JP2004042260A - Thermal displacement correcting method for machine tool, and device thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a thermal displacement correcting method for a machine tool wherein thermal displacement is accurately corrected, and also to provide a device thereof. <P>SOLUTION: In the thermal displacement correcting method for a machine tool 1, the temperature changes of a machine body 10 at least at two places where the temperature changes with different time constants in response to the influence of a heating source are detected by temperature sensors S1, S2 , each of the detected temperature changes is compounded to calculate the compounded temperature change having time constant almost same as that of the thermal displacement of the machine tool 1. A machining error is corrected based on the thermal displacement which changes corresponding to the compounded temperature change. <P>COPYRIGHT: (C)2004,JPO

Description

ここで、ｋ：任意の係数（ここではｋ＝１０とした）である。
【００４３】
なお、図５による前記手順と同様にしてＹ軸方向の熱変位の時定数τ_Ｙ＝１．１５〔ｈ〕を算出すると、図６に示すようにＹ軸方向に関する合成温度変化Ｔ_Ｙも同様にして演算できる。
この場合のノーズ温度変化Ｔ_Ｎ及びヘッド温度変化Ｔ_Ｈの各温度混合比Ｍ_ＮＹ，Ｍ_ＨＹは、次式で算出される。

ここで、ｋ：任意の係数（ここではｋ＝１０とした）である。
【００４４】
合成温度演算手段３１では、式（７），（８）で算出された温度混合比Ｍ_ＮＺ，Ｍ_ＨＺと各温度変化Ｔ_Ｎ，Ｔ_Ｈとに基づいて、次式により合成温度変化Ｔ_Ｚを演算している（ステップ１０３）。式（１１）は温度変換式であり、Ｍ_ＮＺ，Ｍ_ＨＺはその係数になる。
Ｔ_Ｚ＝Ｍ_ＮＺ・Ｔ_Ｎ＋Ｍ_ＨＺ・Ｔ_Ｈ＝１．２４×Ｔ_Ｎ＋０．５３×Ｔ_Ｈ　……（１１）
【００４５】
図７の横軸はノーズ温度変化Ｔ_Ｎを、図８の横軸はヘッド温度変化Ｔ_Ｈをそれぞれ示しており、図７，図８の縦軸はＺ軸方向の熱変位を示している。図示するように、それぞれの温度変化Ｔ_Ｎ，Ｔ_ＨとＺ軸熱変位とは、比例などの対応関係はない。
【００４６】
これに対して、図９に示すように、合成温度変化Ｔ_ＺとＺ軸熱変位とは、傾斜α（α＝３．８８）を有する直線４２で代表されるリニアな相関を持つ領域が生じる。この傾斜αは、合成温度変化Ｔ_Ｚから即時応答成分ΔＺ_１を算出する際の比例定数であり、結局次式が成立する。
ΔＺ_１＝α・Ｔ_Ｚ　　　……（１２）
熱変位演算手段３２では、式（１２）を用いて、合成温度変化Ｔ_Ｚに対応して変化する即時応答成分ΔＺ_１を算出する（ステップ１０４）。
【００４７】
次に、遅れ応答成分ΔＺ_２を考慮するか否かを判別し（ステップ１０５）、考慮しない場合には、熱変位演算手段３２で演算した結果に基づいて、補正手段３３で加工誤差を補正する。具体的には、例えば直交座標系の原点位置をオフセットする（ステップ１０６）。
その後、補正を終了させるか否か判別し（ステップ１０７）、終了させる場合にはＭＣ１を停止して（ステップ１０８）、全体の手順が終了する。補正が終了しない場合にはステップ１０２に戻る。
【００４８】
一方、ステップ１０５の判断において、遅れ応答成分ΔＺ_２を考慮する場合には、ヘッド温度センサＳ_２で検出されたヘッド位置の温度を、遅れ温度演算手段３４に入力する。
コラム３等は、その質量が大きく、また主な発熱源である主軸６とも離れているので、温度変化が遅れるヘッド位置よりさらに遅れて温度変化が表われる。この遅れ温度変化は、例えば図９では長時間経過した領域Ｄで、合成温度変化Ｔ_ＺとＺ軸熱変位とのリニアな相関関係に対して、誤差を与えている。
【００４９】
遅れ温度変化を演算する手法としては「ダミー手法」があり、ヘッド温度変化Ｔ_Ｈよりさらに遅れて表われる遅れ温度変化Ｙの挙動を、ダミーの熱容量Ｃを設定して見込む。
具体的には、微分方程式（１３）の近似解で得られる。
Ｃ・ｄＹ／ｄｔ＋Ｙ＝Ｔ_Ｈ　　……（１３）
【００５０】
図１０の横軸は時間、縦軸は遅れ温度変化を示しており、図１０より次式が得られる。
Ｙ＝Ｙ_０＋（ｄＹ_０／ｄｔ＋ｄＹ／ｄｔ）／２・Δｔ　　……（１４）
【００５１】
式（１３）からｄＹ／ｄｔ及びｄＹ_０／ｄｔを算出して、式（１４）に代入すると、遅れ温度変化Ｙを算出する式は、次式（１５）になる。この式（１５）は温度変換式であり、符号Ｃはその係数に相当する。式（１５）で演算される遅れ温度変化も仮想の温度変化であり、先の式（１１）による合成温度変化と同じく創成温度変化の一種である。
Ｙ＝［Ｔ_Ｈ０＋Ｔ_Ｈ＋（Ｃ／Δｔ）・Ｙ_０−Ｙ_０］／［（Ｃ／Δｔ）＋１］…（１５）
ここで、Δｔ：演算インターバル
Ｔ_Ｈ：ヘッド温度変化入力
Ｔ_Ｈ０：前回のヘッド温度変化入力
Ｙ　：遅れ温度変化出力
Ｙ_０：前回の遅れ温度変化出力
Ｃ　：ダミー熱容量
である。
なお、遅れ温度変化出力Ｙの単位は〔℃〕なので、遅れ応答成分ΔＺ_２　を算出するには、温度と熱変位の変換係数を意味する内部補正係数ｅを用いる。
ΔＺ_２＝ｅ・Ｙ　　　　　……（１６）
【００５２】
図１１の横軸は時間、左側の零から上の縦軸はヘッド温度変化、右側の零から下の縦軸は遅れ応答成分である。図中には、ヘッド温度変化の時系列データ４３と、図９においてＺ軸熱変位から直線４２の縦軸の値を差し引いた誤差４４と、ヘッド温度変化の時系列データ４３を用いて式（１５）と式（１６）を介して得た演算結果４５とが、示されている。前記誤差４４と前記演算結果４５は、それぞれ遅れ応答成分ΔＺ_２の実データと演算データに該当する。
即ち、式（１５）に含まれるダミー熱容量Ｃ、及び式（１６）の内部補正係数ｅを適宜選択することにより、演算データを実データに近づけることが可能になり、最良値がそれぞれ決定できる。ここで決まる熱容量Ｃと係数ｅの値は工作機械毎に固有の値であり、主軸の回転数等運転条件が変わってもその値は変わらないので、この作業は一度行なっておけばよい。
例えば、前記ヘッド温度変化の時系列データ４３に基づいて、繰り返し演算により遅れ応答成分ΔＺ_２を算出すれば、Ｃ＝９００，ｅ＝−４として表１の結果を得る。表１の遅れ応答成分ΔＺ_２の値は、コラム３等の遅れ応答成分であり、図９におけるＺ軸熱変位と直線４２との縦軸方向の誤差に相当する。
【００５３】
【表１】

【００５４】
遅れ温度演算手段３４では、ダミー熱容量Ｃが確定した式（１５）を用いて、ヘッド温度変化Ｔ_Ｈに対応する遅れ温度変化Ｙを演算する（ステップ１０９）。次いで、この遅れ温度変化Ｙを、内部補正係数ｅが確定した式（１６）に代入すると、遅れ応答成分ΔＺ_２が得られる（ステップ１１０）。
先に、式（１２）を用いて即時応答成分ΔＺ_１を算出した熱変位演算手段３２では、このようにして算出した遅れ応答成分ΔＺ_２を即時応答成分ΔＺ_１に加算して、Ｚ軸熱変位ΔＺを演算する（ステップ１１１）。
【００５５】
図１２の縦軸は図５に示したものと同じ実測されたＺ軸熱変位であり、横軸は温度センサＳ_１，Ｓ_２で検出した温度データを用いて、ステップ１１１までの手順を経て見積もったＺ軸熱変位ΔＺである。この熱変位ΔＺの演算には、式（２）の全体補正係数“ａ”を１として、即時応答成分ΔＺ_１を、式（１１）と式（１２）で展開した次式を用いている。
ΔＺ＝４．８×Ｔ_Ｎ＋２．１×Ｔ_Ｈ＋ΔＺ_２　　……（１７）
【００５６】
この式（１７）の、ノーズ温度変化Ｔ_Ｎの項の係数、及びヘッド温度変化Ｔ_Ｈの項の係数は、先の式（５）の内部補正係数ｂ，ｃにそれぞれ該当するもので、次式で算出される。これらの係数の値は、機体１０から検出した温度を用いて、即時応答成分を演算する場合の、当該工作機械の熱特性を集約したものになる。
ｂ＝α・Ｍ_ＮＺ＝３．８８×１．２４＝４．８　　　……（１８）
ｃ＝α・Ｍ_ＨＺ＝３．８８×０．５３＝２．１　　　……（１９）
また、式（１７）の遅れ応答成分ΔＺ_２は表１に示した結果を用いている。
【００５７】
図１２では、縦軸の実測されたＺ軸熱変位と、式（１７）で求めた横軸のＺ軸熱変位ΔＺの値とは、４５度の傾きの直線４６上で略一致する。これは両者が同じ値であることを意味しており、したがって、機体１０から検出した比較的少ない温度データを用いた演算により、Ｚ軸熱変位を十分高精度に予測することができる。
このようにして、ステップ１１１で演算されたＺ軸熱変位ΔＺに基づいて補正手段３３で加工誤差を補正することにより熱変位の補正がなされて（ステップ１１２）、工作物９を高精度で切削加工することができる。
その後、補正終了か否かを判別し（ステップ１０７）、終了させる場合にはＭＣ１を停止して（ステップ１０８）、全体の手順が終了する。補正が終了しない場合にはステップ１０２に戻る。
【００５８】
次に、第２実施例の手順を説明する。なお、第２実施例では、熱変位あるいは温度変化の時定数を予め算出しておく必要はない。
図２及び図４に示すように、ＭＣ１を起動して工具７により工作物９の切削加工を開始する（ステップ２０１）。第１の箇所例えばノーズ位置の温度を検出して（ステップ２０２）、この検出信号を第１の遅れ温度演算手段３４ａに入力する。
次いで、先の「ダミー手法」により、ノーズ温度変化Ｔ_Ｎより遅れて表れるＺ軸熱変位の時定数と同じ時定数を有する第１の遅れ温度変化Ｙ_１の挙動を、ダミーの熱容量Ｃ_１を設定して見込む。
【００５９】
先の式（１３），（１４）と同様の展開を行なうと、第１の遅れ温度変化Ｙ_１を算出する式は次式になる。

ここで、Δｔ：演算インターバル
Ｔ_Ｎ：ノーズ温度変化入力
Ｔ_ＮＯ：前回のノーズ温度変化入力
Ｙ_１：第１の遅れ温度変化出力
Ｙ_１０：前回の第１の遅れ温度変化出力
Ｃ_１：ダミー熱容量
である。
【００６０】
図１３に示すように、Ｚ軸熱変位と第１の遅れ温度変化Ｙ_１とは傾斜ｅを有する直線４７で代表されるリニアな相関を持つ領域が生じる。この傾斜ｅは、第１の遅れ温度変化Ｙ_１に対応するＺ軸熱変位を算出する際の比例定数であり、ここで算出されるＺ軸熱変位は、先の式（６）の第１項に該当する。
この第１項の繰り返し演算の結果が、先の図５に示したＺ軸熱変位の時系列データに一致するように、ダミー熱容量Ｃ_１と内部補正係数ｅの値が適宜選択される。ここで決まる熱容量Ｃ_１と係数ｅの値は工作機械毎に固有の値であり、この作業は一度行なっておけばよい。
第１の遅れ温度演算手段３４ａでは、ダミー熱容量Ｃ_１の確定した式（２０）を用いて、ノーズ温度変化Ｔ_Ｎに対応する第１の遅れ温度変化Ｙ_１を演算する（ステップ２０３）。
熱変位演算手段３２では、この遅れ温度変化Ｙ_１を，内部補正係数ｅが確定した式（６）の第１項に代入して、第１の遅れ応答成分を算出する（ステップ２０４）。
【００６１】
次に、第２の遅れ応答成分を考慮するか否かを判別し（ステップ２０５）、考慮しない場合には、熱変位演算手段３２で演算した結果に基づいて、補正手段３３で加工誤差を補正する（ステップ２０６）。
その後、補正を終了させるか否かを判別し（ステップ２０７）、終了させる場合にはＭＣ１を停止して（ステップ２０８）、全体の手順が終了する。補正が終了しない場合にはステップ２０２に戻る。
【００６２】
一方、ステップ２０５の判断において、第２の遅れ応答成分を考慮する場合には、温度センサＳ_３で第２箇所例えばコラム位置の温度変化Ｔ_Ｃを検出して（ステップ２０９）、第２の遅れ温度演算手段３４ｂに入力する。
コラム位置に表れる温度変化Ｔ_Ｃは、例えば図１３の長時間経過した領域Ｄで、第１の遅れ温度変化Ｙ_１とＺ軸熱変位とのリニアな相関関係に誤差を与える。前述の「ダミー手法」を再度用いて、コラム温度変化Ｔ_Ｃより第２の遅れ温度変化Ｙ_２の挙動を、ダミーの熱容量Ｃ_２を設定して見込む。
第２の遅れ温度変化Ｙ_２を算出する式は次式になる。

