JP4243377B2 - Surface acoustic wave filter device and method of forming weighted electrode in the same device - Google Patents

Description

【０００７】
【数２】
Ｅ(Ω)＝Ｗ(Ω)｛Ｄ(Ω)−Ｆ(Ω)｝
ここで、上記数２中、Ｗ(Ω)は近似誤差に対する重み関数であり、Ｄ(Ω)は目的とする特性である。
【０００８】
【発明が解決しようとする課題】
しかし、ミニマックス法を用いて設計した弾性表面波フィルタ装置においては、交差部の交差長の中心が電極パターンの端部にいくに従って傾き、かつ電極パターンの端部の交差長が大きくなるために、すなわち図２３のようになるために、チップサイズが大きくならざるを得ない。
【０００９】
一方、これを回避するために、本出願人は特願平９−１９６１９９号で提案したように、ダブル電極を構成する一対の電極指の長さを異ならせるようにした（図２４参照）。これによれば、弾性表面波フィルタ装置の電極パターンは、図２５に示すように、交差長の中心が傾き、かつ端部の交差長がある程度大きくなっても、チップサイズを小さく抑えることができた。
【００１０】
また、上述した全ての弾性表面波フィルタ装置は正規型電極として電極指を偶数にした場合であり、本願発明者はこの正規型電極の電極指数を奇数にすることも考えた。これによれば、図２６のように、電極パターンの交差長の中心の傾きがなくなり、端部の交差長が大きくなっても、チップサイズを小さく抑えることができた。
【００１１】
しかしながら、上述したこれらの弾性表面波フィルタ装置にあっても、端部における交差長が大きいために、弾性表面波の回折が大きく現れ、周波数特性が悪化し好ましくない。
【００１２】
【発明の概要】
本発明は上記問題に対処するためになされもので、その目的は、上記周波数特性の変化を極力小さく抑えた上で、重み付け電極の端部における交差長を小さくして、弾性表面波の回折を減少させるようにした弾性表面波フィルタ装置を提供することにある。
【００１３】
上記目的を達成するために、本発明の特徴は、圧電基板上に弾性表面波の伝播方向に沿って平行にかつ相対向して延設された第１及び第２連結部と、これら第１及び第２連結部から直角方向にそれぞれ延設されて互いに交差する複数対の第１及び第２電極指により構成した櫛歯状の重み付け電極を備えた弾性表面波フィルタ装置において、前記重み付け電極の端部（弾性表面波の伝播方向の端部）にて隣り合う前記複数対の第１及び第２電極指の各交差部の交差長を弾性表面波の伝播方向に向けて小さくするとともに、前記隣り合う各交差部の交差長の差を一定に保つようにしたことにある。なお、前記隣り合う各交差部の交差長は、周波数特性の最適化を図る設計法（ミニマックス）法によって設定するのが望ましい。また、重み付け電極をダブル電極で構成するとよい。
【００１４】
詳しくは発明の実施の形態に記載の理論説明にて後述するが、最適化法、例えばミニマックス法で設計した隣り合う交差部の交差長の差を一定に保つように両交差部の交差長をそれぞれ変更しても、これらの交差部により得られる周波数成分は同一に保たれる。例えば、図１(Ａ)に示すように最適化法で設計した隣り合う交差部の交差長ａ_i＝０．３，ａ_i+2＝０．２を、ａ_i＝１．０，ａ_i+2＝０．０に変更しても、前記隣り合う交差部により得られる周波数成分は同一に保たれる。したがって、本願発明のように、重み付け電極の端部にて隣り合う交差部の交差長を周波数特性の最適化を図る設計法で設定した交差長よりそれぞれ小さくするとともに、隣り合う交差部の交差長の差を周波数特性の最適化を図る設計法で設定した同交差長の差と等しく保つようにしても、最適化法で設計した弾性表面波フィルタ装置と同様な周波数特性を得ることができる。その結果、本願発明によれば、周波数特性の変化を極力小さく抑えた上で、重み付け電極の端部における交差長を小さくすることができるので、弾性表面波の回折を減少させることができる。特に、前記のように重み付け電極をダブル電極で構成することにより電極指端で生じる弾性表面波の音響的反射を打ち消すことができる。
【００１５】
また、本発明の他の特徴は、
圧電基板上に弾性表面波の伝播方向に沿って平行にかつ相対向して延設された第１及び第２連結部と、これら第１及び第２連結部から直角方向にそれぞれ延設されて互いに交差する複数対の第１及び第２電極指により構成した櫛歯状の重み付け電極を備えた弾性表面波フィルタ装置において、前記重み付け電極の端部（弾性表面波の伝播方向の端部）にて隣り合う前記複数対の第１及び第２電極指の各交差部の交差方向を互いに逆にするとともに、前記各交差部の互いに異なる交差方向を正負として、前記隣り合う交差部の交差長の差を周波数特性の最適化を図る設計法で設定した同交差長の差と等しく保つようにしたことにある。好ましくは、周波数特性の最適化を図る設計法として、ミニマックス法を採用するとよい。また、重み付け電極をダブル電極で構成するとよい。
【００１６】
詳しくは発明の実施の形態に記載の理論説明にて後述するが、最適化法、例えばミニマックス法で設計した隣り合う交差部の交差長の差を一定に保つように両交差部の交差長をそれぞれ変更しても、これらの交差部により得られる周波数成分は同一に保たれる。例えば、図１(Ａ)に示すように最適化法で設計した隣り合う交差部の交差長ａ_i＝０．３，ａ_i+2＝０．２を、図１(Ｂ)に示すようにａ_i＝０．０５，ａ_i+2＝−０．０５に変更しても、前記隣り合う交差部により得られる周波数成分は同一に保たれる。したがって、本願発明のように、重み付け電極の端部にて隣り合う交差部の交差方向を互いに逆にするととともに、交差部の互いに異なる交差方向を正負として隣り合う交差部の交差長の差を最適化法で設定した同交差長の差と等しく保つようにしても、最適化法で設計した弾性表面波フィルタ装置と同様な周波数特性を得ることができる。その結果、本願発明によれば、周波数特性の変化を極力小さく抑えた上で、重み付け電極の端部における交差長を小さくすることができるので、弾性表面波の回折を減少させることができる。特に、前記のように重み付け電極をダブル電極で構成することにより電極指端で生じる弾性表面波の音響的反射を打ち消すことができる。
【００１７】
【発明の実施の形態】
［理論説明］
弾性表面波フィルタ装置においては、各櫛歯状電極をダブル電極にて設計するのが一般的なので、まず、ダブル電極型の重み付け電極の設計について考える。図１(Ａ)に示すように、隣り合う電極指が交差する各交差部の交差長を中心から外側に向かっていくにしたがってａ₀，ａ₂，ａ₄，…ａ_i，ａ_i+2，…と表すと、重み付け電極の周波数特性は下記数３のように表される。
【００１８】
【数３】

【００１９】
ただし、ｉ，ｎは偶数であり、Ωは所望の通過帯域の中心周波数ｆ₀を０．５として０〜２ｆ₀の周波数範囲を０〜１で表した場合の規格化周波数である（図２参照）。
【００２０】
ここで、本願発明者らは、ｉ番目とｉ＋２番目の交差部が周波数特性に与える影響に着目し、前記数３の右辺のうちａ_i及びａ_i+2を含む部分だけ抜き出すと、ｉ番目及びｉ＋２番目の交差部により得られる周波数成分の和ｙを下記数４のように表される。
