JP3841273B2 - Scanning probe calibration apparatus, calibration program, and calibration method - Google Patents

Description

となる。
【００４０】
そこで、倣いプローブ２３０のＸ軸測定範囲の全域に渡って、この誤差ｄｘ、ｄｙ、ｄｚを収集する。ここで、例えばピッチングの角度が決まれば、ピッチングＸ軸誤差とピッチングＺ軸誤差が一義的に決まることは言うまでもない。従って、変数としては、Ｘ軸方向の指示誤差Ｘ（ｘ）、Ｘ軸方向のＹ軸真直度誤差Ｘ（ｙ）、Ｘ軸方向のＺ軸真直度誤差Ｘ（ｚ）、ピッチング角度、ローリング角度、ヨーイング角度の６つがあるが、式が３つしかないので、このままでは解を求めることはできない。この解法としては、例えば線形多重回帰問題として、ＱＲ分解等を利用した最小自乗法を用いる。
Ｙ軸変位、Ｚ軸変位に対する誤差についても同様に求めて、各誤差要素を求める（Ｓ４０、Ｓ５０）。
【００４１】
次に、各軸の直交誤差を求める（Ｓ６０）。
図９は倣いプローブ２３０のセンサのＰＸ軸とＰＹ軸の直交誤差を求める方法を模式的に示したものである。ここで、図中のＸ、Ｙは校正測定装置２１０のＸ軸スケール２４４とＹ軸スケール２４５の軸方向を示し、両者間の直交誤差は許容範囲内に補正されている。ＰＸ、ＰＹは倣いプローブ２３０のＸ軸センサとＹ軸センサの軸方向を示しており、それぞれ、ＸとＹに対してθｘ、θｙの角度誤差を持っている。
まず、校正測定装置２１０を駆動して接触球２３３をＰ１点へ位置決めし、その時のＸ軸センサとＹ軸センサの出力を記憶する。次に接触球２３３をＰ２点へ位置決めし、その時のＸ軸センサとＹ軸センサの出力を記憶する。ここで、Ｐ１点からＰ２点までの長さは、正確にＬとなるようにＰ１点とＰ２点を選定する。この時、ＰＸ軸上でのＰ１点からＰ２点までの計数値をＡ、ＰＹ軸上でのＰ１点からＰ２点までの計数値をＢとする。
【００４２】
次に、校正測定装置２１０を駆動して接触球２３３をＰ３点へ位置決めし、その時のＸ軸センサとＹ軸センサの出力を記憶する。次に接触球２３３をＰ４点へ位置決めし、その時のＸ軸センサとＹ軸センサの出力を記憶する。ここで、Ｐ３点からＰ４点までの長さは、正確にＬとなるようにＰ３点とＰ４点を選定する。この時、ＰＸ軸上でのＰ３点からＰ４点までの計数値をＣ、ＰＹ軸上でのＰ３点からＰ４点までの計数値をＤとする。
この結果から、次の式を解いてθｘとθｙを得る。
Ｌ^２＝Ａ^２・cos^２θｘ＋Ｂ^２・cos^２θｙ …（１０）
Ｌ^２＝Ｃ^２・cos^２θｘ＋Ｄ^２・cos^２θｙ …（１１）
これによって、Ｘ軸センサとＹ軸センサの直交誤差を求めることが出来るが、同様にしてＹ軸センサとＺ軸センサの直交誤差およびＸ軸センサとＺ軸センサの直交誤差を求める。
【００４３】
なお、前記指示誤差の算出時には、この直交誤差の影響を加味しておらず、図５のＳ３０、Ｓ４０、Ｓ５０で求めた各軸の指示誤差には、この直交誤差成分が含まれている。そのため、指示誤差から直交誤差を分離しておく。より具体的には、補正前指示誤差成分から、前記直交誤差成分（例えばＸ軸の場合は、角度θｘによる一次相関成分）を除外して、指示誤差と直交誤差を分離する。
【００４４】
以上の測定、計算処理によって各種の誤差データを算出できたので、これらの誤差を容易に補正することができるように、補正データ算出用の関数を求める（Ｓ７０）。これは、倣いプローブ２３０の測定範囲全域に渡って誤差データを関数近似することで求めることが出来る。この結果、Ｘ軸、Ｙ軸、Ｚ軸用の補正データ算出関数はそれぞれ、Ｆｘ（ｘ、ｙ、ｚ）、Ｆｙ（ｘ、ｙ、ｚ）、Ｆｚ（ｘ、ｙ、ｚ）となる。つまり、一般的には、例えばＸ軸の補正データは、Ｘ軸のみならず、Ｙ軸とＺ軸の変位位置の影響を受ける。
【００４５】
この関数近似に当たって、必要に応じて指示誤差のみを除外して関数近似を行っても良い。
この場合、指示誤差は指示誤差データ列として、別途処理を行う目的で分離される。
これらの結果（近似された関数データ、指示誤差データ列）は、被校正倣いプローブの補正データとして出力する。すなわち、校正装置の表示装置に表示、印字装置で印字、あるいは出力装置を経由して三次元測定機に直接記憶させて格納したり、あるいはＩＣメモリやフレキシブルディスク等の記憶媒体に記憶させて、倣いプローブを使用する時点で三次元測定機に補正データ（関数データ）を転送する。また、倣いプローブ内に格納した不揮発性メモリなどに記憶して、三次元測定機にプローブを装着した時点で、補正データを倣いプローブから読み出すようにしても良い。このようにすれば、補正データ自体の取り扱いの間違いが生じにくく、倣いプローブのポータビリティが向上して使い勝手がよくなる。
【００４６】
次にプローブオフセットを求める（Ｓ８０）。
一般に倣いプローブは、被測定物の形状に応じた各種の測定子が用いられる。従って、原理的には、倣いプローブに応じた測定プログラム（パートプログラム）が作成される必要がある。同一測定物であってもプローブ毎に専用の測定プログラムを用意するということは実務的には極めて煩雑となる。この煩雑さを解消して、測定プログラムの共用化をはかるために、プローブオフセットの考え方が用いられる。
そのために三次元測定機は通常、基準点を有し、この基準点から倣いプローブの測定子の接触球の中心位置までのオフセット量を「プローブオフセット」として保持しておき、使用されるプローブが決定されると、該当のプローブオフセットが選択されて、この基準点の位置にプローブオフセットが加算されて、接触球の位置が算出される。
【００４７】
三次元測定機の基準点は、Ｚ軸スピンドルの下端中心位置に設けられることが多い。三次元測定機は、機械原点の位置とＸ軸、Ｙ軸、Ｚ軸の各スケールの値から、この基準点位置を知ることが出来るので、前述の通りプローブオフセットを加算すれば、接触球の中心座標を現在位置座標として算出することができ、これによって各軸の駆動制御を行うことが出来る。
プローブオフセットは、例えば図１４に示すような三次元測定機１００に該当する倣いプローブ１１８を装着し、三次元測定機のテーブルの一定位置に設置されている基準球１２０を測定して、その中心座標を求める。このようにして求められた基準球１２０の中心座標は、プローブオフセット分だけ座標値がシフトするので、このシフト量をプローブオフセットとすれば良い。このプローブオフセットは前述の補正データと同様に保存される。
これらによって補正データの算出を終了する（Ｓ９０）。
【００４８】
次に、これらの補正データ（関数データ）を用いて倣いプローブ２３０の出力を補正する方法を図１０のフローチャートによって説明する。
補正データは、倣いプローブが三次元測定機に装着された段階で、前もって誤差補正を行うデータ処理装置等に前記いずれかの方法によって格納しておき、Ｓ１００によって補正処理を開始する。
まず、倣いプローブ２３０のセンサ出力を入力する（Ｓ１１０）。倣いプローブ２３０から出力されたＸ軸センサ２５１、Ｙ軸センサ２５２、Ｚ軸センサ２５３の出力は、図２に示したカウンタ（Ｘ軸Ｐカウンタ、Ｙ軸Ｐカウンタ、Ｚ軸Ｐカウンタ）に類似のＰカウンタによって計数されので、測長センサの値としては、このＰカウンタの内容を入力することになる。また、倣いプローブが温度センサを内蔵している場合は、温度センサ出力も入力する。
【００４９】
次に、比例誤差補正の要否をチェックする（Ｓ１２０）。
比例誤差補正とは、測長センサの出力に対して比例係数的に補正を行うもので、ここでは温度補正と指示誤差の多点補正を行う。
通常、測長センサ（Ｘ軸センサ、Ｙ軸センサ、Ｚ軸センサ）は一定の線膨張係数を有し、温度によってセンサのスケール自体が伸縮する。そこで、前述の温度センサによって、測長センサの温度を測定することによってスケールの伸縮を補正する。
また、指示誤差の多点補正は、倣いプローブの測長センサの指示誤差を近似関数の形で与えて補正を行うのに代えて、複数位置での指示誤差データを基にして補正を行うもので、より高精度な補正が行える。この場合は、図５のＳ７０では、指示誤差の関数化を行わず、指示誤差データ列を生成しておく。指示誤差データ列は倣いプローブの測定範囲の全領域に渡って、一定距離毎の誤差データをデータ列としたものである。
【００５０】
これらの比例誤差補正が必要ない場合は、Ｓ１５０へ処理を移行する。
比例誤差補正を行う場合は、まず比例誤差を算出する（Ｓ１３０）。この場合の温度誤差算出は具体的には次のように行う。
倣いプローブの測長センサの線膨張係数をβ、温度センサの出力を摂氏Ｔｃ度とする。温度誤差は、通常、標準温度として摂氏２０度が用いられることから、この摂氏２０度に対する温度の差Ｔｗ＝Ｔｃ−２０を求め、これに線膨張係数βを乗算して求める。すなわち、一例として倣いプローブのＸ軸測長センサの出力をＸとした場合、温度誤差Ｘｔは次のように求める。
Ｘｔ＝Ｘ・β・（２０−Ｔｃ） …（１２）
Ｙ軸とＺ軸も同様に求める。
【００５１】