ここで、Δｔ：演算インターバル
Ｔ_Ｃ：コラム温度変化入力
Ｔ_ＣＯ：前回のコラム温度変化入力
Ｙ_２：第２の遅れ温度変化出力
Ｙ_２０：前回の第２の遅れ温度変化出力
Ｃ_２：ダミー熱容量
である。
【００６３】
なお、この第２の遅れ温度変化Ｙ_２に対応するＺ軸熱変位は、温度と熱変位の変換係数ｆを含んだ先の式（６）の第２項に該当する。
この第２項の繰り返し演算の結果が、図１３において、Ｚ軸熱変位から直線４７を差し引いた誤差に一致するように、ダミー熱容量Ｃ_２と係数ｆの値が適宜選択される。ここで決まる熱容量Ｃ_２と係数ｆの値は工作機械毎に固有の値であり、この作業は一度行なっておけばよい。
第２の遅れ温度演算手段３４ｂでは、ダミー熱容量Ｃ_２が確定した式（２１）を用いて、コラム温度変化Ｔ_Ｃに対応する第２の遅れ温度変化Ｙ_２を演算する（ステップ２１０）。次いで、内部補正係数ｆが確定した式（６）の第２項に第２の遅れ温度変化Ｙ_２を代入して、第２の遅れ応答成分を算出する（ステップ２１１）。
先に式（６）の第１項に該当する第１の遅れ応答成分を算出した熱変位演算手段３２では、このようにして算出した第２の遅れ応答成分を加算して、Ｚ軸熱変位ΔＺ_２を演算する（ステップ２１２）。
【００６４】
ステップ２１２で演算されたＺ軸熱変位ΔＺ_２に基づいて補正手段３３で加工誤差を補正することにより熱変位に対する補正がなされて（ステップ２１３）、工作物９を高精度で加工することができる。
その後、補正終了か否かを判別し（ステップ２０７）、終了させる場合にはＭＣ１を停止して（ステップ２０８）、全体の手順が終了する。補正が終了しない場合にはステップ２０２に戻る。
【００６５】
図１４乃至図１７はＭＣ１を実機運転した場合の実測データである。
図１４はＺ軸熱変位のデータを示すグラフ、図１５はＺ軸熱変位の他のデータを示すグラフ、図１６はＹ軸熱変位のデータを示すグラフ、図１７はＹ軸熱変位の他のデータを示すグラフである。
いずれも主軸６の回転数Ｓは、Ｓ＝１０，０００〔ｍｉｎ　^−１〕であり、図１４及び図１６は連続運転の場合を示している。図１５及び図１７は、タイムチャート線５１に示すように、約５０分間回転、１０分間停止（途中一回７０分間の停止あり）の場合の熱変位を示している。図１４乃至図１７の横軸は時間である。
【００６６】
図１４及び図１６に示すように、図中実線４８，５２で示す補正前の熱変位は最大約４０〔μｍ〕であった。これに対して、本発明では、補正後の熱変位の目標値を零に近づけることができる。即ち、本発明による熱変位の補正を行なった場合は、図中破線４９，５３で示すように±５〔μｍ〕以下にまで残留熱変位を小さくすることができる。
なお、補正前の熱変位が１００〔μｍ〕以上の場合も、本発明によれば補正後の熱変位を±５〔μｍ〕以下にまで小さくできることが確認されている。このように、本発明は熱変位の補正を高精度で行なうことができる。
また、図１５及び図１７に示すように断続運転を行なった場合も、本発明によれば、破線５０，５４に示すように補正後の残留熱変位を±５〔μｍ〕以下にすることができる。
【００６７】
（第３実施例）
図１８乃至図２８は第３実施例を説明するための図である。
例えば、図１８に示すように、ＭＣ１ａは、工具７が装着される主軸６と、主軸受２０及び上部軸受（他の軸受）２２を介して主軸６を回転自在に支持する主軸頭５ａとを備えている。主軸６を回転駆動するビルトインモータ２１は、両軸受２０，２２の間に配設されて、両軸受とともに主軸頭５ａに内蔵されている。主軸受２０は主軸６を中心軸方向に位置決めし、上部軸受２２は熱変位で伸縮する主軸６を中心軸方向に摺動可能に保持している。したがって、上部軸受２２及びモータ２１が回転により発熱しても、主軸６は上方に伸びるので工具７には影響を与えない。
その結果、ＭＣ１ａの場合には主軸受２０のみを発熱源と考えて熱変位補正をすればよいことになる。主軸受２０の温度変化を検出する温度検出手段としてのヘッド温度センサＳ_２を、主軸頭５ａに取付けている。
なお、発熱源である主軸受２０による温度変化が表われる場所であれば、ヘッド位置以外の例えばノーズまたはコラムの温度を検出してもよい。また、主軸が工作物を把持するタイプの工作機械の場合であってもよい。
なお、第１，第２実施例と同一または相当部分には同一符号を付してその説明を省略する。
【００６８】
ここで、第３実施例における熱変位補正の原理を説明する。
第１，第２実施例と同様にＺ軸方向の補正を例にとって説明する。第３実施例におけるＺ軸方向の熱変位の演算式は次式で示される。
ΔＺ＝ａ・（ΔＺ_３＋ΔＺ_４）　　……（２２）
ここで、ΔＺ　：Ｚ軸熱変位
ａ　：全体補正係数（式（２）のものに同じ）
ΔＺ_３：Ｚ軸熱変位の創成変位成分
ΔＺ_４：Ｚ軸熱変位の遅れ応答成分
である。
即ち、演算式（２２）は、温度変化から熱変位の時定数と同じ時定数を有する創成温度変化を基に演算した創成変位成分ΔＺ_３と、温度変化に対し遅れを伴って熱変位が表れる遅れ応答成分ΔＺ_４とを含んでいる。
ここで扱われる温度変化は、各温度センサから出力される温度と基準温度との差で算出する。基準温度には、先の第１，第２実施例で考慮されたものが同じく採用される。
【００６９】
なお、本発明は、式（２３）に示すように創成変位成分ΔＺ_３のみに基づく演算式を使用することもできる。
ΔＺ＝ａ・ΔＺ_３　　　　……（２３）
創成変位成分ΔＺ_３は、次式により算出される。
ΔＺ_３＝ｇ・Ｙ_３　　　　……（２４）
ここで、Ｙ_３：創成温度変化〔℃〕
ｇ　：内部補正係数〔±μｍ／℃〕
である。
【００７０】
第３実施例で使用される式（２４）は、一箇所に設置された温度センサの出力により機体１０の熱変位を演算する式である。そして、ヘッド温度センサＳ_２で検出された温度の温度変化Ｔを展開して得た創成温度変化Ｙ_３から創成変位成分ΔＺ_３を演算することになる。
なお、温度センサの設置箇所は少なくとも一箇所あればよいが、発熱源の数に応じて適宜追加される。また、温度センサの設置箇所は、発熱源の発熱の影響を受ける箇所ならば、主軸頭５ａ以外の場所であってもよい。
【００７１】
一方、遅れ応答成分ΔＺ_４を演算する式は下記の通りである。
ΔＺ_４＝ｈ・Ｙ_４　　　　……（２５）
ここで、Ｙ_４：遅れ温度変化〔℃〕
ｈ　：内部補正係数〔±μｍ／℃〕
である。
式（２５）では、ヘッド温度センサＳ_２で検出された温度の温度変化Ｔに遅れを見込んで演算した遅れ温度変化Ｙ_４から、遅れ応答成分ΔＺ_４を演算することになる。
なお、温度センサの設置箇所は少なくとも一箇所あればよいが、発熱源の数に応じて適宜追加される。
【００７２】
図１８は、本発明の第３実施例を示すブロック図である。
第３実施例の熱変位補正装置１２ｂは、温度センサＳ_２で検出された温度を展開して熱変位の時定数と略同じ時定数を有する仮想の位置Ｐ_１における温度変化を演算する創成温度演算手段３１ａと、この創成温度演算手段３１ａで演算された創成温度変化に対応して変化する創成変位成分ΔＺ_３を演算する熱変位演算手段３２と、この熱変位演算手段３２で算出された熱変位に基づいて加工誤差を補正する補正手段３３とを備えている。
【００７３】
好ましい態様として、熱変位補正装置１２ｂは遅れ温度演算手段３４を更に備えている。遅れ温度演算手段３４は、温度センサＳ_２　で検出された温度の温度変化より遅れて表れる遅れ温度変化を、前記温度変化に遅れを見込んで演算する。
熱変位演算手段３２は、遅れ温度演算手段３４で演算された遅れ温度変化に対応して変化する遅れ応答成分ΔＺ_４を算出し、先の創成変位成分ΔＺ_３に加算する。この加算された熱変位に基づいて、補正手段３３で加工誤差が補正され、その信号が出力される。
本第３実施例においては、先の第１，第２実施例と同一又は相当機能部分の説明は省略する。
【００７４】
以下に、第３実施例の具体的な手順を図１９乃至図２８により説明する。
図１９は第３実施例の動作を示すフローチャート、図２０は、Ｚ軸熱変位と、ヘッド位置で検出された温度の温度変化Ｔの代表例（サンプル温度変化）との経時変化を示すグラフ、図２１はサンプル温度変化とＺ軸熱変位との標準的関係を示すグラフである。
図２２は、サンプル温度変化と、サンプル温度変化より時定数の小さい温度変化“Ａ”と、サンプル温度変化より時定数の大きい温度変化“Ｂ”とを示すグラフである。図２３は図２２に示した温度変化に加えて、サンプル温度変化を用いて創成した創成温度変化Ｙ_３Ａ，Ｙ_３Ｂ（図中「○」印）を示すグラフである。図２４は創成温度変化に対するＺ軸熱変位を示すグラフ、図２５は、Ｚ軸熱変位が遅れ応答成分を含んでいる場合の、創成温度変化に対するＺ軸熱変位を示すグラフである。
図２６はサンプル温度変化と、サンプル温度変化より遅れて表れる遅れ温度変化Ｃと、サンプル温度変化を用いて創成した遅れ温度変化Ｙ_４（図中「○」印）とを示すグラフ、図２７は遅れ温度変化に対する遅れ応答成分を示すグラフである。図２８は創成温度変化及び遅れ温度変化から見積もった熱変位に対するＺ軸熱変位を示すグラフである。
【００７５】
第３実施例では、予めＺ軸方向の熱変位を検出する。これと同時に、ヘッド温度センサＳ_２で検出される温度の温度変化のデータに基づいて、それぞれの時定数を算出しておく。時定数を算出する手順は、先の第１実施例においてＺ軸方向の熱変位の時定数を算出したものと同じである。
図２０は、主軸が一定回転（回転数Ｓ＝１０，０００〔ｍｉｎ　^−１〕）の下で、ヘッド温度センサＳ_２で検出した温度の温度変化Ｔの代表例（サンプル温度変化）と、２つの例に係るＺ軸熱変位（熱変位“Ａ，Ｂ”）の時系列データを示している。
Ｚ軸熱変位の時定数がサンプル温度変化Ｔの時定数τ_Ｓより小さい場合、即ち熱変位が早く表れる場合には、熱変位“Ａ”（時定数τ_Ａ）となる。一方、前記熱変位時定数が前記時定数τ_Ｓより大きい場合、即ち、ゆっくりと熱変位が表れる場合には、熱変位“Ｂ”（時定数τ_Ｂ）となる。したがって、ＭＣ１ａの熱変位特性により、実際には熱変位“Ａ，Ｂ”のうちどちらか一方のデータになる。
ここで抽出されたＺ軸熱変位の時定数とサンプル温度変化Ｔの時定数との値のバランスは、工作機械毎に固有の熱特性を表すものであり、主軸回転数等運転条件が変わっても変化が少ない。したがって、一度両時定数を算出する作業を行なっておけばよい。
【００７６】
次に、図１８及び図１９に示すように、ＭＣ１ａを起動して工具７により工作物９の切削を開始する（ステップ３０１）。また、ヘッド位置の温度を検出して（ステップ３０２）、創成温度演算手段３１ａに入力する。
【００７７】
しかしながら、例えば図２０のデータに基づいて、サンプル温度変化Ｔと熱変位“Ａ”との関係、及びサンプル温度変化Ｔと熱変位“Ｂ”との関係を表すと、図２１に示すように、それぞれ弓状曲線５５，５６になる。
即ち、これらサンプル温度変化ＴとＺ軸熱変位とが単純なリニアの関係にならないので、ヘッド位置から随時検出した温度の温度変化から直ちに熱変位を見込むことができない。
【００７８】
そこで、「リニアライズ手法」により、ヘッド位置から検出した温度の温度変化を用いて、熱変位の時定数と略同じ時定数を有する創成温度変化を演算する。図２２は、時定数をτ_Ｓとするサンプル温度変化５７と、熱変位“Ａ”と同じ時定数τ_Ａを有する温度変化“Ａ”の模擬例５８と、熱変位“Ｂ”と同じ時定数τ_Ｂを有する温度変化“Ｂ”の模擬例５９とを示している。
曲線５７乃至５９で示す温度変化は、いずれも値Ｔ_ｍａｘで飽和するようになっており、各時定数の一例を以下に示す。
τ_Ａ＝　５〔ｍｉｎ〕
τ_Ｓ＝１０〔ｍｉｎ〕
τ_Ｂ＝１５〔ｍｉｎ〕
【００７９】
また、各温度変化５７乃至５９の挙動関係は、下記の微分方程式（２６）乃至（２８）でそれぞれ表現できる。
τ_Ｓ・ｄＴ／ｄｔ＋Ｔ＝Ｘ　　　……（２６）
τ_Ａ　・ｄＹ_３Ａ／ｄｔ＋Ｙ_３Ａ＝Ｘ　……（２７）
τ_Ｂ・ｄＹ_３Ｂ／ｄｔ＋Ｙ_３Ｂ＝Ｘ　……（２８）
なお、式（２７），（２８）は微分方程式の一般式（２９）で表わすことができる。
τ_Ｚ・ｄＹ_３／ｄｔ＋Ｙ_３＝Ｘ　……（２９）
ここで、Ｔ　：ヘッド温度センサＳ_２で検出される温度変化〔℃〕
Ｘ　：発熱部の温度変化〔℃〕
Ｙ_３：創成温度変化〔℃〕
Ｙ_３Ａ：温度変化“Ａ”を創成する創成温度変化〔℃〕
Ｙ_３Ｂ：温度変化“Ｂ”を創成する創成温度変化〔℃〕
τ_Ｚ：Ｚ軸熱変位の時定数〔ｍｉｎ〕
である。
【００８０】
式（２６）によれば、サンプル温度変化Ｔから発熱部の温度変化Ｘが分かるので、この値Ｘを式（２９）に代入する。すると、サンプル温度変化の時定数τ_Ｓとは異なった時定数τ_Ｚを有する創成温度変化Ｙ_３が得られる。時定数τ_Ｚは、ＭＣ１ａの熱特性で定まるものであり、工作機械毎に固有の値である。
【００８１】
実際に創成温度演算手段３１ａ（図１８）で演算する場合には、微分方程式（２６），（２９）をそれぞれ離散化した次式（３０），（３１）の繰り返し演算で解を算出することになる（ステップ３０３）。
Ｘ　＝τ_Ｓ・（Ｔ−Ｔ_０）／Δｔ＋Ｔ_０　　……（３０）
Ｙ_３＝（２Ｘ＋２τ_Ｚ・Ｙ_３０／Δｔ−Ｙ_３０）／（２τ_Ｚ／Δｔ＋１）……（３１）
ここで、Δｔ：演算インターバル〔ｍｉｎ〕
Ｔ_０：前回のサンプル温度変化Ｔ入力〔℃〕
Ｙ_３０：前回の創成温度変化Ｙ_３出力〔℃〕
である。
【００８２】
図２３は、サンプル温度変化Ｔ及び温度変化“Ａ，Ｂ”に加えて、式（３０），（３１）によりサンプル温度変化Ｔを用いて創成した創成温度変化Ｙ_３　（具体的には、創成温度変化Ｙ_３Ａ又はＹ_３Ｂ）を「○」印で表示している。この「○」印は、測定間隔即ち演算インターバルΔｔが１．０〔ｍｉｎ〕　の場合を示している。
このように式（３０），（３１）を用いた繰り返し演算により、任意の時定数を有する熱変位と略同じ時定数を有する創成温度変化を創成することができる。
【００８３】
創成温度変化Ｙ_３は、熱変位と同じ時定数を有しているので、図２４の直線６０に示すように、Ｚ軸熱変位とリニアの関係になる。直線の傾きｇが、創成温度変化Ｙ_３と熱変位の相関を表している。熱変位演算手段３２では、式（２４）を用いて創成温度変化Ｙ_３から熱変位（即ち、創成変位成分ΔＺ_３）を算出する（ステップ３０４）。
【００８４】
以上述べたように、リニアライズ手法は、基本的には時定数の小さい敏感な熱変位を、例えば発熱源から離れたヘッド位置で検出した温度の温度変化Ｔから見込むものである。この手法により算出される創成変位成分ΔＺ_３は、式（２）の即時応答成分ΔＺ_１に相当するものであり、この手法単独でも精度のよい熱変位補正ができる。
次に、遅れ応答成分ΔＺ_４を考慮するか否かを判別し（ステップ３０５）、考慮しない場合には、熱変位演算手段３２で演算した結果に基づいて、補正手段３３で加工誤差を補正する（ステップ３０６）。
その後、補正を終了するか否かを判別し（ステップ３０７）、終了させる場合にはＭＣ１ａを停止して（ステップ３０８）、全体の手順が終了する。補正が終了しない場合にはステップ３０２に戻る。
【００８５】
一方、ステップ３０５の判断において遅れ応答成分ΔＺ_４を考慮する場合には、ヘッド温度センサＳ_２で検出されたヘッド位置の温度を、遅れ温度演算手段３４に入力する。
コラム３等は、その質量が大きく、また主な発熱源である主軸６とも離れているので、温度変化が遅れるヘッド位置よりさらに遅れて温度変化が表れる。この遅れ温度変化は、第１及び第２実施例の図９，図１３で示したものと同様に、図２５の長時間経過した領域Ｄで、創成温度変化Ｙ_３とＺ軸熱変位とのリニアな相関に誤差を与える。
【００８６】
遅れ応答成分ΔＺ_４を考慮する場合には、前述の「ダミー手法」を用いて、ヘッド温度センサＳ_２で検出した温度の温度変化Ｔより遅れて表れる遅れ温度変化Ｙ_４の挙動を、ダミーの熱容量Ｃ_４を設定して見込む。
遅れ温度変化Ｙ_４を算出する式は次式になる。

