JP2004219321A - Position displacement sensor using semiconductor magnetoresistive element - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a position displacement sensor using a semiconductor magnetoresistive element. <P>SOLUTION: In the position displacement sensor, at least a pair of impedance elements, where an AC voltage E is applied, is arranged on a straight line or a circumference at an interval and impedance in each element changes according to the position relationship with a movable magnetic body while the movable magnetic body is displaced from a position corresponding to one of the pair of impedance elements to a position corresponding to the other along the straight line or the circumference, thus containing a change constituent corresponding to the half cycle of sinusoidal waves in the position displacement sensor. The semiconductor magnetoresistive elements MR1 and MR3 should be used as the impedance elements. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【００１５】
この方式を位相差変換と呼び、図５に動作波形及び解析手順を示す。図５の分図（Ａ）は印加電圧ＡＳｉｎωｔと、Ｓｉｎωｔを搬送波とするθ変位に対応した（１）、（２）式の各右辺の電圧波形図、同（Ｂ）はＡＳｉｎωｔのグラフと特定の位相（θ，ｘ）を含む右辺の正弦波、この場合は、ａＳｉｎ（ωｔ−θ，ｘ）のグラフとの時間関係を示す波形図であり、更に同（Ｃ）はこれらの波形図の根拠となる数式的推移を示したものである。なお、進み位相波ａＳｉｎ（ωｔ＋θ，ｘ）の場合、そのグラフは破線で示した遅れ位相波ａＳｉｎ（ωｔ−θ，ｘ）のグラフと対称的に、実線で示したＡＳｉｎωｔのグラフよりも（θ，ｘ）だけ時間的に先行した正弦波として想起できるであろう。
【００１６】
図５Ｂのグラフから明らかなとおり、θ又はｘを求めるにはＡＳｉｎωｔグラフのゼロクロス点から、ａＳｉｎ（ωｔ−θ，ｘ）又はａＳｉｎ（ωｔ＋θ，ｘ）グラフのゼロクロス点までの時間をカウントすればよいことが理解されるであろう。このカウントは進相、遅相いずれの位相で求めてもよいが、コイルＬ１〜Ｌ４等に温度変化によるインピーダンス変化が生じている場合，それによる位相変動誤差±ｄはωｔに付随するため、ここでは詳述しないが上記の進相、遅相の両正弦波グラフの双方を用いて処理することにより、この±ｄを容易に消去することができる。
【００１７】
【発明が解決しようとする課題】
さて、インピーダンス型において、位置変位を上記のように交流振幅の正弦／余弦変位として検出するセンサの場合、コイルＬ１〜Ｌ４で相当のインピーダンス値（Ｚ）を得て、そのインピーダンスによる十分な交流電圧の降下（コイル端間電圧）を生ずるためにはＺ＝２πｆＬ（Ω）を与える周波数ｆを高くするか、インダクタンスＬ（Ｈ）自体を大きくする必要がある。
【００１８】
しかしながら、周波数ｆを高くすることに対しては回路の応答性による制約があり、逆に１周期内のクロック数（θ分解能）を大きくしたいという要請からは周波数ｆを低下すべきであるため、その調和点としてｆ＝１０ｋＨｚ程度が採用され、それ以上はθ分解能を上げられないという問題がある。（この場合のθ分解能とは、交流の１周期を２πラジアンという尺度で見た場合の表現であり、直線型の場合にはｘ分解能に対応する。）、また相当のインダクタンスを得るには１コイル当たり約２００〜８００ターンを要するため、図６に示すとおり、直線型の場合のコイルペアＬ１− Ｌ３又はＬ２ − Ｌ４の間隔（ 中心間隔）を約４．１ｍｍ以下にはできないという設計的理由による実寸分解能の制約がある。
【００１９】
本発明は、位置変位を交流振幅の正弦／余弦変位として検出するインピーダンス型センサにおける上記のような問題点を克服して、高い分解能の位置変位センサを提供しようとするものである。
【００２０】
【課題を解決するための手段】