倣いプローブでは、各軸のスライド機構が近接して配置されているので、通常、温度センサは代表的な１ヶ所のみで測定すれば良いが、必要に応じて各軸毎に温度センサを個別に配置し、各軸毎に異なる温度を用いて温度誤差算出を行っても良い。また、倣いプローブの各軸の測長センサの線膨張係数が異なる場合は、軸毎の固有の線膨張計数を用いて、軸毎に温度誤差を求めても良い。
倣いプローブの測長センサの指示精度の補正のための誤差算出は次のように行う。
【００５２】
図１１は誤差データ列の様子を示したもので、この例では、Ｘ軸の全測長範囲に渡って、Ｘ１からＸ５の５点の誤差データ列から構成されている。それぞれの誤差値は、Ｅ１、Ｅ２、、、Ｅ５となっている。今、Ｘ軸センサ出力がＸｎで、Ｘ１を超え、かつＸ２以下である場合に、その誤差値Ｅｎは次のようにして算出する。
Ｅｎ＝Ｅ１・Ｘｎ・（Ｅ２−Ｅ１）／（Ｘ２−Ｘ１）…（１３）
つまり、多点補正は複数箇所の離散誤差データから、求める位置での誤差値を内挿によって求める。Ｙ軸とＺ軸も同様に求める。
なお、指示誤差が指示誤差データ列で与えられず、近似関数データの形で与えられる場合は、以下に述べるＳ１６０において指示誤差を関数誤差として算出する。
【００５３】
これらの比例誤差を求めた後、比例誤差を補正する（Ｓ１４０）。
すなわち、同様にＸ軸の場合は、温度誤差Ｘｔと指示誤差Ｅｎを測長センサの出力Ｘに加算して比例誤差補正結果Ｘｃを得る。Ｙ軸とＺ軸も同様に求める。
Ｘｃ＝Ｘ＋Ｘｔ＋Ｅｎ …（１４）
次に、関数誤差補正要否をチェックする（Ｓ１５０）。
ここで求める関数誤差には、回転誤差、真直度誤差、直交誤差があり、これらの誤差補正が不要な場合はＳ１８０へ移行して補正処理を終了する。
【００５４】
関数誤差補正が必要な場合は、まず、関数誤差を算出する（Ｓ１６０）。
関数Ｆｘ、Ｆｙ、Ｆｚは既に図５のＳ７０で与えられているので、これを用いて例えばＸ軸の関数誤差Ｘｆを次のように求める。Ｙ軸とＺ軸も同様に求める。
Ｘｆ＝Ｆｘ（Ｘｃ、Ｙｃ、Ｚｃ） …（１５）
その後、Ｘ軸の関数誤差補正結果Ｘｐを次のように求めて、倣いプローブのＸ軸測長センサの補正結果を得る。Ｙ軸とＺ軸も同様に求める。
Ｘｐ＝Ｘｃ＋Ｘｆ …（１６）
以上の処理によって補正処理を終了する（Ｓ１８０）。
実際にこの倣いプローブを使用して求める最終座標値は、これらの結果を用いて、さらに以下の計算処理によって算出する。
【００５５】
三次元測定機の駆動機構の変位を測定するリニヤスケールの値（Ｘｓ、Ｙｓ、Ｚｓ）は前述の通り、基準点（通常はＺ軸スピンドルの下端中心位置）位置を示しているので、これにプローブオフセット（Ｘ軸オフセットＳｘ、Ｙ軸オフセットＳｙ、Ｚ軸オフセットＳｚ）を加算し、さらに倣いプローブの測定子の変位位置を加算して、最終的に測定子の接触球の中心座標値Ｘｏ、Ｙｏ、Ｚｏを求める。
Ｘｏ＝Ｘｓ＋Ｓｘ＋Ｘｐ …（１７）
Ｙｏ＝Ｙｓ＋Ｓｙ＋Ｙｐ …（１８）
Ｚｏ＝Ｚｓ＋Ｓｚ＋Ｚｐ …（１９）
被測定物の測定にあたっては、三次元測定機の各軸の駆動機構によって倣いプローブを移動させ、被測定物の表面をトレースし、順次、接触球の中心座標値を求めることによって、被測定物の寸法や形状等を求めることが出来る。
【００５６】
次に図１２は、非接触倣いプローブ２９０を校正するときの校正装置２００の構成を示す。
図１と相違する点は、倣いプローブ２９０が非接触測定を行う点と連結体５００がスピンドル２１７のみに固定される点のみで、その他の構成は図１と同一である。
非接触倣いプローブ２９０としては、例えば画像光学系を含む焦点検出機能付撮像プローブなどがあり、非接触で測定対象の倣い測定が可能である。この非接触倣いプローブ２９０の本体２９１はプローブ支持体２２０に着脱可能に固定される。
【００５７】
連結体５００は図１３に示すように、スピンドル２１７にのみ固定され、十字ヘアーラインなどの測定目標５３０を備えている。非接触倣いプローブ２９０はこの測定目標５３０を倣い測定し、測長センサの座標値（Ｘ軸センサ、Ｙ軸センサ、Ｚ軸センサの値）を出力する。例えば、非接触倣いプローブ２９０が焦点検出機能付二次元ＣＣＤ撮像プローブの場合には、焦点検出結果がＺ軸センサの出力となり、二次元ＣＣＤで検出された測定目標５３０の測定位置（十字ヘアーラインの交差位置）がＸ軸センサ、Ｙ軸センサの出力となる。
【００５８】
この場合、図５と同様な校正を行えば、画像光学系の収差（真直度誤差、回転誤差）、二次元ＣＣＤの個別センサーの配列ピッチの誤差（指示誤差）、二次元ＣＣＤの個別センサーの直交誤差（直交誤差）、焦点検出機能の誤差（真直度誤差、回転誤差、指示誤差）などを校正することが出来るので、その結果を用いて、図１０の手順によって誤差補正を行うことが出来ることは言うまでもない。
【００５９】
以上、本発明について好適な実施形態を挙げて説明したが、本発明は、この実施形態に限られるものではなく、本発明の要旨を逸脱しない範囲での変更が可能である。
たとえば、図１の実施形態では、被校正倣いプローブ２３０の本体２３１をプローブ支持体２２０に着脱自在に固定したが、プローブ支持体２２０に代えてほぼ同一形状の基準体２２１を備え、スピンドル２１７に被校正倣いプローブ２３０の本体２３１を着脱自在に固定し、その測定子２３２と基準体２２１を連結体３００で結合しても良い。要するに被校正倣いプローブの測定子を正確に相対真直運動させることが出来ればいずれでも良い。
【００６０】
また、図１２の実施形態においては、プローブ支持体２２０に非接触倣いプローブ２９０の本体２９１を着脱可能に固定したが、プローブ支持体に代えてほぼ同一形状の基準体２２１を備え、スピンドル２１７に被校正非接触倣いプローブ２９０の本体２９１を着脱可能に固定し、連結体５００を基準体２２１に固定して被校正非接触倣いプローブ２９０で測定目標５３０を倣い測定するようにしても良い。要するに被校正非接触倣いプローブと連結体５００を正確に相対真直運動させることが出来ればいずれでも良い。
さらに、三次元倣いプローブに限って説明したが、輪郭形状測定に用いられる二次元倣いプローブであっても良い。
【００６１】
また、倣いプローブに温度センサを備える形態の説明を行ったが、この温度センサは倣いプローブの近傍に設けても良いし、さらに恒温室中に校正装置が設置される場合は、その恒温室温度を用いても良い。あるいは、校正精度面の要求がそれほど厳しくなく、温度センサを設けない倣いプローブの場合は、温度誤差の校正を省略しても良い。
また、実際の校正のための測定は、手動操作によって各軸測定点へ位置決めするほか、プログラムによって校正測定装置が自動位置決めを行って自動測定されるものであっても良い。
【００６２】
【発明の効果】
以上述べたように、校正測定装置と被校正倣いプローブの出力座標を比較することによって、被校正倣いプローブの幾何学的偏差を収集し、その誤差を算出して補正データを求めることができるので、倣いプローブの誤差補正が容易になる。
【図面の簡単な説明】
【図１】本発明に係る実施形態の校正装置における校正測定装置の構成を示す図である。
【図２】同校正装置の構成を示すブロック図である。
【図３】同校正装置における連結体を示す図である。
【図４】同校正装置における他の連結体を示す図である。
【図５】倣いプローブの校正方法を示すフローチャートである。
【図６】倣いプローブの回転誤差を示す模式図である。
【図７】倣いプローブの回転誤差を示す他の模式図である。
【図８】倣いプローブの回転誤差を示す他の模式図である。
【図９】倣いプローブの直交誤差を求める説明図である。
【図１０】倣いプローブの誤差補正方法を示すフローチャートである。
【図１１】指示誤差の補正方法を示す説明図である。
【図１２】本発明に係る他の実施形態の校正装置における校正測定装置の構成を示す図である。
【図１３】同校正装置における連結体を示す図である。
【図１４】三次元測定機におけるプローブの使用状態を示す図である。
【図１５】倣いプローブの構成例を示す図である。
【図１６】倣いプローブの案内機構の回転誤差を示す模式図である。
【符号の説明】
２００ 校正装置
２１０ 校正測定装置
２２０ プローブ支持体
２３０、２９０ 被校正倣いプローブ
２６０ 駆動装置
２７０ 計算機
３００、４００、５００ 連結体[0001]
The present invention relates to a scanning probe calibration apparatus, a calibration program, and a calibration method, and more particularly, a calibration apparatus and a calibration program for calibrating various errors of a scanning probe that measures surface properties such as dimensions, shape, waviness, and roughness of an object to be measured. And calibration methods.