ここで、Δｔ：演算インターバル〔ｍｉｎ〕
Ｔ　：ヘッド温度センサＳ_２で検出される温度の温度変化入力
Ｔ_０：前回の温度変化Ｔ入力〔℃〕
Ｙ_４：遅れ温度変化出力〔℃〕
Ｙ_４０：前回の遅れ温度変化Ｙ_４出力〔℃〕
Ｃ_４：ダミー熱容量〔ｍｉｎ〕
である。
【００８７】
図２６は、先のヘッド温度センサＳ_２で検出された温度の温度変化Ｔの代表例であるサンプル温度変化５７と、遅れ応答成分と同じ熱的挙動を示す遅れ温度変化の模擬例６１（温度変化Ｃ、時定数τ_Ｃ）とを示している。更に、図２６には、式（３２）によりサンプル温度変化Ｔを用いて創成した遅れ温度変化Ｙ_４を、「○」印で表示している。
温度変化Ｃは長時間経過すると、飽和値Ｔ_ｍａｘでサンプル温度変化Ｔに等しくなることを前提にしており、「○」印は、測定間隔即ち演算インターバルΔｔが１．０〔ｍｉｎ〕の場合を示している。
このように、式（３２）の熱容量Ｃ_４を適宜選択した繰り返し演算により、任意の時定数τ_Ｃ（τ_Ｃ＞τ_Ｓ）の温度変化Ｃと略同じ温度変化の挙動をする遅れ温度変化Ｙ_４を創成することができる。
【００８８】
この遅れ温度変化Ｙ_４と遅れ応答成分とは、図２７に示すようにリニアの関係の直線６２になるので、先の式（２５）が成立する。実際には、温度変化Ｃを抽出する手順を踏む必要は特にはない。例えば、サンプル温度変化Ｔを用いて、式（３２）のダミー熱容量Ｃ_４及び式（２５）の内部補正係数ｈを適宜選択した繰り返し演算結果が、先の図２５におけるＺ軸熱変位（領域Ｄを含む線）から直線６０を差し引いた誤差に一致するように、熱容量Ｃ_４と係数ｈの最良値が決定される。ここで決まる熱容量Ｃ_４と係数ｈの値は工作機械毎に固有の値であり、この作業は一度行なっておけばよい。
遅れ温度演算手段３４では、ダミー熱容量Ｃ_４が確定した式（３２）を用いて、ヘッド温度センサＳ_２で検出される温度の温度変化Ｔに対応する遅れ温度変化Ｙ_４を演算する（ステップ３０９）。次いで、この遅れ温度変化Ｙ_４を、内部補正係数ｈが確定した式（２５）に代入すると、遅れ応答成分ΔＺ_４が得られる（ステップ３１０）。
先に、式（２４）を用いて創成変位成分ΔＺ_３を算出した熱変位演算手段３２では、このようにして算出した遅れ応答成分ΔＺ_４を創成変位成分ΔＺ_３に加算して、Ｚ軸熱変位ΔＺを演算する（ステップ３１１）。
【００８９】
図２８の縦軸は実測されたＺ軸熱変位であり、横軸はサンプル温度変化Ｔを用いてステップ３１１までの手順を経て見積もったＺ軸熱変位ΔＺである。この熱変位ΔＺの演算には、全体補正係数“ａ”を１とした次式を用いている。
ΔＺ＝ｇ・Ｙ_３＋ｈ・Ｙ_４　　……（３３）
図２８に示す実測された縦軸のＺ軸熱変位と、式（３３）で求めた横軸のＺ軸熱変位ΔＺの値とは、４５°の傾きの直線６３上で略一致する。これは両者が同じ値であることを意味している。
したがって、機体１０の主軸受２０の発熱の影響を受ける箇所に設置したヘッド温度センサＳ_２の温度データにより、Ｚ軸熱変位を十分高精度に予測することができる。
【００９０】
このようにして、ステップ３１１で演算されたＺ軸熱変位ΔＺに基づいて補正手段３３で加工誤差を補正することにより熱変位補正がなされて（ステップ３１２）、工作物９を高精度で切削加工することができる。
その後、補正終了か否かを判別し（ステップ３０７）、終了させる場合にはＭＣ１ａを停止して（ステップ３０８）、全体の手順が終了する。補正が終了しない場合にはステップ３０２に戻る。
【００９１】
（第４実施例）
次に、リニアライズ手法を応用した第４実施例を、図２９乃至図３２により説明する。
前記第１乃至第３実施例では、主として、熱変位の原因となる発熱源の数が１個の例を説明したが、本発明では、熱変位の原因となる発熱源が複数あって、且つ発熱源の影響が互いに独立していると想定される場合にも、展開可能な式の構成になっている。
本第４実施例は、複数の発熱源（いわゆる多熱源）が互いに影響し合いながら熱変位を発生させる工作機械の場合であり、先の各実施例では説明されていなかった部分を補足する。
【００９２】
図２９はブロック図、図３０は主軸台断面図である。
ＮＣ旋盤６４は、機体としての主軸台６６を発熱源とする工作機械である。図示するように、ＮＣ旋盤６４は、チャック６５及び爪６７を介して工作物６８を把持する主軸６９と、主軸台６６と、ビルトインモータ７０とを備えている。主軸台６６は、主軸６９を軸支する加工位置側の前軸受７１及び反加工位置側の後軸受７２を介して主軸６９を回転自在に支持している。
ロータ７０ａを含むモータ７０は、前，後の軸受７１，７２の間に配設されるとともに主軸台６６に内蔵されて、主軸６９を回転駆動する。
後軸受７２には、主軸６９を中心軸方向０_１に対して位置決めするアンギュラ玉軸受が使用されている。熱変位で伸縮する主軸６９は、前軸受７１内を中心軸方向Ｏ_１に伸縮可能になっている。工作物６８に近い前軸受７１には大きな荷重がかかるので、この荷重に耐えて切削性能を向上させるために、前軸受７１には定格荷重の大きい複列円筒ころ軸受が使用されている。
なお、ここで示した構造は、主軸６９をベルトで駆動する構造のものと比べて、ビルトインモータ７０で主軸６９を直接駆動しているので高速回転が可能である。また、主軸６９の振動を抑えることができるので高精度な切削加工ができる。
【００９３】
ＮＣ旋盤６４を起動してビルトインモータ７０のロータ７０ａを回転させ、前，後の軸受７１，７２及びロータ７０ａがそれぞれ発熱すると、主軸６９は前方（図中右方）に伸びて工作物６８を中心軸方向Ｏ_１に移動させて加工精度を低下させることになる。
そのため、熱変位補正装置１２ｂでは、発熱源となる前，後の軸受７１，７２，及びモータ７０のステータの各近傍にそれぞれ位置するように３本の温度センサｓ_１，ｓ_２，　ｓ_３を、主軸台６６に取付けている。
【００９４】
このようにＺ軸方向（即ち、中心軸方向Ｏ_１）の熱変位に影響する発熱源が複数ある場合も、例えば前記リニアライズ手法を応用することによって高精度な熱変位補正ができる。
温度検出手段としての温度センサｓ_１，ｓ_２，　ｓ_３の各出力信号は、回路３６ａ，３６ｂ，３６ｃを介してＡ／Ｄ変換器１３に入力し、Ａ／Ｄ変換器１３からの出力信号は、創成温度演算手段３１ａ及び遅れ温度演算手段３４に入力される。なお、その他の構成については第３実施例と同様であるので、説明を省略する。
【００９５】
多熱源の場合、各発熱源の影響による熱変位を、熱変位演算手段３２により個別に演算する。Ｚ軸方向の総合熱変位Δｚは次の一般式で示される。
Δｚ＝ａ・（Δｚ_１＋Δｚ_２＋……＋Δｚ_ｎ）　……（３４）
Δｚ_１＝β_１・Ｙ_Ａ１＋γ_１・Ｙ_Ｂ１
Δｚ_２＝β_２・Ｙ_Ａ２＋γ_２・Ｙ_Ｂ２
…
…
Δｚ_ｎ＝β_ｎ・Ｙ_Ａｎ＋γ_ｎ・Ｙ_Ｂｎ
ここで、Δｚ：Ｚ軸総合熱変位
Δｚ_１乃至Δｚ_ｎ：第１番目乃至第ｎ番目の発熱源によるＺ軸方向
の熱変位
ａ　：全体補正係数（式（２）のものに同じ）
β_１乃至β_ｎ：創成温度変化に係る内部補正係数
γ_１乃至γ_ｎ：遅れ温度変化に係る内部補正係数
Ｙ_Ａ１乃至Ｙ_Ａｎ：創成温度変化
Ｙ_Ｂ１乃至Ｙ_Ｂｎ：遅れ温度変化
である。
したがって、第３実施例の式（３３）は、式（３４）の第１項の熱変位Δｚ_１に相当する。また、第３実施例では発熱源が１個なので、式（３４）の第２項以下の項を零として演算したことになる。
【００９６】
本第４実施例では発熱源が３個なのでｎ＝３となり、式（３４）は以下のように展開できる。
Δｚ＝ａ・（Δｚ_１＋Δｚ_２＋Δｚ_３）　……（３５）
Δｚ_１＝Ｋ_１・Δｚ
Δｚ_２＝Ｋ_２・Δｚ　　　　　　　　……（３６）
Δｚ_３＝Ｋ_３・Δｚ
Ｋ_１＝Ｐ・Ｔ_１／（Ｐ・Ｔ_１＋Ｑ・Ｔ_２＋Ｒ・Ｔ_３）
Ｋ_２＝Ｑ・Ｔ_２／（Ｐ・Ｔ_１＋Ｑ・Ｔ_２＋Ｒ・Ｔ_３）　…（３７）
Ｋ_３＝Ｒ・Ｔ_３／（Ｐ・Ｔ_１＋Ｑ・Ｔ_２＋Ｒ・Ｔ_３）
ここで、Ｔ_１：温度センサｓ_１で検出された温度の温度変化
Ｔ_２：温度センサｓ_２で検出された温度の温度変化
Ｔ_３：温度センサｓ_３で検出された温度の温度変化
Ｐ，Ｑ，Ｒ：内部補正係数
である。
【００９７】
前記三つの式（３７）における内部補正係数Ｐ，Ｑ，Ｒは、各温度変化Ｔ_１乃至Ｔ_３の重みを意味しており、３回以上の発熱条件を変えたテストにおけるサンプル温度変化の飽和値の違いから下式により決定される。
ここで決まる係数Ｐ，Ｑ，Ｒの値のバランスは、工作機械のタイプ毎に固有の熱特性を表すものであり、主軸回転数等の運転条件が変わっても変化が少ないので、この作業は一度行なっておけばよい。
Ｐ・Ｔ_１Ｓ＋Ｑ・Ｔ_２Ｓ＋Ｒ・Ｔ_３Ｓ＝Δｚ　……（３８）
ここで、Ｔ_１Ｓ：サンプル温度変化Ｔ_１の飽和値
Ｔ_２Ｓ：サンプル温度変化Ｔ_２の飽和値
Ｔ_３Ｓ：サンプル温度変化Ｔ_３の飽和値
である。
【００９８】
内部補正係数Ｐ，Ｑ，Ｒの値を式（３７）に代入することによって、係数Ｋ_１，Ｋ_２，Ｋ_３の値が定まるので、三つの式（３６）は図３１のように表現できる。この結果、発熱源７１，７２，７０ａの影響による各熱変位Δｚ_１，Δｚ_２，Δｚ_３と各サンプル温度変化との相関関係は、リニアライズ手法等によってリニアにすることが可能になる。図３１は各発熱源のＺ軸熱変位を示すグラフである。
ＮＣ旋盤など旋削工作機械は、通常は熱容量が小さいためＭＣと比べて熱変位が敏感に表われるので、温度センサの取付け位置の制約のない第３実施例の手法は特に有効である。
また、第３実施例の手法を用いれば、一つの発熱源につき一個の温度センサを設ければよいので、多熱源を有するＮＣ旋盤６４等の場合に温度センサの数を減らすことができる。
【００９９】
図３２は、リニアライズ手法とダミー手法を組合せてＮＣ旋盤６４を実機運転した場合の、Ｚ軸熱変位の実測データを示すグラフである。
図示するように、図中実線７３，７４で示す補正前の熱変位は約７０〔μｍ〕であった。これに対して、本発明に係る熱変位の補正を行なった場合は、図中破線７３ａ，７４ａで示すように、熱変位は±１０〔μｍ〕以下にまで小さくなる。なお、図中の符号Ｓは主軸回転数を意味する。
【０１００】
（第５実施例）
次に、第５実施例を図３３乃至図３７により説明する。
第５実施例では、本発明に係る前述の各手法のいずれかを、複数の主軸を有する工作機械に適用して熱変位補正を行なっている。
主軸は工作物及び工具のいずれか一方を把持している。この主軸の温度を温度調節装置により調節して各主軸の熱変位を略均等にし、温度検出手段により少なくとも一つの主軸について機体の温度変化を検出して、熱変位補正を行なっている。
なお、第５実施例において、前記各実施例と同一又は相当部分には同一符号を付してその説明を省略する。
【０１０１】
例えば、図３３に示す工作機械は多軸ヘッドを有する立形のＭＣ７５であり、４個の工作物を４本の工具で同一形状に同時加工するのに使用される。ベッド７６に固定され床面上に立設しているコラム７７の上部には、Ｙ軸方向即ち水平方向に向けて配設されたクロスレール７８が固定されている。
クロスレール７８には、サドル７９がＹ軸方向に移動可能に取付けられており、クロスレール７８に設けられたＹ軸サーボモータ８０によりサドル７９は往復移動する。
サドル７９には主軸頭８１がＺ軸方向に移動可能に取付けられている。主軸頭８１は、サドル７９に設けられたＺ軸サーボモータ８２によりサドル７９に対してＺ軸方向に往復移動し、クロスレール７８に対してはサドル７９とともにＹ軸方向に往復移動する。
主軸頭８１には、Ｚ軸方向を向いている複数（例えば４本）の主軸６ａ乃至６ｄが並設されており、各主軸の先端には工具７ａ乃至７ｄがそれぞれ装着されている。
【０１０２】
ベッド７６上には、主軸と同数（例えば４個）の工作物９ａ乃至９ｄを載置するためのテーブル８３が、Ｘ軸方向に移動可能に取付けられている。テーブル８３は、ベッド７６に設けられたＸ軸サーボモータ８４により往復移動する。
コラム７７の横には、冷却油供給装置８５が設置されている。この冷却油供給装置８５は、主軸６ａ乃至６ｄの温度を調節して各主軸の熱変位を略均等にするための温度調節装置としての主軸冷却装置を構成している。
【０１０３】
図３４は主軸冷却装置８７の説明図である。
図示するように、主軸６ａ乃至６ｄの軸受近傍には、冷却油を流して軸受を冷却するための流路８５ａ乃至８５ｄがそれぞれ形成されている。冷却油供給装置８５から供給された冷却油は、配管８８ａ乃至８８ｄをそれぞれ流れ、手動又は自動で操作される流量調整弁８９ａ乃至８９ｄで流量を調整される。これにより、各主軸６ａ乃至６ｄの温度が個別に調節される。冷却油は、流路８５ａ乃至８５ｄを流れ、主軸６ａ乃至６ｄの軸受を冷却したのち、供給装置８５に戻ってここで冷却されて再び循環使用される。
なお、流量調節の代わりに又はこれに加えて冷却油温度を供給装置８５で調節することにより、主軸温度を主軸毎に調節するようにしてもよい。また、冷却油の代わりにクーラント（切削油剤）又は水を使用してもよい。
【０１０４】
ところで、４本の主軸が回転した場合の熱変位を比較すると、図中左右両側の主軸６ａ，６ｄの熱変位の方が、内側の主軸６ｂ，６ｃよりも通常は小さい。これは各主軸の取付け場所の違いによるためである。即ち、両側の主軸６ａ，６ｄの軸受で発生した熱は、主軸頭８１に速やかに伝導して、この主軸６ａ，６ｄの温度上昇を抑えるからである。
そこで、熱変位の最も小さい代表主軸６ａの発熱部近傍に位置するように主軸頭８１に温度検出手段としてのノーズ温度センサＳ_１を取付けている。このセンサＳ_１で機体８６の温度変化を検出することにより、前記各実施例における熱変位補正装置１２，１２ａ，１２ｂを用いた手法で、代表主軸６ａに対する熱変位補正を行なっている。
そして、流量調整弁８９ａ乃至８９ｄで流量調節（及び／又は、油の温度調節）をして各主軸を冷却することによって、他の３本の主軸６ｂ乃至６ｄの熱変位を代表主軸６ａの熱変位に略一致させ、これにより、主軸間の熱変位のばらつきをなくしている。
なお、代表主軸６ａは熱変位が最小なので、この主軸６ａの冷却はしないか又はわずかに冷却し、他の主軸６ｂ乃至６ｄの冷却油を調節すれば、供給装置８５からの油の供給量が全体として少なくなるので好ましい。
【０１０５】
図３５は、主軸冷却装置８７に代えて主軸加熱装置９０を温度調節装置として用いた場合を示している。
図示するように、主軸６ａ乃至６ｄの近傍に、ヒータ９１ａ乃至９１ｄなど加熱部材をそれぞれ配設している。各ヒータ９１ａ乃至９１ｄに流れる電流を電流制御装置９１により個別に制御することにより、ヒータ９１ａ乃至９１ｄの発熱量を調節している。電流制御の方が、冷却油の量や温度の制御より容易で応答も早く、しかも主軸加熱装置全体がコンパクトになるので好ましい。
この場合には、熱変位が最大の主軸例えば６ｂを代表主軸にしてこの主軸６ｂの近傍にノーズ温度センサＳ_１を取付けて、機体８６の温度変化を検出している。そして、電流制御装置９１で電流を制御して各主軸を加熱することによって、他の３本の主軸６ａ，６ｃ，６ｄの熱変位を代表主軸６ｂの熱変位に略一致させて、主軸間の熱変位のばらつきをなくしている。
なお、代表主軸６ｂを加熱しないか又はわずかに加熱して、他の主軸の温度を調節すれば、主軸加熱装置９０全体の電流が少なくなるので好ましい。
【０１０６】
本実施例では、センサＳ_１とは別に、主軸位置より離れた任意の位置（例えば、主軸頭８１の適当箇所）に配置されてこの位置の機体８６の温度変化を検出するヘッド温度センサＳ_２が、必要に応じて取付けられる。したがって、ミックス手法単独又はミックス手法とダミー手法とを組合せた前述の各方法による熱変位補正もできる。
各センサＳ_１，Ｓ_２の出力信号は、熱変位補正装置１２，１２ａ，１２ｂのＡ／Ｄ変換器１３に入力したのち、前記各実施例と同様に処理される。
なお、図３４，図３５に鎖線で示すように、代表主軸以外の各主軸の近傍にも温度センサＳ_１０をそれぞれ取付け、各主軸の温度を検出することが好ましい。センサＳ_１０で検出された温度の温度変化は熱変位補正には用いないが、この温度変化から各主軸の熱変位を推定して、主軸冷却装置８７又は主軸加熱装置９０により、主軸間の熱変位のばらつきをなくする管理をすることができる。
【０１０７】
図３６は、主軸冷却装置８７を用いた場合の本実施例の手順を示すフローチャートである。なお、下記説明中のカッコ内には、主軸加熱装置９０を用いた場合を記載している。
まず初めに、調節弁８９ａ乃至８９ｄを操作して、各主軸６ａ乃至６ｄに流れる冷却油の流量（又は、ヒータ９１ａ乃至９１ｄの電流）を必要最少限に絞る（ステップ４０１）。次いで、全主軸の同期回転を開始する（ステップ４０２）。一定の回転数で運転して所定時間経過後、各主軸の伸びによる主軸先端部におけるＺ軸方向の熱変位を実測する（ステップ４０３）。
各主軸の熱変位即ち伸び量が略同一になるように、各主軸の流路８５ａ乃至８５ｄに流れる流量を調節弁８９ａ乃至８９ｄにより（又は、ヒータ９１ａ乃至９１ｄの電流を電流制御装置９１により）調節して、この調節量を設定する（ステップ４０４）。
全主軸の回転を停止し、機体８６全体が十分に放熱するまで運転を停止する（ステップ４０５）。
【０１０８】
その後、全主軸の同期回転を再開し、ステップ４０４で設定された調節量に従って冷却油（又は電流）を各主軸にそれぞれ流す（ステップ４０６）。
次に、代表主軸６ａ（又は６ｂ）の先端部におけるＺ軸方向の熱変位を時系列データとして実測するとともに、機体８６の温度変化をセンサＳ_１，Ｓ_２の一方又は両方により検出する（ステップ４０７）。
こうして検出された温度変化を用い、前記各実施例と同様にして、ノーズ温度時定数τ_Ｎ，ヘッド温度時定数τ_Ｈ，サンプル温度時定数τ_Ｓ等を抽出し、さらにダミー熱容量及び内部補正係数を算出する。そして、これらの値を熱変位補正装置１２，１２ａ，１２ｂにセットする（ステップ４０８）。
ステップ４０９で補正を開始し、代表主軸による熱変位補正を実行する（ステップ４１０）。ステップ４１１で補正を終了する場合は全主軸の同期回転を停止し（ステップ４１２）、全体の手順が終了する。停止しない場合にはステップ４１０に戻る。
【０１０９】
図３３に示すように、補正装置１２，１２ａ，１２ｂで算出して補正された代表主軸の熱変位は、プログラマブルコントローラ１５を介して数値制御装置１６に送信され、Ｚ軸サーボモータ８２にフィードバックされる。これにより、Ｚ軸サーボモータ８２が、主軸頭８１をＺ軸方向に微小距離移動させて位置補正をする。
主軸冷却装置８７（又は、主軸加熱装置９０）により、代表主軸と他の主軸との熱変位を略均等にしているので、４個の工作物９ａ乃至９ｄを主軸６ａ乃至６ｄに装着した工具７ａ乃至７ｄにより高精度に同時加工することができる。
【０１１０】
図３７は複数の主軸を有する工作機械の平面構造を含むブロック図で、第５実施例の応用例である。
図示する工作機械は多軸ＮＣ旋盤９２であり、第４実施例と同様の構造の主軸台６６及び主軸６９を２組備えている。したがって、このＮＣ旋盤９２は多熱源及び多軸を有していることになる。
ベッド９３上には、２台の主軸台６６が並設され、且つ、サドル９４がＺ軸方向に移動可能に取付けられている。サドル９４はＺ軸サーボモータ９５により往復移動する。
サドル９４上には、クロススライド９６がＸ軸方向に移動可能に取付けられており、Ｘ軸サーボモータ９７により往復移動する。クロススライド９６上には、工具９８を有するブロック９９が複数取付けられている。チャック６５等を介して工作物を把持する主軸６９が回転することにより、工作物を工具９８で切削加工する。
ベッド９３には、図３４に示す主軸冷却装置８７と同様の原理の主軸冷却装置を構成する冷却油供給装置１００が設置されている。この主軸冷却装置は、２本の主軸６９の温度を個別に調節して、両主軸の熱変位を略均等にするためのものである。
【０１１１】
両主軸６９の前，後の軸受７１，７２及びビルトインモータ７０（図３０）の近傍には、冷却油を流して主軸の伸びを抑えるための流路が主軸台６６内に形成されている。主軸冷却装置は、図３４と同様の配管及び流量調節弁を有しており、各主軸６９への流量の調節を個別にできるようになっている。
一方の主軸台６６には、センサｓ_１，ｓ_２，ｓ_３が第４実施例と同じように取付けられているので、熱変位補正装置１２ｂにより第４実施例と同様にして熱変位補正がなされる。センサｓ_１，ｓ_２，ｓ_３が取付けられている方の主軸６９を代表主軸とし、主軸冷却装置によりこの代表主軸６９と他方の主軸６９の熱変位を略一致させて、主軸間の熱変位のばらつきをなくしている。
図３６に示す手順と同様にして代表主軸の熱変位補正をすれば、代表主軸６９と他方の主軸６９にそれぞれ把持されている工作物を、各工具９８により高精度で同時加工することができる。
なお、第５実施例では代表主軸が１本の場合を示したが、同時加工を行なわない場合には、複数の代表主軸をそれぞれ独立して熱変位補正してもよい。また、ペルチェ効果を応用した冷却装置又は加熱装置を温度調節装置として使用してもよい。
【０１１２】
ところで、多軸形の従来の工作機械には、発熱による主軸の伸びを少なくするために、多量の冷却油等を各主軸軸受近傍の流路に流して強力に冷却するものもある。この手法は、各主軸の熱変位を物理的に零に近づけることにより、各主軸の熱変位とばらつきとを同時に吸収しようとするものである。
しかしながら、この手法での熱変位の吸収には限界があり、熱変位を＋１０〔μｍ〕以下にすることは不可能である。また、多量の冷却油を循環させるので、大容量の冷却装置が必要であり、大量のエネルギが無駄になる。また、強い冷却効果により軸受にひずみが生じて主軸が焼付く虞もある。
これに対して、第５実施例は、各主軸の熱変位を物理的に零に近づけるのではなく、各主軸間の熱変位のばらつきをなくするように、代表主軸と他の主軸の熱変位を略一致させるとともに、代表主軸の熱変位補正を行なっている。したがって、補正後の加工誤差を零に近づけることができ、主軸冷却装置８５は小型のもので十分であり、エネルギ消費も少ない。また、冷却効果が弱いので軸受が焼付くこともない。
【０１１３】
なお、第１乃至第５実施例では繰り返し演算を行なっている。したがって、図１，図２，図１８，図２９，図３７に示すように、前回の演算結果を記憶し、工作機械の電源をオフして再度オンするまでの間の時間も記憶する記憶手段３５を、熱変位補正装置１２，１２ａ，１２ｂに設けることが好ましい。
記憶手段３５は、遅れ温度演算手段３４，３４ａ，３４ｂ，創成温度演算手段３１ａとの間でデータの授受を行なうことになる。このようにすれば、電源をオフした場合でも、熱変位補正の演算の経歴が保存されるので繰り返し演算が有効になる。
【０１１４】
また、ミックス手法と組合せるダミー手法，又はリニアライズ手法と組合せるダミー手法を用いる場合に、工作機械のコラム，ベッド，クロスレール等に別途設けた温度センサにより機体の温度変化を検出してもよい。
なお、本発明における温度検出手段としては温度センサの代わりに、温度変化による機体の伸縮を検出するひずみゲージ（Ｓｔｒａｉｎ　ｇａｕｇｅ）を使用してもよい。即ち、機体の温度変化を温度センサにより直接検出する代りに、温度変化と同様の出力特性を有するひずみゲージを機体に取付ける。そして、このゲージの出力信号をＡ／Ｄ変換器１３に入力させれば、実質的に温度変化を検出するのと同じことになり同様の作用効果を奏する。
ところで、各実施例における相関は一定の対応関係があればよく、一次の相関以外の場合でもよい。
【０１１５】
本発明は、従来のような機体構成部分の長さを使用していないので、機体構造上の長さの制約がなく、また、機体構成部分の長さ測定や回転数を種々変えてデータの実測作業をする必要はない。
したがって、回転数の測定は１回のみでよいことになり、実機を用いた熱変位特性抽出の実測作業が簡略化される。また、機体構成材料の線膨張係数の確認作業も不要である。
【０１１６】
また、温度センサは任意の位置に取付けてよいので、温度センサの取付位置の制約が緩和されると同時に、少数（例えば、一つの発熱源について１本又は２本）の温度センサのみで熱変位を精度よく見込むことができる自由度の高いものにすることができる。
また、機体の温度に基づいて補正をしており、室温を直接検出していない。したがって、例えば冬季に部屋の扉を開けたり夏季にクーラーを運転するなどして室温が急激に変化しても、室温による影響がなくなり、補正の精度を高精度に維持することができる。
また、本発明の熱変位補正方法及びその装置は、熱変位が機械の精度や性能に悪影響を与える他の種類の機械、例えば印刷機械，プレス，レーザ加工機等の自動制御機械に適用しても、同様の作用効果を奏する。この自動制御機械は、ＮＣ装置等の自動制御装置によって制御されている。
なお、各図中同一符号は同一又は相当部分を示す。
【０１１７】
【発明の効果】
本発明は、上述のように時定数の概念を用いて、発熱源の影響を受けて互いに時定数の異なる温度変化をする少なくとも二箇所における機体の温度変化を検出し、この検出された各温度変化を合成して、工作機械の熱変位の時定数と略同じ時定数を有する合成温度変化を演算し、この合成温度変化に対応して変化する熱変位に基づいて加工誤差を補正するように構成したので、合成温度変化と熱変位はリニアな相関になって、合成温度変化から熱変位を容易に見積もることが可能になり、熱変位に対する補正を高精度で行なうことができる。
【図面の簡単な説明】
【図１】本発明の第１実施例を示すブロック図である。
【図２】本発明の第２実施例を示すブロック図である。
【図３】第１実施例の動作を示すフローチャートである。
【図４】第２実施例の動作を示すフローチャートである。
【図５】Ｚ軸熱変位の経時変化を示すグラフである。
【図６】ノーズ位置とヘッド位置でそれぞれ検出された温度の温度変化と、合成温度変化を示すグラフである。
【図７】ノーズ温度変化とＺ軸熱変位との関係を示すグラフである。
【図８】ヘッド温度変化とＺ軸熱変位との関係を示すグラフである。
【図９】合成温度変化に対するＺ軸熱変位を示すグラフである。
【図１０】遅れ温度変化を演算する手法を説明するグラフである。
【図１１】温度変化から遅れ応答成分を演算する手法を示すグラフである。
【図１２】演算されたＺ軸熱変位と実測されたＺ軸熱変位との関係を示すグラフである。
【図１３】遅れ温度変化に対するＺ軸熱変位を示すグラフである。
【図１４】Ｚ軸熱変位の実測データを示すグラフである。
【図１５】Ｚ軸熱変位の他の実測データを示すグラフである。
【図１６】Ｙ軸熱変位の実測データを示すグラフである。
【図１７】Ｙ軸熱変位の他の実測データを示すグラフである。
【図１８】本発明の第３実施例を示すブロック図である。
【図１９】第３実施例の動作を示すフローチャートである。
【図２０】サンプル温度変化及びＺ軸熱変位を示すグラフである。
【図２１】サンプル温度変化とＺ軸熱変位との関係を示すグラフである。
【図２２】サンプル温度変化及び温度変化“Ａ，Ｂ”を示すグラフである。
【図２３】サンプル温度変化，温度変化“Ａ，Ｂ”，及び創成温度変化を示すグラフである。
【図２４】創成温度変化とＺ軸熱変位との関係を示すグラフである。
【図２５】Ｚ軸熱変位が遅れ応答成分を含んでいる場合の、創成温度変化とＺ軸熱変位との関係を示すグラフである。
【図２６】サンプル温度変化，遅れ温度変化，及び創成した遅れ温度変化を示すグラフである。
【図２７】遅れ温度変化と遅れ応答成分との関係を示すグラフである。
【図２８】創成温度変化及び遅れ温度変化から見積もった熱変位と、Ｚ軸熱変位との関係を示すグラフである。
【図２９】第４実施例を示すブロック図である。
【図３０】ＮＣ旋盤の主軸台の断面図である。
【図３１】各発熱源のＺ軸熱変位を示すグラフである。
【図３２】Ｚ軸熱変位の実測データを示すグラフである。
【図３３】第５実施例を示すブロック図である。
【図３４】主軸冷却装置の説明図である。
【図３５】主軸加熱装置の説明図である。
【図３６】第５実施例の動作を示すフローチャートである。
【図３７】第５実施例の応用例を示す図で、複数の主軸を有する工作機械の平面構造を含むブロック図である。
【符号の説明】
１，１ａ，７５　　立形マシニングセンタ（工作機械）
５，５ａ，８１　　主軸頭
６，６ａ乃至６ｄ，６９　　主軸
７，７ａ乃至７ｄ，９８　　工具
９，９ａ乃至９ｄ，６８　　工作物
１０，６６，８６　　　　　機体
１２，１２ａ，１２ｂ　　　熱変位補正装置
２０　　主軸受
２１　　ビルトインモータ
２２　　上部軸受（他の軸受）
３１　　合成温度演算手段（温度演算手段）
３１ａ　創成温度演算手段（温度演算手段）
３２　　熱変位演算手段
３３　　補正手段
３４　　遅れ温度演算手段（温度演算手段）
３４ａ　第１の遅れ温度演算手段（温度演算手段）
３４ｂ　第２の遅れ温度演算手段（温度演算手段）
６４　　ＮＣ旋盤（工作機械）
６６　　主軸台
７０　　ビルトインモータ
７１　　前軸受
７２　　後軸受
９２　　多軸ＮＣ旋盤（工作機械）
Ｏ_１　　中心軸方向
Ｓ_１　　ノーズ温度センサ（温度検出手段）
Ｓ_２　　ヘッド温度センサ（温度検出手段）
Ｓ_３　　温度センサ（温度検出手段）
ｓ_１乃至ｓ_３　温度センサ（温度検出手段）[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method and a device for correcting thermal displacement of a machine tool.
[0002]
[Prior art]
A machine tool has a heat source in each part of a machine body. For example, there are many heat sources such as rolling friction heat of a bearing of a main shaft and heat generated from a cutting portion. These heats are conducted to each part of the machine body to deform the machine body, and the deformation of the machine body has a great influence on the processing accuracy.
Therefore, various correction methods and devices have been proposed in which the thermal displacement of the airframe due to these various causes is predicted, and the error due to the thermal displacement is fed back to the servo system for correction.
[0003]
In a machine tool having such a correction function, it is important how to accurately estimate thermal displacement accompanying the operation of the machine, and various attempts have been made for that purpose. For example, there is a type in which thermal displacement is predicted from operating conditions such as the number of revolutions of the main shaft, or a type in which thermal displacement is directly detected by a displacement sensor incorporated in the body.
The present applicant has proposed in Japanese Patent Publication No. Hei 6-22779 and Japanese Patent Laid-Open Publication No. Hei 3-79256 a thermal displacement correction method for a machine tool that calculates a thermal displacement from a body temperature. The calculation of the thermal displacement in this method is basically based on the principle of the following equation (1).
ΔL = L × linear expansion coefficient × temperature change (1)
Here, ΔL: thermal displacement of the body components
L: Length of airframe components
It is.
[0004]
[Patent Document 1]
Japanese Patent Publication No. Hei 6-22779
[Patent Document 2]
JP-A-3-79256
[Patent Document 3]
JP-A-58-109250
[Patent Document 4]
JP-A-60-9634
[Patent Document 5]
JP-A-5-84628
[0005]
[Problems to be solved by the invention]
The limit of the processing accuracy after correction in the conventional technique is about 20 to 30 [μm]. However, in recent years, machine tool users generally demand that the accuracy after correction be suppressed to a processing error of 10 [μm] or less. This is because it is necessary to process a new material such as a ceramic material, or a more miniaturized workpiece or the like with high precision.
[0006]
Further, in the above calculation method, the length L of the component is estimated from the configuration of the machine body, and the temperature change is detected from the center position of the length L, so that the mounting position of the temperature sensor is limited. Furthermore, in order to accurately predict thermal displacement, it is necessary to divide the body into small components, and a large number of temperature sensors are required to calculate a temperature change in each portion. In addition, it was necessary to measure the length L of the fuselage component and check the linear expansion coefficient of each fuselage component.
These have been obstacles to mounting a thermal displacement correction device for a machine tool that calculates thermal displacement from the temperature of the machine.
[0007]
On the other hand, Japanese Patent Application Laid-Open No. 58-109250 discloses a method of controlling the temperature of cooling air by using a metal piece that is thermally similar to a machine tool and assuming the temperature to be a temperature representative of the machine tool. Accordingly, a thermal displacement compensating device for compensating for thermal displacement of a machine tool has been proposed. However, in this case, a thermally similar metal piece had to be separately prepared.
Further, Japanese Patent Application Laid-Open No. Sho 60-9634 proposes a thermal displacement compensator using a temperature sensor having a thermal time constant adapted to the Y-axis thermal displacement characteristics. However, in this correction device, details of a temperature sensor having a thermal time constant matched to the characteristics of thermal displacement are not disclosed.
[0008]
By the way, in the case of a machine tool having a plurality of spindles, the extension of each spindle due to variations in the preload applied to the bearings of the spindle, differences in how the temperature is transmitted depending on the mounting location of the spindle, and the state of bearing lubrication, etc. Often there are differences.
Therefore, for example, after a plurality of workpieces attached to the main spindle are simultaneously rough-finished to the same shape, the number of main spindles used in the finishing is limited to one, and the other main spindles are stopped. Thus, heat generation of the spindle is suppressed, and thermal displacement correction is performed by paying attention only to thermal deformation of the spindle to be used, and finishing is performed. In this case, in order to prevent the tool held by the stopped spindle from interfering with the workpiece, it is necessary to remove the stopped tool in advance. Therefore, the work efficiency of the finishing process was extremely poor.
Japanese Patent Application Laid-Open No. Hei 5-84628 proposes a thermal displacement correcting device for a machine tool having a plurality of spindles. However, there is a limit to the thermal displacement correction by this correction device, and it has been difficult to make the processing error after correction as close to zero as possible.
[0009]
The present invention has been made in order to solve such a problem, and an object of the present invention is to provide a method and a device for correcting a thermal displacement of a machine tool, which can highly accurately correct a processing error caused by heat. .
Another object of the present invention is to eliminate the need for measuring the length of a body component of a machine tool and for confirming the linear expansion coefficient of the body component material, and for measuring the thermal displacement characteristics using an actual machine. Is to simplify.
Still another object of the present invention is to provide a method of correcting a thermal displacement of a machine tool having a high degree of freedom, which can greatly reduce a restriction on a mounting position of a temperature sensor and can accurately estimate a thermal displacement with a small number of temperature sensors. It is to provide the device.
Still another object of the present invention is to simultaneously process a plurality of workpieces with high accuracy and greatly improve work efficiency in a case of a machine tool having a plurality of spindles.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, the present invention detects a thermal displacement due to elongation in a predetermined axial direction or a thermal displacement due to inclination of a main shaft when an arbitrary main shaft rotation speed is given to a machine tool, and at the same time, detects a temperature of an appropriate portion of a machine body. The change is detected by a temperature sensor.
If this temperature change and the thermal displacement are the same in a time series, the temperature change and the thermal displacement are simply linearly correlated, so it is assumed that the thermal displacement can be easily estimated from the temperature change. ing.
However, the time constant of the temperature change detected from an appropriate part of the body is not necessarily the same as the time constant of the thermal displacement in the predetermined axial direction. Therefore, there is a need for a method of appropriately processing the data of the temperature change to match the time constant of the thermal displacement.
Therefore, the thermal displacement correction method according to the present invention detects a temperature change of the machine body affected by the heat source, and uses the detected temperature change to set a time constant substantially equal to the time constant of the thermal displacement of the machine tool. The calculated temperature change is calculated, and the processing error is corrected based on the calculated thermal displacement obtained by using a function that determines the relationship between the calculated temperature change and the thermal displacement.
The “thermal displacement of a machine tool” is ideally a thermal displacement at a processing point by a tool, but in reality, for example, an appropriate one of a spindle tip or a test bar temporarily attached to the spindle tip. Thermal displacement at a point.
A thermal displacement correction device for realizing the correction method includes: a temperature detection unit configured to detect a temperature change of a body affected by a heat source; and a temperature change unit that detects a temperature change detected by the temperature detection unit. Temperature calculating means for calculating a temperature change having substantially the same time constant as the time constant of the thermal displacement of the machine; and a thermal displacement using a function which determines the relationship between the temperature change and the thermal displacement calculated by the temperature calculating means. And a correction means for correcting a processing error based on the thermal displacement calculated by the thermal displacement calculation means.
[0011]
As an example of a data processing method for matching the detected temperature change data with the time constant of the thermal displacement, a “mixing method” and a “dummy method” are used.
In the mixing method, first, a thermal displacement when an arbitrary spindle speed is given to a machine tool is detected. At the same time, a temperature change at a portion having a time constant of a temperature change smaller than the time constant of the thermal displacement and a temperature at a portion having a time constant of a temperature change larger than the time constant of the thermal displacement are detected. Then, these at least two temperature changes having different time constants are combined to create a combined temperature change having the same time constant as the thermal displacement time constant.
In the mixing method, when creating the synthetic temperature change, a plurality of synthetic temperature changes are once created, and the plurality of synthetic temperature changes are further combined to have a time constant substantially equal to the time constant of thermal displacement. Synthetic temperature changes may be created.
By the way, in the mixing method, when creating the synthetic temperature change, data obtained from the temperature sensor every moment is directly used as the temperature data. Therefore, although the accuracy of the synthesized temperature change is highly reliable, only the time constant of the thermal displacement located between the time constants of the respective temperature changes can be synthesized.
[0012]
On the other hand, the dummy method detects thermal displacement when a machine tool is given an arbitrary spindle rotation speed, and at the same time, detects a temperature change at an appropriate point having a time constant of a temperature change smaller than the time constant of the thermal displacement. To detect. In this method, an imaginary delayed temperature change having a time constant substantially equal to the time constant of the thermal displacement appearing later than the detected temperature change is created by a repetitive calculation that allows for a delay in the temperature change.
In the dummy method, when the delay temperature change is created, another delay temperature change is once created, and a further delay in the delay temperature change is anticipated. Temperature changes may be created.
As described above, in the dummy method, the delay temperature change is calculated by a repetitive calculation that allows for a delay in the temperature change. Therefore, while the operation formula is simple, it is a rough approximation calculation, and the reliability of accuracy is slightly lacking.
[0013]
The linear correlation between the generated temperature change and the thermal displacement, which have the same time constant as the thermal displacement of the machine tool, is due to the effect that the heat generated at the head of the spindle accompanying the rotation of the spindle is transmitted to the column, or other heat sources such as room temperature. May gradually break down due to the delayed response component of the thermal displacement due to the influence of the above.
Therefore, in order to maintain the linear correlation between the generated temperature change and the thermal displacement for a long time, the thermal displacement estimated by the temperature change created by the mixing method or the dummy method is delayed by the delayed response of the gradually appearing thermal displacement. The estimation is performed by adding the components.
For example, using the temperature data detected from at least two points where the temperature constants differ from each other under the influence of the heat source, a synthetic temperature change is created by a mixing method, and a linear correlation between the synthetic temperature change and The resulting thermal displacement is calculated. Then, using the temperature data detected from one of the previous temperature data or a separately detected relatively gentle temperature change, a delay correlation between the temperature change and the delay temperature change created with sufficient delay can be obtained. Calculate the delay response component of the displacement.
Estimation of the thermal displacement due to a temperature change created by a combination method of the mixing method and the dummy method is preferable because the reliability of the accuracy increases.
In addition, a first delay temperature change is created by a dummy method using temperature data detected from a place where the temperature near the heat source changes rapidly and greatly, and a heat obtained by a linear correlation with the delay temperature change. Calculate the displacement.
Then, using the previous temperature data or the temperature data detected from a relatively gentle temperature change detected separately, a second correlation between the temperature change and the second delay temperature change, which is created by sufficiently considering the delay of the temperature change, is obtained. The delay response component of the displacement is calculated. The estimation of the thermal displacement due to the temperature change created by the combination method of the dummy method and the dummy method is somewhat less accurate than the estimation by the temperature change created by the combination method of the mix method and the dummy method.
[0014]
For the mixing method and the dummy method, or the method combining the dummy method and the dummy method, a machining center (hereinafter, referred to as MC) using the head of the spindle as a heat source, or a numerically controlled lathe (hereinafter, referred to as MC) having a heat source inside the headstock NC lathe) will be described in detail.
First, in the combination of the mixing method and the dummy method, the MC or the like is operated to detect a thermal displacement when an arbitrary spindle speed is given. At the same time, the temperature change of the body at the nose position having a time constant of the temperature change smaller than the time constant of the thermal displacement and the head position having the time constant of the temperature change larger than the time constant of the thermal displacement are respectively detected. Then, a combined temperature change having the same time constant as the time constant of the thermal displacement is created by the mixing method, and the thermal displacement that changes corresponding to the combined temperature change is calculated.
Further, in the combination of the dummy method and the dummy method, the MC or the like is operated to detect a thermal displacement when an arbitrary spindle speed is given. At the same time, the temperature change of the airframe at the nose position having the time constant of the temperature change smaller than the time constant of the thermal displacement is detected. Then, a lag temperature change having the same time constant as the time constant of the thermal displacement is created by the dummy method, and the thermal displacement that changes corresponding to the lag temperature change is calculated.
Next, in each case of the combination of the mix method and the dummy method and the combination of the dummy method and the dummy method, the following dummy method is further added.
That is, in consideration of the delay in the temperature change detected at the spindle head position such as MC, the actual thermal displacement and the thermal displacement previously calculated using the combined temperature change or the delayed temperature change gradually shift. A lag temperature change having substantially the same aging characteristics as the lag response component of the displacement is generated by repeated calculation. A delay response component that changes in response to the delay temperature change is calculated.
[0015]
The “Linearize” method is an extension of a data processing method that makes the detected temperature change data substantially coincide with the time constant of the thermal displacement, and the present invention also uses this method.
In the linearization method, a thermal displacement when an arbitrary spindle speed is given to a machine tool is detected. At the same time, a temperature change in an appropriate portion of the body that changes its temperature under the influence of the heat generated by the heat source is detected. Using this detected temperature change, a temperature change in the heat source is calculated. Then, using this temperature change of the heat source, a generated temperature change having substantially the same time constant as the time constant of the thermal displacement is calculated.
In the linearization method, when calculating the generating temperature change, another generating temperature change is once calculated using the temperature change at the heat source, and the generated temperature change is used to calculate the same as the time constant of the thermal displacement. A creation temperature change having a time constant may be calculated.
In the linearization method, it is not always necessary to detect a temperature change at a location having a time constant of a temperature change smaller than the time constant of the thermal displacement. For this reason, although the degree of freedom of the position where the temperature sensor is provided is high, the procedure for calculating the generated temperature change is slightly complicated.
[0016]
In the method combining the linearization method and the dummy method, a thermal displacement when an arbitrary spindle speed is given to the machine tool is detected. At the same time, a temperature change in an appropriate portion of the machine body that changes in temperature under the influence of heat generated by the heat source is detected. Then, a generation temperature change having substantially the same time constant as the time constant of the thermal displacement is calculated by a linearization method, and a thermal displacement that changes corresponding to the generation temperature change is calculated.
Then, in anticipation of a delay in the detected temperature data or a temperature data detected from a point where a relatively gentle temperature change is detected separately, the actual thermal displacement and the thermal displacement previously calculated using the generated temperature change A delay temperature change having substantially the same aging characteristics as a delay response component of a displacement that gradually shifts is generated by repeated calculation. A delay response component that changes in response to the delay temperature change is calculated.
[0017]
In the mixing method, it is a condition that one of the time constants of the detected temperature change is smaller than the time constant of the thermal displacement. In the dummy method, the time constant of the detected temperature change must be smaller than the time constant of the thermal displacement. Therefore, when estimating the thermal displacement based on the temperature change created by these methods, there is a restriction on the position where the temperature change is detected.
On the other hand, in the linearization method, there is no condition for the magnitude of the time constant of the detected temperature change, and only one detected temperature is required for one heat source. Therefore, the present invention is advantageous for calculating the generating temperature in an MC or NC lathe having a spindle head whose thermal displacement is more sensitive to a temperature change, or in a machine tool having a plurality of heat sources that affect machining accuracy.
For example, an NC lathe includes a spindle that grips a workpiece or a tool, and a headstock that rotatably supports the spindle through a front bearing on a machining position and a rear bearing on a non-machining position that supports the spindle. A built-in motor disposed between the front and rear bearings, built in the headstock, and rotationally driving the spindle.
In the NC lathe, the rear bearing positions the main shaft with respect to the center axis direction, and the front bearing holds the main shaft that expands and contracts due to thermal displacement so as to be slidable in the center axis direction, and serves as a heat source. Three temperature sensors for detecting temperature changes in the vicinity of the front and rear bearings and the built-in motor, respectively, are attached to the headstock.
When the linearization method is applied to a machine tool having such multiple heat sources, a temperature change at each heat source is calculated using a temperature change detected by each temperature sensor. Then, using this temperature change, the respective generated temperature changes having the same time constant as the time constant of the thermal displacement due to the heat generation effect of each heat source are calculated. Each of the generated temperature changes calculated in this way has a linear correlation with the thermal displacement due to the influence of each heat source.
[0018]
With the rotation of the spindle of the machine tool, heat is generated from a heat source such as a spindle bearing or a spindle drive motor, and this heat is transmitted to the components of the machine body, resulting in a temperature change. Ordinary machine tools mainly use cast iron or steel as a structural material.
Therefore, when there is a temperature change, a thermal displacement proportional to the linear expansion coefficient of these structural materials occurs in each part. The thermal displacement of each of these parts is added to reduce the processing accuracy of the machine tool.
The temperature change accompanying the rotation of the spindle of the machine tool appears earlier in the vicinity of the heat source, but appears more distant from the heat source such as the head (spindle head), the head mounting portion, and the column, and the time change of the temperature change, respectively. The characteristics are different. For this reason, the temperature change and the thermal displacement of an arbitrary portion of the fuselage are not usually directly linked.
[0019]
However, the time series data of the thermal displacement when an arbitrary spindle speed is given to the machine tool, and the time series data of the temperature change detected from an appropriate part of the machine body affected by the heat generated by the heat source are approximated. By applying the step input response function of a single first-order lag element, it is possible to extract the time constant until the change is saturated. The balance between the time constant of the thermal displacement and the time constant of the temperature change is representative of the thermal characteristics of the machine tool which are common over a wide range of the spindle speed.
