JP3561100B2

JP3561100B2 - Optical encoder

Info

Publication number: JP3561100B2
Application number: JP28081996A
Authority: JP
Inventors: 淳家城; 一博羽根
Original assignee: Okuma Corp
Current assignee: Okuma Corp
Priority date: 1996-10-23
Filing date: 1996-10-23
Publication date: 2004-09-02
Anticipated expiration: 2016-10-23
Also published as: JPH10122909A

Description

【０００１】
【発明の属する技術分野】
本発明は、工作機械や半導体製造装置の位置計測に利用される光学式エンコーダに関する。
【０００２】
【従来の技術】
光学式エンコーダは、光束を発する発光ユニットと２つの回折格子の後方に光電変換素子を配し、２つの回折格子を透過した光を光電変換素子により検出し、２つの回折格子の相対移動によって生じる光の強度の変化を基に、移動量を検出するものである。
図７には、前記２つの回折格子２０の一例が示されている。回折格子２０は、光を透過させる透過部及び透過させない非透過部が、交互に配置されている、いわゆる振幅変調型の格子を有している。この配列ピッチが格子ピッチと呼ばれ、透過部、非透過部の各々の幅は、格子ピッチＰの２分の１となっている。
前記の回折格子２０を２つ用いたエンコーダにおいては、それぞれの透過部が一致したとき、透過光量が最大となり、光電変換素子で検出される出力も最大となる。また、一方の透過部と他方の非透過部の位置が一致した場合、透過光量は最小となり、光電変換素子で検出される出力も最小となる。光電変換素子から出力される電気信号は、２つの回折格子２０の相対変位量に対して、前記最大値、最小値の間を増減する。２つの回折格子２０を等速で相対変位させた場合に、光電変換素子で得られる出力信号は、理想的には、周期Ｐの三角波信号となる。しかし、実際は回折の影響によって前記三角波形状はひずんでいる。そして、この関数を正弦波として取扱い相対変位の検出がなされていた。
【０００３】
【発明が解決しようとする課題】
回折格子２０を用いた光学式エンコーダは、出力信号を正弦波として取り扱って相対位置を求めている。しかしながら、この出力信号に種々のひずみ成分を含んでおり、そのため、格子ピッチより細かな位置データを求める際に、ひずみ成分による正弦波からのずれ量により、位置検出値に大きな誤差が含まれていた。この実際の出力信号と正弦波のずれにより生じる誤差は分割誤差と呼ばれている。
さらに、上述した従来の光学式エンコーダで得られる変位信号のひずみ率は、第１回折格子と第２回折格子との間隔が変化すると大きく変動してしまう。そのため、誤差を一定値以内に抑えるためには、第１回折格子と第２回折格子との間隔を適切な間隔で一定に保つ必要があり、非常に厳しい取り付け精度が要求されるという問題があった。
特開平３−４８１２２号公報には、出力信号のひずみ成分を除去するために回折格子が、隣合った透過部の間隔が等しくなく、透過部を所定の位相差をもって配置することで、３次や５次の高調波成分を除去する光学式エンコーダが開示されている。しかし、このような構成において、例えば３次及び５次の高調波歪成分を除去する場合には、少なくとも４つのスリットを必要とする。スリットの本数が十分多い場合は、照射光束の不均一さがあっても十分な平均化効果が発揮されるが、スリットの本数が少ない場合、例えば４本〜数本程度の場合は平均化効果が低下し、３次及び５次の高調波歪成分の除去能力が薄れるという問題があった。
【０００４】
また、従来の構成の光学式エンコーダでは、図８に示すような第２回折格子２１が用いられており、位置を検出する場合、各相の検出場所が離れており、各相の検出が異なった場所の第１格子において行なわれているため、スケールの汚れやキズや誤差の影響により各信号にアンバランスが生じ、それが誤差になってしまうという問題があった。そして、光源の平行性や強度のばらつきによっても、各相の信号にばらつきが生じるという問題があった。
また、各相の光のクロストークを防ぐために、第２の格子や受光部３０の受光素子各相の間にスペースを設ける必要があり、装置の大型化をもたらしていた。
【０００５】
本発明は上述した事情から成されたものであり、本発明の目的は、より歪の少ない変位信号を充分な平均化効果をもって安定して出力し、かつ、各相の検出をできる限り同一の場所で行うことができる光学式エンコーダを提供することにある。
【０００６】
【課題を解決するための手段】
前述の課題を解決するために、本発明にかかる光学式エンコーダは、スケールと、このスケールと相対変位する受光素子とを備え、前記受光素子からの所定位相の位相差を有する信号をもとに、前記相対変位を検出する光学式エンコーダにおいて、前記スケールの長手方向のパターン幅ならびにパターン周期に対応した受光素子が前記長手方向に複数並設されて成る受光素子群であって、複数の前記所定位相の信号のそれぞれに対応した受光素子群を備え、一つの前記受光素子群と他の前記受光素子群に属する受光素子が長手方向に入り混じって配置され、一つの前記受光素子群と他の前記受光素子群は、一つの前記受光素子群で得られる信号と他の前記受光素子群で得られる信号とが、所定の位相差を有するように配置されており、前記受光素子は、除去する高次フーリェ成分をａ次およびｂ次としたときに、前記受光素子の幅が下に示す式によって表され、
受光素子パターン幅Ｗ＝Ｐ×（ｎ／ａ±（１＋２ｍ）／２ｂ）
ただし、ｎ：−∞〜∞ ｍ：−∞〜∞
一つの相の信号に対応した受光素子群は、それに含まれる受光素子の幅が、前記受光素子パターンの配置分布に、ａ次およびｂ次の高次フーリェ成分を含まないように決定されているものであり、或いは、スケールと、このスケールと相対変位する受光素子とを備え、前記受光素子からの所定位相の位相差を有する信号をもとに、前記相対変位を検出する光学式エンコーダにおいて、前記スケールの長手方向のパターン幅ならびにパターン周期に対応した受光素子が前記長手方向に複数並設されて成る受光素子群であって、複数の前記所定位相の信号のそれぞれに対応した受光素子群を備え、一つの前記受光素子群と他の前記受光素子群に属する受光素子が長手方向に入り混じって配置され、一つの前記受光素子群と他の前記受光素子群は、一つの前記受光素子群で得られる信号と他の前記受光素子群で得られる信号とが、所定の位相差を有するように配置されており、一つの相の信号に対応した受光素子群は、それに含まれる受光素子の幅が、前記受光素子パターンの配置分布に、高次フーリェ成分を含まないように決定されており、前記受光素子は、除去する高次フーリェ成分をａ次およびｂ次としたときに、前記受光素子の幅が下に示す式によって表され、