【００２１】
【数４】
ｙ＝ａ_iｃｏｓ(ｉπΩ)＋ａ_i+2ｃｏｓ((ｉ＋２)πΩ)
この数４において通過帯域の周波数特性を近似するものとしてΩ＝０．５を採用して、前記数４に代入すると、前記数４は下記数５のように表される。
【００２２】
【数５】

【００２３】
この数５から隣り合う交差部の交差長の差を一定にすれば、ｉ番目及びｉ＋２番目の交差部の通過帯域の周波数成分をほぼ同一に維持できる、すなわちこれらの和である重み付け電極の通過帯域の周波数特性をほぼ同一に維持できることが解った。したがって、隣り合う交差部の交差長を小さくするとともに隣り合う交差部の交差長の差を一定に保ちさえすれば、一対の交差長をそれぞれ変更しても重み付け電極の周波数特性を同一に維持することができる。例えば、ミニマックス法による設計データとして、図１(Ａ)に示すようにａ_i＝０．３，ａ_i+2＝０．２を得たとしても（ミニマックス法を用いると、通常、このように同一方向に大きな交差長を有するものとなる）、ａ_i＝０．１，ａ_i+2＝０．０に変更することができ、さらに、図１(Ｂ)に示すようにａ_i＝０．０５，ａ_i+2＝−０．０５に変更することができる。なお、前記交差長の数値は最大交差長を「１．０」とした場合の各交差長の比率を表す値であり、図１(Ａ)，(Ｂ)の下部に示されている。したがって、このことを使えば、図２６のＸに示すような電極パターンの端部における交差長を短くすることができ、同端部の大きい交差長を有する交差部による回折の問題を解消することができる。
【００２４】
次に、シングル電極型の重み付け電極の設計について考える。ダブル電極型の重み付け電極において対となる電極指を１つの電極指とするとともに、シングル電極型の重み付け電極の各交差部の交差長の分布（電極パターン）をダブル電極型の重み付け電極の電極パターンと同一に設定すれば、シングル電極型の重み付け電極はダブル電極型の重み付け電極と同一の周波数特性が得られることがわかっている（図３(Ａ)(Ｂ)参照）。図３(Ｂ)に示すように、隣り合う電極指が交差する各交差部の交差長を中心から外側に向かっていくにしたがってｂ₀，ｂ₁，ｂ₂，…ｂ_I，ｂ_I+1，…と表すと、各交差長ｂ₀，ｂ₁，ｂ₂，…ｂ_I，ｂ_I+1，…は、図３(Ａ)に示す各交差長ａ₀，ａ₂，ａ₄，…ａ_i，ａ_i+2，…に相当するので、シングル電極型においても、上述したダブル電極型と同様に、隣り合う交差部の交差長の差を一定に保ちさえすれば、一対の交差長をそれぞれ変更しても中心周波数付近の周波数特性を同一に維持することができる。したがって、シングル型においても、前記一対の交差長をそれぞれ変更することにより、電極パターンの端部における交差長を短くすることができ、同端部の大きい交差長を有する交差部による回折の問題を解消することができる。
【００２５】
［具体的実施形態］
以下、本発明の一実施形態に係る弾性表面波フィルタ装置を図面を用いて説明すると、図４は同弾性表面波フィルタ装置を上面から見て概略的に示している。
【００２６】
この弾性表面波フィルタ装置は、ニオブ酸リチウム(LiNbO₃)、タンタル酸リチウム(LiTaO₃)、水晶などの圧電材料で厚さの薄い直方体状に形成した圧電基板１０を備えている。圧電基板１０上には、櫛歯状の重み付け電極で構成した入力電極２０と、櫛歯状の正規型電極で構成した出力電極３０とが弾性表面波の伝搬方向に並設されている。
【００２７】
入力電極（重み付け電極）２０は、図１(Ｂ)及び図４に示すように、導電体帯により形成された第１及び第２連結部２１，２２、並びに複数の第１及び第２電極指２３，２４を備えている。第１及び第２連結部２１，２２は、所定距離だけ隔てて平行かつ相対向して弾性表面波の伝搬方向（図示左右方向）に沿って延設されている。複数の第１及び第２電極指２３，２４は、第１及び第２連結部２１，２２から直角方向にそれぞれ延設されて互いに交差している。各第１電極指２３は、電極指端で生じる弾性表面波の音響的反射を打ち消すために、第１及び第２連結部２１，２２の延設方向に２つに分割されて同じ長さに設定された一対の分割電極指２３ａ，２３ｂによりそれぞれ構成されている。各第２電極指２４も、前記と同じ理由のために、第１及び第２連結部２１，２２の延設方向に２つに分割されて同じ長さに設定された一対の分割電極指２４ａ，２４ｂによりそれぞれ構成されている。なお、図１(Ｂ)は図４に示す重み付け電極の一部を抜き出して簡略して示しており、実際には電極指の幅及び隣り合う電極指間の距離は共に等しく設定されている（例えば、λ／８に設定されている。λは弾性表面波の波長を表す）。本願明細書においては、一対の分割電極指で構成する電極指をダブル電極といい、１つの電極指で構成するものをシングル電極という。
【００２８】
この場合、第１電極指２３と第２電極指２４との各交差部の交差長ａ_i（ｉは偶数）、すなわち第１電極指２３の分割電極指２３ａと第２電極指２４の分割電極指２４ｂとの各交差部の交差長ａ_i（ｉは偶数）、及び第１電極指２３の分割電極指２３ｂと第２電極指２４の分割電極指２４ａとの各交差部の交差長ａ_i（ｉは偶数）は、所定の周波数特性を得るためにミニマックス法により計算した値にそれぞれ設定してある。なお、交差長ａ_iは、隣り合う電極指の交差部における同電極指の延設方向に沿った長さだけではなく、同延設方向（図示上下方向）に沿った交差方向を正負で表している。すなわち、左側に位置する第１電極２３及び右側に位置する第２電極２４からなる交差部の交差長は図示上向きであり（これを正とする）、左側に位置する第２電極２４及び右側に位置する第１電極２３からなる交差部の交差長は図示下向きである（これを負とする）。
【００２９】
ミニマックス法により弾性表面波フィルタ装置を設計すると、図２６に示すように、電極パターンの端部Ｘは、同方向に比較的大きな交差長を有する隣り合う一対の交差部からなる複数の交差対Ｘ１〜Ｘ５を有することとなる。しかし、本実施形態では、上記理論説明で述べた方法を採用することにより、隣り合う交差部の交差長の差を一定に保つように両交差部の交差長をそれぞれ変更するので、図５に示すように、交差対Ｘ１を互いに逆方向かつ小さな交差長を有する交差部に変更した電極パターンに設定されている。また、３つの交差対Ｘ１〜Ｘ３を互いに逆方向かつ小さな交差長を有する交差部に変更して、図６に示すような電極パターンに設定してもよい。さらに、５つの交差対Ｘ１〜Ｘ５を互いに逆方向かつ小さな交差長を有する交差部に変更して、図７に示すような電極パターンに設定してもよい。なお、図５〜図７及び図２６は電極パターンの右半分のみを示しており、その左半分は右半分の点対称の関係にある。
【００３０】
出力電極（正規型電極）３０は、図４に示すように、重み付け電極２０と同様に導電体帯により形成された第１及び第２連結部３１，３２、並びに複数の第１及び第２電極指３３，３４を備えており、これら第１及び第２電極指３３，３４も、同じ長さに設定された一対の分割電極指３３ａ，３３ｂによりそれぞれ構成されている。このとき、第１及び第２電極指３３，３４の総数は奇数に設定されている。
【００３１】