[0002]
[Background]
A three-dimensional measuring machine for measuring the three-dimensional shape of the object to be measured, a contour shape measuring machine and an image measuring machine for measuring a two-dimensional contour shape, a roundness measuring machine for measuring roundness, and a surface of the object to be measured. 2. Description of the Related Art A surface texture measuring instrument that measures the contour shape, roughness, undulation, etc. of the surface of an object to be measured, such as a surface roughness measuring instrument that measures swell and roughness, is known. Many of these include a contact type or non-contact type sensor and a guide mechanism that relatively moves the object to be measured.
[0003]
Many of these guide mechanisms include a guide, a feed screw, and a nut screwed to the feed screw, and a slider coupled to the nut is moved, and the movement of the slider is measured with a linear scale or the like. In addition, there are some which are not necessarily provided with a feed screw and are composed of a guide and a slider, and read a displacement amount of the slider which is manually moved by a linear scale or the like. Usually, one or more types of sensors such as probes and CCD cameras are attached to the slider.
Probes used for these applications include a touch signal probe and a scanning probe.
[0004]
FIG. 14 shows an example in which such a touch signal probe or a scanning probe is used by being attached to the tip of the spindle 117 of the coordinate measuring machine.
The coordinate measuring machine 100 is configured as follows.
A surface plate 112 is placed on the vibration isolation table 111 so as to coincide with a horizontal surface with the upper surface as a base surface, and at the upper ends of beam supports 113 a and 113 b erected from both ends of the surface plate 112. A beam 114 extending in the X-axis direction is supported. The lower end of the beam support 113a is driven in the Y-axis direction by the Y-axis drive mechanism 115. The lower end of the beam support 113b is supported on the surface plate 112 by an air bearing so as to be movable in the Y-axis direction. The current movement positions of the beam supports 113a and 113b are detected by a Y-axis scale (not shown).
[0005]
The beam 114 supports a column 116 extending in the vertical direction (Z-axis direction). The column 116 is driven along the beam 114 in the X-axis direction. The current movement position of the column 116 is detected by an X-axis scale (not shown). The column 116 is provided with a spindle 117 so as to be driven along the column 116 in the Z-axis direction. The current movement position of the spindle 117 is detected by a Z-axis scale (not shown).
A probe 118 having a contact-type measuring element 119 is attached to the lower end of the spindle 117. The probe 118 measures an object to be measured placed on the surface plate 112. For example, an optical linear scale is used for the X-axis scale, the Y-axis scale, and the Z-axis scale.
[0006]
The touch signal probe reads the value of the linear scale at the moment when the measuring element contacts the object to be measured, and obtains the measurement position of the object to be measured. JP-A-10-73429 is known as an example of such a touch signal probe. This is a structure in which a probe with a spherical contact at the tip can always be returned to a fixed position by the seating mechanism. When the contact comes into contact with the object to be measured, the probe is displaced and detached from the seating mechanism. The contact is released and a touch signal is output.
This touch signal probe basically obtains the coordinate position of one point of the object to be measured, and in order to measure a plurality of locations of the object to be measured, each measurement operation is required. If the contour data of the measurement object is to be obtained closely, it is necessary to perform many positioning and measurement operations, which increases the overall measurement time, and as a result, is affected by environmental fluctuations such as temperature changes. Therefore, it is not necessarily suitable for high-precision measurement.
[0007]
On the other hand, since the scanning probe can continuously measure the position of the object to be measured, it can easily determine the contour data densely and at high speed by measuring multiple points of the object to be measured. There is a possibility that high-accuracy measurement is possible as a whole because it is not easily affected by environmental changes.
As such a scanning probe, there is a probe disclosed in Japanese Patent Laid-Open No. 5-256640 (see FIG. 15). In this probe, a stylus is supported via an X-axis slider, a Y-axis slider, and a Z-axis slider that are movable in a direction orthogonal to the base, and sliding between the base and the three sliders. A guide mechanism with very little friction is configured by sending pressurized air to the section to form an air bearing. The three sensors of the base, the Z-axis slider, the Z-axis slider and the Y-axis slider, and the Z-axis sensor, the Y-axis sensor, and the X-axis sensor that detect the relative displacement of each of the Y-axis slider and the X-axis slider. The three-dimensional displacement amount of the stylus can be obtained by these three sensors.
[0008]
For these sensors, for example, an absolute optical linear scale is used. Therefore, if the scanning probe is moved in the direction of the surface of the object to be measured while the stylus (measurement element) of the scanning probe is in contact with the surface of the object to be measured, Since it is displaced along the contour shape of the surface of the object to be measured, the contour shape data of the object to be measured can be continuously collected. In this case, the contour shape data is obtained by combining the outputs of the three sensors output from the scanning probe and the value of the linear scale for measuring the displacement of the driving mechanism of the coordinate measuring machine.
Note that the normal stop position (return position) of the X-axis slider, Y-axis slider, and Z-axis slider of the scanning probe when the stylus is not in contact with the object to be measured is the origin position of each absolute sensor.
In this way, the scanning probe has the feature that data can be collected at high speed, but on the other hand, it has a complicated structure due to the built-in guide mechanism and sensor compared with the touch signal probe, and sufficient measurement is possible. There is a problem that it is difficult to ensure accuracy.
[0009]
These guide mechanisms are inevitably subject to machining errors and deformations caused by environmental fluctuations, and as a result, the Z-axis slider, Y-axis slider, and X-axis slider cannot move correctly. The data indicating the displacement of the stylus measured by the Z-axis sensor, the Y-axis sensor, and the X-axis sensor that measure the displacement includes an error.
For example, taking the schematic diagram of the X-axis guide mechanism in FIG. 16A as an example, when the X-axis slider is displaced in the X-axis direction, the X-axis slider rotates around the X-axis, Pitching that rotates around the Y axis and yawing that the X axis slider rotates around the Z axis occur.
In addition to errors due to these rotational motions, there is a straightness error due to the linearity of the motion of the X-axis guide mechanism. This includes a Y-axis straightness error that is displaced in the Y-axis direction and a Z-axis straightness error that is displaced in the Z-axis direction when the X-axis slider is displaced in the X-axis direction.
[0010]
Further, the X-axis sensor itself has an error in the indicated value with respect to the displacement amount (indicated error) over the entire measurement area. That is, there are six types of errors per axis, but in addition to these errors, there are orthogonal errors due to orthogonality. In the case of a three-dimensional scanning probe, the X axis and the Y axis, the Y axis and the Z axis Since there are three types of orthogonal errors of the Z axis and the X axis, there is a possibility that at least 21 types of geometric errors may occur.
That is, the error caused by the rotational motion is A (x) = X-axis roll error, A (y), where A is the rotation around the X axis, B is the rotation around the Y axis, and C is the rotation around the Z axis. = Y-axis pitch error, A (z) = Z-axis pitch error, B (x) = X-axis pitch error, B (y) = Y-axis roll error, B (z) = Z-axis yaw error, C (x) = X-axis yaw error, C (y) = Y-axis yaw error, and C (z) = Z-axis roll error.