受光素子パターン幅Ｗ＝Ｐ×（ｎ／ａ±（１＋２ｍ）／２ｂ）
ただし、ｎ：−∞〜∞ ｍ：−∞〜∞
さらに除去すべき高次フーリェ成分をｃ次およびｄ次としたときに、一つの相の信号に対応した受光素子群は、それに含まれる受光素子の間隔が一定でなく、一つの受光素子を基準としたとき、他の受光素子が、ｋ×Ｐ（ｋは正の整数、Ｐは信号の周期）に加えて、Ｐ／（２・ｃ）、およびＰ／（２・ｄ）およびＰ／（２・ｃ）＋Ｐ／（２・ｄ）だけシフトして配置されているものである。
【０００７】
【作用】
この構成によれば、受光素子の幅により高次の歪成分が除去されるようになっており、少ない本数の受光素子によりより歪の少ない変位信号を得ることができるので、信号が、充分な平均化効果をもって安定して出力できる。また、各相用の受光素子が光電変換素子上に格子状に混在して形成されているので、各相の検出をほぼ同一場所で行なうことができる。そのため、スケールの汚れやキズなどがあっても、各相への影響が同一で、各出力信号はアンバランスが生じず、それによる誤差は発生しないという利点を有する。そして、これらは、光源の平行性や強度のばらつき、スケールの誤差に対しても、同様に影響を受けにくという利点を有する。
【０００８】
【発明の実施の形態】
以下、本発明の好適な実施の形態（以下、実施形態と記す）を図面に従って説明する。
図１は本発明の第１の実施形態のエンコーダの斜視構造図であり、本実施形態のエンコーダは、光源からの光を最初に受ける第１回折格子１と、図示矢印方向に相対移動し、第１回折格子１を透過した光を受けてこの光量に応じた電気信号を出力する受光素子などからなる受光部３を有している。受光部３の受光素子は、図２に示すように格子状に設けられている。図２に示す受光部３の受光素子３０は、平均すると数μｍ〜数百μｍの間隔で配置されている。この配置パターンは、４相の信号ａ，ｂ，ａ／，ｂ／（０゜、９０゜、１８０゜、２７０゜）を出力する光学式エンコーダに用いられるものであり、各々受光素子群３ａ、３ｂ、３ａ／、３ｂ／より出力するようになっている。さらにこれら受光素子群は、３１ａと３２ａ、３１ｂと３２ｂ、３１ａ／と３２ａ／、３１ｂ／と３２ｂ／とにわかれている。
本実施例では、第１格子１のピッチをＰとすると、受光素子は、各受光素子群の信号が、所定の位相差を有するために、その中心の間隔が、３Ｐ／４となっている。受光素子３１ａの幅は１３Ｐ／３０、受光素子３２ｂの幅は７Ｐ／３０となっている。そして、ある受光素子を受光素子３１ａとしたとき、その隣の受光素子をｂ信号用の受光素子３２ｂ、その隣の受光素子をａ／信号用の受光素子３１ａ／、その隣の受光素子をｂ／信号用の受光素子３２ｂ／とし、その隣には受光素子３２ａ、３１ｂ、３２ａ／、３１ｂ／が配置される。配置パターンは、これらを周期として繰り返す。
受光素子で得られる光強度の信号は、図１に示すように第１回折格子１を透過した光を直接格子状の受光素子で受光する場合、例えば受光素子３ａで得られる信号Ｉａは、数１で示される。
【数１】
Ｉａ（ｘ） ∝ Σ １Ｃｋ２Ｃｋ ×ｃｏｓ（πＭｋ＾２）×ｃｏｓ（２πｋｘ／ｐ）
ただし、１Ｃｋ：第１回折格子１の透過率のフーリエ係数２Ｃｋ：受光素子パターンの配置分布のフーリエ係数ｋ：整数、Ｍ：λＺ／Ｐ＾２、λ：光源の波長、Ｚ：第１格子１と受光部３との距離、Ｐ：第１格子１の格子ピッチ、ｘ：相対変位である。なお、受光素子パターンの配置分布は、長手方向の位置ｘに対する関数であって、例えば、受光素子パターンの受光する部分を１、受光しない部分を０とした場合の分布関数である。２Ｃｋは、この受光素子パターンの分布関数のフーリェ係数である。
上式から、この信号Ｉａは、１Ｃｋや２Ｃｋに比例することが示される。従って、信号Ｉａは、第１回折格子１の透過率１Ｃｋや、受光素子パターンのフーリエ係数２Ｃｋが偶数次の成分を含まない場合は、偶数次成分を含まず、ｋは奇数のみをとる。従来の構成では、２枚の回折格子の回折により信号に偶数次成分が含まれていたが、本発明の構成では、偶数次成分が含まれない信号を得ることができるという長所を有する。
【０００９】
ここで、信号Ｉａに含まれる奇数次の高調波を除去するには、例えば、３次高調波を除去するためには、上式より、１Ｃ３もしくは２Ｃ３が０となるようにすればよい。本発明においては、受光素子パターンの３次のフーリエ係数２Ｃ３が０となるように、本実施例の受光素子が、７Ｐ／３０及び１３Ｐ／３０の幅で構成されているが、この受光素子のパターンには、３次及び５次の歪成分が含まれない。このパターン幅は以下の方針で決定される。
２種の受光素子のパターン幅を２Ｌとすると、パターンの３次及び５次の歪は下記数２で示される。
【数２】
３次の歪＝２／π×１／３×ｓｉｎ（２π３Ｌ／Ｐ）×ｃｏｓ（３Ｘ）
５次の歪＝２／π×１／５×ｓｉｎ（２π５Ｌ／Ｐ）×ｃｏｓ（５Ｘ）
＝ −２／π×１／５×ｓｉｎ（２π５Ｌ／Ｐ＋π（１＋２ｎ′′））×ｃｏｓ（５Ｘ）
歪を消すには、パターン幅が２Ｌの受光素子による歪成分と、パターン幅が２Ｌ′の受光素子による歪成分との和が”０”となればよいので、下式となる。
【数３】
ｓｉｎ（２π３Ｌ／Ｐ）＋ｓｉｎ（２π３Ｌ′／Ｐ＋２πｎ′）＝０
ｓｉｎ（２π５Ｌ／Ｐ）− ｓｉｎ（２π５Ｌ／Ｐ＋π（１＋２ｎ′′））＝０
ただし、ｎ′、ｎ′′は整数。
これより、パターン幅を計算すると、以下のようになる。
【数４】
２Ｌ＝ｐ（ｎ／３＋（１＋２ｎ′）／１０）
２Ｌ′ ＝ｐ（ｎ／３−（１＋２ｎ′）／１０）
結局、上記関係式より求めた２種のパターン幅の受光素子を用いた構成により、３次と５次の２つの歪の除去が可能である。なお、受光素子パターン幅２Ｌ及び２Ｌ′の値は、ｎとｎ′を変えていくと無限に得られるが、受光素子パターン幅２Ｌ及び２Ｌ′に採択可能な値を用いればよい。図３にその例を示す。
これにより例えば受光素子群３ａから得られる信号は、３次と５次の歪成分を含まない。３ｂ、３ａ／、３ｂ／についても、同様である。
受光素子３ｂ、３ａ／、３ｂ／からは数５〜数７の信号Ｉｂ、Ｉａ／、Ｉｂ／が得られる。
【００１０】
【数５】
Ｉｂ（ｘ）＝Ｉａ（ｘーπ／２）
【００１１】
【数６】
Ｉａ／（ｘ）＝Ｉａ（ｘーπ）
【００１２】
【数７】
Ｉｂ／（ｘ）＝Ｉａ（ｘー３π／２）
したがって、受光素子群３ａ、３ｂ、３ａ／、３ｂ／から、各々位相が９０゜異なった信号が得られる。
【００１３】