上記のように構成した実施形態においては、重み付け電極２０の端部に位置する交差長を小さく設定したので、同交差長による弾性表面波の回折を減少させ、周波数特性の悪化を防ぐことができる。
【００３２】
このことを図８に示すような測定回路を用いて実験により確認する。この実験には、下記表１に示す仕様の弾性表面波フィルタ装置を使用する。
【００３３】
【表１】
圧電基板 タンタル酸リチウム(LiTaO₃) Ｘカット
重み付け電極本数 ８３６本
正規型電極本数 ２１本
交差長 ４０００μｍ
電極中心間距離 １４８６０μｍ
電極櫛幅（Ｗ／Ｐ） ０．４
なお、交差長は重み付け電極の最大交差長（本実施形態においては、正規型電極の交差長と等しい）を表し、電極中心間距離は重み付け電極２０の中心と正規型電極３０の中心との間の距離を表し、電極櫛幅は電極指幅（Ｗ）の隣り合う電極指の中心間距離（Ｐ）に対する比を表している。
【００３４】
弾性表面波フィルタ装置の入力端４１には、抵抗５１を介して信号発生器５２が接続されており、その出力端４２には、抵抗５３及び測定器５４が接続されている。弾性表面波フィルタ装置の他の端子４３，４４と、信号発生器５２、抵抗５３及び測定器５４はそれぞれアースされている。入力端４１は入力電極（重み付け電極）２０の第１連結部２１（又は第２連結部２２）に接続されており、出力端４２は出力電極（正規型電極）３０の第１連結部３１（又は第２連結部３２）に接続されている。他の端子４３，４４は第２連結部２２，３２（又は第１連結部２１，３１）に接続されている。
【００３５】
この測定回路により図２６、図５〜７にそれぞれ示す電極パターンを有する重み付け電極２０の周波数特性を測定すると、図９〜１２及び図１４〜１７に示す実験データをそれぞれ得ることができる。図９〜１２は、通過帯域の振幅特性を示しており、特に中心周波数ｆ₀を中心とする平坦な部分を拡大したもので振幅リップルをよく表すものをグラフ下部中央に示している。これらをまとめると、図１３に示すように、電極パターンの端部において交差長を小さく変更する交差対の数を増やすにしたがって、すなわち大きな交差長を有する交差対を減らすにしたがって振幅リップルが小さくなり振幅特性が改善されることが確認できる。さらに、図１４〜１７は、通過帯域の群遅延特性を示しており、これらをまとめると、図１８に示すように、電極パターンの端部において交差長を小さく変更する交差対の数を増やすにしたがって、すなわち大きな交差長を有する交差対を減らすにしたがって遅延リップルが小さくなり群遅延特性が改善されることが確認できる。
【００３６】
上述したように、本実施形態によれば、周波数特性の変化を極力小さく抑えた上で、重み付け電極２０の端部における交差対の交差長を小さくすることができるので、弾性表面波の回折を減少させることができる。特に、前記のように重み付け電極２０をダブル電極で構成することにより電極指端で生じる弾性表面波の音響的反射を打ち消すことができる。
【００３７】
［変形例］
次に、上記実施形態の変形例について説明する。この変形例は、上記従来技術に記載した図２３に示す電極パターンを有する重み付け電極と、偶数本の電極指を有する正規型電極とで構成した弾性表面波フィルタ装置に本発明を適用したものである。すなわち、図２３に示す電極パターンの端部において、隣り合う交差部の交差長の差を一定に保つように両交差部の交差長をそれぞれ変更するので、図１９に示すように、端部の複数の交差対を互いに逆方向かつ小さな交差長を有する交差部に変更するとともに、交差長の中心の傾きを抑えた電極パターンに設定する。これによっても、重み付け電極の端部に位置する交差長を小さく設定したので、同交差長による弾性表面波の回折を減少させ、周波数特性の悪化を防ぐことができる。
【００３８】
さらに、他の変形例について説明する。この変形例は、上記従来技術に記載した図２５に示す電極パターンを有する重み付け電極と、偶数本の電極指を有する正規型電極とで構成した弾性表面波フィルタ装置に本発明を適用したものである。すなわち、図２５に示す電極パターンの端部において、隣り合う交差部の交差長の差を一定に保つように両交差部の交差長をそれぞれ変更するので、図２０に示すように、端部の複数の交差対を互いに逆方向かつ小さな交差長を有する交差部に変更するとともに、交差長の中心の傾きを抑えた電極パターンに設定する。なお、重み付け電極を構成する電極指は互いに異なる長さの分割電極指で構成されている。これによっても、前記と同様な作用・効果を期待できる。
【００３９】
なお、上記実施形態及び変形例においては、重み付け電極及び正規型電極をダブル電極で構成するようにしたが、このダブル電極の代わりに、シングル電極で構成するようにしてもよい。
【００４０】
なお、上記実施形態及び変形例においては、最適化法の例としてミニマックス法についてのみ説明したが、同最適化法には、線形計画法、非線形計画法もあり、これら設計法にも本発明は適用できる。
【図面の簡単な説明】
【図１】 (Ａ)はミニマックス法で設計した重み付け電極を示す概略図であり、(Ｂ)は本発明の一実施形態に係る弾性表面波フィルタ装置の重み付け電極を示す概略図である。
【図２】 規格化周波数を示すグラフである。
【図３】 (Ａ)は図１に示す重み付け電極の一部分を拡大した図であり、(Ｂ)は(Ａ)に示す重み付け電極と同一の電極パターンを有し、かつシングル電極型で構成した重み付け電極の一部分を拡大した図である。
【図４】 本発明の一実施形態に係る弾性表面波フィルタ装置を上面から見た概略図である。
【図５】 図４の重み付け電極の端部にて１つの交差対を変更した電極パターンの右半分を示す図である。
【図６】 図４の重み付け電極の端部にて３つの交差対を変更した電極パターンの右半分を示す図である。
【図７】 図４の重み付け電極の端部にて５つの交差対を変更した電極パターンの右半分を示す図である。
【図８】 弾性表面波フィルタ装置を構成する重み付け電極の周波数特性を測定するための測定回路を示す図である。
【図９】 図２６に示す電極パターンを有する重み付け電極の振幅特性を示すグラフである。
【図１０】 図５に示す電極パターンを有する重み付け電極の振幅特性を示すグラフである。
【図１１】 図６に示す電極パターンを有する重み付け電極の振幅特性を示すグラフである。
【図１２】 図７に示す電極パターンを有する重み付け電極の振幅特性を示すグラフである。
【図１３】 交差対変更数と振幅リップルとの関係を示すグラフである。
【図１４】 図２６に示す電極パターンを有する重み付け電極の群遅延特性を示すグラフである。
【図１５】 図５に示す電極パターンを有する重み付け電極の群遅延特性を示すグラフである。
【図１６】 図６に示す電極パターンを有する重み付け電極の群遅延特性を示すグラフである。
【図１７】 図７に示す電極パターンを有する重み付け電極の群遅延特性を示すグラフである。
【図１８】 交差対変更数と遅延リップルとの関係を示すグラフである。
【図１９】 本発明の変形例に係る弾性表面波フィルタ装置の重み付け電極の電極パターンの右半分を示す図である。
【図２０】 本発明の他の変形例に係る弾性表面波フィルタ装置の重み付け電極の電極パターンの右半分を示す図である。