上記の課題を解決するため、本発明は交流電圧が印加される少なくとも一対のインピーダンス素子が所定の経路、特に直線経路上又は円周経路上において間隔を置いて配列され、可動磁性体が前記一対のインピーダンス素子の一方に対応する位置から他方に対応する位置まで前記所定の経路に沿って変位する間に、各素子のインピーダンスが前記可動磁性体との位置関係に応じて変化することにより、正弦波の半周期に対応する変化成分を含むようにした位置変位センサにおいて、前記インピーダンス素子としてフィールド磁極に近接した半導体磁気抵抗素子を用いたことを特徴とするものである。
【００２１】
本発明の構成によれば、上記の通りインピーダンス素子として半導体磁気抵抗素子を用いたため、そのインピーダンス（抵抗値）は印加交流電圧の周波数とは無関係となり、十分な低周波数とすること、及び前述したコイルペアの間隔に対応する素子配列ピッチも十分に小さくすることができる。また、素子間隔がある程度大きくても、いわゆるθ分解能を十分に高めることにより、計測精度を上げ、高分解能を得ることができる。
【００２２】
【発明の実施の形態】
【００２３】
一般に、ＭＲと称される半導体磁気抵抗素子（以下「ＭＲ素子」と略称する）は温度特性が大きいため、従来の直流印加方式においても、図７に示すように一対のＭＲ素子を直列接続し、温度特性が互いに相殺されるように差動出力を取出す構成としている。
【００２４】
本発明において、インピーダンス素子として従来のコイルペア、例えば，Ｌ１− Ｌ３に代えて用いられるＭＲ素子であるＭＲ１、ＭＲ３は前述した通り計測動作上のインピーダンス変化が逆位相関係となり、図８に示すごとく、差動増幅器７等を用いて基本的に差動出力を取出す構成となっているため、両者の温度特性もまた当然に相殺されるようになっている。直線型センサ及び回転型センサの具体的な構成例は、それぞれ図９及び図１０に示す通りである。
【００２５】
【実施例】
図９Ａは直線変位検出用センサ１０の実施例を示す模式的な断面略図である。センサ本体にはＮまたはＳ磁極を有するフィールド磁石１２が配置され、例えば、４個のＭＲ素子、ＭＲ１〜ＭＲ４が等間隔で磁極横断方向に間隔を置いて順次配列される。これらのＭＲ配列に沿って磁性体配列棒１３が配置され、その軸方向（ＭＲ配列）に移動可能として被計測体の直線変位を測定できるようになっている。
【００２６】
磁性体配列棒１３は強磁性体１４と、反磁性体１５とをＭＲ１〜ＭＲ４の配列のほぼ全長（正確にはＭＲ間ピッチの４倍）を１ピッチとして交互に配列したものであり、１個のＭＲ素子（例えば、ＭＲ２）が強磁性体１４の中央部に対応するときは、それと対をなす１個飛んだ位置のＭＲ素子（例えば、ＭＲ４）が反磁性体１５の中央部に対応するようになっている。なお、反磁性体１５に代えて，図１に示した従来の誘導方式（直線型）における弱磁性体等の心棒のみにし、空隙を拡大する形で強磁性体１４よりも顕著に磁気抵抗（リラクタンス）を大きくする形式としてもよい。
【００２７】
図９Ｂに示すとおり、各Ｒ− ＭＲ回路に並列的に交流電圧Ｅ＝ＡＳｉｎωｔが印加されると、この場合の素子ＭＲ１，ＭＲ２，ＭＲ３，ＭＲ４は、前述した誘導方式においてコイルインダクタンスの変化を与えたと同様な、磁性体配列棒３との関係位置に応じた抵抗値変化を示すため、それらＭＲ素子両端に現れる正弦波電圧の振幅がその抵抗値変化に応じて変化する。このような各ＭＲ素子間電圧は、ＭＲ１− ＭＲ３の組，及びＭＲ２− ＭＲ４の組におけるセンサ出力として、演算増幅器ＯＡにより差動的に取出されるものである。
【００２８】
図９Ａにおける磁性体配列棒１３の軸方向位置（ＭＲ２− 強磁性体１４対応、ＭＲ４− 反磁性体１５対応位置）をｘ＝０とし、且つ同棒の、例えば図の左方への移動を正方向とすれば、図９Ｂの回路では、ＭＲ１− ＭＲ３用演算増幅器ＯＡ（１− ３）の出力Ｘが、基本信号としての正弦波Ｓｉｎωｔに、変動する係数すなわち振幅の変位ａＳｉｎｘを掛けた値となり、ＭＲ２− ＭＲ４用演算増幅器ＯＡ（２− ４）の出力Ｙが、基本信号としての正弦波Ｓｉｎωｔに、変動する係数すなわち振幅の変位ａＣｏｓｘを掛けた値となるように構成されている。したがって、これらの正弦波では磁性体配列棒１３の１ピッチ変位（実寸）をｘ（ｐ）＝２π［ｒａｄ］ （３６０度）として処理しうることが理解されるであろう。
【００２９】
図１０Ａは回転変位検出用センサ１０’の実施例を示す模式的な断面略図であり、円環状センサ本体（ヨーク）内において中心軸に向かって突出した４磁極（この場合、Ｎ極）が９０度間隔で配置され、これらの磁極端面に、それぞれ素子ＭＲ１，ＭＲ２，ＭＲ３，ＭＲ４が装着され、各素子の突出端が更に中心軸を向くようになっている。センサ本体内には更に、素子ＭＲ１〜ＭＲ４よりも内側で中心軸の周りで回転するようにした強磁性体からなる偏心磁性体１６が配置され、各Ｎ極から出た磁束は素子ＭＲ１〜ＭＲ４を通り，空隙から偏心磁性体１６の周囲面に入り、同１６のいずれかの部位から何らかの磁路１７を経てセンサ本体のＳ極に戻るようになっている。
【００３０】