ここで、各相の受光素子は、ほとんど同一場所に混入されて置かれているので、照射される光強度は各相に均等であり、受光素子の特性も揃っており、第１回折格子１の汚れやキズなどの影響を受けにくく安定に高精度な位置検出が可能となる。またＩａ、とＩａ／、ならびにＩｂとＩｂ／の差動をとるように受光素子どうしを接続してもよく、その場合、従来の構成では各相信号のアンバランスにより除去しきれなかったＤＣ成分が相殺され、精度よくＡＣ成分のみを得ることができる。隣り合った受光素子は、その中心の間隔が３Ｐ／４となっているが、５Ｐ／４やその他の間隔でもよい。また、信号は、２相を得るようにしてもよいし、１２０ｄｅｇ等の位相差の３相信号を得るようにしてもよい。
【００１４】
なお、上記数４を一般化して、ａ次及びｂ次の歪成分を除去する場合には、下式で示されるパターン幅とすればよい。
【数８】
２Ｌ＝ｐ（ｎ／ａ＋（１＋２ｎ′）／２ｂ）
２Ｌ′ ＝ｐ（ｎ／ａ−（１＋２ｎ′）／２ｂ）
【００１５】
上記関係式より、２種のパターン幅の受光素子を用いた構成により、任意の２つの歪の除去が可能である。なお、受光素子パターン幅２Ｌ及び２Ｌ′の値は、ｎとｎ′を変えていくと無限に得られるが、受光素子パターン幅２Ｌ及び２Ｌ′に採択可能な値を用いればよい。一般的には受光素子パターン幅を２Ｐまでに限定すればよい。あまり大きな受光素子パターン幅を選択すると、受光素子の本数の密度が低下して平均化効果が劣るためである。
図３に示す例の場合、その中から、さらにＰ以下の受光素子パターン幅のみを選択すると、１７Ｐ／３０及び２３Ｐ／３０、１１Ｐ／３０及び２９Ｐ／３０、７Ｐ／３０及び１３Ｐ／３０、Ｐ／３０及び１９Ｐ／３０の４種の組合せが存在する。図２に示した例は、１３Ｐ／３０及び７Ｐ／３０を用いているが、この他にも、これらパターン幅の受光素子を設ければよい。
また、３次、５次以外の歪成分を除去する場合も同様にして、上記数８のａ，ｂにその次数を当てはめて得られる幅のパターンを設ければよい。これによれば、例えば２次及び３次、３次及び７次、５次及び１１次など、任意の２つの次数の歪成分を２個の受光素子により同時に除去することが可能である。従来の技術では、２つの歪成分を除去する場合には、４個の受光素子を必要とするので、本発明の効果は大きい。
【００１６】
続いて、第２実施形態について説明する。本実施形態の光学式エンコーダは、第１実施形態の受光部３を図４に示す受光部３０に入れ替えたものである。
本実施例では、受光部３０に含まれる受光素子のパターンの幅が１７Ｐ／３０及び２３Ｐ／３０の受光素子と、７Ｐ／３０及び１３Ｐ／３０の受光素子とで構成されている。この受光素子パターン幅は、上述の４種の組合せのうちの２種を選んで含ませたものである。この受光素子パターンにおいては、受光素子の平均幅がＰ／２となっている。この場合、受光素子パターンのフーリエ係数２Ｃｋが、偶数次成分を含まないという長所を有する。図２の実施例に比べて、受光素子の面積率が高く、受光効率の向上がはかれる。
【００１７】
また、図示しないが、図４に示す受光部３０の代わりに、受光素子のパターンの幅が１７Ｐ／３０及び２３Ｐ／３０の受光素子と、１１Ｐ／３０及び２９Ｐ／３０の２組の受光素子とを用いると３次及び５次の歪成分が含まれない上に、図２に示すような１組の幅の受光素子の組合せの場合よりも７次の歪成分も低下させることができる。従って、この受光素子で受光される回折光における７次の回折光強度は減少するので、この受光素子より構成される受光部３０を光学式エンコーダに用いた場合、受光部３０で得られる受光信号に含まれる７次の歪成分も低下させることができる。さらに、受光部３０に含まれる受光素子のパターンの幅が１７Ｐ／３０及び２３Ｐ／３０の受光素子の組と、１１Ｐ／３０及び２９Ｐ／３０の受光素子の格子の組が、２：１の割合の数で設けるとさらに、７次の歪成分も低下させることができる。これらの組み合わせは、パターンのフーリエ係数を数値計算して選択された。なお、このような効果が得られれば、他のパターン幅の受光素子の組合せを用いてもよいし、他の次数（例えば１１次等）の歪成分を低下させる受光素子の組合せを用いてもよい。
【００１８】
上述の図４の例を更に改善し第３実施形態について、図５を参照して説明する。
図４の実施例は、受光素子の中心の間隔が全て３Ｐ／４である。これに対して図５の実施例では、図４と同様にパターン幅が１７Ｐ／３０及び２３Ｐ／３０の受光素子と、７Ｐ／３０及び１３Ｐ／３０の受光素子とで構成されているが、そのパターンの配置される間隔が一定ではない。
受光素子３３ａ−１、３４ａ−１は、図中の受光素子３１ａ−１、３２ａ−１が配置されている位置からＰの整数倍の位置に対して、Ｐ／１４だけシフトして配置されている。つまり、受光素子３１ａ−１、３２ａ−１の群に対して、受光素子３３ａ−１、３４ａ−１の群とが、位相差Ｐ／１４を有している。この位相差Ｐ／１４は、受光素子パターン３１ａ−１、３２ａ−１の７次の歪成分と、受光素子パターン３３ａ−１、３４ａ−１の７次の歪成分の位相が逆相となって相殺されるように受光素子パターンを配置している。そして、受光素子の幅による効果と合わせて、３次、５次、７次の各歪成分が除去されたａ相信号が得られる。ｂ，ａ／，ｂ／についても同様である。なお、７次の歪成分に限らず、任意のＣ次の歪成分が除去可能であり、そのためには受光素子３１ａ−１、３２ａ−１の群と受光素子３３ａ−１、３４ａ−１の群との間にＰ／（２・Ｃ）だけの位相差を持たせればよい。また、受光素子の左右の配置順序などは本実施例に限定されるものではなく、これらの組み合わせを含んでいればどのように行ってもよい。また、受光素子の幅も前記数８を満足すればよく、任意の２つの次数の歪成分を同時に除去することができる。
【００１９】
さらに図５の実施例を改善した第４実施形態について、図６を参照して説明する。
図６の実施例では、受光素子間の間隔がさらに異なっている。受光素子３１ａ−１、３２ａ−１に対して、受光素子３３ａ−１、３４ａ−１は、位相差Ｐ／１４を有している。同様に、受光素子３１ａ−２、３２ａ−２と受光素子３３ａ−２、３４ａ−２も、位相差Ｐ／１４を有しているが、さらに、これらの受光素子３１ａ−２、３２ａ−２、３３ａ−２、３４ａ−２の群が、受光素子３１ａ−１、３２ａ−１、３３ａ−１、３４ａ−１の群に対して１２Ｐ＋Ｐ／２２だけ隔てて配置されている。この位相差Ｐ／２２は、受光素子パターン３１ａ−１、３２ａ−１、３３ａ−１、３４ａ−１の１１次の歪成分と、受光素子パターン３１ａ−２、３２ａ−２、３３ａ−２、３４ａ−２の１１次の歪成分の位相が逆相となって相殺されるように受光素子パターンを配置している。
これによって、受光素子の幅によって３次と５次（ａ＝３、ｂ＝５）の歪成分が、受光素子の間隔によって７次と１１次（ｃ＝７、ｄ＝１１）の歪成分が除去された信号が得られる。ここで、除去する歪成分は、これに限らず、ａ次、ｂ次、ｃ次、ｄ次には任意の次数が選択できる。