【図２１】 最小二乗法を用いて設計した弾性表面波フィルタ装置の周波数特性を示すグラフである。
【図２２】 ミニマックス法を用いて設計した弾性表面波フィルタ装置の周波数特性を示すグラフである。
【図２３】 従来の弾性表面波フィルタ装置の重み付け電極の電極パターンの右半分を示す図である。
【図２４】 従来の弾性表面波フィルタ装置の重み付け電極の一部を拡大した一例を示す図である。
【図２５】 図２４に示す重み付け電極の電極パターンの右半分を示す図である。
【図２６】 従来の弾性表面波フィルタ装置の重み付け電極の電極パターンの右半分を示す図である。
【符号の説明】
１０…圧電基板、２０…入力電極（重み付け電極）、２１，３１…第１連結部、２２，３２…第２連結部、２３，３３…第１電極指、２４，３４…第２電極指、２３ａ，２３ｂ，２４ａ，２４ｂ，３３ａ，３３ｂ，３４ａ，３４ｂ…分割電極指、３０…出力電極（正規型電極）、Ｘ…電極パターンの端部、Ｘ１〜Ｘ５…交差対、ａ₀，ａ₂，ａ₄，ａ_i，ａ_i+2，ｂ₀，ｂ₁，ｂ₂，ｂ_I，ｂ_I+1，…交差長。[0001]
BACKGROUND OF THE INVENTION
The present invention provides a design method for optimizing the frequency characteristics of a weighted electrode, for example, a surface acoustic wave designed by a minimax method, in which weighted electrodes and regular electrodes are arranged side by side on a piezoelectric substrate in the propagation direction of the surface acoustic wave. The present invention relates to a filter device.
[0002]
[Prior art]
Conventionally, in the design of this type of surface acoustic wave filter device, in order to obtain a desired frequency characteristic, an electrode pattern (distribution of intersections where adjacent electrode fingers intersect) has been designed using the method of least squares. .
[0003]
However, the frequency characteristic of the surface acoustic wave filter device designed using the least square method is as shown in FIG. 21, and the gain is attenuated with a slope on both sides of the desired passband. In this case, by increasing the gain slope on both sides of the passband, the gain on both sides of the passband can be suppressed in order to obtain a desired characteristic, but in order to increase the gain slope, the chip size Need to be larger.
[0004]
On the other hand, if a surface acoustic wave filter device is designed using a design method (hereinafter referred to as an optimization method) that optimizes frequency characteristics (amplitude characteristics and delay characteristics), the frequency characteristics of the apparatus are as shown in FIG. As a result, the gain suddenly decreases on both sides of the desired pass band, and a constant magnitude is maintained. In the surface acoustic wave filter device using the optimization method, the gains on both sides of the passband can be rapidly reduced without increasing the chip size so much. Therefore, in recent years, optimization methods have been used in the design of surface acoustic wave filter devices.
[0005]
The optimization method is a method for optimizing the frequency characteristics (amplitude characteristics and delay characteristics) of the surface acoustic wave filter device, and the minimax method, which is a representative example of the optimization method, is defined as follows. The In other words, the minimax method is also known as weighted equiripple approximation (or Chebyshev approximation), and defines the filter's amplitude characteristics F (Ω) as shown in the following equation 1, and within the specified frequency band. The coefficient a _i is determined so as to minimize the maximum value of the weighted error | E (Ω) | defined by the following equation (2).