Therefore, a method of appropriately processing the data of the temperature change and calculating a generated temperature change having a time constant substantially matching the time constant of the thermal displacement is used. Then, since a linear correlation is established between the generated temperature change and the thermal displacement, it is possible to indirectly fairly accurately estimate the thermal displacement from the temperature change.
Assuming that the thermal displacement phenomenon is an action due to a temperature change of a single first-order lag element, a thermal displacement estimation error occurs due to a difference from the complexity of the actual airframe configuration. Therefore, the same operation can be repeatedly applied by approximating the error component as an effect due to a temperature change of a single different first-order lag element.
Therefore, the thermal displacement can be estimated with sufficiently high accuracy from the temperature change data detected from an appropriate portion of the body.
[0020]
Specifically, first, in order to extract the thermal characteristics of the machine tool, a thermal displacement when an arbitrary spindle speed is given in a preliminary test is detected using an electric micrometer or the like. At the same time, a temperature change in an appropriate portion of the body, which changes its temperature under the influence of heat generated by the heat source, is detected using a thermistor temperature sensor or the like.
Next, the time constant data is extracted by applying a step input response function of a first-order lag element to each time-series data until the change is substantially saturated. Using this temperature change data, select one of the mixing method, dummy method, linearization method, or a combination of these methods, which is prepared to create a temperature change having the same time constant as the thermal displacement time constant. Then, the coefficients of the temperature conversion formula determined in each method are calculated.
When the temperature conversion equation in the selected method is rewritten by giving the time series data of the previous temperature change, the generated temperature change has the same time constant as the thermal deformation. A linear correlation is established between the generated temperature change and the previous thermal deformation data, and the slope becomes a proportional constant for calculating the thermal displacement from the generated temperature change.
In the thermal displacement correction at the time of machine operation, the data of the temperature change detected from time to time from the point where the temperature change was previously detected is converted from time to time to the creation temperature change using the temperature conversion formula in the previously selected method. . Next, this generated temperature change is multiplied by the previously calculated proportionality constant to determine the thermal displacement to be corrected.
[0021]
BEST MODE FOR CARRYING OUT THE INVENTION
As described above, the method for correcting thermal displacement of a machine tool according to the embodiment of the present invention detects a temperature change of a machine body affected by a heat source, and uses the detected temperature change to determine a thermal displacement of the machine tool. A temperature change having substantially the same time constant as the time constant is calculated, a thermal displacement that changes in accordance with the calculated temperature change is calculated, and a temperature change in an appropriate portion of the body is detected. In anticipation of a delay in the temperature change, a delay temperature change having substantially the same aging characteristic as a delay response component in which the thermal displacement of the machine tool and the calculated thermal displacement gradually shift is calculated. The processing error is corrected based on the total value obtained by adding the delay response component that changes correspondingly to the calculated thermal displacement.
The method for correcting thermal displacement of a machine tool according to another embodiment includes detecting a temperature change of an airframe at at least two places where temperature changes with time constants different from each other under the influence of a heat source, and detecting each detected temperature change. Are combined to calculate a combined temperature change having substantially the same time constant as the time constant of the thermal displacement of the machine tool, and the processing error is corrected based on the thermal displacement that changes in response to the combined temperature change.
Further, the method for correcting thermal displacement of a machine tool according to another embodiment detects a temperature change of a machine body near a heat source, anticipates a delay in the detected temperature change, and compares a time constant of a thermal displacement of the machine tool with a time constant. A delay temperature change having substantially the same time constant is calculated, and a processing error is corrected based on a thermal displacement that changes in response to the delay temperature change.
In addition, the method for correcting thermal displacement of a machine tool according to another embodiment detects temperature changes of a body at least at two locations where temperature changes are different from each other under the influence of a heat source, and each of the detected temperature changes is detected. By combining the temperature changes, a combined temperature change having substantially the same time constant as the time constant of the thermal displacement of the machine tool is calculated, and a thermal displacement that changes corresponding to the combined temperature change is calculated, and an appropriate portion of the machine body is calculated. The temperature change of the machine tool is detected, and the delay having the same time-dependent characteristic as the delay response component in which the thermal displacement of the machine tool and the calculated thermal displacement are gradually shifted in anticipation of a delay in the detected temperature change. A temperature change is calculated, and a processing error is corrected based on a total value obtained by adding a delay response component that changes according to the delay temperature change to the calculated thermal displacement.
Further, the method for correcting thermal displacement of a machine tool according to another embodiment detects a temperature change of a machine body near a heat source, anticipates a delay in the detected temperature change, and compares a time constant of a thermal displacement of the machine tool with a time constant. A first delay temperature change having substantially the same time constant is calculated, a thermal displacement that changes corresponding to the first delay temperature change is calculated, and a temperature change at an appropriate portion of the body is detected. In anticipation of a delay in the temperature change, a second delay temperature change having substantially the same aging characteristic as a delay response component in which the thermal displacement of the machine tool and the calculated thermal displacement gradually shift is calculated. The processing error is corrected based on a total value obtained by adding a delay response component that changes in response to the second delay temperature change to the calculated thermal displacement.
In order to realize the method, the thermal displacement correction device for a machine tool according to the embodiment includes a temperature detecting unit that detects a temperature change of a body affected by a heat source, and a temperature change detected by the temperature detecting unit. A temperature calculating means for calculating a temperature change having substantially the same time constant as the time constant of the thermal displacement of the machine tool, and calculating a thermal displacement changing corresponding to the temperature change calculated by the temperature calculating means. A thermal displacement calculating means, a temperature detecting means for separately detecting a temperature change in an appropriate portion of the machine as needed, and a delay in the temperature change detected by any one of the temperature detecting means, A delay temperature calculating means for calculating a delay temperature change having substantially the same aging characteristics as a delay response component in which the thermal displacement of the thermal displacement and the output of the thermal displacement calculating means gradually shift, and the thermal displacement calculating means A delay response component that changes in response to a change in temperature is calculated, and the delay response component is added to the calculated thermal displacement to calculate a total value. The error has been corrected.
A thermal displacement compensating device for a machine tool according to another embodiment includes a temperature detecting unit that detects temperature changes of an airframe at at least two places where the temperature changes with time constants different from each other under the influence of a heat source; Means for calculating a combined temperature change having substantially the same time constant as the time constant of the thermal displacement in the predetermined axial direction by combining the respective temperature changes detected by the means; and calculating by the combined temperature calculation means. A thermal displacement calculating means for calculating a thermal displacement that changes in accordance with the combined temperature change; and a correcting means for correcting a processing error based on the thermal displacement in the predetermined axial direction calculated by the thermal displacement calculating means. It has.
In addition, the thermal displacement correction device for a machine tool according to another embodiment includes a temperature detection unit that detects a temperature change of a body near a heat source, and in anticipation of a delay in the temperature change detected by the temperature detection unit, A delay temperature calculating means for calculating a delay temperature change having substantially the same time constant as the time constant of the thermal displacement in the predetermined axial direction; and a thermal displacement changing corresponding to the delay temperature change calculated by the delay temperature calculation means. Thermal displacement calculating means for calculating, and correcting means for correcting a processing error based on the thermal displacement in the predetermined axial direction calculated by the thermal displacement calculating means.
In addition, a thermal displacement correction device for a machine tool according to another embodiment includes a temperature detecting unit that detects a temperature change of an airframe at at least two places where temperature changes are performed with different time constants under the influence of a heat source. Combining temperature change detected by the temperature detecting means to calculate a combined temperature change having substantially the same time constant as the time constant of the thermal displacement in the predetermined axial direction; and calculating the combined temperature change by the combined temperature calculating means. A thermal displacement calculating means for calculating a thermal displacement that changes in response to the synthesized temperature change, a temperature detecting means for separately detecting a temperature change in an appropriate portion of the airframe as needed, and any one of the temperature detections In view of the delay in the temperature change detected by the means, the delay temperature change having substantially the same aging characteristic as the delay response component in which the thermal displacement in the predetermined axial direction and the output of the thermal displacement calculating means gradually shift. And a delay temperature calculating means for calculating a delay response component that changes in response to the delay temperature change, and adding the delay response component to the calculated thermal displacement. Then, a total value is calculated, and a processing error is corrected by the correction means based on the total value.
In addition, the thermal displacement correction device for a machine tool according to another embodiment includes a temperature detection unit that detects a temperature change of a body near a heat source, and in anticipation of a delay in the temperature change detected by the temperature detection unit, A first delay temperature calculating means for calculating a delay temperature change having substantially the same time constant as the time constant of the thermal displacement in the predetermined axial direction; and a delay temperature change means for calculating the delay temperature change calculated by the first delay temperature calculation means. A thermal displacement calculating means for calculating a thermal displacement that changes with the temperature change, a temperature detecting means for separately detecting a temperature change of an appropriate portion of the body as needed, and a temperature change detected by any one of the temperature detecting means. Second delay temperature calculating means for calculating a delay temperature change having substantially the same aging characteristics as a delay response component in which the thermal displacement in the predetermined axial direction and the output of the thermal displacement calculating means gradually shift in anticipation of the delay. Comprising the above, The displacement calculating means calculates a delay response component that changes in accordance with the delay temperature change calculated by the second delay temperature calculating means, and adds the delay response component to the calculated thermal displacement. The total value is calculated based on the total value, and the processing means corrects the processing error based on the total value.
In the method and the apparatus for correcting thermal displacement, it is preferable that the machine tool is one of a machining center using a spindle head as the heat source and an NC lathe using a headstock as the heat source.
Further, in the thermal displacement correction method and the apparatus, the machine tool may include a main shaft that holds one of a workpiece and a tool, a main bearing that supports the main shaft on a processing position side, and a main shaft that supports the main shaft. A spindle head rotatably supporting the spindle via the bearings, and a built-in motor disposed between the two bearings and built in the spindle head to rotationally drive the spindle. The main shaft is positioned with respect to the center axis direction, the other bearing holds the main shaft which expands and contracts due to thermal displacement so as to be slidable in the center axis direction, and the temperature change at a head position affected by the heat source. It is preferable that a head temperature sensor for detecting the temperature is mounted on the spindle head.
In the thermal displacement correction method and the apparatus, the machine tool may include a main spindle that holds one of a workpiece and a tool, a front bearing that supports the main spindle, and a rear bearing that is opposite to the machining position. A headstock rotatably supporting the spindle via a bearing; and a built-in motor disposed between the front and rear bearings and built in the headstock to rotationally drive the spindle. The bearing positions the main shaft with respect to the center axis direction, the front bearing holds the main shaft which expands and contracts due to thermal displacement so as to be slidable in the center axis direction, and the front and rear bearings serving as the heat source are provided. It is preferable that three temperature sensors for detecting a temperature change near each of the built-in motors are attached to the headstock.
Further, in the thermal displacement correction method and the apparatus, a plurality of spindles that grip one of a workpiece and a tool and rotate in synchronization with the spindle head are provided, and a heat generation characteristic accompanying the rotation of the plurality of spindles is provided. The difference is controlled by controlling the amount of cooling oil flowing through the jacket provided at the nose portion of each spindle or the temperature of the cooling oil, or controlling the amount of electricity supplied to the heater provided at the nose portion of each spindle to achieve uniform thermal displacement. Preferably, at least one of the spindles is used as the heat source.
In the method and the apparatus for correcting thermal displacement, it is preferable that the temperature change is calculated by subtracting a reference temperature from a detected temperature.
[0022]
Hereinafter, an embodiment of the present invention will be described with reference to FIGS.
(First and second embodiments)
FIGS. 1 to 17 are diagrams for explaining the first and second embodiments of the present invention.
For example, the numerical control (NC) machine tool shown in FIG. 1 is a vertical machining center (MC) 1, but may be another type of NC machine tool other than the MC. A column 3 is provided upright on the bed 2, and a spindle head 5 is mounted on the column 3 so as to be movable in the Z-axis direction. The column 3 can move on the bed 2 in the Y-axis direction.
A spindle 6 is provided on the spindle head 5 in the Z-axis direction, and a tool 7 is mounted on a tip of the spindle 6. The spindle 6 is driven to rotate by a spindle motor 4 attached to the spindle head 5. A workpiece 9 placed on a table 8 provided on the bed 2 is cut by a tool 7. The table 8 moves on the bed 2 in the X-axis direction.
The axis direction of the main shaft 6 is defined as a Z-axis, and directions orthogonal to the Z-axis are defined as an X-axis and a Y-axis.
The MC 1 is provided with a temperature detecting means for detecting a temperature change of the body 10. In the first embodiment, a nose temperature sensor S for detecting the temperature at the nose position of the spindle head 5 at the front end side of the spindle.₁And a head temperature sensor S arranged at an arbitrary position away from the nose position to detect the temperature of the spindle head 5₂Are respectively attached. Temperature sensor S as temperature detecting means₁, S₂May be any type, but a thermistor temperature sensor that is strong against disturbance is desirable.
Nose temperature sensor S₁Is close to the main bearing of the main shaft 6, which is a main heat source, so that a temperature change appears immediately, and the time constant is small. On the other hand, the head temperature sensor S₂Since is farther from the main bearing, the temperature change appears slowly, and thus the time constant is large.
In the second embodiment, the nose temperature sensor S₁And a temperature sensor S as temperature detecting means for detecting the temperature of a portion where the influence of the heat generated by the heat source slowly affects the body 10.₃, Respectively.
[0023]
Here, the principle of the thermal displacement correction in the present invention will be described.
According to the present invention, the thermal displacement in each of the X, Y, and Z axial directions can be corrected. For example, in the X axis direction, the column 3 and the spindle head 5 have a symmetrical structure with respect to the X axis. Therefore, correction in the X-axis direction is not normally necessary. In the following description, an example will be described in which correction is mainly performed in the Z-axis direction among the Y-axis and Z-axis.
[0024]
The equation for calculating the thermal displacement in the Z-axis direction is given by the following equation.
ΔZ = a · (ΔZ₁+ ΔZ₂) ... (2)
Here, ΔZ: Z-axis thermal displacement
a: Overall correction coefficient
(The coefficient “a” is a coefficient for correcting the difference between the result of the operation formula (2) and the actual accuracy)
ΔZ₁: Immediate response component of Z-axis thermal displacement
ΔZ₂: Delay response component of Z-axis thermal displacement
It is.
[0025]
That is, the arithmetic expression (2) is an immediate response component ΔZ for which the thermal displacement can be immediately predicted from the temperature change.₁And a delay response component ΔZ in which thermal displacement appears with a delay₂And The temperature change is calculated as a temperature difference obtained by subtracting the reference temperature from the temperature detected and output by each temperature sensor.
As the reference temperature, the output of the first temperature sensor when the power of the MC1 is turned on, the average value of the outputs obtained by adding the outputs a plurality of times, or an absolute reference such as 20 [° C.] is used. The reference temperature for each temperature sensor is stored in the RAM 11.
By the way, when the temperature change in the environment where the machine tool is installed is relatively slow, the thermal deformation of the entire machine tool due to the change in room temperature changes to a substantially similar shape including the tool and the workpiece. That is, since processing error does not occur with such a slow change in room temperature, a prediction of thermal deformation based on a temperature change including the room temperature change is different from actual thermal deformation.
Therefore, in this case, the instantaneous temperature detected by a temperature sensor separately provided on a bed or the like of the machine tool is adopted as a reference temperature, and the temperature obtained by subtracting this reference temperature from the temperature output from each temperature sensor is used as the temperature. Use as a change. In this way, accurate thermal displacement correction can be performed even when there is a change in room temperature.
[0026]
It should be noted that the present invention provides an immediate response component ΔZ as shown in equation (3).₁It is also possible to use an arithmetic expression based only on the above. Or, as shown in equation (4), the delay response component ΔZ₂It is also possible to use an arithmetic expression based only on the above.
ΔZ = a · ΔZ₁…… (3)
ΔZ = a · ΔZ₂…… (4)
[0027]
Instant response component ΔZ₁Is calculated by the following equation.
ΔZ₁= B · ΔT₁+ C · ΔT₂…… (5)
Where ΔT₁: Temperature sensor S₁Temperature change obtained by subtracting the reference temperature from the output of [° C]
ΔT₂: Temperature sensor S₂Temperature change obtained by subtracting the reference temperature from the output of [° C]
b: Internal correction coefficient [± μm / ° C]
c: Internal correction coefficient [± μm / ° C]
It is.
[0028]
Equation (5) used in the first embodiment is an equation for calculating the thermal displacement of the body 10 based on the outputs of the temperature sensors installed at two places. And the nose temperature sensor S₁Temperature change ΔT calculated from the temperature detected in step₁And the head temperature sensor S₂Temperature change ΔT calculated from the temperature detected in step₂From the immediate response component ΔZ₁Is calculated.
Note that the immediate response component ΔZ₁In the calculation of (1), at least two places for installing the temperature sensor are required, but they are added as appropriate according to the number of heat sources. The temperature sensor may be installed at a location other than the nose position and the head position as long as the location is affected by the heat generated by the heat source.
[0029]
On the other hand, the delay response component ΔZ₂Is calculated as follows.
ΔZ₂= EY₁+ FY₂…… (6)
Here, e: internal correction coefficient [± μm / ° C.]
f: Internal correction coefficient [± μm / ° C]
Y₁: First delay temperature change [° C]
Y₂: Second delay temperature change [° C]
It is.
Equation (6) is an equation for calculating the thermal displacement of the body 10 based on the first and second delay temperature changes.
[0030]
In the first embodiment, since there is only one delay temperature change, the value of the internal correction coefficient f becomes zero. Then, the head temperature sensor S₂Temperature change ΔT of the temperature detected at₂Delay temperature change Y calculated with the delay taken into account₁From the delay response component ΔZ₂Is calculated.
In the second embodiment, two delay temperature changes are used. Nose temperature sensor S₁Temperature change ΔT of the temperature detected at₁First delay temperature change Y calculated with a delay taken into account₁And the column temperature sensor S₃Temperature change ΔT of the temperature detected at₂Delay temperature change Y calculated with the delay taken into account₂From the above, the delay response component ΔZ₂Is calculated.
Note that the delay response component ΔZ₂The temperature information used in the calculation of (1) may be one or two, but is appropriately added according to the number of heat sources.
[0031]
FIG. 1 is a block diagram showing a first embodiment of the present invention.
As shown, each temperature sensor S₁, S₂Are input to the A / D converter 13 of the thermal displacement compensator 12 via the circuits 36 and 37, and the input analog signal is converted into a digital signal here. The digital signal from the A / D converter 13 is input to the operation storage unit 14, where the thermal displacement is calculated.
The processing error is corrected by the correction means 33 based on the calculated thermal displacement. The output signal of the correction means 33 is transmitted to the numerical controller 16 via the programmable controller 15 and fed back to the servo system to correct the position.
That is, the correction unit 33 outputs the calculation result to an external offset unit that externally applies an offset to the movement command value of the numerical controller 16. As a result, for example, the origin position of the rectangular coordinate system is offset, and the numerical controller 16 controls the trajectory of the tool 7 of the MC 1.
Note that the programmable controller 15 manages the operation sequence of the MC 1 in response to a command from the numerical controller 16.
[0032]
Each temperature sensor S₁, S₂Is calculated by the operation storage unit 14 via the A / D converter 13 and the temperature sensor S in the RAM 11₁, S₂The data is written to the memory address specified for. Further, each temperature sensor S₁, S₂Stored temperature data sampled at regular intervals. This temperature data is displayed on the display of the numerical controller 16.
The ROM 17 stores a program for calculating the thermal displacement according to the present invention, a correction coefficient, and the like. The clock 18 is a normal clock, and each of the temperature sensors S₁, S₂Is to determine the detection time.
[0033]
The thermal displacement correction device 12 includes a temperature sensor S₁, S₂A virtual position P having substantially the same time constant as the time constant of thermal displacement using the temperature change of the temperature detected at₁, And an immediate response component ΔZ that changes in response to the combined temperature change calculated by the combined temperature calculating unit 31.₁That is, there are provided a thermal displacement calculating means 32 for calculating a thermal displacement, and a correcting means 33 for correcting a processing error based on the thermal displacement calculated by the thermal displacement calculating means 32.
[0034]
As a preferred embodiment, the thermal displacement compensator 12 of the first embodiment includes a temperature sensor S₂Further, there is further provided a delay temperature calculating means 34 for calculating a delay temperature change appearing later than the temperature change of the temperature detected in consideration of the delay in the temperature change.
The thermal displacement calculator 32 calculates a delay response component ΔZ that changes in response to the delay temperature change calculated by the delay temperature calculator 34.₂Is calculated, and this delayed response component is replaced with the immediate response component ΔZ.₁Is added to. The correction means 33 performs a calculation for correcting a processing error based on the added total value, that is, the thermal displacement, and outputs the result.
The storage means 35, which simultaneously stores the final calculation result of the delay temperature calculation means 34 and the power-off time of the machine tool, provides compensation when the calculation of the delay temperature is interrupted.
[0035]
Instant response component ΔZ₁The thermal displacement correction method of the present invention in which only the above is considered is a method based on Expression (3). Instant response component ΔZ₁The immediate response component ΔZ₂The thermal displacement correction method based on the equation (2) in consideration of the above is preferable because the correction can be performed with higher accuracy.
[0036]
FIG. 2 is a block diagram showing a second embodiment of the present invention. The thermal displacement correction method of the second embodiment uses the delay response component ΔZ₂This is a method based on Expression (4) when only the above is considered.
The thermal displacement compensator 12a according to the second embodiment includes a temperature sensor S₁In anticipation of a delay in the temperature change of the temperature detected in, a virtual position P having substantially the same time constant as the time constant of the thermal displacement₁And a delay response component ΔZ that changes in response to the first delay temperature change calculated by the first delay temperature calculation means 34a.₂And a correcting means 33 for correcting a processing error based on the thermal displacement calculated by the thermal displacement calculating means 32.
[0037]
As a preferred mode, the thermal displacement compensator 12a of the second embodiment includes a temperature sensor S installed at a location where the influence of the heat generated by the heat source slowly affects the body 10.₃Further, there is provided a second delay temperature calculating means 34b for calculating a delay temperature change appearing later than the temperature change of the temperature detected in the above-mentioned manner in consideration of the delay in the temperature change.
The thermal displacement calculation means 32 calculates a second delay response component that changes in response to the delay temperature change calculated by the second delay temperature calculation means 34b, and uses this second delay response component as the previous delay response component. Component ΔZ₂Is added to. The correction means 33 corrects a processing error based on the added value, that is, the thermal displacement, and outputs the result.
In the second embodiment, one temperature sensor S₁The thermal displacement can be corrected only by the output of the temperature sensor S.₃The thermal displacement correction method taking into account the output of (1) is preferable because the correction can be performed with higher accuracy.
In the second embodiment, description of the same or equivalent functional portions as those in the first embodiment will be omitted.
[0038]
Hereinafter, a specific procedure of the first and second embodiments will be described with reference to FIGS.
FIG. 3 is a flowchart showing the operation of the first embodiment, FIG. 4 is a flowchart showing the operation of the second embodiment, FIG. 5 is a graph showing the change with time of the Z-axis thermal displacement, and FIG. 4 is a graph showing a temperature change of a synthesized temperature and a synthesized change of a synthesized temperature.
7 is a graph showing the relationship between the nose temperature change and the Z-axis thermal displacement, FIG. 8 is a graph showing the relationship between the head temperature change and the Z-axis thermal displacement, and FIG. 9 is a graph showing the Z-axis thermal displacement with respect to the combined temperature change. It is.
FIG. 10 is a graph illustrating a method of calculating a delay temperature change, FIG. 11 is a graph illustrating a method of calculating a delay response component from a temperature change, and FIG. 12 is a calculated Z-axis thermal displacement and an actually measured Z-axis thermal displacement. FIG. 13 is a graph showing the Z-axis thermal displacement with respect to the delayed temperature change.
[0039]
In the first embodiment, first, the time constant of the thermal displacement in the MC1, for example, in the Z-axis direction is calculated in advance based on the data in FIG. The horizontal axis in FIG. 5 is time, and the vertical axis is thermal displacement in the Z-axis direction. When calculating the time constant of the thermal displacement in the Z-axis direction, MC1 is set to the spindle rotational speed S (for example, S = 10,000 [min]^-1]) To operate continuously. Then, the thermal displacement in the Z-axis direction at the tip of the spindle 6 or at an appropriate place of the test bar temporarily attached to the tip of the spindle is actually measured as time-series data 39.
In the case where the main shaft is tilted due to heat generation, it is preferable to actually measure the thermal displacement at, for example, the root portion and the tip portion of the test bar.
Since the data 39 (marked by “○” in the figure) usually includes the effect of the change in room temperature, the room temperature correction having a saturation value 40 (for example, 43 [μm]) corrected for the effect of the change in room temperature. The data 41 (indicated by “●” in the figure) is calculated.
The “time constant” is “the time required for the output to reach 63.2% of the saturation value when a step-like input is applied in a linear first-order lag system”. Therefore, by applying the step input response function of the first-order lag element to the room temperature correction data 41 by the least square method, the thermal displacement time constant τ in the Z-axis direction_z(For example, τ_z= 0.57 [h]).
[0040]
Further, the nose temperature sensor S simultaneously detects the thermal displacement in the Z-axis direction.₁And head temperature sensor S₂The time constant of each temperature change is calculated based on the data on the temperature change of the temperature detected in step (1). The procedure for calculating the time constant is the same as that for calculating the time constant of the thermal displacement in the Z-axis direction.
The horizontal axis in FIG. 6 is time, and the vertical axis is temperature change. As shown, the nose temperature change T close to the heat source_NQuickly reaches the saturation temperature change “A” (A = 6.5 [° C.]), and its nose temperature time constant τ_NIs as small as 0.39 [h].
On the other hand, head temperature change T at a position farther from the heat source_HChanges its temperature with a delay. Therefore, it takes time to reach the saturation temperature change “B” (B = 3.7 [° C.]), so that the head temperature time constant τ_HIs as large as 1.31 [h].
[0041]
Next, as shown in FIG. 1 and FIG. 3, the MC 1 is activated to start the cutting of the workpiece 9 by the tool 7 (step 101). The temperatures of the nose position and the head position are detected (step 102), and the detection results are input to the combined temperature calculating means 31.
Next, the nose temperature change T_NAnd head temperature change T_HAnd the time constant τ of the thermal displacement data in the Z-axis direction shown in FIG._Z(Τ_Z= 0.57 [h]), a virtual temperature change T having the same time constant as_ZIs calculated.
[0042]
Nose temperature change T when synthesizing synthesis temperature_NAnd head temperature change T_HEach temperature mixing ratio M_NZ, M_HZIs calculated by the following equation, for example. The following formula is one example of the mixing method, and there are various other mixing methods. These temperature mixing ratios are values unique to each machine tool, and the values do not change even when the operating conditions such as the spindle rotation speed change. Therefore, the operation of calculating the temperature mixing ratio only needs to be performed once.