【００２０】
さらに、１３次の歪成分を除去する場合は、ｅ＝１３として、Ｐ／（２・１３）の位相差、つまり、上述の受光素子３１ａ−１、３２ａ−１、３３ａ−１、３４ａ−１、３１ａ−２、３２ａ−２、３３ａ−２、３４ａ−２の群に対して、Ｐ／２６だけ隔てて受光素子群を配置すればよい。さらに、１７次や２３次の歪成分を除去する場合は、同様に、ｆ＝１７、ｇ＝２１として、Ｐ／（２・１７）や、Ｐ／（２・２３）の位相差をもって受光素子群を配置すればよい。。なお、これらの場合、９次や１５次、２１次の歪成分を除去するためのパターンを設ける必要がない。これは、９次や１５次、２１次が３次を因数とする歪成分であり、３次の歪成分を除去するためのパターンで除去されるからである。
【００２１】
また、本発明の構成では、信号には、偶数次の歪成分は含まれない。数４からわかるように信号の歪成分は、パターンのフーリエ係数１Ｃｋ、２Ｃｋに比例する。上述の例では、フーリエ係数１Ｃｋ、２Ｃｋとも偶数次成分を含んでおらず、信号にも偶数次の歪成分は含まれない。
利得の減衰も無い状態で偶数次の成分が全く含まれないため、逆相信号との差分を行なうまでもなく望ましい信号が得られ、受光素子パターンによる歪成分除去との相乗効果が得られる。従って、上述のようにａ＝３、ｂ＝５、ｃ＝７、ｄ＝１１、ｅ＝１３として除去すれば、２次から１５次までの成分がすべて除去された信号が得られる。また、それ以上の次数の偶数次成分や、ａ、ｂ、ｃ、ｄ、ｅ次の倍数の奇数次成分も除去される。これは、本発明のように、第１回折格子１を透過した光を直接格子状の受光素子で受光する場合の大きな長所である。
【００２２】
ここで、パターンは、できるだけパターン間のスペースが均等に近くなるように設計するとよい。これにより、隣接するパターンの間隔がいずれの場所でも均等に近く一定以上に保てるため、受光素子の製作が簡易になると共に、受光素子間のクロストークが減少するなど光学特性も向上し、その効果は大きい。図６のパターン周期は、正確には１相につき８個のパターン配置で１周期であるが、全体の配置は１周期分でも良いし、２周期以上でもよい。
また、受光部３は、耐環境性のために樹脂モールドやキャンパッケージ化してもよいし、セラミックやメタルなどのパッケージに格納して、受光面をガラスや樹脂などで保護してもよい。
【００２３】
また、これまでの実施形態では、変位信号のひずみ成分を除去する方法として、受光部３の幅や周期を変調する例を示したが、これを第１回折格子に設けてもよい。その際、第１回折格子は、振幅変調格子でも、位相格子でもよい。また、これまで、パターン部の幅や周期を変調する例を示したが、非パターンの幅や周期を変調してもよい。変調は、受光素子の形状やその他の方式で行なってもよい。受光素子パターンの変調による高次ひずみ成分の除去は、本実施例で説明した次数以外の歪成分を除去するようにしてもよく、３、５、７、１１、１３次等の奇数次ひずみ成分に対して複数の組み合わせで行なってもよいしある次数に対してのみ行なってもよいし、２次や偶数次のひずみ成分に対して行なってもよい。偶数次の成分については、変位信号の位相が１８０ｄｅｇことなる信号を生成し、その反転信号の差をとることで除去できるので、このように変位信号レベルで除去してもよい。第１回折格子１は反射型の回折格子を用いてもよい。発光される光束は、レーザダイオード等のコヒーレントなものでもよいし、ＬＥＤなどのインコヒーレントなものを用いてもよい。この光束は、平行光束でも、非平行でもよい。
また、本発明は、直線式でも回転式のエンコーダでも用いることができる。円筒部材の周囲に位相格子を設けたトルボット干渉を利用したエンコーダにも用いることができる。第１の格子のピッチと受光部の受光素子の平均的なピッチはほぼ同じでも、１：２等の異なったピッチでも本発明は適用可能であり、本発明は上記実施形態に限定されるものではない。
【００２４】
【発明の効果】
以上のように本発明によれば、歪成分が除去された精密な信号を、少ない格子状の受光素子により得ることができるので、より平均化効果の高い位置検出を行なうことができる。また、第１回折格子を透過した光を直接格子状の受光素子で受光するすることで偶数次の歪成分を含まないため、さらに相乗的に歪のない信号を得ることができる。さらに、各相用の受光素子が、光電変換素子上に格子状に混在して形成されているので平均化効果もより相乗的に向上し、スケールの汚れやキズや誤差の影響、光源の平行性や強度のばらつきがあっても、精度の高い検出が可能となる。従って、高精度の位置検出を安定して行うことができるので、精度の高い加工を容易に行うことが可能となり、生産効率の向上を図ることができる。
【図面の簡単な説明】
【図１】本発明の光学式エンコーダの実施例を示す斜視構造図である。
【図２】図１に示す実施例の受光部の第１の例を示す図である。
【図３】図１の実施例の格子部の他の例のパターンの格子パターン幅に関するデータ例を示す図である。
【図４】図１に示す実施例の受光部の第２の例を示す図である。
【図５】図１に示す実施例の受光部の第３の例を示す図である。
【図６】図１に示す実施例の受光部の第４の例を示す図である。
【図７】従来の光学式エンコーダの回折格子の一例を示す図である。
【図８】従来の光学式エンコーダの第２格子の一例を示す図である。
【符号の説明】
１第１回折格子
２第２回折格子
３受光部[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an optical encoder used for position measurement of a machine tool or a semiconductor manufacturing device.
[0002]
[Prior art]
The optical encoder has a light emitting unit that emits a light beam and a photoelectric conversion element disposed behind the two diffraction gratings, detects light transmitted through the two diffraction gratings by the photoelectric conversion element, and is generated by the relative movement of the two diffraction gratings. The movement amount is detected based on a change in light intensity.
