by a skilled artisan to allow magnetoresistive element 5P to have enhanced detected characteristics.

(414) Furthermore, while a configuration is shown in which an arrangement is made such that a bias magnetic field vector of an angle of 45 degrees is applied as a bias magnetic field vector for magnetoresistive element MR, this arrangement or angle is also able to be appropriately redesigned by a skilled artisan to allow magnetic sensor 5P to have enhanced detection characteristics. For example, it is also possible to design such that a bias magnetic field vector with an angle of 30 degrees is applied to magnetoresistive element MR.

(415) FIG. 43 illustrates a circuit configuration of magnetic sensor 5R.

(416) As shown in FIG. 43, magnetic sensor 5R preferably includes a bridge structure including four magnetoresistive elements MR1 to MR4.

(417) Magnetoresistive elements MR1 and MR2 are connected in series between a power supply voltage Vcc and a ground voltage GND. In parallel with magnetoresistive elements MR1 and MR2, magnetoresistive elements MR3 and MR4 are connected in series between power supply voltage Vcc and ground voltage GND.

(418) A signal V is output from a connection node of magnetoresistive elements MR3 and MR4 and a signal V+ is output from a connection node of magnetoresistive elements MR1 and MR2, and a difference V between signals V+ and V is output.

(419) As indicated in this example, bias magnetic field vector VA0 is applied to magnetoresistive elements MR3 and MR4. Bias magnetic field vector VB0 is applied to magnetoresistive elements MR1 and MR2.

(420) The vectors vary in direction in response to an external magnetic field with respect to the ascending/descending direction of float 20.

(421) Magnetic sensor 5R detects variation of the bias magnetic field vectors and outputs a signal (potential difference V) corresponding to the result of the detection.

(422) For example, bias magnetic field vectors VA0 and VB0 in response to an external magnetic field (in a right to left direction) result in signal V+ being decreased and signal V being increased. Accordingly, difference V decreases.

(423) In contrast, bias magnetic field vectors VA0 and VB0 in response to an external magnetic field (in a left to right direction) result in signal V+ being increased and signal V being decreased. Accordingly, difference V increases.

(424) Magnetic sensor 5R also allows the position of float 20 to be detected in the above method.

(425) Furthermore, the two bias magnetic field vectors are preferably set to have a direction in a horizontal direction perpendicular to the ascending/descending direction, and if float 20 is displaced, the bias magnetic field vectors' rotation angles are the same, as has been described above, and an error in angular detection is able to be significantly reduced or prevented, and positional detection with high accuracy is able to be performed.

(426) The magnetoresistive element described in the above example can also be a magnetoresistive element including an exchange coupling film in which an antiferromagnetic layer and a ferromagnetic layer are exchange-coupled.

(427) While preferred embodiments of the present invention have been described above, it is to be understood that variations and modifications will be apparent to those skilled in the art without departing from the scope and spirit of the present invention. The scope of the present invention, therefore, is to be determined solely by the following claims.