図１０Ｂに示すとおり、素子ＭＲ１〜ＭＲ４は、適当な抵抗を介して並列接続されて交流電源Ｅに接続されることにより、直線型の場合と同じく同相で印加され、偏心磁性体１６周面との間隔に応じたインダクタンスの変化によって異なる両端間電圧を具現するようになっている。この場合も、各ＭＲ素子間電圧は、ＭＲ１− ＭＲ３の組，及びＭＲ２− ＭＲ４の組におけるセンサ出力として、演算増幅器ＯＡにより差動的に取出されるようになっている。
【００３１】
図１０Ａにおける偏心磁性体１６の角度位置（最下位偏心位置）をθ＝０とし、且つ図の時計回りの回転を正方向とすれば、図１０Ｂでは、ＭＲ１− ＭＲ３用演算増幅器ＯＡ（１− ３）の出力Ｘが、基本信号としての正弦波Ｓｉｎωｔに、係数すなわち振幅の変位ａＳｉｎθを掛けた値となり、ＭＲ２− ＭＲ４用演算増幅器ＯＡ（２− ４）の出力Ｙが、基本信号としての正弦波Ｓｉｎωｔに、係数すなわち振幅の変位ａＣｏｓθを掛けた値となるように差動回路を構成したことを示している。したがって、偏心磁性体１６の１回転は、θの３６０度変位に対応することが理解されるであろう。
【００３２】
かくして、基本信号の周波数を十分に低くし、且つ直線型においては測定ピッチを１ｍｍ程度としても、上記のような出力Ｘ及びＹを正確に得ることができ、これらを前述した図５の過程に従ってＸ式のＳｉｎωｔをＣｏｓωｔに変換したＸ’式として、Ｘ’＋Ｙ＝ａＳｉｎ（ωｔ±θ，ｘ）を得るならば、例えば、１ピッチ（θ＝３６０度、ｘ＝１ｍｍ）を２^１６＝６５５３６分割し，角度の場合なら３６０／６５５３６＝０．００５５［ 度］ の分解能、１ｍｍピッチの直線変位の場合なら１０００／６５５３６＝０．０１５［ μｍ］ という、所謂ナノレベルの分解能を得ることができる。
【００３３】
【発明の効果】
以上述べたとおり、本発明は位置変位を交流振幅の正弦／余弦変位として検出するインピーダンス型センサにおけるインピーダンス素子として半導体磁気抵抗素子、いわゆるＭＲ素子を用いたことにより、所謂ナノテクノロジーにも対応できる高分解能の位置変位センサを提供するものであることが明らかである。
【図面の簡単な説明】
【図１】従来の誘導方式による直線変位検出型センサの構造原理を模式的に示す斜視図である。
【図２】従来の誘導方式による回転変位検出型センサの構造原理を模式的に示す斜視図である。
【図３】図１及び図２の構造がそれぞれ対象とする角度及び直線変位検出の態様を、一括して回路的に示したものである。
【図４】図３の誘導方式におけるトランス構造に代えて、実質上自己インダクタンスＬ１〜Ｌ４のみからなるインピーダンス素子を用い，同様の関係が得られるようにしたセンサ回路図である。
【図５】直線変位検出型及び回転変位検出型センサの動作波形及び解析手順を示す図であり、（Ａ）は印加電圧ＡＳｉｎωｔと、Ｓｉｎωｔを搬送波とするθ変位を含む電圧波形図、（Ｂ）はＡＳｉｎωｔのグラフと特定の位相（θ，ｘ）を含む正弦波、ａＳｉｎ（ωｔ−θ，ｘ）のグラフとの時間関係を示す波形図、（Ｃ）はこれらの波形図の根拠となる数式的推移を示した図である。
【図６】従来の直線型センサの場合におけるコイルペアＬ１− Ｌ３又はＬ２ − Ｌ４の間隔（ 中心間隔）を約８．２ｍｍ以下にはできないことを示すためのコイル断面図である。という設計的理由による実寸分解能の制約がある。
【図７】従来の直流印加方式において、一対のＭＲ素子を直列接続し温度特性が互いに相殺されるように差動出力を取出す構成とした回路図である。
【図８】本発明の直線型センサ及び回転型センサの部分要素として、差動増幅器等を用いて基本的に差動出力を取出すように構成した部分回路図である。
【図９】本発明のセンサに適用される好ましい直線変位検出センサの実施例を示す図であって、その構造の模式的な断面略図（Ａ）、及び素子ＭＲ１− ＭＲ３の組，及びＭＲ２− ＭＲ４の組の差動接続と演算増幅器より構成されるそれらの並列的な差動回路（Ｂ）を示したものである。
【図１０】本発明のセンサに適用される好ましい回転変位検出センサの実施例を示す図であって、その構造の模式的な断面略図（Ａ）、及び素子ＭＲ１− ＭＲ３の組，及びＭＲ２− ＭＲ４の組の差動接続と演算増幅器より構成されるそれらの並列的な差動回路（Ｂ）を示したものである。
【符号の説明】
７ 差動増幅器
１０ 直線変位検出用センサ
１２ フィールド磁石
ＭＲ１〜ＭＲ４ 半導体磁気抵抗素子
１３ 磁性体配列棒
１４ 強磁性体
１５ 反磁性体
１６ 偏心磁性体
１７ 磁路[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a position displacement sensor using a semiconductor magnetoresistive element, and more particularly to a sine / cosine displacement detection type sensor by applying an alternating current.
[0002]
[Prior art]