Here, k is an arbitrary coefficient (here, k = 10).
[0043]
The time constant τ of the thermal displacement in the Y-axis direction is the same as the procedure shown in FIG._Y= 1.15 [h], the combined temperature change T in the Y-axis direction as shown in FIG._YCan be similarly calculated.
Nose temperature change T in this case_NAnd head temperature change T_HEach temperature mixing ratio M_NY, M_HYIs calculated by the following equation.

Here, k is an arbitrary coefficient (here, k = 10).
[0044]
The composite temperature calculating means 31 calculates the temperature mixture ratio M calculated by the equations (7) and (8)._NZ, M_HZAnd each temperature change T_N, T_HAnd the resultant temperature change T_ZIs calculated (step 103). Equation (11) is a temperature conversion equation, and M_NZ, M_HZIs the coefficient.
T_Z= M_NZ・ T_N+ M_HZ・ T_H= 1.24 × T_N+ 0.53 × T_H…… (11)
[0045]
The horizontal axis in FIG. 7 is the nose temperature change T._NThe horizontal axis in FIG._H7, and the vertical axis in FIGS. 7 and 8 indicates the thermal displacement in the Z-axis direction. As shown, each temperature change T_N, T_HAnd Z-axis thermal displacement have no correspondence such as proportionality.
[0046]
On the other hand, as shown in FIG._ZAnd a Z-axis thermal displacement, a region having a linear correlation represented by a straight line 42 having an inclination α (α = 3.88) occurs. This slope α is the synthetic temperature change T_ZInstantaneous response component ΔZ₁Is calculated, and the following equation is eventually established.
ΔZ₁= Α · T_Z… (12)
The thermal displacement calculating means 32 uses the equation (12) to calculate the resultant temperature change T_ZResponse component ΔZ that changes in response to₁Is calculated (step 104).
[0047]
Next, the delay response component ΔZ₂Is determined (step 105), and if not, the processing error is corrected by the correction unit 33 based on the result calculated by the thermal displacement calculation unit 32. Specifically, for example, the origin position of the rectangular coordinate system is offset (step 106).
Thereafter, it is determined whether or not the correction is to be terminated (step 107). When the correction is to be terminated, MC1 is stopped (step 108), and the entire procedure is terminated. If the correction is not completed, the process returns to step 102.
[0048]
On the other hand, in the determination of step 105, the delay response component ΔZ₂Is considered, the head temperature sensor S₂The temperature at the head position detected in step (1) is input to the delay temperature calculating means 34.
Since the column 3 and the like have a large mass and are separated from the main shaft 6 which is a main heat source, the temperature change appears after the head position where the temperature change is delayed. This delayed temperature change is, for example, in the region D where a long time has passed in FIG._ZAn error is given to the linear correlation between the pressure and the Z-axis thermal displacement.
[0049]
There is a “dummy method” as a method of calculating the delay temperature change, and the head temperature change T_HThe behavior of the delayed temperature change Y that appears even further later is expected by setting the dummy heat capacity C.
Specifically, it is obtained as an approximate solution of the differential equation (13).
C · dY / dt + Y = T_H…… (13)
[0050]
The horizontal axis in FIG. 10 indicates time, and the vertical axis indicates the delay temperature change. The following equation is obtained from FIG.
Y = Y₀+ (DY₀/ Dt + dY / dt) / 2 · Δt (14)
[0051]
From equation (13), dY / dt and dY₀When / dt is calculated and substituted into the equation (14), the equation for calculating the delay temperature change Y is the following equation (15). This equation (15) is a temperature conversion equation, and the symbol C corresponds to the coefficient. The delay temperature change calculated by the equation (15) is also a virtual temperature change, and is a kind of the creation temperature change similarly to the synthetic temperature change by the equation (11).
Y = [T_H0+ T_H+ (C / Δt) · Y₀-Y₀] / [(C / Δt) +1] (15)
Here, Δt: calculation interval
T_H: Head temperature change input
T_H0: Previous head temperature change input
Y: Delayed temperature change output
Y₀: Previous delay temperature change output
C: Dummy heat capacity
It is.
Since the unit of the delay temperature change output Y is [° C.], the delay response component ΔZ₂To calculate, an internal correction coefficient e which means a conversion coefficient between temperature and thermal displacement is used.
ΔZ₂= EY ・ (16)
[0052]
In FIG. 11, the horizontal axis is time, the vertical axis from zero on the left is the head temperature change, and the vertical axis from zero on the right is the delay response component. In the drawing, the time series data 43 of the head temperature change, the error 44 obtained by subtracting the value of the vertical axis of the straight line 42 from the Z-axis thermal displacement in FIG. 15) and the calculation result 45 obtained through equation (16) are shown. The error 44 and the calculation result 45 are respectively a delay response component ΔZ₂Correspond to actual data and operation data.
In other words, by appropriately selecting the dummy heat capacity C included in the equation (15) and the internal correction coefficient e in the equation (16), the calculated data can be made closer to the actual data, and the best value can be determined. The values of the heat capacity C and the coefficient e determined here are unique values for each machine tool, and the values do not change even when the operating conditions such as the rotation speed of the spindle change, so that this operation may be performed only once.
For example, based on the time series data 43 of the head temperature change, the delay response component ΔZ is repeatedly calculated.₂Is calculated, the results in Table 1 are obtained assuming that C = 900 and e = -4. Delay response component ΔZ in Table 1₂Is a delay response component of the column 3 and the like, and corresponds to an error in the vertical axis direction between the Z-axis thermal displacement and the straight line 42 in FIG.
[0053]
[Table 1]

