200839242 九、發明說明 【發明所屬之技術領域】 本發明是關於使用靜電容量來檢測加速度的靜電容量 型加速度感測器。 【先前技術】 作爲檢測加速度的感測器，例如有靜電容量型加速度 感測器。該靜電容量型加速度感測器是由固定電極，及藉 由施加有加速度（G)進行搖動的可動電極（砝碼）所構成， 藉由檢測固定電極與可動電極之間的靜電容量的變化，可 求出加速度。 作爲此種靜電容量型加速度感測器，有加工矽基板形 成可動電極的複數砝碼部，演算對應於此些砝碼部所受到 的加速度的物理量變化，而分離成沿著3個互相地正交的 座標軸的方向的加速度的構造的3軸用靜電容量型加速度 感測器（專利文獻1)。 專利文獻1 :日本特表200 5 -5 3 48 97號公報 【發明內容】 然而，在上述3軸用靜電容量型加速度感測器中，因 在各軸方向的檢測値中包含其他軸方向的成分，因此算出 互相地正交的座標軸方向的加速度的方法變成複雜，使得 感測器的訊號處理電路會變大。一方面，也可考量排設在 各座標軸方向具有感度的3個1軸用的靜電容量型加速度 感測器，惟有感測器成大型的，感測器構造變成複雜的問 -5- 200839242 題。 本發明是鑑於此些事項而創作者，其目的是在於提供 感測器構造簡單而可實現小型化，而且訊號處理電路的負 載小的3軸用的靜電容量型加速度感測器。 本發明的靜電容量型加速度感測器，屬於分別獨立X 軸方向、Y軸方向及Z軸方向的加速度，而從功能作爲砝 碼的可動電極與固定電極之間的靜電容量的變化所檢測的 靜電容量型加速度感測器，其特徵爲：具備··具有3個可 動電極的第1基板；及對於各該可動電極，至少其一方隔 著所定間隔相對，將上述靜電容量檢測作爲差動容量所用 的各該一對檢測用電極對與具有作爲上述固定電極的上述 第1基板的一方主面接合的第2基板；及與上述第1基板 的另一方主面接合的第3基板，各該可動電極是具有在上 述第1基板的厚度方向相對的一對面，而且被密閉於藉由 上述第1基板，第2基板及第3基板所形成的各該空腔內 ，X軸方向用及Y軸方向用的可動電極爲對於上述第1基 板藉由扭轉梁被支撐成可搖動，而Z軸方向用的可動電極 爲對於上述第1基板藉由彎曲梁被支撐成可昇降，上述扭 轉梁及上述彎曲梁是沿著各該可動電極的一方的面所形成 依照該構成，可獨立X軸方向，Y軸方向及Z軸方 向的加速度而進行檢查。亦即，他軸感度低之故，因而在 沒有他軸方向的加速度成分的干擾之狀態下可檢測各軸方 向的加速度。藉由此，加速度檢測所用的演算較簡單，而 -6 - 200839242 可減小訊號電路的負載。又，各軸方向的加速度，是以差 動方式進行算出，因而可精度優異地求出。 在本發明的靜電容量型加速度感測器中，各該空腔間 的內壓爲被設定成大約相同，而可動電極與檢測用電極之 間的靜電容量的變化量爲在各軸成爲大約相同的範圍內的 方式，設定有Z軸方向用的檢測用電極的面積及可動電極 的變位量較佳。藉由此，施加加速度時，可得到如所期望 的相同阻尼特性，而在各軸方向以大約相同感度可檢測加 速度。 在本發明靜電容量型加速度感測器中，上述X軸方 向用及 Y軸方向用的可動電極的最大變位量爲上述所定 間隔的±20%以內，上述Z軸方向用的可動電極的最大變 位量爲上述所定間隔的± 1 0%以內較佳。依照該構成，在 線性優異的範圍內，可檢測加速度。 在本發明的靜電容量型加速度感測器中，上述檢測用 電極都形成在上述第2基板的同一面上較佳。依照該構成 ，以1個工程就可形成差動容量檢測用電極，而可簡化製 程。 在本發明的靜電容量型加速度感測器中，各軸用的可 動電極是被排設在上述第1基板，而都具有大約同一形狀 較佳。依照該構成，可簡化製程。 在本發明的靜電容量型加速度感測器中，z軸方向用 的檢測用電極的一方，是被密閉於與X軸、γ軸、z軸方 向的各該可動電極的空腔不相同的獨立的空腔內，合計4 200839242 個空腔爲在上述第2基板的平面視’排列成大約正方形較 佳。