In recent years, sensors that detect position displacement as sine / cosine displacement of AC amplitude have been developed, and these sensors are classified into two types, an inductive type and an impedance type, based on a driving principle. In either type, there are a linear displacement detection method in which a plurality of impedance elements or coils are linearly arranged, and a rotation angle detection method in which the impedance elements or coils are arranged in the circumferential direction.In the induction type, a secondary coil of a transformer is used. These arrays are formed (for example, see Patent Documents 1 and 2).
[0003]
[Patent Document 1]
Japanese Utility Model Registration No. 3036285 (see FIG. 2)
[0004]
[Patent Document 2]
JP-A-10-176925 (refer to FIGS. 1 to 3)
[0005]
On the other hand, as an element for detecting a displacement of an object from an impedance change due to a change in a magnetic field, there has recently been used an element using a semiconductor magnetoresistive element (for example, see Patent Document 3). It is merely to measure the change in DC resistance (DC voltage drop), and is not intended to perform a precise measurement by trigonometric function processing by adding a sine wave AC as in the present invention.
[0006]
[Patent Document 3]
JP-A-2000-292113 (see abstract)
[0007]
Therefore, in order to provide a basic understanding of the position displacement sensor based on the sine / cosine displacement of the AC amplitude, which is the object of the present invention, first, the operation principle of the linear type displacement sensor will be outlined with reference to FIG. In FIG. 1, reference numeral 1 denotes a solenoid type primary coil, and reference numeral 2 denotes a secondary coil row coaxially arranged inside the coil. Here, four coils L1, L2, L3, L4 of the same standard are arranged at equal intervals. I have. Into this coil arrangement, a magnetic substance arrangement rod (movable magnetic substance) 3 is inserted coaxially with the coil axis and movable in the axial direction. A plurality of cylindrical ferromagnetic bodies 4 are arranged on the magnetic array bar 3 with one pitch equal to four times the coil arrangement interval (distance between adjacent coil centers). The rod axis of the magnetic array rod 3 is made of a metal or plastic material that does not show remarkable magnetism.