[0054]
The delay temperature calculating means 34 uses the equation (15) in which the dummy heat capacity C is determined, and calculates the head temperature change T_HIs calculated (step 109). Next, when this delay temperature change Y is substituted into Expression (16) in which the internal correction coefficient e is determined, the delay response component ΔZ₂Is obtained (step 110).
First, using the equation (12), the immediate response component ΔZ₁Is calculated, the delay response component ΔZ calculated in this manner is₂To the immediate response component ΔZ₁To calculate the Z-axis thermal displacement ΔZ (step 111).
[0055]
The vertical axis in FIG. 12 is the same measured Z-axis thermal displacement as that shown in FIG. 5, and the horizontal axis is the temperature sensor S₁, S₂Is the Z-axis thermal displacement ΔZ estimated through the procedure up to step 111 using the temperature data detected in step (1). In the calculation of the thermal displacement ΔZ, the instantaneous response component ΔZ is set by setting the overall correction coefficient “a” of the equation (2) to 1.₁Is developed by the following equations (11) and (12).
ΔZ = 4.8 × T_N+ 2.1 × T_H+ ΔZ₂…… (17)
[0056]
The nose temperature change T of this equation (17)_NAnd the head temperature change T_HThe coefficients of the term correspond to the internal correction coefficients b and c in the above equation (5), respectively, and are calculated by the following equation. The values of these coefficients are obtained by integrating the thermal characteristics of the machine tool when calculating the immediate response component using the temperature detected from the machine body 10.
b = α · M_NZ= 3.88 × 1.24 = 4.8 (18)
c = α · M_HZ= 3.88 × 0.53 = 2.1 (19)
Further, the delay response component ΔZ of the equation (17)₂Uses the results shown in Table 1.
[0057]
In FIG. 12, the actually measured Z-axis thermal displacement on the vertical axis substantially coincides with the value of the Z-axis thermal displacement ΔZ on the horizontal axis obtained by Expression (17) on a straight line 46 having a 45-degree inclination. This means that both values are the same, and therefore, the Z-axis thermal displacement can be predicted with sufficiently high accuracy by calculation using relatively small temperature data detected from the airframe 10.
In this manner, the thermal displacement is corrected by correcting the processing error by the correcting means 33 based on the Z-axis thermal displacement ΔZ calculated in step 111 (step 112), and the workpiece 9 is cut with high precision. Can be processed.
Thereafter, it is determined whether or not the correction has been completed (step 107). When the correction is to be terminated, the MC1 is stopped (step 108), and the entire procedure ends. If the correction is not completed, the process returns to step 102.
[0058]
Next, the procedure of the second embodiment will be described. In the second embodiment, it is not necessary to calculate the time constant of the thermal displacement or the temperature change in advance.
As shown in FIGS. 2 and 4, the MC 1 is activated to start the cutting of the workpiece 9 by the tool 7 (Step 201). The temperature at a first location, for example, the nose position is detected (step 202), and this detection signal is input to the first delay temperature calculating means 34a.
Next, the nose temperature change T_NA first delayed temperature change Y having the same time constant as the time constant of the Z-axis thermal displacement appearing later₁Behavior of the dummy heat capacity C₁Set and expect.
[0059]
By performing the same expansion as the above equations (13) and (14), the first delay temperature change Y₁Is calculated by the following equation.

Here, Δt: calculation interval
T_N: Nose temperature change input
T_NO: Previous nose temperature change input
Y₁: First delay temperature change output
Y₁₀: Previous first delay temperature change output
C₁： Dummy heat capacity
It is.
[0060]
As shown in FIG. 13, the Z-axis thermal displacement and the first delayed temperature change Y₁A region having a linear correlation represented by a straight line 47 having a slope e is generated. This slope e is equal to the first delay temperature change Y₁Is a proportionality constant when calculating the Z-axis thermal displacement corresponding to the above equation, and the Z-axis thermal displacement calculated here corresponds to the first term of the equation (6).
The dummy heat capacity C is set such that the result of the repetitive calculation of the first term coincides with the time-series data of the Z-axis thermal displacement shown in FIG.₁And the value of the internal correction coefficient e are appropriately selected. Heat capacity C determined here₁And the value of the coefficient e are values unique to each machine tool, and this operation may be performed once.
In the first delay temperature calculating means 34a, the dummy heat capacity C₁The nose temperature change T is calculated using the equation (20)_NLagged temperature change Y corresponding to₁Is calculated (step 203).
In the thermal displacement calculating means 32, the delay temperature change Y₁Is substituted into the first term of the equation (6) in which the internal correction coefficient e is determined, to calculate a first delay response component (step 204).
[0061]
Next, it is determined whether or not the second delay response component is to be considered (step 205). (Step 206).
Thereafter, it is determined whether or not to end the correction (step 207). If it is to be ended, the MC1 is stopped (step 208), and the entire procedure ends. If the correction is not completed, the process returns to step 202.
[0062]
On the other hand, when the second delay response component is considered in the determination in step 205, the temperature sensor S₃And the temperature change T at the second location, eg, the column_CIs detected (step 209) and input to the second delay temperature calculating means 34b.
Temperature change T at column position_CIs the first delayed temperature change Y, for example, in a region D where a long time has passed in FIG.₁Gives an error to the linear correlation between the pressure and the Z-axis thermal displacement. By using the above “dummy method” again, the column temperature change T_CMore second delay temperature change Y₂Behavior of the dummy heat capacity C₂Set and expect.
Second delay temperature change Y₂Is calculated by the following equation.

Here, Δt: calculation interval
T_C: Column temperature change input
T_CO: Previous column temperature change input
Y₂: Second delay temperature change output
Y₂₀: Previous second delay temperature change output
C₂： Dummy heat capacity
It is.
[0063]
Note that this second delay temperature change Y₂Corresponds to the second term of the above equation (6) including the conversion coefficient f between the temperature and the thermal displacement.
In FIG. 13, the dummy heat capacity C is set so that the result of the repetitive calculation of the second term matches the error obtained by subtracting the straight line 47 from the Z-axis thermal displacement in FIG.₂And the value of the coefficient f are appropriately selected. Heat capacity C determined here₂And the value of the coefficient f are values unique to each machine tool, and this operation may be performed once.
In the second delay temperature calculating means 34b, the dummy heat capacity C₂Using the equation (21) in which the column temperature change T_CDelay temperature change Y corresponding to₂Is calculated (step 210). Next, the second delay temperature change Y is added to the second term of the equation (6) in which the internal correction coefficient f is determined.₂To calculate a second delay response component (step 211).
The thermal displacement calculating means 32, which has calculated the first delay response component corresponding to the first term of the equation (6), adds the second delay response component calculated in this way to obtain the Z-axis thermal displacement. ΔZ₂Is calculated (step 212).
[0064]
Z-axis thermal displacement ΔZ calculated in step 212₂By correcting the processing error by the correction means 33 based on the above, the thermal displacement is corrected (step 213), and the workpiece 9 can be processed with high accuracy.
Thereafter, it is determined whether or not the correction has been completed (step 207). When the correction is to be terminated, MC1 is stopped (step 208), and the entire procedure is terminated. If the correction is not completed, the process returns to step 202.
[0065]
14 to 17 show actual measurement data when the MC1 is operated in a real machine.
FIG. 14 is a graph showing data of the Z-axis thermal displacement, FIG. 15 is a graph showing other data of the Z-axis thermal displacement, FIG. 16 is a graph showing data of the Y-axis thermal displacement, and FIG. 6 is a graph showing data of the first embodiment.
In any case, the rotational speed S of the main shaft 6 is S = 10,000 [min].^-114 and FIG. 16 show the case of continuous operation. FIGS. 15 and 17 show the thermal displacement in the case of rotating for about 50 minutes and stopping for 10 minutes (stopping once for 70 minutes in the middle), as indicated by the time chart line 51. FIG. The horizontal axis in FIGS. 14 to 17 is time.
[0066]
As shown in FIGS. 14 and 16, the maximum thermal displacement before correction indicated by solid lines 48 and 52 in the figures was about 40 [μm]. On the other hand, in the present invention, the target value of the corrected thermal displacement can be made closer to zero. That is, when the thermal displacement is corrected according to the present invention, the residual thermal displacement can be reduced to ± 5 [μm] or less as shown by broken lines 49 and 53 in the figure.
It has been confirmed that according to the present invention, even when the thermal displacement before correction is 100 [μm] or more, the thermal displacement after correction can be reduced to ± 5 [μm] or less. As described above, according to the present invention, thermal displacement can be corrected with high accuracy.
Further, according to the present invention, even when the intermittent operation is performed as shown in FIGS. 15 and 17, the residual heat displacement after correction can be made ± 5 [μm] or less as shown by broken lines 50 and 54. it can.
[0067]
(Third embodiment)
18 to 28 are views for explaining the third embodiment.
For example, as shown in FIG. 18, the MC 1 a includes a main shaft 6 on which the tool 7 is mounted, and a main shaft head 5 a that rotatably supports the main shaft 6 via a main bearing 20 and an upper bearing (another bearing) 22. Have. A built-in motor 21 for rotating the main shaft 6 is disposed between the bearings 20 and 22, and is built in the main shaft head 5a together with the both bearings. The main bearing 20 positions the main shaft 6 in the center axis direction, and the upper bearing 22 holds the main shaft 6 that expands and contracts due to thermal displacement so as to be slidable in the center axis direction. Therefore, even if the upper bearing 22 and the motor 21 generate heat by rotation, the main shaft 6 extends upward, so that the tool 7 is not affected.
As a result, in the case of the MC 1a, only the main bearing 20 should be considered as a heat source to perform the thermal displacement correction. Head temperature sensor S as temperature detecting means for detecting a temperature change of main bearing 20₂Are attached to the spindle head 5a.
As long as the temperature changes due to the main bearing 20 as a heat source, a temperature other than the head position, for example, the temperature of the nose or the column may be detected. Further, the machine tool may be of a type in which the spindle holds a workpiece.
The same or corresponding parts as in the first and second embodiments are denoted by the same reference numerals, and description thereof will be omitted.
[0068]
Here, the principle of the thermal displacement correction in the third embodiment will be described.
A description will be given by taking correction in the Z-axis direction as an example, as in the first and second embodiments. The equation for calculating the thermal displacement in the Z-axis direction in the third embodiment is shown by the following equation.
ΔZ = a · (ΔZ₃+ ΔZ₄) …… (22)
Here, ΔZ: Z-axis thermal displacement
a: Overall correction coefficient (same as that of equation (2))
ΔZ₃: Creation displacement component of Z-axis thermal displacement
ΔZ₄: Delay response component of Z-axis thermal displacement
It is.
That is, the arithmetic expression (22) is based on the created displacement component ΔZ calculated based on the created temperature change having the same time constant as the thermal displacement time constant from the temperature change.₃And a delay response component ΔZ in which a thermal displacement appears with a delay with respect to a temperature change.₄And
The temperature change handled here is calculated by the difference between the temperature output from each temperature sensor and the reference temperature. The reference temperature used in the first and second embodiments is also employed.
[0069]
It should be noted that the present invention provides a method for generating displacement component ΔZ as shown in equation (23).₃It is also possible to use an arithmetic expression based only on the above.
ΔZ = a · ΔZ₃…… (23)
Generating displacement component ΔZ₃Is calculated by the following equation.
ΔZ₃= GY₃…… (24)
Where Y₃： Creation temperature change [℃]
g: Internal correction coefficient [± μm / ° C]
It is.
[0070]
Formula (24) used in the third embodiment is a formula for calculating the thermal displacement of the body 10 based on the output of the temperature sensor installed at one location. Then, the head temperature sensor S₂Temperature change Y obtained by developing the temperature change T of the temperature detected at₃From the displacement component ΔZ₃Is calculated.
It is sufficient that at least one temperature sensor is installed, but the temperature sensor is appropriately added according to the number of heat sources. The temperature sensor may be installed at a location other than the spindle head 5a as long as the location is affected by the heat generated by the heat source.
[0071]
On the other hand, the delay response component ΔZ₄Is calculated as follows.
ΔZ₄= HY₄…… (25)
Where Y₄: Delayed temperature change [° C]
h: Internal correction coefficient [± μm / ° C]
It is.
In equation (25), the head temperature sensor S₂Delayed temperature change Y calculated with a delay in temperature change T of temperature detected in₄From the delay response component ΔZ₄Is calculated.
It is sufficient that at least one temperature sensor is installed, but the temperature sensor is appropriately added according to the number of heat sources.
[0072]
FIG. 18 is a block diagram showing a third embodiment of the present invention.
The thermal displacement compensator 12b of the third embodiment includes a temperature sensor S₂A virtual position P having the same time constant as the time constant of the thermal displacement by developing the temperature detected in₁And a generating displacement component ΔZ that changes in response to the generating temperature change calculated by the generating temperature calculating means 31a.₃And a correcting means 33 for correcting a processing error based on the thermal displacement calculated by the thermal displacement calculating means 32.
[0073]
In a preferred embodiment, the thermal displacement compensator 12b further includes a delay temperature calculator 34. The delay temperature calculating means 34 includes a temperature sensor S₂A delayed temperature change appearing later than the temperature change of the temperature detected in is calculated in consideration of the delay in the temperature change.
The thermal displacement calculator 32 calculates a delay response component ΔZ that changes in response to the delay temperature change calculated by the delay temperature calculator 34.₄Is calculated, and the generated displacement component ΔZ is calculated.₃Is added to. On the basis of the added thermal displacement, the correction means 33 corrects the processing error and outputs the signal.
In the third embodiment, description of the same or equivalent functional portions as those in the first and second embodiments will be omitted.
[0074]
Hereinafter, a specific procedure of the third embodiment will be described with reference to FIGS.
FIG. 19 is a flowchart showing the operation of the third embodiment, and FIG. 20 is a graph showing the temporal change of the Z-axis thermal displacement and a representative example of the temperature change T of the temperature detected at the head position (sample temperature change). FIG. 21 is a graph showing a standard relationship between a sample temperature change and a Z-axis thermal displacement.
FIG. 22 is a graph showing a sample temperature change, a temperature change “A” having a smaller time constant than the sample temperature change, and a temperature change “B” having a larger time constant than the sample temperature change. FIG. 23 shows the created temperature change Y created using the sample temperature change in addition to the temperature change shown in FIG._3A, Y_3BIt is a graph which shows ("O" mark in a figure). FIG. 24 is a graph showing the Z-axis thermal displacement with respect to the generating temperature change, and FIG. 25 is a graph showing the Z-axis thermal displacement with respect to the generating temperature change when the Z-axis thermal displacement includes a delay response component.
FIG. 26 shows a sample temperature change, a delay temperature change C appearing later than the sample temperature change, and a delay temperature change Y created using the sample temperature change.₄FIG. 27 is a graph showing a delay response component with respect to a delay temperature change. FIG. 28 is a graph showing the Z-axis thermal displacement with respect to the thermal displacement estimated from the generated temperature change and the delayed temperature change.
[0075]
In the third embodiment, the thermal displacement in the Z-axis direction is detected in advance. At the same time, the head temperature sensor S₂The respective time constants are calculated based on the data on the temperature change of the temperature detected in step (1). The procedure for calculating the time constant is the same as that for calculating the time constant of the thermal displacement in the Z-axis direction in the first embodiment.
FIG. 20 shows that the main shaft rotates at a constant speed (rotation speed S = 10,000 [min]).^-1]), The head temperature sensor S₂5 shows time-series data of Z-axis thermal displacements (thermal displacements “A, B”) according to two typical examples of the temperature change T of the temperature detected at (1) (sample temperature change).
The time constant of the Z-axis thermal displacement is the time constant τ of the sample temperature change T._SIf it is smaller, that is, if the thermal displacement appears earlier, the thermal displacement “A” (time constant τ_A). On the other hand, the thermal displacement time constant is_SIf it is larger, that is, if the thermal displacement appears slowly, the thermal displacement “B” (time constant τ_B). Therefore, depending on the thermal displacement characteristics of the MC 1a, the data actually becomes either one of the thermal displacements “A, B”.
The balance between the extracted time constant of the Z-axis thermal displacement and the time constant of the sample temperature change T represents a thermal characteristic peculiar to each machine tool. Also little change. Therefore, the work of calculating both time constants may be performed once.
[0076]
Next, as shown in FIGS. 18 and 19, the MC 1a is activated to start cutting the workpiece 9 by the tool 7 (step 301). Further, the temperature of the head position is detected (step 302), and is input to the generating temperature calculating means 31a.
[0077]
However, for example, based on the data of FIG. 20, the relationship between the sample temperature change T and the thermal displacement “A” and the relationship between the sample temperature change T and the thermal displacement “B” are expressed as shown in FIG. They become arcuate curves 55 and 56, respectively.
That is, since the sample temperature change T and the Z-axis thermal displacement do not have a simple linear relationship, the thermal displacement cannot be immediately estimated from the temperature change of the temperature detected from the head position as needed.
[0078]
Therefore, using the "linearization technique", the temperature change of the temperature detected from the head position is used to calculate the generated temperature change having a time constant substantially equal to the time constant of the thermal displacement. FIG. 22 shows that the time constant is τ_SSample temperature change 57 and the same time constant τ as the thermal displacement “A”_ASimulated example 58 of a temperature change “A” having the same time constant τ as the thermal displacement “B”_BAnd a simulation example 59 of the temperature change “B” having
The temperature changes shown by the curves 57 to 59 all have the value T_maxAnd an example of each time constant is shown below.
τ_A= 5 [min]
τ_S= 10 [min]
τ_B= 15 [min]
[0079]
The behavioral relationship between the temperature changes 57 to 59 can be expressed by the following differential equations (26) to (28), respectively.
τ_SDT / dt + T = X (26)
τ_A・ DY_3A/ Dt + Y_3A= X ... (27)
τ_B・ DY_3B/ Dt + Y_3B= X (28)
Expressions (27) and (28) can be represented by a general expression (29) of a differential equation.
τ_Z・ DY₃/ Dt + Y₃= X (29)
Here, T: head temperature sensor S₂Temperature change detected in [℃]
X: Temperature change of heating part [° C]
Y₃： Creation temperature change [℃]
Y_3A: Creation temperature change to create temperature change "A" [° C]
Y_3B: Creation temperature change to create temperature change "B" [° C]
τ_Z： Time constant of Z axis thermal displacement [min]
It is.
[0080]
According to the equation (26), the temperature change X of the heat generating part can be found from the sample temperature change T, and this value X is substituted into the equation (29). Then, the time constant τ of the sample temperature change_STime constant τ different from_ZTemperature change Y having₃Is obtained. Time constant τ_ZIs determined by the thermal characteristics of the MC 1a and is a value unique to each machine tool.
[0081]
When actually calculating by the generating temperature calculating means 31a (FIG. 18), a solution is calculated by repeating the following equations (30) and (31) obtained by discretizing the differential equations (26) and (29), respectively. (Step 303).
X = τ_S・ (TT₀) / Δt + T₀…… (30)
Y₃= (2X + 2τ_Z・ Y₃₀/ Δt-Y₃₀) / (2τ_Z/ Δt + 1) (31)
Here, Δt: calculation interval [min]
T₀: Previous sample temperature change T input [° C]
Y₃₀: Previous generation temperature change Y₃Output [° C]
It is.
[0082]
FIG. 23 shows the created temperature change Y created by using the sample temperature change T by the equations (30) and (31) in addition to the sample temperature change T and the temperature change “A, B”.₃(Specifically, the creation temperature change Y_3AOr Y_3B) Are indicated by “○”. The mark “」 ”indicates a case where the measurement interval, that is, the calculation interval Δt is 1.0 [min].
As described above, by the repetitive calculation using the equations (30) and (31), it is possible to generate a generated temperature change having substantially the same time constant as the thermal displacement having an arbitrary time constant.
[0083]
Creation temperature change Y₃Has the same time constant as the thermal displacement, and therefore has a linear relationship with the Z-axis thermal displacement, as shown by the straight line 60 in FIG. The slope g of the straight line is the creation temperature change Y₃And shows the correlation between the thermal displacement. The thermal displacement calculating means 32 uses the equation (24) to calculate the generated temperature change Y₃From the thermal displacement (ie, the generated displacement component ΔZ₃) Is calculated (step 304).
[0084]
As described above, the linearization method basically allows a sensitive thermal displacement having a small time constant to be estimated from a temperature change T of a temperature detected at a head position away from a heat source, for example. Creation displacement component ΔZ calculated by this method₃Is the immediate response component ΔZ of equation (2).₁This method alone enables accurate thermal displacement correction.
Next, the delay response component ΔZ₄Is determined (step 305). If not, the processing error is corrected by the correction unit 33 based on the result calculated by the thermal displacement calculation unit 32 (step 306).
Thereafter, it is determined whether or not to end the correction (step 307). When the correction is to be ended, the MC 1a is stopped (step 308), and the entire procedure ends. If the correction is not completed, the process returns to step 302.
[0085]
On the other hand, the delay response component ΔZ₄Is considered, the head temperature sensor S₂The temperature at the head position detected in step (1) is input to the delay temperature calculating means 34.
The column 3 and the like have a large mass and are distant from the main shaft 6 which is a main heat source, so that the temperature change appears later than the head position where the temperature change is delayed. This delayed temperature change is similar to that shown in FIGS. 9 and 13 of the first and second embodiments, and the generated temperature change Y in the region D where a long time has passed in FIG.₃Gives an error to the linear correlation between the pressure and the Z-axis thermal displacement.
[0086]
Delay response component ΔZ₄Is taken into consideration, the head temperature sensor S₂Temperature change Y appearing later than temperature change T of temperature detected in₄Behavior of the dummy heat capacity C₄Set and expect.
Delay temperature change Y₄Is calculated by the following equation.