依照該構成，在第2基板形成有電極的領域成爲均等 ，而對於熱性應力的影響成爲相等。 依照本發明的靜電容量型加速度感測器’屬於分別獨 立X軸方向、Y軸方向及Z軸方向的加速度’而從功能 作爲砝碼的可動電極與固定電極之間的靜電容量的變化所 檢測的靜電容量型加速度感測器，其特徵爲：具備：具有 3個可動電極的第1基板；及對於各該可動電極，至少其 一方隔著所定間隔相對，將上述靜電容量檢測作爲差動容 量所用的各該一對檢測用電極對與具有作爲上述固定電極 的上述第1基板的一方主面接合的第2基板；及與上述第 1基板的另一方主面接合的第3基板，各該可動電極是具 有在上述第1基板的厚度方向相對的一對面，而且被密閉 於藉由上述第1基板，第2基板及第3基板所形成的各該 空腔內，X軸方向用及Y軸方向用的可動電極爲對於上述 第1基板藉由扭轉梁被支撐成可搖動，而Z軸方向用的可 動電極爲對於上述第1基板藉由彎曲粱被支撐成可昇降， 上述扭轉梁及上述彎曲梁是沿著各該可動電極的一方的面 所形成之故。因而可實現感測器構造簡單而可實現小型化 ’而且訊號處理電路的負載小的3軸用的靜電容量型加速 度感測器。 【實施方式】 以下，參照所附圖式詳述本發明的實施形態。 -8 - 200839242 第1圖是表示本發明的實施形態的靜電容量型加速度 感測器的俯視圖。又，第2(a)圖是表示沿著第1圖的IIA-IIA線的斷面圖，對2(b)圖是表示沿著第1圖的ΠΒ-ΙΙΒ 線的斷面圖。 表示於第1圖的靜電容量型加速度感測器，是接合著 具有分別獨立X軸方向，Y軸方向及Z軸方向的加速度 而用以檢測的3個可動電極的砝碼部12a，12b，12c的第 1基板的矽製基板1 1的一方的主面，及對於各該砝碼部 12a，12b，12c隔著所定間隔而具有將靜電容量的變化檢 測作爲差動容量的各該一對檢測用電極對14a，14b，14c ，14d，14e，14f的第2基板的玻璃基板13所構，成在矽 製基板1 1的另一方的主面，接合有第3基板的玻璃基板 15，成爲構成將砝碼部12a，12b予以搖動，並將砝碼部 12c予以昇降的領域（空腔）16。 在表示於第1圖的靜電容量型加速度感測器中，在Y 軸方向的加速度具有感度的可動電極爲砝碼部12a，而在 X軸方向的加速度具有感度的可動電極爲砝碼部1 2b，又 在Z軸方向的加速度具有感度的可動電極爲砝碼部1 2c。 對於Y軸方向用砝碼部1 2a的一對檢測用電極對是固定電 極14a，14b，而對於X軸方向用砝碼部12b的一對檢測 用電極對是固定電極14c，14d，又對於Z軸方向用砝碼 部12c的一對檢測用電極對是固定電極14e，14f。 Y軸方向用砝碼部1 2 a，是在平面視具有大約矩形狀 在相對的邊，藉由扭轉粱1 1 a對於矽製基板1 1被支撐成 -9- 200839242 可搖動，而X軸方向用砝碼部1 2b，是在平面視具有大約 矩形狀，在相對的邊，藉由扭轉梁1 1 b對於矽製基板11 被支撐成可搖動。各該扭轉梁11a，lib，是在平面視， 設於砝碼部12a，12b的各該相對的邊的中央附近。一方 面，Z軸方向用砝碼部1 2c，是在平面視具有大約矩形狀 ，其周圍爲藉由彎曲梁1 1 c對於矽製基板1 1被支撐成可 昇降。 在玻璃基板13上，形成有3對檢測用電極對。對於 Y軸方向砝碼部12a的固定電極14a，14b，是具有大約相 同面積，由第1圖可知，在平面視，位於Y軸方向用砝 碼部1 2a的下方，而以通過扭轉梁1 1 a的中央部分作爲境 界分割所形成（在第1圖中爲上下分割）。合倂兩個固定電 極14a，14b的面積與Y軸方向砝碼部12a的面積成爲大 約相等。 對於X軸方向砝碼部12b的固定電極14c，14d，是 具有大約相同面積，由第1圖可知，在平面視，位於X 軸方向用砝碼部1 2b的下方，而以通過扭轉梁1 1 b的中央 部分作爲境界分割所形成（在第1圖中爲左右分割）。