FIG. 7 shows an example of the two diffraction gratings 20. The diffraction grating 20 has a so-called amplitude modulation type grating in which transmitting portions that transmit light and non-transmitting portions that do not transmit light are alternately arranged. This arrangement pitch is called a grating pitch, and the width of each of the transmitting portion and the non-transmitting portion is half of the grating pitch P.
In the encoder using the two diffraction gratings 20, when the transmission portions match, the amount of transmitted light becomes maximum, and the output detected by the photoelectric conversion element also becomes maximum. Further, when the position of one transmissive part and the position of the other non-transmissive part coincide, the amount of transmitted light becomes minimum and the output detected by the photoelectric conversion element also becomes minimum. The electric signal output from the photoelectric conversion element increases or decreases between the maximum value and the minimum value with respect to the relative displacement of the two diffraction gratings 20. When the two diffraction gratings 20 are relatively displaced at a constant speed, the output signal obtained by the photoelectric conversion element is ideally a triangular wave signal having a period P. However, actually, the triangular wave shape is distorted due to the influence of diffraction. This function was treated as a sine wave to detect relative displacement.
[0003]
[Problems to be solved by the invention]
The optical encoder using the diffraction grating 20 determines the relative position by treating the output signal as a sine wave. However, this output signal contains various distortion components. Therefore, when position data finer than the lattice pitch is obtained, a large error is included in the position detection value due to the amount of deviation from the sine wave due to the distortion components. Was. An error caused by a deviation between the actual output signal and the sine wave is called a division error.
Further, the distortion rate of the displacement signal obtained by the above-described conventional optical encoder greatly changes when the distance between the first diffraction grating and the second diffraction grating changes. Therefore, in order to keep the error within a certain value, it is necessary to keep the interval between the first diffraction grating and the second diffraction grating constant at an appropriate interval, and there is a problem that extremely strict mounting accuracy is required. Was.
Japanese Patent Application Laid-Open No. 3-48122 discloses that, in order to remove a distortion component of an output signal, a diffraction grating has a third order by arranging adjacent transmission portions with unequal intervals and disposing the transmission portions with a predetermined phase difference. An optical encoder that removes the fifth and fifth harmonic components is disclosed. However, in such a configuration, for example, when removing the third and fifth harmonic distortion components, at least four slits are required. When the number of slits is sufficiently large, a sufficient averaging effect is exhibited even when the irradiation light beam is non-uniform, but when the number of slits is small, for example, when the number of slits is about four to several, the averaging effect is obtained. And the ability to remove the third and fifth harmonic distortion components is reduced.
[0004]
Further, in the optical encoder having the conventional configuration, the second diffraction grating 21 as shown in FIG. 8 is used, and when detecting the position, the detection positions of the respective phases are distant, and the detection of the respective phases is different. Since the measurement is performed on the first grating at the place where the signals are located, there is a problem that imbalance occurs in each signal due to the influence of dirt, flaws and errors on the scale, which results in errors. Further, there is a problem that the signals of the respective phases also vary due to variations in the parallelism and intensity of the light source.
Further, in order to prevent crosstalk of light of each phase, it is necessary to provide a space between the second grating and each phase of the light receiving element of the light receiving section 30, which has led to an increase in the size of the device.
[0005]
The present invention has been made in view of the above circumstances, and an object of the present invention is to output a displacement signal with less distortion stably with a sufficient averaging effect, and to detect each phase as much as possible. It is to provide an optical encoder that can be performed on site.
[0006]
[Means for Solving the Problems]
In order to solve the above-described problem, an optical encoder according to the present invention includes a scale, and a light receiving element that is displaced relative to the scale, based on a signal having a predetermined phase difference from the light receiving element. An optical encoder for detecting the relative displacement, wherein a plurality of light receiving elements corresponding to a pattern width and a pattern period in a longitudinal direction of the scale are arranged in parallel in the longitudinal direction; A light receiving element group corresponding to each of the phase signals is provided, and light receiving elements belonging to one light receiving element group and another light receiving element group are arranged in a mixed manner in the longitudinal direction, and one light receiving element group and another The light receiving element group is arranged such that a signal obtained by one light receiving element group and a signal obtained by another light receiving element group have a predetermined phase difference,The light-receiving element, when the higher-order Fourier components to be removed are a-order and b-order, the width of the light-receiving element is represented by the following equation,
Light receiving element pattern width W = P × (n / a ± (1 + 2m) / 2b)
However, n: -∞ ~ ∞ m: -∞ ~ ∞
The light-receiving element group corresponding to one phase signal has a light-receiving element width included in the light-receiving element pattern arrangement distribution,a-order and b-orderDetermined to not include higher-order Fourier componentsOr, a scale, comprising a light receiving element that is relatively displaced from the scale, based on a signal having a phase difference of a predetermined phase from the light receiving element, in an optical encoder that detects the relative displacement, A light receiving element group in which a plurality of light receiving elements corresponding to a pattern width and a pattern period in the longitudinal direction of the scale are arranged in parallel in the longitudinal direction, and a light receiving element group corresponding to each of a plurality of the signals of the predetermined phase. A light-receiving element belonging to one light-receiving element group and another light-receiving element group are arranged in a mixed manner in the longitudinal direction, and one light-receiving element group and another light-receiving element group are one light-receiving element group. The signal obtained in the above and the signal obtained in the other light receiving element group are arranged so as to have a predetermined phase difference, and the light receiving element group corresponding to one phase signal is included therein. The width of the light-receiving element is determined so that the arrangement distribution of the light-receiving element pattern does not include a higher-order Fourier component, and the light-receiving element has a higher-order Fourier component to be removed and an a-order and a b-order. The width of the light receiving element is represented by the following equation,
Light receiving element pattern width W = P × (n / a ± (1 + 2m) / 2b)
However, n: -∞ ~ ∞ m: -∞ ~ ∞
When the higher-order Fourier components to be further removed are the c-order and the d-order, the light-receiving element group corresponding to one phase signal has an irregular distance between the light-receiving elements included therein, and is based on one light-receiving element. , P / (2 · c), P / (2 · d), and P / () in addition to k × P (k is a positive integer, P is a signal period). 2 · c) + P / (2 · d)Things.
[0007]
[Action]
According to this configuration, higher-order distortion components are removed depending on the width of the light receiving element, and a displacement signal with less distortion can be obtained with a small number of light receiving elements. It can output stably with an averaging effect. In addition, since the light receiving elements for each phase are formed in a lattice on the photoelectric conversion elements, the detection of each phase can be performed at substantially the same place. Therefore, even if the scale is dirty or flawed, there is an advantage that the influence on each phase is the same, each output signal is not unbalanced, and no error is caused thereby. They also have the advantage that they are similarly unaffected by light source parallelism and intensity variations and scale errors.
[0008]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, preferred embodiments of the present invention (hereinafter, referred to as embodiments) will be described with reference to the drawings.