Here, Δt: calculation interval [min]
T: Head temperature sensor S₂Temperature change input of temperature detected by
T₀： Temperature change T input [° C]
Y₄: Delayed temperature change output [° C]
Y₄₀: Previous delay temperature change Y₄Output [° C]
C₄: Dummy heat capacity [min]
It is.
[0087]
FIG. 26 shows the head temperature sensor S₂A sample temperature change 57, which is a typical example of the temperature change T of the temperature detected in the above, and a simulated example 61 of a lag temperature change showing the same thermal behavior as the lag response component (temperature change C, time constant τ_C). Further, FIG. 26 shows the delay temperature change Y created using the sample temperature change T by the equation (32).₄Are indicated by “O” marks.
When the temperature change C has passed for a long time, the saturation value T_maxIs assumed to be equal to the sample temperature change T, and a mark “○” indicates a case where the measurement interval, that is, the calculation interval Δt is 1.0 [min].
Thus, the heat capacity C of the equation (32)₄Can be set to any time constant τ_C(Τ_C> Τ_S), A delayed temperature change Y that behaves substantially the same as the temperature change C₄Can be created.
[0088]
This delay temperature change Y₄27 and the delay response component, a straight line 62 having a linear relationship as shown in FIG. Actually, there is no particular need to take a procedure for extracting the temperature change C. For example, using the sample temperature change T, the dummy heat capacity C₄And the heat capacity C so that the repetition calculation result of appropriately selecting the internal correction coefficient h in the equation (25) matches the error obtained by subtracting the straight line 60 from the Z-axis thermal displacement (the line including the region D) in FIG.₄And the best value of the coefficient h are determined. Heat capacity C determined here₄And the value of the coefficient h are values unique to each machine tool, and this operation may be performed once.
In the delay temperature calculating means 34, the dummy heat capacity C₄Using the equation (32) in which the head temperature sensor S₂Temperature change Y corresponding to the temperature change T of the temperature detected at₄Is calculated (step 309). Next, this delay temperature change Y₄Into the equation (25) in which the internal correction coefficient h is determined, the delay response component ΔZ₄Are obtained (step 310).
First, using equation (24), the creation displacement component ΔZ₃Is calculated, the delay response component ΔZ calculated in this manner is₄Create displacement component ΔZ₃To calculate the Z-axis thermal displacement ΔZ (step 311).
[0089]
The vertical axis of FIG. 28 is the actually measured Z-axis thermal displacement, and the horizontal axis is the Z-axis thermal displacement ΔZ estimated through the procedure up to step 311 using the sample temperature change T. In the calculation of the thermal displacement ΔZ, the following equation with the overall correction coefficient “a” being 1 is used.
ΔZ = g · Y₃+ HY₄…… (33)
The actually measured Z-axis thermal displacement on the vertical axis shown in FIG. 28 substantially coincides with the value of the Z-axis thermal displacement ΔZ on the horizontal axis obtained by equation (33) on a straight line 63 having a 45 ° inclination. This means that they have the same value.
Therefore, the head temperature sensor S installed at a location affected by the heat generated by the main bearing 20 of the body 10₂, The Z-axis thermal displacement can be predicted with sufficiently high accuracy.
[0090]
In this manner, the thermal displacement is corrected by correcting the processing error by the correcting means 33 based on the Z-axis thermal displacement ΔZ calculated in step 311 (step 312), and the workpiece 9 is cut with high precision. can do.
Thereafter, it is determined whether or not the correction has been completed (step 307). When the correction is to be terminated, the MC 1a is stopped (step 308), and the entire procedure ends. If the correction is not completed, the process returns to step 302.
[0091]
(Fourth embodiment)
Next, a fourth embodiment to which the linearization method is applied will be described with reference to FIGS.
In the first to third embodiments, an example in which the number of heat sources that cause thermal displacement is one has been mainly described. However, in the present invention, there are a plurality of heat sources that cause thermal displacement, and Even when it is assumed that the influences of the heat sources are independent of each other, the structure is of an expandable type.
The fourth embodiment is a case of a machine tool in which a plurality of heat sources (so-called multiple heat sources) generate thermal displacement while affecting each other, and supplement a portion that has not been described in each of the above embodiments.
[0092]
FIG. 29 is a block diagram, and FIG. 30 is a sectional view of a headstock.
The NC lathe 64 is a machine tool that uses a headstock 66 as a machine body as a heat source. As shown in the drawing, the NC lathe 64 includes a spindle 69 that grips a workpiece 68 via a chuck 65 and a pawl 67, a headstock 66, and a built-in motor 70. The headstock 66 rotatably supports the spindle 69 via a front bearing 71 on the processing position side that supports the spindle 69 and a rear bearing 72 on the non-processing position side.
The motor 70 including the rotor 70 a is disposed between the front and rear bearings 71 and 72 and is built in the headstock 66 to drive the main shaft 69 to rotate.
In the rear bearing 72, the main shaft 69 is set in the center axis direction 0.₁Angular contact ball bearings are used. The main shaft 69 which expands and contracts due to thermal displacement moves inside the front bearing 71 in the central axis direction O₁It can be expanded and contracted. Since a large load is applied to the front bearing 71 near the workpiece 68, a double-row cylindrical roller bearing having a large rated load is used as the front bearing 71 in order to withstand this load and improve cutting performance.
In the structure shown here, the main shaft 69 is directly driven by the built-in motor 70, so that high-speed rotation is possible as compared with the structure in which the main shaft 69 is driven by a belt. In addition, since vibration of the main shaft 69 can be suppressed, high-precision cutting can be performed.
[0093]
When the NC lathe 64 is activated to rotate the rotor 70a of the built-in motor 70, and the front and rear bearings 71, 72 and the rotor 70a generate heat, respectively, the main shaft 69 extends forward (to the right in the figure) to move the workpiece 68. Central axis direction O₁To lower the machining accuracy.
Therefore, in the thermal displacement compensator 12b, the three temperature sensors s are located near the bearings 71, 72 before and after the heat source and the stator of the motor 70, respectively.₁, S₂, S₃Are attached to the headstock 66.
[0094]
Thus, the Z-axis direction (that is, the central axis direction O)₁In the case where there are a plurality of heat sources that affect the thermal displacement of (1), highly accurate thermal displacement correction can be performed by applying the linearization method, for example.
Temperature sensor s as temperature detecting means₁, S₂, S₃Are input to the A / D converter 13 via the circuits 36a, 36b and 36c, and the output signal from the A / D converter 13 is input to the generating temperature calculating means 31a and the delay temperature calculating means 34. Is done. The other configuration is the same as that of the third embodiment, and the description is omitted.
[0095]
In the case of multiple heat sources, the thermal displacement due to the influence of each heat source is individually calculated by the thermal displacement calculating means 32. The total thermal displacement Δz in the Z-axis direction is represented by the following general formula.
Δz = a · (Δz₁+ Δz₂+ ... + Δz_n)… (34)
Δz₁= Β₁・ Y_A1+ Γ₁・ Y_B1
Δz₂= Β₂・ Y_A2+ Γ₂・ Y_B2
…
…
Δz_n= Β_n・ Y_An+ Γ_n・ Y_Bn
Here, Δz: Z-axis total thermal displacement
Δz₁Or Δz_n: Z-axis direction by first to n-th heat sources
Thermal displacement
a: Overall correction coefficient (same as that of equation (2))
β₁Or β_n： Internal correction coefficient related to change in generating temperature
γ₁Or γ_n: Internal correction coefficient for delay temperature change
Y_A1Or Y_An： Creation temperature change
Y_B1Or Y_Bn: Delayed temperature change
It is.
Therefore, equation (33) of the third embodiment is obtained by calculating the thermal displacement Δz of the first term of equation (34).₁Is equivalent to Further, in the third embodiment, since there is one heat source, the calculation is performed by setting the second and subsequent terms of the equation (34) to zero.
[0096]
In the fourth embodiment, since there are three heat sources, n = 3, and the equation (34) can be expanded as follows.
Δz = a · (Δz₁+ Δz₂+ Δz₃) …… (35)
Δz₁= K₁・ Δz
Δz₂= K₂・ Δz ... (36)
Δz₃= K₃・ Δz
K₁= PT₁/ (PT₁+ Q · T₂+ RT₃)
K₂= QT₂/ (PT₁+ Q · T₂+ RT₃)… (37)
K₃= RT₃/ (PT₁+ Q · T₂+ RT₃)
Where T₁: Temperature sensor₁Temperature change of temperature detected in
T₂: Temperature sensor₂Temperature change of temperature detected in
T₃: Temperature sensor₃Temperature change of temperature detected in
P, Q, R: Internal correction coefficient
It is.
[0097]
The internal correction coefficients P, Q, and R in the above three equations (37) are expressed by the respective temperature changes T₁Or T₃And is determined by the following formula from the difference in the saturation value of the sample temperature change in the test in which the heat generation condition is changed three or more times.
The balance of the values of the coefficients P, Q, and R determined here represents a thermal characteristic peculiar to each type of machine tool. Even if the operating condition such as the spindle speed changes, the change is small. You only need to do it once.
PT_1S+ Q · T_2S+ RT_3S= Δz (38)
Where T_1S: Sample temperature change T₁Saturation value of
T_2S: Sample temperature change T₂Saturation value of
T_3S: Sample temperature change T₃Saturation value of
It is.
[0098]
By substituting the values of the internal correction coefficients P, Q, and R into equation (37), the coefficient K₁, K₂, K₃Are determined, the three equations (36) can be expressed as shown in FIG. As a result, each thermal displacement Δz due to the influence of the heat sources 71, 72, 70a₁, Δz₂, Δz₃Can be made linear by a linearization method or the like. FIG. 31 is a graph showing the Z-axis thermal displacement of each heat source.
Since a turning machine tool such as an NC lathe usually has a small heat capacity and is more sensitive to thermal displacement than an MC, the method of the third embodiment, which has no restriction on the mounting position of the temperature sensor, is particularly effective.
In addition, if the method of the third embodiment is used, one temperature sensor may be provided for one heat source, so that the number of temperature sensors can be reduced in the case of the NC lathe 64 having multiple heat sources.
[0099]
FIG. 32 is a graph showing the measured data of the Z-axis thermal displacement when the NC lathe 64 is operated on an actual machine by combining the linearizing method and the dummy method.
As shown, the thermal displacement before correction indicated by solid lines 73 and 74 in the figure was about 70 [μm]. On the other hand, when the thermal displacement is corrected according to the present invention, the thermal displacement is reduced to ± 10 [μm] or less as shown by broken lines 73a and 74a in the figure. In addition, the symbol S in the figure means the spindle speed.
[0100]
(Fifth embodiment)
Next, a fifth embodiment will be described with reference to FIGS.
In the fifth embodiment, any one of the above-described methods according to the present invention is applied to a machine tool having a plurality of spindles to perform thermal displacement correction.
The spindle holds one of a workpiece and a tool. The temperature of the spindle is adjusted by a temperature controller to make the thermal displacement of each spindle substantially uniform, and the temperature detection means detects a temperature change of the body of at least one spindle to thereby correct the thermal displacement.
In the fifth embodiment, the same or corresponding parts as those in the above embodiments are denoted by the same reference numerals, and the description thereof will be omitted.
[0101]
For example, the machine tool shown in FIG. 33 is a vertical MC75 having a multi-axis head, and is used to simultaneously machine four workpieces to the same shape with four tools. A cross rail 78 fixed in the Y-axis direction, that is, in the horizontal direction is fixed to an upper portion of a column 77 fixed to the bed 76 and standing on the floor.
A saddle 79 is attached to the cross rail 78 so as to be movable in the Y-axis direction. The saddle 79 reciprocates by a Y-axis servo motor 80 provided on the cross rail 78.
A spindle head 81 is attached to the saddle 79 so as to be movable in the Z-axis direction. The spindle head 81 reciprocates in the Z-axis direction with respect to the saddle 79 by a Z-axis servomotor 82 provided on the saddle 79, and reciprocates in the Y-axis direction with the saddle 79 with respect to the cross rail 78.
A plurality (for example, four) of spindles 6a to 6d oriented in the Z-axis direction are provided side by side on the spindle head 81, and tools 7a to 7d are mounted on the tips of the respective spindles.
[0102]
On the bed 76, a table 83 for mounting the same number (for example, four) of the workpieces 9a to 9d as the main spindle is mounted movably in the X-axis direction. The table 83 is reciprocated by an X-axis servo motor 84 provided on the bed 76.
A cooling oil supply device 85 is provided beside the column 77. The cooling oil supply device 85 constitutes a spindle cooling device as a temperature adjustment device for adjusting the temperatures of the spindles 6a to 6d to make the thermal displacement of each spindle substantially uniform.
[0103]
FIG. 34 is an explanatory diagram of the spindle cooling device 87.
As shown, flow paths 85a to 85d for cooling the bearing by flowing cooling oil are formed near the bearings of the main shafts 6a to 6d, respectively. The cooling oil supplied from the cooling oil supply device 85 flows through the pipes 88a to 88d, respectively, and the flow rate is adjusted by flow control valves 89a to 89d that are manually or automatically operated. Thereby, the temperature of each of the spindles 6a to 6d is individually adjusted. The cooling oil flows through the flow paths 85a to 85d to cool the bearings of the main shafts 6a to 6d, and then returns to the supply device 85, where it is cooled and reused.
Instead of or in addition to the flow rate adjustment, the cooling oil temperature may be adjusted by the supply device 85 to adjust the spindle temperature for each spindle. Further, a coolant (cutting oil) or water may be used instead of the cooling oil.
[0104]
By the way, comparing the thermal displacement when the four main shafts rotate, the thermal displacement of the main shafts 6a, 6d on the left and right sides in the figure is usually smaller than that of the inner main shafts 6b, 6c. This is due to the difference in the mounting location of each spindle. That is, the heat generated in the bearings of the main shafts 6a and 6d on both sides is quickly conducted to the main shaft head 81 to suppress the temperature rise of the main shafts 6a and 6d.
Therefore, a nose temperature sensor S as a temperature detecting means is provided on the spindle head 81 so as to be located near the heat generating portion of the representative spindle 6a having the smallest thermal displacement.₁Is installed. This sensor S₁By detecting the temperature change of the fuselage 86, the thermal displacement of the representative main shaft 6a is corrected by the method using the thermal displacement correcting devices 12, 12a, and 12b in the above-described embodiments.
Then, by controlling the flow rate (and / or adjusting the temperature of the oil) by the flow control valves 89a to 89d to cool the respective spindles, the thermal displacement of the other three spindles 6b to 6d is reduced by the heat of the representative spindle 6a. The displacement is made substantially coincident with the displacement, thereby eliminating the variation in thermal displacement between the spindles.
Since the representative main shaft 6a has the smallest thermal displacement, if the main shaft 6a is not cooled or slightly cooled and the cooling oil of the other main shafts 6b to 6d is adjusted, the supply amount of the oil from the supply device 85 is reduced. It is preferable because the total amount decreases.
[0105]
FIG. 35 shows a case where a spindle heating device 90 is used as a temperature adjusting device instead of the spindle cooling device 87.
As shown in the figure, heating members such as heaters 91a to 91d are arranged near the main shafts 6a to 6d, respectively. The amount of heat generated by the heaters 91a to 91d is adjusted by individually controlling the current flowing through each of the heaters 91a to 91d by the current control device 91. The current control is preferred because it is easier and quicker to control the amount and temperature of the cooling oil, and the whole spindle heating device becomes compact.
In this case, the main shaft having the largest thermal displacement, for example, 6b is set as a representative main shaft, and a nose temperature sensor S is provided near the main shaft 6b.₁Is attached to detect a temperature change of the airframe 86. Then, the current is controlled by the current control device 91 to heat each of the main spindles, so that the thermal displacements of the other three main spindles 6a, 6c, and 6d substantially coincide with the thermal displacement of the representative main spindle 6b. Eliminates variations in thermal displacement.
It is preferable to adjust the temperature of the other main shafts without heating or slightly heating the representative main shaft 6b because the current of the entire main shaft heating device 90 is reduced.
[0106]
In this embodiment, the sensor S₁Separately, a head temperature sensor S is disposed at an arbitrary position (for example, an appropriate position of the spindle head 81) away from the spindle position and detects a temperature change of the body 86 at this position.₂Is attached as needed. Therefore, the thermal displacement can also be corrected by the above-described methods combining the mixing method alone or combining the mixing method and the dummy method.
Each sensor S₁, S₂Is input to the A / D converter 13 of the thermal displacement compensator 12, 12a, 12b, and then processed in the same manner as in each of the above embodiments.
As shown by a chain line in FIGS. 34 and 35, temperature sensors S are also provided in the vicinity of each of the spindles other than the representative spindle.₁₀It is preferable to detect the temperature of each of the spindles. Sensor S₁₀The temperature change of the temperature detected in the above is not used for the thermal displacement correction, but the thermal displacement of each spindle is estimated from the temperature change, and the variation of the thermal displacement between the spindles is calculated by the spindle cooling device 87 or the spindle heating device 90. Can be managed.
[0107]
FIG. 36 is a flowchart showing the procedure of this embodiment when the spindle cooling device 87 is used. The case in which the spindle heating device 90 is used is shown in parentheses in the following description.
First, the control valves 89a to 89d are operated to reduce the flow rate of the cooling oil (or the currents of the heaters 91a to 91d) flowing through each of the spindles 6a to 6d to the minimum necessary (step 401). Next, synchronous rotation of all the spindles is started (step 402). After a lapse of a predetermined time after operation at a constant rotation speed, the thermal displacement in the Z-axis direction at the tip of the spindle due to the extension of each spindle is actually measured (step 403).
The flow rates flowing through the flow paths 85a to 85d of the respective spindles are adjusted by the control valves 89a to 89d (or the currents of the heaters 91a to 91d are adjusted by the current control device 91) so that the thermal displacements, that is, the elongation amounts of the respective spindles become substantially the same. The adjustment is performed to set the adjustment amount (step 404).
The rotation of all the spindles is stopped, and the operation is stopped until the entire body 86 radiates heat sufficiently (step 405).
[0108]
Thereafter, the synchronous rotation of all the spindles is restarted, and the cooling oil (or current) flows through each spindle according to the adjustment amount set in Step 404 (Step 406).
Next, the thermal displacement in the Z-axis direction at the tip of the representative spindle 6a (or 6b) is actually measured as time-series data, and the temperature change of the airframe 86 is measured by the sensor S.₁, S₂(Step 407).
Using the temperature change thus detected, the nose temperature time constant τ_N, Head temperature time constant τ_H, Sample temperature time constant τ_SAnd the like, and further calculate a dummy heat capacity and an internal correction coefficient. Then, these values are set in the thermal displacement correction devices 12, 12a, 12b (step 408).
The correction is started in step 409, and the thermal displacement is corrected using the representative spindle (step 410). When the correction is to be ended in step 411, the synchronous rotation of all the spindles is stopped (step 412), and the entire procedure ends. If not, the process returns to step 410.
[0109]
As shown in FIG. 33, the thermal displacement of the representative spindle calculated and corrected by the correction devices 12, 12a, and 12b is transmitted to the numerical control device 16 via the programmable controller 15 and fed back to the Z-axis servo motor 82. You. Thus, the Z-axis servo motor 82 performs position correction by moving the spindle head 81 by a small distance in the Z-axis direction.
Since the thermal displacement between the representative spindle and the other spindles is made substantially uniform by the spindle cooling device 87 (or the spindle heating device 90), the tool 7a in which four workpieces 9a to 9d are mounted on the spindles 6a to 6d. Through 7d, simultaneous processing can be performed with high precision.
[0110]
FIG. 37 is a block diagram including a planar structure of a machine tool having a plurality of spindles, and is an application example of the fifth embodiment.
The illustrated machine tool is a multi-axis NC lathe 92, and includes two sets of a headstock 66 and a main spindle 69 having the same structure as in the fourth embodiment. Therefore, this NC lathe 92 has multiple heat sources and multiple axes.
On the bed 93, two headstocks 66 are juxtaposed, and a saddle 94 is mounted movably in the Z-axis direction. The saddle 94 is reciprocated by a Z-axis servo motor 95.
A cross slide 96 is mounted on the saddle 94 so as to be movable in the X-axis direction, and reciprocates by an X-axis servo motor 97. A plurality of blocks 99 each having a tool 98 are mounted on the cross slide 96. The workpiece is cut by the tool 98 by rotating the main shaft 69 that grips the workpiece via the chuck 65 and the like.
The bed 93 is provided with a cooling oil supply device 100 constituting a spindle cooling device having the same principle as the spindle cooling device 87 shown in FIG. This spindle cooling device is for individually adjusting the temperature of the two spindles 69 to make the thermal displacement of both spindles substantially uniform.
[0111]
In the vicinity of the bearings 71, 72 before and after the two spindles 69 and the built-in motor 70 (FIG. 30), a flow passage for flowing cooling oil to suppress the extension of the spindle is formed in the spindle head 66. The spindle cooling device has the same piping and flow rate control valve as in FIG. 34, and can adjust the flow rate to each spindle 69 individually.
One headstock 66 has a sensor s₁, S₂, S₃Are mounted in the same manner as in the fourth embodiment, so that the thermal displacement is corrected by the thermal displacement correcting device 12b in the same manner as in the fourth embodiment. Sensors₁, S₂, S₃The main shaft 69 to which the is attached is used as a representative main shaft, and the main shaft cooling device makes the thermal displacements of the representative main shaft 69 and the other main shaft 69 substantially coincide with each other, thereby eliminating variations in the thermal displacement between the main shafts.
If the thermal displacement of the representative spindle is corrected in the same manner as in the procedure shown in FIG. 36, the workpiece held by the representative spindle 69 and the other spindle 69 can be simultaneously machined with high accuracy by the tools 98. .
In the fifth embodiment, the case where the number of the representative spindles is one is shown. However, when the simultaneous machining is not performed, the thermal displacement may be corrected independently for each of the plurality of representative spindles. Further, a cooling device or a heating device using the Peltier effect may be used as the temperature control device.
[0112]
By the way, there is a conventional multi-shaft type machine tool in which a large amount of cooling oil or the like is caused to flow through a flow path near each main shaft bearing to cool the main shaft in order to reduce elongation of the main shaft due to heat generation. In this method, the thermal displacement of each spindle is physically approached to zero, thereby simultaneously absorbing the thermal displacement and variation of each spindle.
However, there is a limit to absorption of thermal displacement by this method, and it is impossible to reduce the thermal displacement to +10 [μm] or less. Further, since a large amount of cooling oil is circulated, a large-capacity cooling device is required, and a large amount of energy is wasted. In addition, there is a possibility that the bearing is distorted due to the strong cooling effect and the main shaft is seized.
On the other hand, in the fifth embodiment, the thermal displacement of the main spindle and the other main spindles is reduced so that the thermal displacement of each main spindle does not physically approach zero, but the variation of the thermal displacement between the main spindles is eliminated. And the thermal displacement of the representative spindle is corrected. Therefore, the machining error after the correction can be made close to zero, and a small-sized spindle cooling device 85 is sufficient, and the energy consumption is small. Also, since the cooling effect is weak, the bearing does not seize.
[0113]
In the first to fifth embodiments, repeated calculations are performed. Therefore, as shown in FIG. 1, FIG. 2, FIG. 18, FIG. 29, and FIG. 37, storage means for storing the previous calculation result and also for storing the time from when the power of the machine tool is turned off to when it is turned on again. 35 is preferably provided in the thermal displacement compensator 12, 12a, 12b.
The storage means 35 exchanges data with the delay temperature calculating means 34, 34a, 34b and the generating temperature calculating means 31a. In this way, even when the power is turned off, the history of the calculation of the thermal displacement correction is stored, so that the repeated calculation becomes effective.
[0114]
In addition, when using the dummy method combined with the mixing method or the dummy method combined with the linearization method, even if the temperature change of the body is detected by a temperature sensor separately provided on a column, a bed, a cross rail, etc. of the machine tool. Good.
In addition, as the temperature detecting means in the present invention, a strain gauge (Strain @ gauge) for detecting expansion and contraction of the body due to a temperature change may be used instead of the temperature sensor. That is, instead of directly detecting the temperature change of the body by the temperature sensor, a strain gauge having output characteristics similar to the temperature change is attached to the body. Then, when the output signal of this gauge is input to the A / D converter 13, it is substantially the same as detecting a temperature change, and the same operation and effect can be obtained.
Incidentally, the correlation in each embodiment only needs to have a certain correspondence, and may be a case other than the primary correlation.
[0115]
The present invention does not use the length of the fuselage components as in the prior art, so there is no restriction on the length of the fuselage structure. There is no need for actual measurement work.
Therefore, the number of rotations needs to be measured only once, and the actual measurement operation of extracting the thermal displacement characteristics using the actual machine is simplified. Further, it is not necessary to confirm the linear expansion coefficient of the fuselage constituent material.
[0116]
Further, since the temperature sensor may be mounted at an arbitrary position, the restriction on the mounting position of the temperature sensor is relaxed, and at the same time, the thermal displacement is reduced by only a small number (for example, one or two for one heat source). With a high degree of freedom that can be accurately estimated.
Further, the correction is performed based on the temperature of the airframe, and the room temperature is not directly detected. Therefore, even if the room temperature changes suddenly, for example, by opening the door of the room in winter or operating the cooler in summer, the influence of the room temperature is eliminated, and the accuracy of the correction can be maintained with high accuracy.
The thermal displacement correction method and apparatus of the present invention are applied to other types of machines in which thermal displacement adversely affects the accuracy and performance of the machine, for example, automatic control machines such as printing machines, presses, and laser processing machines. Has the same effect. This automatic control machine is controlled by an automatic control device such as an NC device.
In the drawings, the same reference numerals indicate the same or corresponding parts.
[0117]
【The invention's effect】
The present invention uses the concept of the time constant as described above to detect temperature changes of the airframe at at least two places where the time constants are different from each other under the influence of the heat source, and each of the detected temperature By combining the changes, a combined temperature change having a time constant substantially equal to the time constant of the thermal displacement of the machine tool is calculated, and the processing error is corrected based on the thermal displacement that changes in accordance with the combined temperature change. With this configuration, the combined temperature change and the thermal displacement have a linear correlation, and the thermal displacement can be easily estimated from the combined temperature change, and the thermal displacement can be corrected with high accuracy.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a first embodiment of the present invention.
FIG. 2 is a block diagram showing a second embodiment of the present invention.
FIG. 3 is a flowchart showing the operation of the first embodiment.
FIG. 4 is a flowchart showing the operation of the second embodiment.
FIG. 5 is a graph showing a temporal change of a Z-axis thermal displacement.
FIG. 6 is a graph showing a change in temperature detected at the nose position and a change in temperature detected at the head position, and a change in combined temperature.
FIG. 7 is a graph showing a relationship between a nose temperature change and a Z-axis thermal displacement.
FIG. 8 is a graph showing a relationship between a head temperature change and a Z-axis thermal displacement.
FIG. 9 is a graph showing a Z-axis thermal displacement with respect to a resultant temperature change.
FIG. 10 is a graph illustrating a method of calculating a delay temperature change.
FIG. 11 is a graph showing a method of calculating a delay response component from a temperature change.
FIG. 12 is a graph showing a relationship between a calculated Z-axis thermal displacement and an actually measured Z-axis thermal displacement.
FIG. 13 is a graph showing Z-axis thermal displacement with respect to a delay temperature change.
FIG. 14 is a graph showing measured data of Z-axis thermal displacement.
FIG. 15 is a graph showing other actually measured data of Z-axis thermal displacement.
FIG. 16 is a graph showing measured data of Y-axis thermal displacement.
FIG. 17 is a graph showing other actually measured data of Y-axis thermal displacement.
FIG. 18 is a block diagram showing a third embodiment of the present invention.
FIG. 19 is a flowchart showing the operation of the third embodiment.
FIG. 20 is a graph showing a sample temperature change and a Z-axis thermal displacement.
FIG. 21 is a graph showing a relationship between a sample temperature change and a Z-axis thermal displacement.
FIG. 22 is a graph showing a sample temperature change and a temperature change “A, B”.
FIG. 23 is a graph showing a sample temperature change, a temperature change “A, B”, and a generated temperature change.
FIG. 24 is a graph showing a relationship between a change in generating temperature and a thermal displacement in the Z axis.
FIG. 25 is a graph showing a relationship between a creation temperature change and a Z-axis thermal displacement when the Z-axis thermal displacement includes a delay response component.
FIG. 26 is a graph showing a sample temperature change, a delay temperature change, and a created delay temperature change.
FIG. 27 is a graph showing a relationship between a delay temperature change and a delay response component.
FIG. 28 is a graph showing a relationship between a thermal displacement estimated from a generated temperature change and a delayed temperature change, and a Z-axis thermal displacement.
FIG. 29 is a block diagram showing a fourth embodiment.
FIG. 30 is a sectional view of a headstock of an NC lathe.
FIG. 31 is a graph showing Z-axis thermal displacement of each heat source.
FIG. 32 is a graph showing measured data of Z-axis thermal displacement.
FIG. 33 is a block diagram showing a fifth embodiment.
FIG. 34 is an explanatory diagram of a spindle cooling device.
FIG. 35 is an explanatory diagram of a spindle heating device.
FIG. 36 is a flowchart showing the operation of the fifth embodiment.
FIG. 37 is a view showing an application example of the fifth embodiment, and is a block diagram including a planar structure of a machine tool having a plurality of spindles.
[Explanation of symbols]
1,1a, 75mm vertical machining center (machine tool)
5,5a, 81mm spindle head
6,6a to 6d, 69mm spindle
7,7a to 7d, 98mm tools
9,9a-9d, 68mm work
10,66,86 aircraft
12, 12a, 12b thermal displacement compensator
20mm main bearing
21 built-in motor
22 Upper bearing (other bearings)
31 Synthetic temperature calculating means (temperature calculating means)
31a Creation temperature calculation means (temperature calculation means)
32 ° thermal displacement calculation means
33 ° correction means
34 ° delay temperature calculating means (temperature calculating means)
34a first delay temperature calculating means (temperature calculating means)
34b second delay temperature calculating means (temperature calculating means)
64 NC lathe (machine tool)
66 headstock
70 built-in motor
71 front bearing
72 ° rear bearing
92 multi-axis NC lathe (machine tool)
O₁Center axis direction
S₁Nose temperature sensor (temperature detection means)
S₂Head temperature sensor (temperature detection means)
S₃Temperature sensor (temperature detection means)
s₁Or s₃Temperature sensor (temperature detection means)