合倂 兩個固定電極14c，14d的面積與X軸方向砝碼部12b的 面積成爲大約相等。 對於z軸方向砝碼部1 2 c的固定電極1 4 e，1 4 f，是 分別具有與Z軸方向用砝碼部1 2c大約相同面積，一方的 固定電極14e是被形成在Z軸方向用砝碼部12c的下方， 而另一方的固定電極14f是被形成在其他領域。 -10- 200839242 如此地，藉由採用在玻璃基板1 3的同一面上 有3對檢測用電極對的構成，以一個工程就可都形 用電極（固定電極），可簡化製程之故，因而較佳。 第1圖所示地，藉由將各軸方向用的砝碼部12a， 1 2c排設於矽製基板1 1，而都形成具有大約同一形 簡化製程之故，因而較佳。 又，X軸方向用的固定電極對14c，14d，及’ 向用的固定電極對14a，14b，及Z軸方向方向用 差動容量檢測電極14e，14f的4個電極，在玻璃3 的平面視，排列成大約正方形。如此地，藉由配置 電極，在玻璃基板1 3中形成有電極的領域成爲均 於熱性應力的影響成爲相等之故，因而較佳。 第2(a)圖是表示沿著第1圖的IIA-IIA線的斷 表示有關於Y軸方向砝碼部12a與X軸方向用砝碼 的構成，第2(b)圖是表示沿著第1圖的IIB-IIB線 圖，表示有關於Z軸方向用砝碼部12c的構成。在 圖中，凹部形成於玻璃基板13的一方的主面，而 極14a，14b，14c，14d形成在其凹部的底面。又 2 (a)圖中，表示著對於X軸方向用砝碼部12b的固 14c，14d，但未表示對於Y軸方向用砝碼部12a的 極 14a，14 b 〇 在玻璃基板13，設有在兩主面貫通成露出的 件16a，16b，而此些的一方露出面爲與可動電極 14d分別電性地被連接。又，導電構件16a，16b 都形成 成檢測 又，如 12b， 狀，可 γ軸方 的兩個 S板13 有固定 等，對 面圖， 部12b 的斷面 第 2(a) 固定電 ，在第 定電極 固定電 導電構 14c， 的另一 -11 - 200839242 方露出面形成有抽出電極17a，17b，而導電構件16a， 16b與抽出電極17a，17b分別被電性地連接。又，在第 2 (a)圖雖未被圖示，針對於對於Y軸方向用砝碼部12a的 固定電極1 4a，1 4b，也以同樣的構成設有導電構件及抽 出電極。 在玻璃基板13上，接合有矽製基板11。在此，爲了 將梁部的形成作成容易，作爲矽製基板使用SOI(Silicon On Insulator)基板。又，在砂製基板11上，接合有玻璃 基板1 5。藉由此，形成有Y軸方向用砝碼部12a與對應 於它的固定電極14a，14b所配置的空腔18a，及X軸方 向用砝碼部12b與對應於它的固定電極14c，14d所配置 的空腔1 8b。又，在玻璃基板1 3與矽製基板1 1之間，或 在玻璃基板1 5與矽製基板1 1之間的接合，爲了提高在基 板間所形成的空腔18a，18b的氣密性，而進行陽極接合 較佳。又，在空腔18a內，SOI基板的活性層Ilf成爲扭 轉梁11a，可搖動地支撐Y軸方向用砝碼部12a，而在空 腔1 8b內，S 01基板的活性層1 1 f成爲扭轉梁1 1 b，可搖 動地支撐X軸方向用砝碼部12b。 在第2(b)圖中，在玻璃基板13的一方的主面形成有 凹部，而在其凹部的底面形成有固定電極14e，14f。在玻 璃基板13，設有在兩主面貫通成露出的導電構件16c， 16d，16e，而此些的一方露出面爲與可動電極14e，14f， 電極19分別電性地被連接。又，導電構件1 6c，1 6d， 16e的另一方露出面形成有抽出電極17c，17d，17e而導 -12- 200839242 電構件16e，16d，16e與抽出電極17c，17d，17e分別被 電性地連接。 在玻璃基板1 3上’接合有砂製基板1 1，而在砂製基 板1 1，接合有玻璃基板1 5。