FIG. 1 is a perspective structural view of an encoder according to a first embodiment of the present invention. The encoder according to the present embodiment relatively moves with respect to a first diffraction grating 1 which first receives light from a light source in a direction indicated by an arrow. It has a light receiving section 3 including a light receiving element or the like that receives light transmitted through the first diffraction grating 1 and outputs an electric signal corresponding to the light amount. The light receiving elements of the light receiving section 3 are provided in a lattice shape as shown in FIG. The light receiving elements 30 of the light receiving section 3 shown in FIG. 2 are arranged at intervals of several μm to several hundred μm on average. This arrangement pattern is used for an optical encoder that outputs four-phase signals a, b, a /, b / (0 °, 90 °, 180 °, 270 °). 3b, 3a /, and 3b /. Further, these light receiving element groups are divided into 31a and 32a, 31b and 32b, 31a / and 32a /, 31b / and 32b /.
In this embodiment, assuming that the pitch of the first grating 1 is P, since the signals of the respective light receiving element groups have a predetermined phase difference, the center interval is 3P / 4. . The width of the light receiving element 31a is 13P / 30, and the width of the light receiving element 32b is 7P / 30. When a certain light receiving element is a light receiving element 31a, the next light receiving element is a light receiving element 32b for a signal, the next light receiving element is a / light receiving element 31a / for a signal, and the next light receiving element is b / Light receiving element 32b / for signals, and light receiving elements 32a, 31b, 32a /, 31b / are arranged adjacent thereto. The arrangement pattern repeats these as a cycle.
As shown in FIG. 1, when the light transmitted through the first diffraction grating 1 is directly received by the light receiving element in the form of a lattice, for example, the signal Ia obtained by the light receiving element 3a is It is shown by Equation 1.
(Equation 1)
Ia (x) ∝ Ｃ 1Ck 2Ck × cos (πMk ＾ 2) × cos (2πkx / p)
Here, 1Ck: Fourier coefficient of transmittance of the first diffraction grating 1 2Ck: Fourier coefficient of arrangement distribution of the light receiving element pattern k: integer, M: λZ / P ＾ 2, λ: light source wavelength, Z: first grating 1 , P: the grating pitch of the first grating 1, x: relative displacement. The arrangement distribution of the light receiving element patterns is a function with respect to the position x in the longitudinal direction, and is, for example, a distribution function when a light receiving part of the light receiving element pattern is 1 and a light non-receiving part is 0. 2Ck is a Fourier coefficient of the distribution function of the light receiving element pattern.
The above equation shows that this signal Ia is proportional to 1Ck or 2Ck. Therefore, when the transmittance 1Ck of the first diffraction grating 1 and the Fourier coefficient 2Ck of the light receiving element pattern do not include an even-order component, the signal Ia does not include an even-order component, and k takes only an odd number. In the conventional configuration, an even-order component is included in a signal due to diffraction of two diffraction gratings. However, the configuration of the present invention has an advantage that a signal that does not include an even-order component can be obtained.
[0009]
Here, in order to remove odd-order harmonics contained in the signal Ia, for example, to remove third-order harmonics, 1C3 or 2C3 may be set to 0 according to the above equation. In the present invention, the light receiving element of this embodiment is configured with a width of 7P / 30 and 13P / 30 so that the third-order Fourier coefficient 2C3 of the light receiving element pattern becomes 0. The pattern does not include third-order and fifth-order distortion components. This pattern width is determined by the following policy.
Assuming that the pattern width of the two types of light receiving elements is 2L, the third and fifth order distortions of the pattern are expressed by the following equation (2).
(Equation 2)
Third-order distortion = 2 / π × 1/3 × sin (2π3L / P) × cos (3X)
Fifth-order distortion = 2 / π × 1/5 × sin (2π5L / P) × cos (5X)
= −2 / π × 1/5 × sin (2π5L / P + π (1 + 2n ″)) × cos (5X)
In order to eliminate the distortion, the sum of the distortion component due to the light receiving element having the pattern width of 2L and the distortion component due to the light receiving element having the pattern width of 2L 'should be "0", and the following equation is obtained.
(Equation 3)
sin (2π3L / P) + sin (2π3L ′ / P + 2πn ′) = 0
sin (2π5L / P) −sin (2π5L / P + π (1 + 2n ″)) = 0
However, n 'and n' 'are integers.
From this, the pattern width is calculated as follows.
(Equation 4)
2L = p (n / 3 + (1 + 2n ') / 10)
2L '= p (n / 3- (1 + 2n') / 10)
After all, with the configuration using the light receiving elements of two types of pattern widths obtained from the above relational expression, it is possible to remove the third and fifth order distortions. Although the values of the light receiving element pattern widths 2L and 2L 'can be obtained indefinitely by changing n and n', values that can be adopted as the light receiving element pattern widths 2L and 2L 'may be used. FIG. 3 shows an example.
Thus, for example, a signal obtained from the light receiving element group 3a does not include third and fifth order distortion components. The same applies to 3b, 3a /, 3b /.
From the light receiving elements 3b, 3a /, 3b /, signals Ib, Ia /, Ib / of Equations 5 to 7 are obtained.
[0010]
(Equation 5)
Ib (x) = Ia (x−π / 2)
[0011]
(Equation 6)
Ia / (x) = Ia (x-π)
[0012]
(Equation 7)
Ib / (x) = Ia (x−3π / 2)
Therefore, signals having phases different by 90 ° are obtained from the light receiving element groups 3a, 3b, 3a /, 3b /.
[0013]
Here, since the light receiving elements of each phase are mixed and placed in almost the same place, the light intensity to be irradiated is uniform for each phase, and the characteristics of the light receiving elements are also uniform. This makes it difficult to be affected by dirt, scratches, and the like, and enables stable and accurate position detection. Further, the light receiving elements may be connected to each other so as to obtain the differential between Ia and Ia / and between Ib and Ib /. In this case, the DC component which cannot be completely removed due to the imbalance of each phase signal in the conventional configuration. Are canceled, and only the AC component can be obtained with high accuracy. Adjacent light receiving elements have a center spacing of 3P / 4, but may have a spacing of 5P / 4 or other spacing. The signal may be obtained in two phases, or a three-phase signal having a phase difference of 120 deg or the like may be obtained.
[0014]
Note that when generalizing the above equation 4 to remove the a-order and b-order distortion components, the pattern width may be set to the pattern width represented by the following equation.
(Equation 8)
2L = p (n / a + (1 + 2n ') / 2b)
2L '= p (n / a- (1 + 2n') / 2b)
[0015]
According to the above relational expression, it is possible to remove any two distortions by using a configuration using light receiving elements having two types of pattern widths. Although the values of the light receiving element pattern widths 2L and 2L 'can be obtained indefinitely by changing n and n', values that can be adopted as the light receiving element pattern widths 2L and 2L 'may be used. Generally, the light receiving element pattern width may be limited to 2P. This is because if an excessively large light-receiving element pattern width is selected, the density of the number of light-receiving elements is reduced and the averaging effect is poor.
In the case of the example shown in FIG. 3, if only the light receiving element pattern width smaller than P is selected from among them, 17P / 30 and 23P / 30, 11P / 30 and 29P / 30, 7P / 30 and 13P / 30, P There are four combinations: / 30 and 19P / 30. In the example shown in FIG. 2, 13P / 30 and 7P / 30 are used, but other light receiving elements having these pattern widths may be provided.