Straightness error is X (y) = Y-axis straightness error in X-axis direction, X (z) = Z-axis straightness error in X-axis direction, Y (x) = X-axis straightness error in Y-axis direction, Y (z) = Z-axis straightness error in the Y-axis direction, Z (x) = X-axis straightness error in the Z-axis direction, and Z (y) = Y-axis straightness error in the Z-axis direction.
[0011]
As the instruction error, X (x) = X-axis instruction error, Y (y) = Y-axis instruction error, and Z (z) = Z-axis instruction error.
The orthogonal error is Pyx = Y axis X axis orthogonal error, Pzx = Z axis X axis orthogonal error, and Pzy = Z axis Y axis orthogonal error.
These errors fluctuate the attitude of the probe supported via the guide mechanism, making it difficult to determine the accurate position of the probe, and as a result, degrade the measurement accuracy.
In the case of a scanning probe, even if the guide mechanism is the same, the amount of change in the orientation of the probe changes depending on the length of the probe. The measurement error changes. In general, the measurement error tends to increase when a long probe is used.
[0012]
FIG. 16B schematically shows the position error that the rotation A (X-axis rolling) around the X-axis gives to the contact sphere 233 of the measuring element 232, dY in the Y-axis direction and Z-axis in the Z-axis direction. It can be seen that an error of dZ occurs.
Further, there is a probe offset to be taken into consideration in calibration of the scanning probe. This is an offset from the reference position of the spindle to the center position of the contact sphere at the tip of the probe, and is composed of Sx (X-axis offset), Sy (Y-axis offset), and Sz (Z-axis offset).
[0013]
[Problems to be solved by the invention]
As described above, since the error factors are complicated, there is a problem that a great deal of labor is required to perform the calibration of these scanning probes strictly.
As an example, the measuring instrument itself for the purpose of calibrating the geometric deviation of the scanning probe has a long history, but it is limited in terms of the variety of measuring methods. . In other words, laser interferometers and levels used for calibration are mainly measuring instruments that detect a single geometric deviation in a single function. If you are trying to manage the uncertainty of measurement using these measuring instruments, the handling of the equipment and the installation orientation (alignment) of the measuring instrument prior to each measurement must be performed by an operator who is familiar with the operation. is there. As a result, it was necessary for the technicians to spend a lot of time performing calibration, which resulted in a high-cost labor-intensive work process that could not be labor-saving. On the other hand, the geometric accuracy of the scanning probe reaches several ppm when normalized within its measurable range, and it is difficult to realize a calibration method that is satisfactory in terms of uncertainty from the viewpoint of simply trying to automate. It has become.
[0014]
In addition, the cost of these labor intensive calibration operations raises the cost of the product.
In addition, there is a problem that it is more difficult to secure a calibration worker having an extremely high level of skill to master these calibration devices.
Furthermore, since the calibration in the manufacturing process of the scanning probe is for scanning probes that do not have a history of calibration in advance, the measurement points that cover the measurable area of the scanning probe are arranged at a sufficiently fine interval. It is necessary to carry out calibration equivalent to the inspection of all and all functions. In addition, when correcting the geometric deviation (or measurement error) of the scanning probe using the calibration result, the calibration value is expressed as a kinematically described geometric deviation that can be used for accuracy correction. It is necessary to adopt the calibration method to output. Because of these characteristics, the dependence on skilled workers was very high, and the obstacles to labor saving were great.
[0015]
The present invention has been made to solve such a problem, and an object of the present invention is to provide a calibration apparatus capable of automating calibration of a scanning probe.
It is another object of the present invention to provide a scanning probe calibration program and calibration method.
[0016]
In order to achieve the above object, the present invention provides a calibration apparatus for calibrating a scanning probe for measuring the surface property of a measured object by detecting a displacement amount in at least two axial directions accompanying the contact of the measuring element with the measured object. The calibration device includes a surface plate, a probe support that is erected on the surface plate and detachably fixes the scanning probe, a spindle, a scale that detects a displacement position of the spindle, and the measuring element. And a connecting body that couples the spindle, and the scanning probe is moved based on the displacement position and the displacement amount when the spindle moves and displaces the probe over the entire measurement range. It is characterized by calibration.
Furthermore, the present invention provides a calibration apparatus for calibrating a scanning probe that detects the amount of displacement in at least two axial directions accompanying the contact of the probe with the measurement object and measures the surface property of the measurement object. A surface plate, a reference body standing on the surface plate, a spindle for detachably fixing the scanning probe, a scale for detecting a displacement position of the spindle , the measuring element, and the reference body. A coupling body to be coupled, wherein the scanning probe is calibrated based on the displacement position and the displacement amount when the probe moves to displace the measuring element over the entire measurement range. And
[0017]
Further, in the present invention, wherein the coupling element comprises a first coupling member connected to one of said measuring element or the spindle, and a second coupling member connected to the other of the measuring element or the spindle provided, it is preferable that said first coupling member and the second connecting member are connected so as to have a degree of freedom only in the rotating direction.
Further, said first coupling member and the second connecting member which is preferably driven by each other by a wire.
Further, it is preferably connected to each other by the spherical bearing and the second connection member and the first coupling member.
[0018]
Furthermore, the present invention provides a calibration apparatus for calibrating a scanning probe that detects a surface property of a measurement object in a non-contact manner by detecting a displacement amount in at least two axial directions. The calibration apparatus includes a surface plate and the surface plate. A probe support which is erected on the probe probe and detachably fixes the scanning probe ; a spindle which fixes a coupling body provided with a measurement target to be measured by the scanning probe; and a scale for detecting a displacement position of the spindle ; The scanning probe is moved based on the displacement position and the displacement amount when the spindle is moved while the measurement target is measured by the scanning probe and the measurement target is displaced over the entire measurement range. Is calibrated.
Furthermore, the present invention provides a calibration apparatus for calibrating a scanning probe that detects a surface property of a measurement object in a non-contact manner by detecting a displacement amount in at least two axial directions. The calibration apparatus includes a surface plate and the surface plate. A reference body for fixing a coupling body, which is erected on the scanning probe and has a measurement target to be measured by the scanning probe; a spindle for detachably fixing the scanning probe; and a scale for detecting a displacement position of the spindle. The scanning probe is moved based on the displacement position and the displacement amount when the scanning probe is moved over the entire measurement range by measuring the measurement target with the scanning probe. It is characterized by calibration.
[0019]
Furthermore, the present invention provides a scanning probe calibration program for calibrating the scanning probe, using any of the calibration apparatuses described above, and causing a computer constituting the calibration apparatus to execute the copying apparatus. A preparation step for inputting a calibration condition including at least each axis measurement range of the probe; an error collection step for performing straight measurement in each axis measurement range and collecting an error based on a displacement amount of each axis by the scanning probe; An error calculation step for calculating an indication error, a straightness error, and a rotation error from the errors collected in the error collection step, and two sets of known distances between the two points in each axis measurement range, for a total of four points Orthogonal data collection step for collecting orthogonal data based on the displacement amount of the scanning probe at each of the points, and the orthogonal data collection From the orthogonal data collected in step, a quadrature error calculating step of calculating the quadrature error, characterized in that to execute the computer.
In the present invention, it is preferable that in the error calculation step, the orthogonal error is separated from the calculated instruction error .
Moreover, it is preferable that the error collection in the error collection step is covered at predetermined intervals in each axis measurement range .
[0020]
In the present invention, it is preferable to use a least square method for calculating the indication error, straightness error, and rotation error in the error calculating step.
Furthermore, the present invention provides a scanning probe calibration method for calibrating the scanning probe using any one of the calibration devices described above, wherein the calibration device includes at least each axis measurement range of the scanning probe. Preparatory step for inputting conditions, and orthogonal data collection for collecting the orthogonal data based on the displacement amount of the scanning probe at each of a total of four sets of two known distances between the two points in each axis measurement range Step, an orthogonal error calculating step for calculating an orthogonal error from the orthogonal data collected in the orthogonal data collecting step, straight measurement in each axis measurement range, and based on a displacement amount of each axis by the scanning probe An error collection step for collecting errors, and an indication error, a straightness error, and a rotation from the errors collected in the error collection step. Calculating the difference, further the error calculating step of separating the quadrature error from the indication error, comprising the.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments using the present invention will be described with reference to the drawings. In addition, what attached | subjected the same code | symbol in all the figures represents the same component.
1 according to the first embodiment, Z-axis sensor of the scanning probe, Y-axis sensor, indicates the calibration device 200 that may directly compare measured indicated value of the X-axis sensor, FIG. 2, the calibration device 200 It is a block diagram which shows a structure .
The calibration device 200 includes a calibration measurement device 210 , a driving device 260, and a computer 270. The calibration measurement device 210 has a configuration similar to that of the three-dimensional measurement device shown in FIG. 14, and is specifically configured as follows. ing.