藉由此，形成有Z軸方向用 石去碼部12c與對應的固定電極14e所配置的空腔18c，及 Z軸方向用砝碼部12c的固定電極14f所配置的空腔。藉 由此，Z軸方向用的檢測用電極的一方，被密閉在與X軸 ，Y軸，Z軸方向的各該可動電極的空腔不相同的獨立的 空腔內。又，在玻璃基板13與矽製基板1 1之間，或在玻 璃基板1 5與矽製基板1 1之間的接合，爲了提高在基板間 所形成的空腔1 8c的氣密性而進行陽極接合較佳。又，在 空腔18a內，SOI基板的活性層Ilf成爲彎曲梁11c，而 可昇降地支撐Z軸方向用砝碼部1 2c。 扭轉梁1 1 a，1 1 b及彎曲梁1 1 c，是形成於砝碼部 12a，12b，12c的底面側。亦即，各該砝碼部 12a，12b ，1 2c，是具有在矽製基板1 1的厚度方向相對的一對面 ，而扭轉梁1 1 a，1 1 b及彎曲梁1 1 c是沿著各該砝碼部 12a，12b，12c的一方的面所形成。又，由第1圖可知 ，把扭轉梁1 1 a，1 1 b及彎曲梁1 1 c，是分別通過砝碼部 1 2 a，1 2 b，1 2 c的重心位置。藉由形成此種梁，降低另 軸的感度，成爲獨立各軸方向的加速度而可加以檢測出 〇 各軸方向的砝碼部1 2 a，1 2 b，1 2 C的大小，形狀，配 置，或扭轉梁1 1 a，1 1 b，1 1 c的形狀，形成位置是在以下 -13- 200839242 所檢討的狀態下來決定較佳。在該情形下。爲了從同一基 板以同一工程來形成粱，最好將梁的厚度（T)作成相等。 又，加速度的測定範圍作成與容量測定範圍一致的方式， 形成對應於X軸方向用砝碼部1 2b，Y軸方向用砝碼部 1 2a的扭轉梁1 1 a，1 1 b。又，在該情形下，依據感測器規 格來決定初期容量（加速度爲零的容量檢測値），最小感度 (最小容量檢測値）及最大感度（最大容量檢測値）。 這時候，考慮如下的參數。 -容量電極的面積（S ) •形成容量的間隙的距離（D) -間隙的介質係數（ε ) -在測定範圍的間隙的變位（d) -砝碼的質量（Μ) -砝碼的平面大小（L) •砝碼的厚度（Η) -砝碼材料的比重（ρ ) \ -梁材料的破壞強度 -製品所必需的破壞強度 -共振頻率 從此些參數，初期容量是藉由ε x(S/D)求出，而砝碼 的貝夏（M)是確碼的形狀爲正方形，則藉由p xL2xH求出 ’梁與砝碼的重心爲止的高度，是砝碼的形狀爲主方體， 則以大約H/2求出。又，梁的破壞強度超過製品所必需的 破壞強度般地，決定梁的寬度、厚度、長度。 -14- 200839242 Y軸方向用砝碼部12a的扭轉梁1 la及X軸 碼部1 2b的扭轉粱1 1 b，是僅在各該測定加速度 有感度較佳。爲了此，扭轉梁1 1 a，1 1 b是位 12a，12b的一方的主面側（在第2圖中爲底面側） 該砝碼部12a，12b的重心位置較佳。藉由此， 梁1 1 a，1 1 b，是對於被施加於梁的寬度方向及 的加速度幾乎不會變位。一方面，Z軸方向用砝 的梁是與X軸方向用砝碼部12b及Y軸方向用石去 的扭轉梁不相同的彎曲梁之故，因而必須防止變 量的砝碼部12c的變位（d)變成過大的情形。 爲了正確地檢測各軸方向的加速度，必須考 的變位與靜電容量之間的線性。亦即，藉由將確 的變位與靜電容量之間的線性的領域設定在測定 可更正確地檢測出加速度。 在本發明的靜電容量型加速度感測器，各軸 速度是以差動容量方式求出。依據X軸方向用砝 及Y軸方向用砝碼部1 2a的變位成模式化靜電容 爲如第3 (a)圖所示。例如，γ軸方向用砝碼部π 轉梁1 1 a作爲軸朝第1圖的上下方向搖動（在第 朝正前方側與內深側搖動）。此時，Y軸方向用石去 是與固定電極 14a，14b中的一方之間的距離變 另一方之間的距離變小。