Similarly, when removing distortion components other than the third and fifth orders, a pattern having a width obtained by applying the order to a and b in Equation 8 above may be provided. According to this, it is possible to simultaneously remove distortion components of any two orders, such as the second, third, third, seventh, fifth, and eleventh order, by using two light receiving elements. In the related art, when two distortion components are removed, four light receiving elements are required, so that the effect of the present invention is great.
[0016]
Subsequently, a second embodiment will be described. The optical encoder of the present embodiment is obtained by replacing the light receiving unit 3 of the first embodiment with a light receiving unit 30 shown in FIG.
In the present embodiment, the pattern of the light receiving element included in the light receiving section 30 is constituted by the light receiving elements of 17P / 30 and 23P / 30 and the light receiving elements of 7P / 30 and 13P / 30. The light receiving element pattern width is obtained by selecting and including two of the above four combinations. In this light receiving element pattern, the average width of the light receiving elements is P / 2. In this case, there is an advantage that the Fourier coefficient 2Ck of the light receiving element pattern does not include an even-order component. As compared with the embodiment of FIG. 2, the area ratio of the light receiving element is higher, and the light receiving efficiency is improved.
[0017]
Although not shown, instead of the light receiving unit 30 shown in FIG. 4, two sets of light receiving elements having a pattern width of 17P / 30 and 23P / 30, and 11P / 30 and 29P / 30 are used. Is used, the third-order and fifth-order distortion components are not included, and the seventh-order distortion component can be reduced as compared with the case of a combination of light receiving elements having a single width as shown in FIG. Accordingly, the intensity of the 7th-order diffracted light in the diffracted light received by the light receiving element decreases. When the light receiving section 30 constituted by the light receiving element is used for an optical encoder, the light receiving signal obtained by the light receiving section 30 is obtained. Can also reduce the seventh-order distortion component. Further, the light receiving element pattern included in the light receiving portion 30 has a pattern of light receiving elements having a pattern width of 17P / 30 and 23P / 30 and a light receiving element having a grid pattern of light receiving elements of 11P / 30 and 29P / 30 in a ratio of 2: 1. When the number is set to 7, the seventh-order distortion component can be further reduced. These combinations were selected by numerically calculating the Fourier coefficients of the pattern. If such an effect is obtained, a combination of light receiving elements having other pattern widths may be used, or a combination of light receiving elements that reduce distortion components of other orders (for example, 11th order) may be used. Good.
[0018]
A third embodiment in which the above-described example of FIG. 4 is further improved will be described with reference to FIG.
In the embodiment of FIG. 4, the distance between the centers of the light receiving elements is 3P / 4. On the other hand, in the embodiment of FIG. 5, similarly to FIG. 4, the pattern width is constituted by the light receiving elements of 17P / 30 and 23P / 30 and the light receiving elements of 7P / 30 and 13P / 30. The intervals at which the patterns are arranged are not constant.
The light receiving elements 33a-1 and 34a-1 are arranged by being shifted by P / 14 from a position where the light receiving elements 31a-1 and 32a-1 are arranged in the figure to an integer multiple of P. I have. That is, the group of the light receiving elements 33a-1 and 34a-1 has a phase difference P / 14 with respect to the group of the light receiving elements 31a-1 and 32a-1. This phase difference P / 14 is such that the phase of the seventh-order distortion component of the light receiving element patterns 31a-1 and 32a-1 and the phase of the seventh-order distortion component of the light receiving element patterns 33a-1 and 34a-1 are opposite. The light receiving element patterns are arranged so as to cancel each other. In addition to the effect of the width of the light receiving element, an a-phase signal from which the third-, fifth-, and seventh-order distortion components are removed can be obtained. The same applies to b, a /, and b /. In addition, not only the seventh-order distortion component but also an arbitrary C-order distortion component can be removed. For this purpose, a group of light receiving elements 31a-1 and 32a-1 and a group of light receiving elements 33a-1 and 34a-1 can be removed. And a phase difference of P / (2 · C) only. The arrangement order of the light receiving elements on the left and right is not limited to the present embodiment, and any arrangement may be used as long as a combination of these is included. In addition, the width of the light receiving element only needs to satisfy Expression 8, and any two order distortion components can be removed at the same time.
[0019]
A fourth embodiment in which the example of FIG. 5 is further improved will be described with reference to FIG.
In the embodiment of FIG. 6, the distance between the light receiving elements is further different. The light receiving elements 33a-1 and 34a-1 have a phase difference P / 14 with respect to the light receiving elements 31a-1 and 32a-1. Similarly, the light receiving elements 31a-2, 32a-2 and the light receiving elements 33a-2, 34a-2 also have a phase difference P / 14, and furthermore, these light receiving elements 31a-2, 32a-2, The group of 33a-2 and 34a-2 is arranged at a distance of 12P + P / 22 from the group of light receiving elements 31a-1, 32a-1, 33a-1, and 34a-1. This phase difference P / 22 is obtained by calculating the eleventh-order distortion component of the light receiving element patterns 31a-1, 32a-1, 33a-1, and 34a-1, and the light receiving element patterns 31a-2, 32a-2, 33a-2, and 34a. The light-receiving element patterns are arranged such that the phases of the −11 second-order distortion components are reversed and canceled.
Thus, the third-order and fifth-order (a = 3, b = 5) distortion components depend on the width of the light-receiving element, and the seventh-order and eleventh-order (c = 7, d = 11) distortion components depend on the distance between the light-receiving elements. A removed signal is obtained. Here, the distortion component to be removed is not limited to this, and any order can be selected for the a-order, b-order, c-order, and d-order.
[0020]
Further, when removing the thirteenth-order distortion component, e = 13 and the phase difference of P / (2 · 13), that is, the above-described light receiving elements 31a-1, 32a-1, 33a-1, and 34a-1. , 31a-2, 32a-2, 33a-2, and 34a-2, the light receiving element group may be arranged at a distance of P / 26. Further, when removing the 17th-order and 23rd-order distortion components, similarly, f = 17 and g = 21, and the light receiving element has a phase difference of P / (2 · 17) or P / (2 · 23). What is necessary is just to arrange a group. . In these cases, it is not necessary to provide a pattern for removing the 9th, 15th, and 21st-order distortion components. This is because the ninth, fifteenth, and twenty-first orders are distortion components with the third order as a factor, and are removed by a pattern for removing the third-order distortion component.
[0021]
In the configuration of the present invention, the signal does not include even-order distortion components. As can be seen from Equation 4, the distortion component of the signal is proportional to the Fourier coefficients 1Ck and 2Ck of the pattern. In the above example, neither the Fourier coefficients 1Ck nor 2Ck include even-order components, and the signal does not include even-order distortion components.