[0008]
The secondary coils L1, L2, L3, and L4 induce a sine wave voltage by a mutual induction action from the primary coil 1 to which the sine wave voltage ASinωt is applied as shown in the figure. In the configuration of the secondary circuit, the voltage between the coil ends having different amplitudes is embodied by the change of the magnetic flux according to the relational position of (1), and the axial position (position of L1-L3) of the magnetic array bar 3 in FIG. : Both coils have an equilibrium position in which only the ends facing each other across L2 correspond to or slightly cover the ends of the ferromagnetic material 4. L2-L4 positions: L2 corresponds to the ferromagnetic material 4, If L4 is a position corresponding to the rod axis of the array rod 3 via a large gap between the ferromagnetic bodies 4, x = 0, and the movement of the rod, for example, to the left in the drawing is the positive direction, The L1-L3 differential output has a varying coefficient to the induced sine wave Sinωt. Value · aSinxSinωt multiplied by the ie amplitude of the displacement aSinx
The L2-L4 differential output is a value obtained by multiplying the induced sine wave Sinωt by a fluctuating coefficient, that is, an amplitude displacement aCosx. ACosxSinωt
Therefore, it will be understood that one pitch displacement (actual size) of the magnetic array bar 3 can be processed as x (p) = 2π [rad] (360 degrees) in a sine or cosine function.
[0009]
The operation principle of the rotation method is as shown in FIG. In FIG. 2, each of the coils L1, L2, L3, and L4 of the secondary coil 2 'has a parallel axis in a form inscribed in the large-diameter primary coil 1', and is arranged at an annular interval. Both have pole cores 5 with extreme faces located in the same axial cross section. A ferromagnetic eccentric plate 6 for covering the pole core 5 of the secondary coil in accordance with the rotational position is provided on the shaft sleeve which is arranged coaxially with the central axis of the secondary coil 2 ′, that is, the central axis of the secondary coil array. Supported.
[0010]
The coils L1, L2, L3, and L4 induce a sine wave voltage based on the sine wave voltage ASinωt applied to the primary side by mutual induction with the primary coil 1 ′. Due to the change in magnetic flux according to the condition, a different coil end-to-end voltage is realized in the secondary circuit configuration. It is clear that these coil end-to-end voltages can be doubled and detected by differentially extracting them in the L1-L3 set and the L2-L4 set whose arrangement relationship is 180 degrees (opposite phase). .
[0011]
The angular position of the ferromagnetic eccentric plate 6 in FIG. 2 (the core of the coil L2 is completely covered by the eccentric plate 6, the core of the coil L4 is completely exposed, and the set of coils L1 to L3 is equally and semi-exposed. Is set to θ = 0 and the clockwise rotation in the figure is set to the positive direction, the secondary coil L1-L3 differential output multiplies the induced sine wave Sinωt by a coefficient, that is, an amplitude displacement aSinθ. Value, ie, aSinθSinωt
The L2-L4 differential output is a value obtained by multiplying the induced sine wave Sinωt by a coefficient, that is, a displacement of the amplitude aCosθ, ie, aCosθSinωt
Therefore, it will be understood that one rotation of the ferromagnetic eccentric plate 6 corresponds to a 360 degree displacement of θ.
[0012]
FIG. 3 is a circuit diagram showing these relationships collectively with respect to the angle and the linear displacement, and FIG. 3 shows a similar relationship. In the same relationship, L1 to L4 are impedance elements substantially consisting only of a self-inductance, and the sine wave voltage ASinωt is directly applied. FIG. 4 is a circuit diagram showing that this also occurs when added. The present invention relates to a novel sensor structure for directly detecting a positional displacement as a sine / cosine displacement of an AC amplitude in the latter impedance type.
[0013]
Therefore, also in the impedance type, if the terminal voltages of the coils L1 to L4 obtained in the same manner as above are represented by X as a sine coefficient equation and Y as a cosine coefficient equation, the angle and linear displacement are collectively expressed as follows. Can be expressed as
X = aSin (θ, x) · Sinωt (1)
Y = aCos (θ, x) Sinωt (2)
From these equations, as an arithmetic circuit for calculating the angular displacement θ (the same applies to the linear displacement x, the description will be made only with θ), first, the equations (1) and (2) are calculated from 0 to 0, respectively. By multiplying the cosine function Cosφ and the sine function Sinφ of the digital phase value φ that sequentially increases,
・ Sinφ ・ Cosθ-Cosφ ・ Sinθ = 0 ・ ・ (3)
At this point, there is a known RD conversion method that specifies θ by setting θ = φ, but this method has a problem that a clock delay occurs when φ is tracked and counted, resulting in poor response.