亦即，在第3(a)圖中， 方向用砝碼部12a靠近固定電極14a，則從固定 遠離’而當Y軸方向用砝碼部12a靠近固定電極 方向用砝 的方向具 於联碼部 ，通過各 此些扭轉 厚度方向 碼部 1 2 c :碼部1 2 a 更靜電容 慮砝碼部 保砝碼部 範圍，就 方向的加 碼部12b 量，則成 ia是以扭 2(a)圖中 碼部1 2 a 大，而與 當Y軸 電極14b 14b，貝ij -15- 200839242 從固定電極14a遠離。將此些Y軸方向用砝碼部12a與固 定電極14a，14b之間的變位量及靜電容量變化（◊記號， □記號）的關係表示於第4(a)圖。在第4圖（a)圖中，橫軸 是以初期間隙間隔規格化增減可動電極與固定電極之間的 距離的數値，縱軸是以初期値規格化形成電極對的各固定 電極與可動電極之間的容量，及各容量間的相差分（差動 容量）的數値。又，在第4(a)圖也表示針對於Y軸方向用 砝碼部12a的差動容量（△記號）。又，X軸方向用砝碼部 12b，固定電極14c，14d的關係也成爲與第3(a)圖，第 4(a)圖同樣。 又，針對於以下的第4(b)，第5(a)，（b)，（c)圖，橫 軸，縱軸是與第4(a)圖的情形同樣。 當將依據Z軸方向用砝碼部1 2c的變位的靜電容量以 模式化，成爲表示於第3(b)圖。例如，Z軸方向用砝碼部 12c是藉由彎曲梁11c朝第2(b)圖的上下方向昇降。這時 候，Z軸方向用砝碼部，12c是變更與固定電極14e之間的 距離。一方面，固定電極14f與矽製基板ll(SOI基板的 活性層1 1 Ό之間的距離是一定之故，因而形成有固定容量 。將該固定容量作爲參照容量，俾差動地檢測依變更Z軸 方向用砝碼部12c與固定電極14e之間的距離的靜電容量 的變化。將此種Z軸方向用砝碼部12c與固定電極14e之 間的變位量及靜電容量變化（◊記號，□記號）的關係表示 於第4(b)圖。又’在第4(b)圖也表示有關於Z軸方向用石去 碼部12c的差動容量（△記號）。 -16- 200839242 將表示於第4(a)圖的特性圖的一部分表示於第5(a)圖 。如從第4(a)圖可知地，差動容量表示左右對稱的變化， 而在應泛範圍可直線近似。如此地，依據表示於上述的差 動容量的原理，當考慮線性的觀點，則有關於X軸方向 的加速度檢測及Y軸方向的加速度檢測，差動容量是對 於初期的間隙間隔，來自砝碼部1 2a，1 2b的初期位置（在 未施加有加速度的狀態的位置）的最大變位爲±20%以內較 佳。這時候，當考慮理想性平行平板電極的變位，則確保 9 5 %以上的線性，惟扭轉梁的差動容量並不是平行平板電 極，因平均變位爲最大變位的1/2，因此令平均在±10%的 範圍內進行動作，將差動容量作成所期望的檢測容量較佳 。藉由此，在線性優異的範圍可檢測加速度。 一方面，有關於Z軸方向的加速度檢測，由第4(b)圖 所示地，差動容量未表示左右對稱的變化，可直線近似的 範圍窄小。在有關於Z軸方向的加速度檢測，如上述地爲 了將一方採用作爲固定容量的差動容量，可確保90 %線性 的變位範圍是對於初期的間隙間隔成爲± 1 〇%。爲了作成 與X軸方向的加速度檢測及Y軸方向的加速度檢測相同 程度的感度（線性），將變位範圍對於初期的間隙間隔抑制 成±5 %以內較佳。爲了設計成爲與X軸方向的加速度檢測 及Y軸方向的加速度檢測相同程度的感度’適當地設定 固定電極14e，14f的面積較佳。例如，如第5(b)圖所示 地，將固定電極14e，14f的面積調整成兩倍’或是如第 5(c)圖所示地，將固定電極14e，14f的面積調整成4倍。 -17- 200839242 針對於Z軸方向用砝碼部12c的彎曲梁1 1 C，z軸方 向用砝碼部1 2c的變位爲設定成爲上述範圍，且對於施加 於Z軸方向以外的加速設定成不會變位。爲了達成此，作 成對稱於X軸方向、Y軸方向的形狀’且增大梁的寬度較 佳。