[0022]
On the vibration isolation table 211, a surface plate 212 is placed so as to coincide with a horizontal surface with the upper surface as a base surface, and at the upper ends of beam supports 213a and 213b erected from both ends of the surface plate 212. A beam 214 extending in the X-axis direction is supported. Beam support 213a has its lower end driven in the Y-axis direction by a Y-axis driving mechanism 242, the driving current position is measured by the Y-axis scale 245. The lower end of the beam support 213b is supported on the surface plate 212 by an air bearing so as to be movable in the Y-axis direction. The beam 214 supports a column 216 extending in the vertical direction (Z-axis direction). Column 216 is driven by X-axis driving mechanism 241 along the beam 214 in the X-axis direction, driving current position is measured by the X-axis scale 244. In the column 216, a spindle 217 is provided along the column 116 so as to be driven in the Z-axis direction by the Z-axis drive mechanism 243, and the driven current position is measured by the Z-axis scale 246.
[0023]
A probe support 220 is erected on the upper surface of the surface plate 212 at a substantially central position in the X-axis direction.
A main body 231 of the calibrated scanning probe 230 is detachably fixed to the tip of the probe support 220 with the measuring element 232 facing downward.
Since the measuring element 232 and the spindle 217 are connected and connected by the connecting body 300, when the spindle 217 moves and is displaced in the X-axis, Y-axis, and Z-axis directions, the measuring element 232 is similarly moved along with the movement. The scanning probe 230 is displaced with respect to the main body 231.
As shown in FIG. 2, the scanning probe 230 includes an X-axis sensor, a Y-axis sensor, and a Z-axis sensor, and the displacement amount according to the displacement of the probe 232 in the X-axis, Y-axis, and Z-axis directions. Are output respectively.
[0024]
The drive device 260 includes an X-axis drive circuit 261 that drives the X-axis drive mechanism 241, a Y-axis drive circuit 262 that drives the Y-axis drive mechanism 242, a Z-axis drive circuit 263 that drives the Z-axis drive mechanism 243, and an X-axis X-axis counter 264 that counts the output of scale 244, Y-axis counter 265 that counts the output of Y-axis scale 245, Z-axis counter 266 that counts the output of Z-axis scale 246, and X that counts the output of X-axis sensor 251 An axis P counter 267, a Y axis P counter 268 that counts the output of the Y axis sensor 252, and a Z axis P counter 269 that counts the output of the Z axis sensor 253 are included, and each is connected to a computer 270. Therefore, the X axis, Y axis, and Z axis of the calibration measuring apparatus 210 can be positioned at an arbitrary position at an arbitrary speed by a command from the computer 270. Further, the computer 270 inputs the count values of the counters 264 to 269 and knows the current positions of the X axis, Y axis, and Z axis of the spindle 217 and the current displacement of the probe 232 of the scanning probe 230. It is configured to be able to.
[0025]
The computer 270 is similar to a known computer except that it includes a connection device (not shown) for exchanging information with the drive device 260, and includes a central processing unit, a storage device, an input device, a display device, a printing device, and an output device. In accordance with a program stored in the storage device, calibration of the calibration apparatus 200, such as error correction of the calibration measurement apparatus 210, collection of scanning probe data, calculation of error, display of error, functioning of error, output of correction data, etc. The entire process is automatically controlled or each function is semi-automatically controlled or manually controlled as necessary.
[0026]
Information exchange between the computer 270 and the driving device 260 is usually performed by wired communication using a transmission control procedure such as IEEE488, but wireless communication or optical communication may be used as necessary.
As shown in FIG. 3, the connection body 300 is connected to the spindle 217 of the calibration measuring device 210 by the intermediate connection body 320 and is connected to the probe 232 of the scanning probe 230 by the intermediate connection body 330. Here, the spindle 217 and the measuring element 232 may be directly fixed to the connecting body 300 without using the intermediate connecting bodies 320 and 330.
[0027]
FIG. 4 shows another type of connection body 400, which is composed of a first connection member 410, a second connection member 412, and a piano wire 413. The first connecting member 410 has a recess at the center, and a support arm 411 projects toward the recess. The center of the second connecting member 412 is hollowed out to form a tunnel portion. The support arm 411 is inserted through the tunnel portion, but does not contact the inner wall of the tunnel portion. The first connecting member 410 and the second connecting member 412 are connected by a piano wire 413. The details are as follows.
[0028]
First, in the X-axis direction, the inner wall of the recess and the outer wall of the second connecting member 412 are fastened by two piano wires. Similarly, in the Y-axis direction, the inner wall of the recess and the outer wall of the second connecting member 412 are fastened by two piano wires. In the Z-axis direction, the bottom surface of the recess and the lower bottom surface of the second connecting member 412 are fastened by a piano wire, and the bottom surface of the tunnel portion of the second connecting member 412 and the bottom surface of the tip of the support arm 411 are fastened by a piano wire. . Each of the six piano wires in total is fastened in a pulled state so that no slack occurs. With such a fastening structure, the second connecting member 412 is not displaceable in any of the X axis, Y axis, and Z axis linear directions with respect to the first connecting member 410, but the X axis, Y axis, There is a slight degree of freedom with respect to rotation about the axis direction of the Z axis. As in the case of FIG. 3, the intermediate connector 420 is connected to the spindle 217 of the calibration measuring device 210 and the intermediate connector 430 is connected to the probe 232 of the scanning probe 230. Here, the spindle 217 and the measuring element 232 may be directly fixed to the connecting body 400 without using the intermediate connecting bodies 420 and 430, or the first connecting member 410 may be connected to the measuring element 232, The two connecting members 412 may be connected to the spindle 217.
[0029]
Since this coupling body 400 can absorb the rotational motion such as rolling, pitching, yawing and the like on the spindle 217 of the calibration measuring device 210 and the probe 232 of the probe 230, such a rotational motion can be absorbed. When the structure of the connecting body is used, the rate at which the harmful rotational motion of the spindle 217 or the measuring element 232 is transmitted to the other is reduced, so that more accurate calibration is possible. For such a purpose, as the structure of the coupling body, for example, a spherical bearing as disclosed in JP 2000-304039 may be used, or a flexible coupling may be used. In short, any structure that allows relative rotation while restricting displacement in the linear direction may be used.
[0030]
Incidentally, since the 21 types of geometric deviations of the drive mechanism of each axis and the scale of each axis of the calibration measuring apparatus 210 are corrected in the entire moving space, each detected and corrected from the scale of each axis. The current position of the shaft has necessary and sufficient accuracy, and as a result, the reference position of the tip portion of the spindle 217 driven by the driving mechanism of each shaft is driven straight with necessary and sufficient accuracy.
[0031]
Since the calibration measuring apparatus 210 has such a configuration, the measuring element 232 of the calibration probe 230 fixed to the tip of the probe support 220 can be measured by the spindle 217 in all possible measurement spaces of the calibration probe 230. It is possible to accurately or manually displace it over (the movable range of the measuring element 232). Further, the displacement position can be accurately detected by the current position of each axis detected and corrected by the scales 244, 245, and 246 of each axis of the calibration measuring apparatus 210, so that the calibration probe 230 to be calibrated over the entire measurement space. Can be compared with the outputs of the X-axis sensor 251, Y-axis sensor 252, and Z-axis sensor 253 (count values of the X-axis P counter 267, Y-axis P counter 268, and Z-axis P counter 269). That is, it is possible to easily create a comparison list of the output value of each axis sensor of the calibration probe 230 and its correct current position over the entire measurement space of the probe 232 of the calibration probe 230. Therefore, by using such a calibration apparatus 200, calibration of the to-be-calibrated scanning probe 230 is facilitated, and it is possible to perform calibration work even if it is not an expert.
[0032]
Next, a method for calibrating the scanning probe 230 will be described with reference to FIG.
The calibration processing procedure is started in S10.
First, the to-be-calibrated scanning probe 230 is fixed to the tip of the probe support 220 of the calibration apparatus 200, and each axis measurement range, origin position, measurement resolution of the to-be-calibrated scanning probe, and the contact ball 233 at the tip of the probe 232 from each slider. The length in each axial direction (X-axis, Y-axis, Z-axis direction) to the center position of each, the connection dimension of the coupling body (each axial direction from the reference position of the spindle 217 to the center position of the contact ball 233 [ X-axis, Y-axis, and stores the input to the storage device of the computer 270 by the Z-axis direction] length) inputting a device (S20).
Next, for the X axis, the X axis direction instruction error X (x), the X axis direction Y axis straightness error X (y), the X axis direction Z axis straightness error X (z), and the pitching error B ( x), a rolling error A (x), and a yawing X-axis error C (x) are obtained (S30).
[0033]
For example, when the contact ball 233 of the scanning probe 230 is displaced in the X-axis direction by the spindle 217 using the coupling body 400 shown in FIG. 4, the X-axis sensor 251, the Y-axis sensor 252, and the Z-axis sensor 253 of the scanning probe 230 are used. Output (count values of the X-axis P counter 267, the Y-axis P counter 268, and the Z-axis P counter 269) show the following changes.