[0014]
In the embodiment of the present invention, by converting Sinωt in the equation (1) into Cosωt on the circuit, aSin (θ, x) · Cosωt is obtained, and the addition theorem of the trigonometric function is directly applied. Ie

[0015]
This method is called phase difference conversion, and FIG. 5 shows an operation waveform and an analysis procedure. 5A is a voltage waveform diagram on each right side of equations (1) and (2) corresponding to applied voltage ASinωt and θ displacement using Sinωt as a carrier, and FIG. 5B is a graph of ASinωt. Is a waveform diagram showing the time relationship with the sine wave on the right side including the phase (θ, x) of this example, in this case, the graph of aSin (ωt−θ, x), and FIG. It shows a mathematical transition that serves as a basis. In the case of the leading phase wave aSin (ωt + θ, x), the graph is symmetrical to the graph of the lagging phase wave aSin (ωt−θ, x) shown by the broken line, and is more (θ) than the graph of ASinωt shown by the solid line. , X) could be recalled as a sine wave temporally preceding.
[0016]
As is clear from the graph of FIG. 5B, to obtain θ or x, the time from the zero cross point of the ASin ωt graph to the zero cross point of the aSin (ωt−θ, x) or aSin (ωt + θ, x) graph may be counted. It will be appreciated. This count may be obtained in any of the leading phase and the lagging phase. However, if an impedance change occurs due to a temperature change in the coils L1 to L4 and the like, a phase variation error ± d due to the variation is accompanied by ωt. Although not described in detail, by performing processing using both of the above-described sine wave graphs of the leading phase and the lagging phase, ± d can be easily eliminated.
[0017]
[Problems to be solved by the invention]
Now, in the case of an impedance type sensor that detects a position displacement as a sine / cosine displacement of an AC amplitude as described above, a considerable impedance value (Z) is obtained by the coils L1 to L4, and a sufficient AC voltage due to the impedance is obtained. In order to cause the voltage drop (voltage between coil ends), it is necessary to increase the frequency f at which Z = 2πfL (Ω) or to increase the inductance L (H) itself.
[0018]
However, increasing the frequency f is limited by the responsiveness of the circuit, and conversely, the frequency f should be reduced in response to a request to increase the number of clocks in one cycle (θ resolution). As the harmony point, about f = 10 kHz is adopted, and beyond that, there is a problem that the θ resolution cannot be increased. (The θ resolution in this case is an expression when one cycle of AC is viewed on a scale of 2π radians, and corresponds to x resolution in the case of a linear type.) To obtain a considerable inductance, Since about 200 to 800 turns are required per coil, as shown in FIG. 6, the design is such that the interval (center interval) between the coil pairs L1 to L3 or L2 to L4 cannot be reduced to about 4.1 mm or less in the case of the linear type. There is a restriction on the actual size resolution.
[0019]
An object of the present invention is to provide a high-resolution position displacement sensor by overcoming the above-described problems in an impedance type sensor that detects a position displacement as a sine / cosine displacement of an AC amplitude.
[0020]
[Means for Solving the Problems]
In order to solve the above problems, the present invention provides that at least a pair of impedance elements to which an AC voltage is applied are arranged at intervals on a predetermined path, particularly on a linear path or a circumferential path, and the movable magnetic body is provided with the pair of impedance elements. During the displacement along the predetermined path from the position corresponding to one of the impedance elements to the position corresponding to the other of the impedance elements, the impedance of each element changes according to the positional relationship with the movable magnetic body, so that the sine In a position displacement sensor including a change component corresponding to a half cycle of a wave, a semiconductor magnetoresistive element close to a field magnetic pole is used as the impedance element.
[0021]
According to the configuration of the present invention, since the semiconductor magnetoresistive element is used as the impedance element as described above, its impedance (resistance value) is independent of the frequency of the applied AC voltage, and is set to a sufficiently low frequency. The element arrangement pitch corresponding to the interval between the coil pairs can also be made sufficiently small. Further, even if the element spacing is large to some extent, by sufficiently increasing the so-called θ resolution, the measurement accuracy can be increased and a high resolution can be obtained.
[0022]
BEST MODE FOR CARRYING OUT THE INVENTION
[0023]
In general, since a semiconductor magnetoresistive element called MR (hereinafter abbreviated as “MR element”) has a large temperature characteristic, even in a conventional DC application method, a pair of MR elements are connected in series as shown in FIG. , And a differential output is taken so that the temperature characteristics cancel each other.
[0024]
In the present invention, the conventional coil pairs as the impedance elements, for example, MR1 and MR3, which are MR elements used in place of L1-L3, have the impedance change in the measurement operation in the opposite phase relationship as described above, and as shown in FIG. Since the differential output is basically obtained by using the differential amplifier 7 and the like, the temperature characteristics of the two are naturally canceled. Specific configuration examples of the linear sensor and the rotary sensor are as shown in FIGS. 9 and 10, respectively.