[0006]
[Expression 1]

[0007]
[Expression 2]
E (Ω) = W (Ω) {D (Ω) −F (Ω)}
Here, in the above formula 2, W (Ω) is a weighting function for the approximation error, and D (Ω) is a target characteristic.
[0008]
[Problems to be solved by the invention]
However, in the surface acoustic wave filter device designed using the minimax method, the center of the intersection length of the intersection portion is inclined toward the end portion of the electrode pattern, and the intersection length of the end portion of the electrode pattern is increased. That is, in order to be as shown in FIG. 23, the chip size must be increased.
[0009]
On the other hand, in order to avoid this, the applicant has made the lengths of the pair of electrode fingers constituting the double electrode different as proposed in Japanese Patent Application No. 9-196199 (see FIG. 24). According to this, as shown in FIG. 25, the electrode pattern of the surface acoustic wave filter device can keep the chip size small even if the center of the crossing length is inclined and the crossing length of the end portion is increased to some extent. It was.
[0010]
Further, all the surface acoustic wave filter devices described above are cases where the electrode fingers are even numbers as the normal type electrodes, and the inventor of the present application has also considered that the electrode index of the normal type electrodes is set to an odd number. According to this, as shown in FIG. 26, the inclination of the center of the crossing length of the electrode pattern is eliminated, and the chip size can be kept small even when the crossing length of the end portion becomes large.
[0011]
However, even in these surface acoustic wave filter devices described above, since the intersection length at the end is large, diffraction of the surface acoustic wave appears greatly, and the frequency characteristics deteriorate, which is not preferable.
[0012]
Summary of the Invention
The present invention has been made in order to cope with the above problem, and its purpose is to suppress the change in the frequency characteristic as much as possible and reduce the crossing length at the end of the weighting electrode to diffract the surface acoustic wave. It is an object of the present invention to provide a surface acoustic wave filter device that can be reduced.
[0013]
In order to achieve the above object, the present invention is characterized by the first and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate . And a surface acoustic wave filter device comprising a comb-like weighted electrode formed by a plurality of pairs of first and second electrode fingers extending in a perpendicular direction from the second connecting portion and intersecting each other . The crossing length of each crossing portion of the plurality of pairs of the first and second electrode fingers adjacent to each other at the end portion ( end portion in the propagation direction of the surface acoustic wave) is reduced toward the propagation direction of the surface acoustic wave , and certain difference in cross length of each intersection adjacent to that to keep constant. In addition, it is desirable to set the intersection length of each of the adjacent intersections by a design method (minimax) method for optimizing frequency characteristics. Moreover, it is good to comprise a weighting electrode by a double electrode.
[0014]
Although details will be described later in the theoretical explanation described in the embodiment of the invention, the intersection length of both intersections is kept constant so that the difference between the intersection lengths of adjacent intersections designed by an optimization method, for example, the minimax method is kept constant. Even if each is changed, the frequency components obtained by these intersections are kept the same. For example, as shown in FIG. 1A, the intersection lengths a _i = 0.3 and a _{i + 2} = 0.2 of adjacent intersections designed by the optimization method are set to a _i = 1.0 and a _{i. Even if + 2} = 0.0, the frequency components obtained by the adjacent intersections are kept the same. Therefore, as in the present invention, the crossing length of the adjacent crossing portions at the end of the weighted electrode is made smaller than the crossing length set by the design method for optimizing the frequency characteristics, and the crossing length of the adjacent crossing portions Even if the difference is kept equal to the difference between the crossing lengths set by the design method for optimizing the frequency characteristics, the same frequency characteristics as those of the surface acoustic wave filter device designed by the optimization method can be obtained. As a result, according to the present invention, it is possible to reduce the crossing length at the end of the weighting electrode while minimizing the change in the frequency characteristics, and to reduce the diffraction of the surface acoustic wave. In particular, the acoustic reflection of the surface acoustic wave generated at the electrode finger end can be canceled by configuring the weighting electrode with a double electrode as described above.
[0015]
Another feature of the present invention is that
First and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate , and extending from the first and second connecting portions in a direction perpendicular to each other. in the surface acoustic wave filter device having a comb-shaped weighting electrodes composed of a plurality pair of first and second electrode fingers of which cross each other, the end portion of the weighted electrode (an end of the propagation direction of a surface acoustic wave) while opposite to each other the cross direction of each intersection of the plurality of pairs first and second electrode fingers of the adjacent Te, the different cross direction of the each intersection as a positive or negative, crosses the front Symbol intersection adjacent This is because the difference in length is kept equal to the difference in crossing length set by the design method for optimizing the frequency characteristics. Preferably, the minimax method may be employed as a design method for optimizing the frequency characteristics. Moreover, it is good to comprise a weighting electrode by a double electrode.
[0016]
Although details will be described later in the theoretical explanation described in the embodiment of the invention, the intersection length of both intersections is kept constant so that the difference between the intersection lengths of adjacent intersections designed by an optimization method, for example, the minimax method is kept constant. Even if each is changed, the frequency components obtained by these intersections are kept the same. For example, as shown in FIG. 1B, the intersection lengths a _i = 0.3 and a _{i + 2} = 0.2 of adjacent intersections designed by the optimization method as shown in FIG. Even when a _i = 0.05 and a _{i + 2} = −0.05, the frequency components obtained by the adjacent intersections are kept the same. Therefore, as in the present invention, the crossing directions of the adjacent crossing portions at the end portions of the weighted electrodes are reversed to each other, and the crossing length difference between the adjacent crossing portions is optimal with the crossing directions different from each other as positive and negative. Even if the difference between the crossing lengths set by the optimization method is kept equal, the same frequency characteristic as that of the surface acoustic wave filter device designed by the optimization method can be obtained. As a result, according to the present invention, it is possible to reduce the crossing length at the end of the weighting electrode while minimizing the change in the frequency characteristics, and to reduce the diffraction of the surface acoustic wave. In particular, the acoustic reflection of the surface acoustic wave generated at the electrode finger end can be canceled by configuring the weighting electrode with a double electrode as described above.
[0017]
DETAILED DESCRIPTION OF THE INVENTION
[Theory explanation]
In a surface acoustic wave filter device, since each comb-like electrode is generally designed with a double electrode, first, the design of a double electrode type weighted electrode will be considered. As shown in FIG. 1A, a ₀ , a ₂ , a ₄ ,..., A _i , a _{i + 2 as} the crossing length of each crossing portion where adjacent electrode fingers cross from the center toward the outside. ,..., The frequency characteristic of the weighting electrode is expressed as in the following formula 3.