That is, the output of the X-axis sensor 251 shows almost the same change as the output of the X-axis scale 244 of the calibration measuring apparatus 210, but the X-axis indication error X (x), the X-axis pitch error B (x), An error is caused by the X-axis yaw error C (x).
Further, the output of the Y-axis sensor 252 causes errors due to the Y-axis straightness error X (y) in the X-axis direction, the X-axis roll error A (x), and the X-axis yaw error C (x).
Further, the output of the Z-axis sensor 253 causes errors due to the Z-axis straightness error X (z), the X-axis roll error A (x), and the X-axis pitch error B (x) in the X-axis direction.
[0034]
FIG. 6 shows the state of the X-axis rolling that occurs when the X-axis slider 234 is displaced in the X-axis direction without restraining the contact ball 233 of the scanning probe 230 on the YZ plane. The reference point in the YZ-axis direction of the X-axis slider 234 is the upper left corner of the X-axis slider 234 in FIG. The measuring element 232 is attached to a position offset in the Y-axis direction and the Z-axis direction (the lower right of the X-axis slider 234 in FIG. 6) with respect to the YZ-axis direction reference point of the X-axis slider 234. Have a certain length. The X-axis slider 234 generates a straight error in the Y-axis direction and the Z-axis direction with respect to the movement in the X-axis direction. Further, the X-axis slider 234 has a posture shown by, for example, a broken line in FIG. Take. That is, the Y-axis direction error dy and the Z-axis direction error dz generated at this time are
dy = Y-axis straightness error X (y) in the X-axis direction + rolling Y-axis error (1)
dz = Z-axis straightness error X (z) in the X-axis direction + rolling Z-axis error (2)
It becomes.
[0035]
Conversely, this indicates that when the contact ball 233 is accurately moved straight in the X-axis direction, error outputs of dy and dz are generated in the outputs of the Y-axis sensor 252 and the Z-axis sensor 253.
FIG. 7 shows, on the XZ plane, the state of X-axis pitching that occurs when the X-axis slider 234 is displaced in the X-axis direction without constraining the contact ball 233 of the scanning probe 230. The XZ-axis direction reference point of the X-axis slider 234 is the upper left corner of the X-axis slider 234 in FIG. With respect to the XZ-axis direction reference point of the X-axis slider 234, the measuring element 232 is attached at a position offset in the X-axis direction and the Z-axis direction (lower right of the X-axis slider 234 in FIG. 7). Have a certain length. The X-axis slider 234 generates a straight error in the Z-axis direction with respect to movement in the X-axis direction. Further, the X-axis sensor has an X-axis instruction error X (x) with respect to the X-axis direction. Further, by the pitching of the X-axis slider 234, the posture shown by the broken line in FIG.
[0036]
That is, the X-axis direction error dx and the Z-axis direction error dz generated at this time are
dx = X-axis direction instruction error X (x) + pitching X-axis error (3)
dz = Z-axis straightness error X (z) in the X-axis direction + pitching Z-axis error (4)
It becomes.
Conversely, this indicates that when the contact ball 233 is accurately moved straight in the X-axis direction, error outputs of dx and dz are generated in the outputs of the X-axis sensor 251 and the Z-axis sensor 253.
[0037]
FIG. 8 shows an X-axis yawing that occurs when the X-axis slider 234 is displaced in the X-axis direction without restraining the contact ball 233 of the scanning probe 230 on the XY plane. The XY-axis direction reference point of the X-axis slider 234 is the upper left corner of the X-axis slider 234 in FIG. The measuring element 232 is attached to a position offset in the X-axis direction and the Z-axis direction with respect to the XY-axis reference point of the X-axis slider 234 (in FIG. 8, the lower right of the X-axis slider 234). It has been. The X-axis slider 234 generates a straight error in the Y-axis direction with respect to movement in the X-axis direction. Further, with respect to the X-axis direction, the X-axis sensor has an X-axis instruction error X (x). Further, by the yawing of the X-axis slider 234, the posture shown by the broken line in FIG.
[0038]
That is, the X-axis direction error dx and the Y-axis direction error dy generated at this time are
dx = X-axis direction instruction error X (x) + Yawing X-axis error (5)
dy = Y-axis straightness error X (y) in the X-axis direction + Yawing Y-axis error (6)
It becomes.
Conversely, this indicates that when the contact ball 233 is accurately moved straight in the X-axis direction, error outputs of dx and dy are generated in the outputs of the X-axis sensor 251 and the Y-axis sensor 252.
[0039]
In the above description, errors caused by rolling, pitching, and yawing are individually described. However, since these errors actually occur at the same time, the X-axis sensor 251, the Y-axis sensor 252, and the Z-axis sensor 253 of the scanning probe 230. Each error component is included together in the errors dx, dy, and dz included in.
That is,

It becomes.
[0040]
Therefore, the errors dx, dy, and dz are collected over the entire X-axis measurement range of the scanning probe 230. Here, for example, if the pitching angle is determined, it goes without saying that the pitching X-axis error and the pitching Z-axis error are uniquely determined. Therefore, the variables include the X-axis direction error X (x), the X-axis direction Y-axis straightness error X (y), the X-axis direction Z-axis straightness error X (z), the pitching angle, and the rolling angle. There are six yawing angles, but there are only three equations, so a solution cannot be obtained as it is. As this solution, for example, a least square method using QR decomposition or the like is used as a linear multiple regression problem.
Similarly, the error with respect to the Y-axis displacement and the Z-axis displacement is obtained, and each error element is obtained (S40, S50).
[0041]
Next, the orthogonal error of each axis is obtained (S60).
FIG. 9 schematically shows a method for obtaining the orthogonal error between the PX axis and the PY axis of the sensor of the scanning probe 230. Here, X and Y in the figure indicate the axial directions of the X-axis scale 244 and the Y-axis scale 245 of the calibration measuring apparatus 210, and the orthogonal error between them is corrected within an allowable range. PX and PY indicate the axial directions of the X-axis sensor and the Y-axis sensor of the scanning probe 230, and have angle errors of θx and θy with respect to X and Y, respectively.
First, the calibration measuring device 210 is driven to position the contact ball 233 at the point P1, and the outputs of the X-axis sensor and the Y-axis sensor at that time are stored. Next, the contact ball 233 is positioned at the point P2, and the outputs of the X-axis sensor and the Y-axis sensor at that time are stored. Here, the P1 point and the P2 point are selected so that the length from the P1 point to the P2 point is exactly L. At this time, the count value from the P1 point to the P2 point on the PX axis is A, and the count value from the P1 point to the P2 point on the PY axis is B.
[0042]
Next, the calibration measuring device 210 is driven to position the contact ball 233 to the point P3, and the outputs of the X-axis sensor and the Y-axis sensor at that time are stored. Next, the contact ball 233 is positioned at the point P4, and the outputs of the X-axis sensor and the Y-axis sensor at that time are stored. Here, the points P3 and P4 are selected so that the length from the point P3 to the point P4 is exactly L. At this time, the count value from the P3 point to the P4 point on the PX axis is C, and the count value from the P3 point to the P4 point on the PY axis is D.
From this result, the following equations are solved to obtain θx and θy.
L ² = A ² · cos ² θx + B ² · cos ² θy (10)
L ² = C ² · cos ² θx + D ² · cos ² θy (11)
Thus, the orthogonal error between the X-axis sensor and the Y-axis sensor can be obtained. Similarly, the orthogonal error between the Y-axis sensor and the Z-axis sensor and the orthogonal error between the X-axis sensor and the Z-axis sensor are obtained.
[0043]
Note that when calculating the instruction error, the influence of the orthogonal error is not taken into consideration, and the instruction error of each axis obtained in S30, S40, and S50 of FIG. 5 includes this orthogonal error component. Therefore, the orthogonal error is separated from the instruction error. More specifically, the orthogonal error component (for example, in the case of the X axis, the primary correlation component based on the angle θx) is excluded from the instruction error component before correction, and the instruction error and the orthogonal error are separated.
[0044]
Since various error data can be calculated by the above measurement and calculation processing, a function for calculating correction data is obtained so that these errors can be easily corrected (S70). This can be obtained by function approximation of error data over the entire measurement range of the scanning probe 230. As a result, the correction data calculation functions for the X axis, the Y axis, and the Z axis are Fx (x, y, z), Fy (x, y, z), and Fz (x, y, z), respectively. That is, in general, for example, X-axis correction data is affected by not only the X-axis but also the displacement positions of the Y-axis and the Z-axis.
[0045]
In performing this function approximation, the function approximation may be performed by excluding only the instruction error as necessary.
In this case, the instruction error is separated as an instruction error data string for the purpose of performing a separate process.