[0025]
【Example】
FIG. 9A is a schematic sectional view showing an embodiment of the sensor 10 for detecting a linear displacement. A field magnet 12 having N or S magnetic poles is arranged in the sensor body. For example, four MR elements, MR1 to MR4, are arranged at equal intervals in the transverse direction of the magnetic poles. The magnetic material array rods 13 are arranged along these MR arrays, and can be moved in the axial direction (MR array) so that the linear displacement of the measured object can be measured.
[0026]
The magnetic material array rod 13 is formed by alternately arranging the ferromagnetic material 14 and the diamagnetic material 15 in such a manner that the pitch of the arrangement of MR1 to MR4 is almost the entire length (more precisely, four times the pitch between MRs). When one MR element (for example, MR2) corresponds to the central part of the ferromagnetic body 14, one MR element (for example, MR4) which is one pair away from it corresponds to the central part of the diamagnetic body 15. It is supposed to. Instead of the diamagnetic material 15, only a mandrel such as a weak magnetic material in the conventional induction system (linear type) shown in FIG. (Reluctance) may be increased.
[0027]
As shown in FIG. 9B, when an AC voltage E = ASinωt is applied to each R-MR circuit in parallel, the elements MR1, MR2, MR3, and MR4 in this case give a change in coil inductance in the inductive method described above. In order to show a change in resistance according to the position relative to the magnetic material arrangement rod 3, the amplitude of the sine wave voltage appearing at both ends of the MR element changes in accordance with the change in resistance. Such voltages between the MR elements are differentially extracted by the operational amplifier OA as sensor outputs in the set of MR1 to MR3 and the set of MR2 to MR4.
[0028]
9A, the axial position (position corresponding to the MR2-ferromagnetic substance 14 and position corresponding to the MR4-diamagnetic substance 15) of the magnetic substance array rod 13 is set to x = 0, and the rod is moved to the left in the drawing, for example. 9B, the output X of the operational amplifier OA (1-3) for MR1-MR3 is a value obtained by multiplying a sine wave Sinωt as a basic signal by a fluctuating coefficient, that is, an amplitude displacement aSinx. The output Y of the operational amplifier OA (2-4) for MR2-MR4 is configured to be a value obtained by multiplying a sine wave Sinωt as a basic signal by a fluctuating coefficient, that is, an amplitude displacement aCosx. Therefore, it will be understood that one pitch displacement (actual size) of the magnetic array rod 13 can be processed as x (p) = 2π [rad] (360 degrees) with these sine waves.
[0029]
FIG. 10A is a schematic cross-sectional schematic view showing an embodiment of the rotational displacement detecting sensor 10 ′. In the annular sensor main body (yoke), four magnetic poles (N pole in this case) projecting toward the central axis are 90. The elements MR1, MR2, MR3, and MR4 are mounted on these magnetic pole faces, respectively, and the protruding ends of the elements are further directed to the central axis. An eccentric magnetic body 16 made of a ferromagnetic material that is rotated around a central axis inside the sensors MR1 to MR4 is further disposed in the sensor main body. Through the gap, enters the peripheral surface of the eccentric magnetic body 16, and returns to the S pole of the sensor main body from any part of the eccentric magnetic body 16 via a certain magnetic path 17.
[0030]
As shown in FIG. 10B, the elements MR1 to MR4 are connected in parallel via an appropriate resistor and connected to an AC power source E, so that the elements MR1 to MR4 are applied in the same phase as in the case of the linear type, and are connected to the peripheral surface of the eccentric magnetic body 16. A different voltage between both ends is realized by the change of the inductance according to the interval of. Also in this case, the voltage between the MR elements is differentially extracted by the operational amplifier OA as the sensor output in the set of MR1 to MR3 and the set of MR2 to MR4.
[0031]
If the angle position (lowest eccentric position) of the eccentric magnetic body 16 in FIG. 10A is θ = 0 and the clockwise rotation in the drawing is the positive direction, in FIG. 10B, the operational amplifier OA (1- The output X of 3) becomes a value obtained by multiplying the sine wave Sinωt as the basic signal by the coefficient, that is, the displacement aSinθ of the amplitude, and the output Y of the operational amplifier OA (2-4) for MR2-MR4 becomes the sine wave as the basic signal. This shows that the differential circuit is configured to have a value obtained by multiplying the wave Sinωt by a coefficient, that is, a displacement aCosθ of the amplitude. Therefore, it will be understood that one rotation of the eccentric magnetic body 16 corresponds to a 360 degree displacement of θ.