[0018]
[Equation 3]

[0019]
However, i, n is an even number, Omega is normalized frequency when expressed in 0-1 frequency range 0～2F ₀ the center frequency f ₀ of the desired pass band of 0.5 (Fig. 2 reference).
[0020]
Here, the inventors of the present application pay attention to the influence of the i-th and i + 2th intersections on the frequency characteristics, and extract only the part including a _i and a _{i + 2} from the right side of the equation (3). And the sum y of the frequency components obtained by the (i + 2) th intersection is expressed by the following equation (4).
[0021]
[Expression 4]
y = a _i cos (iπΩ) + a _{i + 2} cos ((i + 2) πΩ)
In Equation 4, when Ω = 0.5 is adopted as an approximation of the frequency characteristic of the pass band and is substituted into Equation 4, Equation 4 is expressed as Equation 5 below.
[0022]
[Equation 5]

[0023]
If the difference between the crossing lengths of the adjacent crossing parts is made constant from Equation 5, the frequency components of the passbands of the i-th and i + 2th crossing parts can be maintained substantially the same, that is, the sum of these passes through the weighting electrode. It was found that the frequency characteristics of the bands can be maintained almost the same. Therefore, the frequency characteristics of the weighting electrodes can be kept the same even if the pair of crossing lengths are changed as long as the crossing lengths of the adjacent crossing portions are reduced and the difference between the crossing lengths of the adjacent crossing portions is kept constant. be able to. For example, even if a _i = 0.3 and a _{i + 2} = 0.2 are obtained as design data by the minimax method as shown in FIG. A _i = 0.1, a _{i + 2} = 0.0, and a _i as shown in FIG. 1 (B). = 0.05, a _{i + 2} = −0.05. In addition, the numerical value of the said intersection length is a value showing the ratio of each intersection length when the maximum intersection length is set to "1.0", and is shown in the lower part of FIG. 1 (A), (B). Therefore, if this is used, the crossing length at the end of the electrode pattern as shown by X in FIG. 26 can be shortened, and the problem of diffraction due to the crossing having a large crossing length at the same end can be solved. Can do.
[0024]
Next, consider the design of a single electrode type weighted electrode. A pair of electrode fingers in a double electrode type weighting electrode is used as one electrode finger, and the distribution of the crossing length (electrode pattern) of each crossing part of the single electrode type weighting electrode is represented by the electrode pattern of the double electrode type weighting electrode. It is known that the single electrode type weighting electrode can obtain the same frequency characteristics as the double electrode type weighting electrode (see FIGS. 3A and 3B). As shown in FIG. 3B, b ₀ , b ₁ , b ₂ ,..., B _I , b _{I + 1 as} the crossing length of each crossing portion where adjacent electrode fingers cross each other goes from the center to the outside. ,..., The crossing lengths b ₀ , b ₁ , b ₂ ,... B _I , b _{I + 1} ,... Are respectively represented by the crossing lengths a ₀ , a ₂ , a ₄ ,. Since it corresponds to a _i , a _{i + 2} ,..., even in the single electrode type, as long as the difference between the cross lengths of adjacent crossing portions is kept constant as in the double electrode type described above, a pair of cross lengths Even if each is changed, the frequency characteristics near the center frequency can be kept the same. Therefore, even in the single type, by changing each of the pair of crossing lengths, the crossing length at the end of the electrode pattern can be shortened. Can be resolved.
[0025]
[Specific Embodiment]
Hereinafter, a surface acoustic wave filter device according to an embodiment of the present invention will be described with reference to the drawings. FIG. 4 schematically shows the surface acoustic wave filter device as viewed from above.
[0026]
The surface acoustic wave filter device includes a piezoelectric substrate 10 formed in a thin rectangular parallelepiped shape using a piezoelectric material such as lithium niobate (LiNbO ₃ ), lithium tantalate (LiTaO ₃ ), or quartz. On the piezoelectric substrate 10, an input electrode 20 composed of comb-shaped weighted electrodes and an output electrode 30 composed of comb-shaped regular electrodes are juxtaposed in the propagation direction of the surface acoustic wave.
[0027]
As shown in FIGS. 1B and 4, the input electrode (weighting electrode) 20 includes first and second connecting portions 21 and 22 formed of a conductor band, and a plurality of first and second electrode fingers. 23, 24. The first and second connecting portions 21 and 22 are extended along the propagation direction (the left-right direction in the drawing) of the surface acoustic wave in parallel and opposite to each other with a predetermined distance therebetween. The plurality of first and second electrode fingers 23 and 24 extend from the first and second connecting portions 21 and 22 in a perpendicular direction and intersect each other. Each first electrode finger 23 is divided into two pieces in the extending direction of the first and second connecting portions 21 and 22 to have the same length in order to cancel the acoustic reflection of the surface acoustic wave generated at the electrode finger ends. The paired divided electrode fingers 23a and 23b are respectively configured. Each second electrode finger 24 is also divided into two in the extending direction of the first and second connecting portions 21 and 22 and set to the same length for the same reason as described above. , 24b, respectively. FIG. 1B is a simplified illustration of a portion of the weighted electrode shown in FIG. 4, and in practice, the width of the electrode fingers and the distance between adjacent electrode fingers are both set equal ( For example, it is set to λ / 8, where λ represents the wavelength of the surface acoustic wave). In the present specification, an electrode finger constituted by a pair of divided electrode fingers is called a double electrode, and an electrode finger constituted by one electrode finger is called a single electrode.
[0028]
In this case, the intersection length a _i (i is an even number) of each intersection of the first electrode finger 23 and the second electrode finger 24, that is, the divided electrode finger 23 a of the first electrode finger 23 and the divided electrode of the second electrode finger 24. cross length a _i of each intersection of the finger 24b (i is an even number), and cross length of each intersection of the split electrode fingers 23b and the divided electrode fingers 24a of the second electrode fingers 24 of the first electrode finger 23 a _i (I is an even number) is set to a value calculated by the minimax method in order to obtain a predetermined frequency characteristic. Note that the intersection length a _i represents not only the length along the extending direction of the electrode fingers at the intersection of adjacent electrode fingers, but also the crossing direction along the extending direction (vertical direction in the figure) in positive and negative directions. ing. That is, the crossing length of the crossing portion composed of the first electrode 23 located on the left side and the second electrode 24 located on the right side is upward in the figure (this is positive), and the second electrode 24 located on the left side and the right side The intersection length of the intersection formed by the first electrodes 23 positioned is downward in the figure (this is negative).