These results (approximate function data, instruction error data string) are output as correction data for the calibrated scanning probe. In other words, display on the display device of the calibration device, printing on the printing device, or directly stored in the coordinate measuring machine via the output device, or stored in a storage medium such as an IC memory or a flexible disk, When the scanning probe is used, correction data (function data) is transferred to the coordinate measuring machine. Further, the correction data may be stored in a non-volatile memory or the like stored in the scanning probe, and the correction data may be read from the scanning probe when the probe is attached to the coordinate measuring machine. In this way, the handling of the correction data itself is unlikely to occur, and the portability of the scanning probe is improved and the usability is improved.
[0046]
Next, a probe offset is obtained (S80).
In general, the scanning probe uses various measuring elements corresponding to the shape of the object to be measured. Therefore, in principle, it is necessary to create a measurement program (part program) corresponding to the scanning probe. Providing a dedicated measurement program for each probe even for the same measurement object is extremely complicated in practice. In order to eliminate this complexity and share the measurement program, the concept of probe offset is used.
For this purpose, the CMM usually has a reference point, and an offset amount from this reference point to the center position of the contact sphere of the tracing probe probe is held as a “probe offset”. When determined, the corresponding probe offset is selected, the probe offset is added to the position of this reference point, and the position of the contact sphere is calculated.
[0047]
In many cases, the reference point of the coordinate measuring machine is provided at the center of the lower end of the Z-axis spindle. The coordinate measuring machine can know the position of this reference point from the position of the machine origin and the values of the scales of the X, Y, and Z axes. If the probe offset is added as described above, the contact sphere The center coordinates can be calculated as the current position coordinates, and thereby the drive control of each axis can be performed.
The probe offset is measured by, for example, mounting a scanning probe 118 corresponding to the coordinate measuring machine 100 as shown in FIG. 14, measuring a reference sphere 120 installed at a fixed position on the table of the coordinate measuring machine, Find the coordinates. Since the coordinate value of the center coordinate of the reference sphere 120 thus determined is shifted by the probe offset, this shift amount may be used as the probe offset. This probe offset is stored in the same manner as the correction data described above.
Thus, the calculation of the correction data is finished (S90).
[0048]
Next, a method for correcting the output of the scanning probe 230 using these correction data (function data) will be described with reference to the flowchart of FIG.
The correction data is stored in a data processing apparatus or the like that performs error correction in advance when the scanning probe is mounted on the coordinate measuring machine, and the correction process is started in S100.
First, the sensor output of the scanning probe 230 is input (S110). The outputs of the X-axis sensor 251, Y-axis sensor 252, and Z-axis sensor 253 output from the scanning probe 230 are similar to the counters (X-axis P counter, Y-axis P counter, and Z-axis P counter) shown in FIG. Since it is counted by the P counter, the content of the P counter is input as the value of the length measuring sensor. If the scanning probe has a built-in temperature sensor, the temperature sensor output is also input.
[0049]
Next, the necessity of proportional error correction is checked (S120).
Proportional error correction is correction in a proportional coefficient with respect to the output of the length measurement sensor. Here, temperature correction and multipoint correction of instruction error are performed.
Usually, length measuring sensors (X-axis sensor, Y-axis sensor, Z-axis sensor) have a constant linear expansion coefficient, and the scale of the sensor itself expands and contracts due to temperature. Therefore, the expansion and contraction of the scale is corrected by measuring the temperature of the length measuring sensor with the above-described temperature sensor.
In addition, the multipoint correction of the indication error is performed based on the indication error data at a plurality of positions, instead of correcting the indication error of the measuring sensor of the scanning probe in the form of an approximate function. Thus, more accurate correction can be performed. In this case, in step S70 of FIG. 5, the instruction error data is not generated and the instruction error data string is generated. The instruction error data string is a data string of error data at fixed distances over the entire measurement range of the scanning probe.
[0050]
If these proportional error corrections are not necessary, the process proceeds to S150.
When performing the proportional error correction, the proportional error is first calculated (S130). The temperature error calculation in this case is specifically performed as follows.
The linear expansion coefficient of the length measuring sensor of the scanning probe is β, and the output of the temperature sensor is Tc degrees. Since the temperature error is normally 20 degrees Celsius as the standard temperature, a temperature difference Tw = Tc−20 with respect to 20 degrees Celsius is obtained, and this is multiplied by the linear expansion coefficient β. That is, as an example, when the output of the X-axis length measurement sensor of the scanning probe is X, the temperature error Xt is obtained as follows.
Xt = X · β · (20−Tc) (12)
The Y axis and Z axis are obtained in the same manner.
[0051]
In the scanning probe, since the slide mechanism of each axis is arranged close to each other, it is usually only necessary to measure the temperature sensor at one representative location, but if necessary, the temperature sensor is individually provided for each axis. The temperature error may be calculated using a different temperature for each axis. When the linear expansion coefficient of the length measuring sensor for each axis of the scanning probe is different, the temperature error may be obtained for each axis by using a specific linear expansion coefficient for each axis.
The error calculation for correcting the indication accuracy of the length measuring sensor of the scanning probe is performed as follows.
[0052]
FIG. 11 shows the state of the error data string. In this example, the error data string is composed of five error data strings X1 to X5 over the entire length measurement range of the X axis. The respective error values are E1, E2,. If the X-axis sensor output is Xn, exceeds X1, and is equal to or less than X2, the error value En is calculated as follows.
En = E1.Xn. (E2-E1) / (X2-X1) (13)
That is, in the multipoint correction, an error value at a position to be obtained is obtained by interpolation from discrete error data at a plurality of locations. The Y axis and Z axis are obtained in the same manner.
When the instruction error is not given in the instruction error data string but in the form of approximate function data, the instruction error is calculated as a function error in S160 described below.
[0053]
After obtaining these proportional errors, the proportional errors are corrected (S140).
That is, similarly, in the case of the X axis, the temperature error Xt and the instruction error En are added to the output X of the length measurement sensor to obtain a proportional error correction result Xc. The Y axis and Z axis are obtained in the same manner.
Xc = X + Xt + En (14)
Next, it is checked whether function error correction is necessary (S150).
The function error obtained here includes a rotation error, a straightness error, and an orthogonal error. If correction of these errors is unnecessary, the process proceeds to S180 and the correction process is terminated.
[0054]
If function error correction is required, first, a function error is calculated (S160).
Since the functions Fx, Fy, and Fz are already given in S70 of FIG. 5, for example, the function error Xf of the X axis is obtained as follows. The Y axis and Z axis are obtained in the same manner.
Xf = Fx (Xc, Yc, Zc) (15)
Thereafter, the X-axis function error correction result Xp is obtained as follows to obtain the correction result of the X-axis measuring sensor of the scanning probe. The Y axis and Z axis are obtained in the same manner.
Xp = Xc + Xf (16)
The correction process is completed by the above process (S180).
The final coordinate value actually obtained using this scanning probe is further calculated by the following calculation process using these results.
[0055]
As described above, the linear scale values (Xs, Ys, Zs) for measuring the displacement of the driving mechanism of the coordinate measuring machine indicate the position of the reference point (usually the lower end center position of the Z-axis spindle). The probe offset (X-axis offset Sx, Y-axis offset Sy, Z-axis offset Sz) is added, and the displacement position of the probe of the scanning probe is added. Finally, the center coordinate value Xo of the contact ball of the probe is obtained. Find Yo and Zo.
Xo = Xs + Sx + Xp (17)
Yo = Ys + Sy + Yp (18)
Zo = Zs + Sz + Zp (19)
In measuring the object to be measured, the scanning probe is moved by the driving mechanism of each axis of the coordinate measuring machine, the surface of the object to be measured is traced, and the center coordinate value of the contact sphere is sequentially obtained, thereby measuring the object to be measured. The size, shape, etc. can be obtained.
[0056]
Next, FIG. 12 shows a configuration of the calibration apparatus 200 when calibrating the non-contact scanning probe 290.
1 is different from FIG. 1 only in that the scanning probe 290 performs non-contact measurement and the connection body 500 is fixed only to the spindle 217, and the other configurations are the same as those in FIG.
As the non-contact scanning probe 290, for example, there is an imaging probe with a focus detection function including an image optical system, and the scanning measurement of a measurement object is possible without contact. The main body 291 of the non-contact scanning probe 290 is detachably fixed to the probe support 220.
[0057]
As shown in FIG. 13, the coupling body 500 is fixed only to the spindle 217 and includes a measurement target 530 such as a cross hairline. The non-contact scanning probe 290 performs scanning measurement on the measurement target 530 and outputs coordinate values of the length measurement sensor (values of the X-axis sensor, Y-axis sensor, and Z-axis sensor). For example, when the non-contact scanning probe 290 is a two-dimensional CCD imaging probe with a focus detection function, the focus detection result is the output of the Z-axis sensor, and the measurement position of the measurement target 530 detected by the two-dimensional CCD (the cross hairline) The intersection position) is the output of the X-axis sensor and Y-axis sensor.