[0032]
Thus, even if the frequency of the basic signal is sufficiently low and the measurement pitch is about 1 mm in the linear type, the outputs X and Y as described above can be accurately obtained. If X ′ + Y = aSin (ωt ± θ, x) is obtained as X ′ expression obtained by converting Sinωt of the X expression into Cosωt, for example, one pitch (θ = 360 degrees, x = 1 mm) is calculated as 2 ¹⁶ = 65536. It is possible to obtain a so-called nano-level resolution of 360/65536 = 0.0055 [degree] in the case of an angle and 1000/65536 = 0.015 [μm] in the case of a 1 mm pitch linear displacement. .
[0033]
【The invention's effect】
As described above, the present invention uses a semiconductor magnetoresistive element, a so-called MR element, as an impedance element in an impedance type sensor for detecting a position displacement as a sine / cosine displacement of an AC amplitude, and thus can respond to a so-called nanotechnology. It is clear that it provides a position displacement sensor with high resolution.
[Brief description of the drawings]
FIG. 1 is a perspective view schematically showing the structure principle of a conventional linear displacement detection type sensor using an induction method.
FIG. 2 is a perspective view schematically showing the structural principle of a conventional rotational displacement detection type sensor using an induction method.
FIG. 3 is a circuit diagram collectively showing the angle and linear displacement detection modes targeted by the structures of FIGS. 1 and 2, respectively.
FIG. 4 is a sensor circuit diagram in which a similar relationship is obtained by using an impedance element having substantially only self-inductances L1 to L4 instead of the transformer structure in the inductive system of FIG.
5A and 5B are diagrams showing operation waveforms and analysis procedures of a linear displacement detection type sensor and a rotational displacement detection type sensor. FIG. 5A is a voltage waveform diagram including an applied voltage ASinωt and a θ displacement using Sinωt as a carrier wave, and FIG. ) Is a waveform diagram showing the time relationship between the graph of ASinωt and a sine wave including a specific phase (θ, x), and the graph of aSin (ωt−θ, x), and FIG. 2C is the basis for these waveform diagrams. It is a figure showing a mathematical transition.
FIG. 6 is a coil cross-sectional view showing that the distance (center distance) between coil pairs L1-L3 or L2-L4 cannot be reduced to about 8.2 mm or less in the case of a conventional linear sensor. There is a restriction on the actual size resolution due to design reasons.
FIG. 7 is a circuit diagram of a conventional DC application method, in which a pair of MR elements are connected in series and a differential output is taken out so that temperature characteristics are offset each other.
FIG. 8 is a partial circuit diagram configured to basically take out a differential output by using a differential amplifier or the like as a partial element of the linear sensor and the rotary sensor of the present invention.
FIG. 9 is a view showing an embodiment of a preferred linear displacement detection sensor applied to the sensor of the present invention, which is a schematic cross-sectional view (A) of the structure, a set of elements MR1-MR3, and MR2- It shows a differential connection (B) composed of a differential connection of MR4 sets and an operational amplifier.
FIG. 10 is a view showing an embodiment of a preferred rotational displacement detection sensor applied to the sensor of the present invention, which is a schematic cross-sectional schematic view (A) of the structure, a set of elements MR1-MR3, and MR2- This shows a differential connection (B) composed of a differential connection of MR4 sets and an operational amplifier.
[Explanation of symbols]
Reference Signs List 7 Differential amplifier 10 Linear displacement detection sensor 12 Field magnets MR1 to MR4 Semiconductor magnetoresistive element 13 Magnetic substance arrangement rod 14 Ferromagnetic substance 15 Diamagnetic substance 16 Eccentric magnetic substance 17 Magnetic path

Claims (3)

At least a pair of impedance elements to which an AC voltage is applied are arranged at intervals along a predetermined path, and the movable magnetic body has a predetermined position from a position corresponding to one of the pair of impedance elements to a position corresponding to the other of the pair of impedance elements. During displacement along the path, the impedance of each element changes in accordance with the positional relationship with the movable magnetic body, so that the position displacement sensor includes a sinusoidal change component. A position displacement sensor using a semiconductor magnetoresistive element close to a magnetic pole. The position displacement sensor according to claim 1, wherein the predetermined path is linear. The position displacement sensor according to claim 1, wherein the predetermined path has a circular shape.

Images