[0029]
When the surface acoustic wave filter device is designed by the minimax method, as shown in FIG. 26, the end portion X of the electrode pattern has a plurality of intersecting pairs composed of a pair of adjacent intersecting portions having a relatively large intersecting length in the same direction. It will have X1-X5. However, in this embodiment, by adopting the method described in the above theoretical explanation, the crossing lengths of both crossing parts are changed so as to keep the difference between the crossing lengths of adjacent crossing parts constant. As shown, the crossed pair X1 is set to an electrode pattern that is changed to a crossing portion having a small crossing length in the opposite direction. Alternatively, the three crossing pairs X1 to X3 may be changed to crossing portions having opposite crossing directions and small crossing lengths to set the electrode pattern as shown in FIG. Further, the five crossing pairs X1 to X5 may be changed to crossing portions having mutually opposite directions and small crossing lengths, and set to electrode patterns as shown in FIG. 5 to 7 and FIG. 26 show only the right half of the electrode pattern, and the left half has a point-symmetric relationship with the right half.
[0030]
As shown in FIG. 4, the output electrode (regular type electrode) 30 includes first and second connecting portions 31 and 32 formed of a conductor band and a plurality of first and second electrodes, like the weighting electrode 20. Fingers 33 and 34 are provided, and the first and second electrode fingers 33 and 34 are also constituted by a pair of divided electrode fingers 33a and 33b set to the same length. At this time, the total number of the first and second electrode fingers 33 and 34 is set to an odd number.
[0031]
In the embodiment configured as described above, since the intersection length located at the end of the weighting electrode 20 is set to be small, the surface acoustic wave diffraction due to the intersection length can be reduced and the deterioration of the frequency characteristics can be prevented. .
[0032]
This is confirmed by experiments using a measurement circuit as shown in FIG. In this experiment, a surface acoustic wave filter device having the specifications shown in Table 1 below is used.
[0033]
[Table 1]
Piezoelectric substrate Lithium tantalate (LiTaO ₃ ) Number of X-cut weighting electrodes 836 Number of regular type electrodes 21 Number of crossing lengths 4000 μm
Distance between electrode centers 14860 μm
Electrode comb width (W / P) 0.4
The crossing length represents the maximum crossing length of the weighted electrodes (in this embodiment, equal to the crossing length of the normal type electrode), and the distance between the electrode centers is between the center of the weighting electrode 20 and the center of the normal type electrode 30. The electrode comb width represents the ratio of the electrode finger width (W) to the distance (P) between the centers of adjacent electrode fingers.
[0034]
A signal generator 52 is connected to the input end 41 of the surface acoustic wave filter device via a resistor 51, and a resistor 53 and a measuring instrument 54 are connected to the output end 42. The other terminals 43 and 44 of the surface acoustic wave filter device, the signal generator 52, the resistor 53, and the measuring instrument 54 are grounded. The input end 41 is connected to the first connecting portion 21 (or the second connecting portion 22) of the input electrode (weighted electrode) 20, and the output end 42 is connected to the first connecting portion 31 ( Or it is connected to the 2nd connection part 32). The other terminals 43 and 44 are connected to the second connecting portions 22 and 32 (or the first connecting portions 21 and 31).
[0035]
When the frequency characteristics of the weighted electrodes 20 having the electrode patterns shown in FIGS. 26 and 5 to 7 are measured by this measuring circuit, the experimental data shown in FIGS. 9 to 12 and FIGS. 14 to 17 can be obtained. FIGS. 9 to 12 show the amplitude characteristics of the pass band. In particular, an enlarged portion of a flat portion centered on the center frequency f ₀ and well representing the amplitude ripple is shown in the lower center of the graph. In summary, as shown in FIG. 13, the amplitude ripple decreases as the number of crossing pairs whose crossing length is changed to be small at the end of the electrode pattern is increased, that is, as the number of crossing pairs having a large crossing length is reduced. It can be confirmed that the amplitude characteristic is improved. Further, FIGS. 14 to 17 show the group delay characteristics of the passband, and when these are summarized, as shown in FIG. 18, the number of crossing pairs that change the crossing length at the end of the electrode pattern is increased. Therefore, it can be confirmed that the delay ripple is reduced and the group delay characteristic is improved as the number of crossing pairs having a large crossing length is reduced.
[0036]
As described above, according to the present embodiment, it is possible to reduce the cross length of the crossed pair at the end of the weighting electrode 20 while suppressing the change in frequency characteristics as much as possible. Can be reduced. In particular, the acoustic reflection of the surface acoustic wave generated at the electrode finger end can be canceled by configuring the weighting electrode 20 with a double electrode as described above.
[0037]
[Modification]
Next, a modification of the above embodiment will be described. In this modification, the present invention is applied to a surface acoustic wave filter device composed of a weighted electrode having the electrode pattern shown in FIG. 23 described in the prior art and a regular electrode having an even number of electrode fingers. is there. That is, at the end of the electrode pattern shown in FIG. 23, the crossing lengths of both crossing parts are changed so as to keep the difference between the crossing lengths of the adjacent crossing parts constant, as shown in FIG. A plurality of cross pairs are changed to cross portions having opposite cross directions and small cross lengths, and are set to electrode patterns in which the inclination of the center of the cross length is suppressed. Also by this, since the intersection length located at the end of the weighting electrode is set to be small, it is possible to reduce the diffraction of the surface acoustic wave due to the intersection length and to prevent the frequency characteristics from deteriorating.
[0038]
Furthermore, another modified example will be described. In this modification, the present invention is applied to a surface acoustic wave filter device composed of a weighted electrode having the electrode pattern shown in FIG. 25 described in the prior art and a regular electrode having an even number of electrode fingers. is there. That is, at the end portion of the electrode pattern shown in FIG. 25, the crossing lengths of both crossing portions are changed so as to keep the difference in crossing length between adjacent crossing portions constant, so that as shown in FIG. A plurality of cross pairs are changed to cross portions having opposite cross directions and small cross lengths, and are set to electrode patterns in which the inclination of the center of the cross length is suppressed. In addition, the electrode finger which comprises a weighting electrode is comprised by the divided electrode finger of a mutually different length. Also by this, the same operation and effect as described above can be expected.
[0039]
In the above-described embodiment and modification, the weighting electrode and the regular electrode are configured by a double electrode, but may be configured by a single electrode instead of the double electrode.
[0040]
In the embodiment and the modification, only the minimax method has been described as an example of the optimization method. However, the optimization method includes a linear programming method and a non-linear programming method. Is applicable.