Claims (8)

Detect temperature changes of the aircraft at at least two places where the temperature constants differ from each other under the influence of the heat source,
By combining the detected temperature changes, a combined temperature change having substantially the same time constant as the time constant of the thermal displacement of the machine tool is calculated,
A method for correcting a thermal displacement of a machine tool, wherein a processing error is corrected based on a thermal displacement that changes in response to the combined temperature change. Calculate the thermal displacement that changes in response to the combined temperature change,
Detecting a temperature change in an appropriate portion of the aircraft,
In anticipation of a delay in the detected temperature change, a delay temperature change having substantially the same aging characteristics as a delay response component in which the thermal displacement of the machine tool and the calculated thermal displacement gradually shift is calculated,
2. The machining according to claim 1, wherein a machining error is corrected based on a total value obtained by adding a delay response component that changes in response to the delay temperature change to the calculated thermal displacement. A method for compensating thermal displacement of a machine. The machine tool is
A spindle for holding either the workpiece or the tool,
A spindle head rotatably supporting the spindle via a main bearing on a machining position and another bearing opposite to the machining position that supports the spindle;
A built-in motor disposed between the two bearings and built into the spindle head to rotationally drive the spindle;
The main bearing positions the main shaft with respect to a central axis direction, and the other bearing holds the main shaft that expands and contracts due to thermal displacement so as to be slidable in the central axis direction,
3. The method according to claim 1, wherein a head temperature sensor for detecting the temperature change at a head position affected by the heat source is mounted on the spindle head. A plurality of spindles that grip one of the workpiece and the tool and rotate in synchronization with the spindle head,
The difference in heat generation characteristics due to the rotation of the plurality of spindles is controlled by controlling the amount or temperature of cooling oil flowing through a jacket provided at the nose portion of each of the spindles, or of a heater provided at the nose portion of each of the spindles. By controlling the amount of electricity, the configuration having the spindle head uniform thermal displacement,
3. The method according to claim 1, wherein at least one of the spindles is used as the heat source. Temperature detecting means for detecting a temperature change of the aircraft at at least two places that have a temperature change different from each other under the influence of the heat source,
Combining temperature changes detected by the temperature detecting means, a combined temperature calculating means for calculating a combined temperature change having substantially the same time constant as the time constant of the thermal displacement in the predetermined axial direction,
Thermal displacement calculating means for calculating a thermal displacement that changes in accordance with the synthetic temperature change calculated by the synthetic temperature calculating means;
A correction means for correcting a processing error based on the thermal displacement in the predetermined axial direction calculated by the thermal displacement calculation means. Temperature detection means for separately detecting a temperature change in an appropriate portion of the aircraft as needed,
In anticipation of a delay in the temperature change detected by any one of the temperature detecting means, the temporal characteristic is substantially the same as a delay response component in which the thermal displacement in the predetermined axial direction and the output of the thermal displacement calculating means gradually shift. A delay temperature calculating means for calculating a delay temperature change having
The thermal displacement calculation means calculates a delay response component that changes in response to the delay temperature change, and adds the delay response component to the calculated thermal displacement to calculate a total value.
6. The thermal displacement compensating device for a machine tool according to claim 5, wherein a machining error is compensated by the compensating means based on the total value. The machine tool is
A spindle for holding either the workpiece or the tool,
A spindle head rotatably supporting the spindle via a main bearing on a machining position and another bearing opposite to the machining position that supports the spindle;
A built-in motor disposed between the two bearings and built into the spindle head to rotationally drive the spindle;
The main bearing positions the main shaft with respect to a central axis direction, and the other bearing holds the main shaft that expands and contracts due to thermal displacement so as to be slidable in the central axis direction,
7. The thermal displacement compensating device for a machine tool according to claim 5, wherein a head temperature sensor for detecting the temperature change at a head position affected by the heat source is attached to the spindle head. A plurality of spindles that grip one of the workpiece and the tool and rotate in synchronization with the spindle head,
The difference in heat generation characteristics due to the rotation of the plurality of spindles is controlled by controlling the amount or temperature of cooling oil flowing through a jacket provided at the nose portion of each of the spindles, or of a heater provided at the nose portion of each of the spindles. By controlling the amount of electricity, the configuration having the spindle head uniform thermal displacement,
7. The thermal displacement compensator for a machine tool according to claim 5, wherein at least one of the spindles is used as the heat source.

Images