[0058]
In this case, if the same calibration as in FIG. 5 is performed , the aberration (straightness error, rotation error) of the image optical system, the array pitch error (indication error) of the individual sensor of the two-dimensional CCD, the individual sensor of the two-dimensional CCD Since it is possible to calibrate orthogonal errors (orthogonal errors), errors in the focus detection function (straightness error, rotation error, indication error), etc., error correction can be performed using the results according to the procedure of FIG. Needless to say.
[0059]
While the present invention has been described with reference to a preferred embodiment, the present invention is not limited to this embodiment, and modifications can be made without departing from the scope of the present invention.
For example, in the embodiment of FIG. 1, the main body 231 of the calibration probe 230 to be calibrated is detachably fixed to the probe support 220, but instead of the probe support 220, a reference body 221 having substantially the same shape is provided. The body 231 of the scanning probe 230 to be calibrated may be detachably fixed, and the measuring element 232 and the reference body 221 may be coupled by the coupling body 300. In short, any method is acceptable as long as the probe of the calibration probe to be calibrated can be accurately moved in a relatively straight line.
[0060]
In the embodiment shown in FIG. 12, the main body 291 of the non-contact scanning probe 290 is detachably fixed to the probe support 220. However, a reference body 221 having substantially the same shape is provided instead of the probe support, The main body 291 of the non-contact scanning probe 290 to be calibrated may be detachably fixed, the coupling body 500 may be fixed to the reference body 221 , and the measurement target 530 may be copied and measured by the non-contact scanning probe 290 to be calibrated. In short, any one can be used as long as the non-contact scanning probe to be calibrated and the coupling body 500 can be accurately moved relative to each other.
Furthermore, the description has been given only for the three-dimensional scanning probe, but a two-dimensional scanning probe used for contour shape measurement may be used.
[0061]
Further, the embodiment has been described in which the scanning probe is provided with a temperature sensor. However, this temperature sensor may be provided in the vicinity of the scanning probe. May be used. Alternatively, calibration of the temperature error may be omitted in the case of a scanning probe that is not so strict in terms of calibration accuracy and does not have a temperature sensor.
In addition, the measurement for actual calibration may be one in which the calibration measuring device performs automatic positioning and automatic measurement by a program in addition to positioning to each axis measurement point by manual operation.
[0062]
【The invention's effect】
As described above, by comparing the output coordinates of the calibration measuring device and the calibration probe to be calibrated, the geometric deviation of the calibration probe to be calibrated can be collected, and the error can be calculated to obtain correction data. The error correction of the scanning probe becomes easy.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a calibration measuring apparatus in a calibration apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a configuration of the calibration apparatus.
FIG. 3 is a view showing a coupling body in the calibration apparatus.
FIG. 4 is a view showing another connector in the calibration apparatus.
FIG. 5 is a flowchart showing a scanning probe calibration method;
FIG. 6 is a schematic diagram showing a rotation error of the scanning probe.
FIG. 7 is another schematic diagram showing a rotation error of the scanning probe.
FIG. 8 is another schematic diagram showing a rotation error of the scanning probe.
FIG. 9 is an explanatory diagram for obtaining a quadrature error of a scanning probe.
FIG. 10 is a flowchart showing an error correction method of the scanning probe.
FIG. 11 is an explanatory diagram showing a method for correcting an instruction error.
FIG. 12 is a diagram showing a configuration of a calibration measurement apparatus in a calibration apparatus according to another embodiment of the present invention.
FIG. 13 is a view showing a coupling body in the calibration apparatus.
FIG. 14 is a diagram illustrating a usage state of a probe in a coordinate measuring machine.
FIG. 15 is a diagram illustrating a configuration example of a scanning probe.
FIG. 16 is a schematic diagram showing a rotation error of the guide mechanism of the scanning probe.
[Explanation of symbols]
200 Calibration apparatus 210 Calibration measurement apparatus 220 Probe support body 230, 290 Scanning probe to be calibrated 260 Drive apparatus 270 Computer 300, 400, 500 Connected body

Claims (12)

In a calibration apparatus for calibrating a scanning probe that detects the amount of displacement in at least two axial directions associated with the contact of the probe with the measurement object and measures the surface property of the measurement object
The calibration device includes a surface plate, a probe support that is erected on the surface plate and removably fixes the scanning probe, a spindle, a scale that detects a displacement position of the spindle , the measuring element, and the measuring device. A coupling body for coupling with the spindle ,
An apparatus for calibrating a scanning probe, wherein the scanning probe is calibrated based on the displacement position and the amount of displacement when the probe is displaced over the entire measuring range . In a calibration apparatus for calibrating a scanning probe that detects the amount of displacement in at least two axial directions associated with the contact of the probe with the measurement object and measures the surface property of the measurement object
The calibration device includes a surface plate, a reference body standing on the surface plate, a spindle for detachably fixing the scanning probe, a scale for detecting a displacement position of the spindle , the measuring element, and the measuring device. and a coupling member for coupling the reference body,
An apparatus for calibrating a scanning probe, wherein the scanning probe is calibrated based on the displacement position and the amount of displacement when the probe is displaced over the entire measuring range . Wherein the coupling element comprises a second connecting member connected to the first connecting member connected to one of said measuring element or the spindle, to the other of the measuring element or the spindle, the first connection calibration apparatus of the scanning probe according to claim 1 or claim 2 and wherein the member second coupling member, characterized in that it is connected so as to have a degree of freedom only in the rotating direction. Calibration apparatus of the scanning probe according to claim 3, characterized in that said first coupling member and the second connecting member which is driven by each other by a wire. Calibration apparatus of the scanning probe according to claim 3 wherein the first coupling member and the second connecting member, characterized in that are interconnected by a spherical bearing. In a calibration apparatus that calibrates a scanning probe that detects a surface property of an object to be measured in a non-contact manner by detecting a displacement amount in at least two axial directions ,
The calibration device includes a surface plate, a probe support body which is erected on the surface plate and detachably fixes the scanning probe, and a spindle which fixes a coupling body provided with a measurement target to be measured by the scanning probe. And a scale for detecting the displacement position of the spindle ,
The scanning probe is calibrated based on the displacement position and the displacement amount when the spindle is moved and the measurement target is displaced over the entire measurement range while measuring the measurement target with the scanning probe. A scanning probe calibration apparatus characterized by the above. In a calibration apparatus that calibrates a scanning probe that detects the amount of displacement in at least two axial directions and measures the surface property of a measurement object in a non-contact manner ,
The calibration device includes a surface plate and a reference body that is fixed on the surface plate and fixes a connection body provided with a measurement target to be measured by the scanning probe.
A spindle for detachably fixing the scanning probe; and a scale for detecting a displacement position of the spindle ,
The scanning probe is calibrated based on the displacement position and the displacement amount when the scanning probe is moved over the entire measurement range while measuring the measurement target with the scanning probe. A scanning probe calibration apparatus characterized by the above. In a scanning probe calibration program for calibrating the scanning probe by using the scanning probe calibration apparatus according to any one of claims 1 to 7 and causing a computer constituting the calibration apparatus to execute the scanning probe calibration program,
A preparation step of inputting calibration conditions including at least each axis measurement range of the scanning probe to the computer ;
In each axis measurement range, an error collecting step of performing a straight measurement and collecting an error based on a displacement amount of each axis by the scanning probe;
An error calculating step for calculating an instruction error, a straightness error, and a rotation error from the errors collected in the error collecting step ;
An orthogonal data collection step for collecting orthogonal data based on the displacement amount of the scanning probe at each of a total of two sets of four points with known distances between the two points in each axis measurement range;
An orthogonal error calculation step for calculating an orthogonal error from the orthogonal data collected in the orthogonal data collection step;
A calibration program for a scanning probe, characterized in that a computer is executed. Separating the orthogonal error from the calculated instruction error in the error calculation step ;
9. The scanning probe calibration program according to claim 8. The error collection in the error collection step covers the measurement range of each axis at predetermined intervals.
  9. The scanning probe calibration program according to claim 8. A least square method is used to calculate the indication error, straightness error, and rotation error in the error calculation step;
  9. The scanning probe calibration program according to claim 8. In the scanning probe calibration method for performing calibration of the scanning probe using the scanning probe calibration apparatus according to any one of claims 1 to 7,
  A preparation step of inputting calibration conditions including at least each axis measurement range of the scanning probe to the calibration device;
  An orthogonal data collection step for collecting orthogonal data based on the displacement amount of the scanning probe at each of a total of two sets of four points with known distances between the two points in each axis measurement range;
  An orthogonal error calculation step for calculating an orthogonal error from the orthogonal data collected in the orthogonal data collection step;
  In each axis measurement range, an error collecting step that performs straight measurement and collects an error based on a displacement amount of each axis by the scanning probe;
  An error calculation step of calculating an instruction error, a straightness error, and a rotation error from the error collected in the error collection step, and further separating the orthogonal error from the instruction error;
  A scanning probe calibration method characterized by comprising:

Images