[Brief description of the drawings]
FIG. 1A is a schematic diagram showing weighted electrodes designed by the minimax method, and FIG. 1B is a schematic diagram showing weighted electrodes of a surface acoustic wave filter device according to an embodiment of the present invention.
FIG. 2 is a graph showing a normalized frequency.
3A is an enlarged view of a part of the weighting electrode shown in FIG. 1, and FIG. 3B is a single electrode type having the same electrode pattern as the weighting electrode shown in FIG. It is the figure which expanded a part of weighting electrode.
FIG. 4 is a schematic view of a surface acoustic wave filter device according to an embodiment of the present invention as viewed from above.
5 is a diagram showing a right half of an electrode pattern in which one intersection pair is changed at the end of the weighted electrode in FIG. 4;
6 is a diagram showing a right half of an electrode pattern in which three intersecting pairs are changed at the end of the weighted electrode in FIG. 4;
7 is a diagram showing a right half of an electrode pattern in which five crossing pairs are changed at the end of the weighted electrode in FIG. 4;
FIG. 8 is a diagram showing a measurement circuit for measuring frequency characteristics of weighting electrodes constituting the surface acoustic wave filter device.
9 is a graph showing amplitude characteristics of a weighted electrode having the electrode pattern shown in FIG.
10 is a graph showing amplitude characteristics of a weighted electrode having the electrode pattern shown in FIG.
11 is a graph showing amplitude characteristics of a weighted electrode having the electrode pattern shown in FIG.
12 is a graph showing amplitude characteristics of a weighted electrode having the electrode pattern shown in FIG.
FIG. 13 is a graph showing the relationship between the number of crossed pairs and amplitude ripple.
14 is a graph showing group delay characteristics of weighted electrodes having the electrode pattern shown in FIG. 26. FIG.
15 is a graph showing group delay characteristics of weighted electrodes having the electrode pattern shown in FIG.
16 is a graph showing group delay characteristics of weighted electrodes having the electrode pattern shown in FIG.
17 is a graph showing group delay characteristics of weighted electrodes having the electrode pattern shown in FIG.
FIG. 18 is a graph showing the relationship between the number of crossed pairs and delay ripple.
FIG. 19 is a diagram showing a right half of an electrode pattern of weighted electrodes of a surface acoustic wave filter device according to a modification of the present invention.
FIG. 20 is a diagram showing a right half of an electrode pattern of weighted electrodes of a surface acoustic wave filter device according to another modification of the present invention.
FIG. 21 is a graph showing frequency characteristics of a surface acoustic wave filter device designed by using the least square method.
FIG. 22 is a graph showing frequency characteristics of a surface acoustic wave filter device designed by using the minimax method.
FIG. 23 is a diagram showing a right half of an electrode pattern of weighted electrodes of a conventional surface acoustic wave filter device.
FIG. 24 is a diagram showing an example in which a part of a weighting electrode of a conventional surface acoustic wave filter device is enlarged.
25 is a diagram showing the right half of the electrode pattern of the weighting electrode shown in FIG. 24. FIG.
FIG. 26 is a diagram showing a right half of an electrode pattern of weighted electrodes of a conventional surface acoustic wave filter device.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Piezoelectric substrate, 20 ... Input electrode (weighting electrode), 21, 31 ... 1st connection part, 22, 32 ... 2nd connection part, 23, 33 ... 1st electrode finger, 24, 34 ... 2nd electrode finger, 23a, 23b, 24a, 24b, 33a, 33b, 34a, 34b ... split electrode fingers, 30 ... output electrode (normal type electrode), an end portion of the X ... electrode pattern, X1 to X5 ... cross pairs, a _0, a ₂ , A ₄ , a _i , a _{i + 2} , b ₀ , b ₁ , b ₂ , b _I , b _{I + 1} ,.

Claims (5)

First and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate , and extending from the first and second connecting portions in a direction perpendicular to each other. In the surface acoustic wave filter device including a comb-like weighted electrode configured by a plurality of pairs of first and second electrode fingers intersecting each other ,
The crossing length of each crossing portion of the plurality of pairs of the first and second electrode fingers adjacent to each other at the end portion of the weighting electrode ( end portion in the propagation direction of the surface acoustic wave) is reduced toward the propagation direction of the surface acoustic wave. In addition, the surface acoustic wave filter device is characterized in that the difference between the crossing lengths of the adjacent crossing portions is kept constant . First and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate , and extending from the first and second connecting portions in a direction perpendicular to each other. In the surface acoustic wave filter device including a comb-like weighted electrode configured by a plurality of pairs of first and second electrode fingers intersecting each other ,
The crossing directions of the crossing portions of the plurality of pairs of first and second electrode fingers adjacent to each other at the end portions of the weighting electrodes ( end portions in the propagation direction of the surface acoustic waves) are reversed, and the crossing portions different cross direction as a positive or negative, before SL that the difference in cross length of the cross section adjacent to keep constant the surface acoustic wave filter device comprising a. The surface acoustic wave filter device according to claim 1, wherein the weighting electrode is a double electrode. First and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate, and extending from the first and second connecting portions in a direction perpendicular to each other. In the surface acoustic wave filter device including a comb-like weighted electrode configured by a plurality of pairs of first and second electrode fingers intersecting each other,
  Minimax method for optimizing the frequency characteristics of the crossing lengths of the crossing portions of the plurality of pairs of first and second electrode fingers adjacent at the end portions of the weighting electrodes (end portions in the propagation direction of the surface acoustic wave) The weighting electrode forming method is characterized in that the difference between the intersection lengths of the plurality of pairs of intersections is kept constant. First and second connecting portions extending in parallel and opposite to each other along the propagation direction of the surface acoustic wave on the piezoelectric substrate, and extending from the first and second connecting portions in a direction perpendicular to each other. In the surface acoustic wave filter device including a comb-like weighted electrode configured by a plurality of pairs of first and second electrode fingers intersecting each other,
  The crossing directions of the crossing portions of the plurality of pairs of first and second electrode fingers adjacent to each other at the end portions of the weighting electrodes (end portions in the propagation direction of the surface acoustic waves) are reversed, and the crossing portions The crossing directions of the plurality of pairs of electrode fingers are set to be small by the minimax method for optimizing the frequency characteristics, and the crossing directions of the plurality of pairs of electrode fingers are set to be small. A method of forming a weighted electrode, characterized in that a difference in crossing length is kept constant.

Images