Since even-order components are not included at all without attenuating the gain, a desired signal can be obtained without making a difference from the opposite-phase signal, and a synergistic effect with the removal of the distortion component by the light receiving element pattern can be obtained. Therefore, as described above, if a = 3, b = 5, c = 7, d = 11, and e = 13 are removed, a signal from which all the components from the second to the fifteenth are removed can be obtained. In addition, even-order components of higher orders and odd-order components of multiples of the a, b, c, d, and e orders are also removed. This is a great advantage when the light transmitted through the first diffraction grating 1 is directly received by the light receiving element in the form of a lattice as in the present invention.
[0022]
Here, the patterns may be designed so that the spaces between the patterns are as close as possible to each other. As a result, the distance between the adjacent patterns can be kept close to a certain value at any location, so that the manufacturing of the light receiving elements is simplified and the optical characteristics such as the reduction of the crosstalk between the light receiving elements are improved. Is big. Although the pattern period in FIG. 6 is exactly one period with eight pattern arrangements per phase, the entire arrangement may be one period or two or more periods.
The light receiving section 3 may be formed into a resin mold or can package for environmental resistance, or may be stored in a package such as ceramic or metal, and the light receiving surface may be protected with glass or resin.
[0023]
Further, in the embodiments described above, as an example of a method of removing the distortion component of the displacement signal, the example in which the width and the period of the light receiving unit 3 are modulated has been described, but this may be provided in the first diffraction grating. At this time, the first diffraction grating may be an amplitude modulation grating or a phase grating. Further, although an example in which the width and the cycle of the pattern portion are modulated has been described, the width and the cycle of the non-pattern may be modulated. The modulation may be performed by the shape of the light receiving element or another method. The removal of higher-order distortion components by modulating the light-receiving element pattern may be performed by removing distortion components other than the orders described in the present embodiment, such as odd-order distortion components such as 3, 5, 7, 11, and 13 orders. May be performed in a plurality of combinations, may be performed only for a certain order, or may be performed for a second-order or even-order distortion component. Even-order components can be removed by generating a signal in which the phase of the displacement signal is 180 deg, and taking the difference between the inverted signals, and thus may be removed at the displacement signal level. The first diffraction grating 1 may use a reflection type diffraction grating. The emitted light beam may be a coherent one such as a laser diode or an incoherent one such as an LED. This light beam may be a parallel light beam or non-parallel light beam.
Further, the present invention can be used with a linear encoder or a rotary encoder. The present invention can also be used for an encoder using a Talbot interference provided with a phase grating around a cylindrical member. The present invention can be applied to the case where the pitch of the first grating and the average pitch of the light receiving elements of the light receiving section are almost the same or different pitches such as 1: 2, and the present invention is limited to the above embodiment. is not.
[0024]
【The invention's effect】
As described above, according to the present invention, a precise signal from which a distortion component has been removed can be obtained with a small number of lattice-shaped light receiving elements, so that position detection with a higher averaging effect can be performed. In addition, since the light transmitted through the first diffraction grating is directly received by the light receiving element having a lattice shape, even-order distortion components are not included, so that a signal having a further synergistic distortion-free signal can be obtained. Furthermore, since the light receiving elements for each phase are formed in a lattice pattern on the photoelectric conversion elements, the averaging effect is also synergistically improved, and the effects of dirt, scratches and errors on the scale, and the parallelism of the light source are improved. Even if there is a variation in the properties and strength, highly accurate detection is possible. Therefore, high-accuracy position detection can be stably performed, so that high-accuracy processing can be easily performed, and production efficiency can be improved.
[Brief description of the drawings]
FIG. 1 is a perspective structural view showing an embodiment of an optical encoder according to the present invention.
FIG. 2 is a diagram showing a first example of a light receiving unit of the embodiment shown in FIG.
FIG. 3 is a diagram showing an example of data relating to a grid pattern width of another example of the grid portion of the embodiment of FIG. 1;
FIG. 4 is a diagram showing a second example of the light receiving unit of the embodiment shown in FIG.
FIG. 5 is a diagram showing a third example of the light receiving unit of the embodiment shown in FIG.
FIG. 6 is a diagram showing a fourth example of the light receiving unit of the embodiment shown in FIG.
FIG. 7 is a diagram illustrating an example of a diffraction grating of a conventional optical encoder.
FIG. 8 is a diagram illustrating an example of a second grating of a conventional optical encoder.
[Explanation of symbols]
1 First diffraction grating
2 Second diffraction grating
3 Receiver

Claims

A scale, comprising a light receiving element relatively displaced from the scale, based on a signal having a phase difference of a predetermined phase from the light receiving element, in an optical encoder for detecting the relative displacement, in the longitudinal direction of the scale A light receiving element group in which a plurality of light receiving elements corresponding to a pattern width and a pattern period are arranged in parallel in the longitudinal direction, the light receiving element group including a plurality of light receiving element groups corresponding to a plurality of the signals of the predetermined phase; A light receiving element belonging to a light receiving element group and another light receiving element group are arranged in a mixed manner in the longitudinal direction, and one light receiving element group and another light receiving element group are a signal obtained by one light receiving element group and another light receiving element group. of the and a signal obtained by the light receiving element group, are disposed so as to have a predetermined phase difference, the light receiving element, a high-order Fourier components to be removed when the a following and b following The width of the light receiving element is represented by a formula shown below, the light receiving element group corresponding to the signal of the one phase, the width of the light receiving elements contained therein, the location distribution of the light receiving element pattern, a next and b following An optical encoder characterized in that it is determined not to include a higher-order Fourier component.
Light receiving element pattern width W = P × (n / a ± (1 + 2m) / 2b)
However, n: -∞ ~ ∞ m: -∞ ~ ∞

A scale, comprising a light receiving element relatively displaced from the scale, based on a signal having a phase difference of a predetermined phase from the light receiving element, in an optical encoder for detecting the relative displacement, in the longitudinal direction of the scale A light receiving element group in which a plurality of light receiving elements corresponding to a pattern width and a pattern period are arranged in parallel in the longitudinal direction, the light receiving element group including a plurality of light receiving element groups corresponding to a plurality of the signals of the predetermined phase; A light receiving element belonging to a light receiving element group and another light receiving element group are arranged in a mixed manner in the longitudinal direction, and one light receiving element group and another light receiving element group are a signal obtained by one light receiving element group and another light receiving element group. And the signals obtained by the light receiving element group are arranged so as to have a predetermined phase difference, and the light receiving element group corresponding to one phase signal has a light receiving element width included therein. Serial placement distribution of the light receiving element pattern, are determined to be free of higher order Fourier components, the light receiving element, a high-order Fourier components to be removed when the a following and b following, the width of the light receiving element Is represented by the following equation, and when the higher-order Fourier components to be removed are c-order and d-order, the light-receiving element group corresponding to one phase signal has a constant interval between the light-receiving elements included therein. Instead, when one light-receiving element is used as a reference, the other light-receiving elements have P / (2 · c) and P / (p) in addition to k × P (k is a positive integer, P is a signal period). (2 · d) and P / (2 · c) + P / (2 · d) only the optical encoder you characterized by shifting to being disposed.
Light receiving element pattern width W = P × (n / a ± (1 + 2m) / 2b)
Where n: -∞ to ∞ m: -∞ to ∞