又，調整感度是考慮砝碼部1 2 c與固定電極1 4 e之間 的間隔的中心値，砝碼部1 2 c的質量，彎曲梁1 1 c的長度 作成最適當化。 X軸方向用砝碼部12a及Y軸方向用砝碼部12b的厚 度（H)是愈厚而感度愈優異，惟針對於Z軸方向用砝碼部 1 2 c，若考慮他軸方向感度，則砝碼部1 2 c的厚度愈小者 較佳。又，Z軸方向用砝碼部1 2c的大小’是爲了容易製 造感測器，並減小各軸方向的加速度檢測的相對偏差’作 成與X軸方向用5去碼部12b及Y軸方向用联碼部12a作 成同一較佳。 表示於第2(a)，（b)圖的空腔18a，198b，18c的內壓 ，是設定成大約相等較佳。藉由此’施加加速度時’則可 得到如所期望的相同的阻尼特性’而在各軸方向以相同感 度可檢測出加速度。 又，針對於共振頻率，爲了限制感測器的頻率特性’ 因應於用途適當地設定成爲適當數値。 在此種構成的靜電容量型加速度感測器中’當施加X 軸方向的加速度’則以扭轉梁1 1 b使得X軸方向用&碼 部1 2 b進行搖動。如此地，藉由砝碼部1 2 b搖動而變位’ 使得相對的固定電極1 4 c ’ 1 4 d之間的距離有變更’而可 -18- 200839242 將依該距離變化的靜電容量的變化檢測出作爲差動容量， 以該靜電容量變化可測定出加速度。又，當施加γ軸方 向的加速度，則以扭轉梁1 1 a使得Y軸方向用砝碼部1 2 a 進行搖動。如此地，藉由砝碼部1 2a搖動而變位，使得相 對的固定電極14a，14b之間的距離有變更，而可將依該 距離變化的靜電容量的變化檢測出作爲差動容量，以該靜 電容量變化可測定出加速度。又，當施加Z軸方向的加速 度，則藉由彎曲梁1 1 c而使得Z軸方向用砝碼部1 2c進行 昇降。如此地，藉由砝碼部1 2c昇降而變位，使得與相對 的固定電極1 4 e之間的距離有變更，而可將依該距離變化 的靜電容量的變化檢測出作爲與固定電極1 4f之差動容量 ，以該靜電容量變化可測定出加速度。 以下，針對於具有上述構成的靜電容量型加速度感測 器的製造方法的一例加以說明。 第 6(a)圖至第 6(c)圖，第 7(a)，（b)，第 8(a)，（b)圖 ，第9(a)圖至第9(c)圖，是用以說明本發明的靜電容量型 加速度感測器的製造方法的圖式。 如第6(a)圖所示地，在矽基板16的一方的主面藉由 微影成像及乾蝕刻形成作爲導電構件1 6c，1 6d，1 6e的突 出部。然後，在矽基板1 6的突出部上載置玻璃基板1 3， 如第6 (b)圖所示地，一面加熱一面推壓而將突出部塡在玻 璃基板13般地接合兩基板。之後’如第6(c)圖所示地， 硏磨所得到的複合體的兩主面’並在兩主面露出導電構件 16c，16d，16e。又，第6圖是依據對應於第2(b)圖的構 -19- 200839242 成所表示，惟對應於第2(a)圖的構成也同時地形成。亦即 ，對於對應於X軸方向用砝碼部1 2b及Y軸方向用砝碼 部12a之固定電極14a，14b，14c，14d的導電構件也同 樣地被形成。 然後，如第7(a)圖所示地，對於表示於第6(c)圖的構 造體，藉由微影成像及乾蝕刻，在一方的主面形成凹部 1 3 c。該凹部1 3 c的深度，是相當於砝碼部與固定電極之 間的間隙。之後，如第7(b)圖所示地，在所露出的導電構 件16c，16d，16e上藉由濺鍍被覆電極材料，而藉由微影 成像及蝕刻，分別形成固定電極14e，14f及電極19。 之後，如第8(a)圖所示地，將活性層Uf，具有絕緣 層lid及基底層lie的SOI基板（矽製基板11)的活性層 1 If藉由微影成像及蝕刻形成彎曲梁1 lc。又，SOI基板 的活性層1 1 f的厚度相當於梁的厚度。然後，如第8 (b)圖 所示地，在形成基底層1 1 e的砝碼部的領域藉由微影成像 及蝕刻形成凹部llg。又，第8圖是依據對應於第2(b)圖 的構成所表示，惟對應於第2(a)圖的構成也同時地被形成 。亦即，對應於X軸方向用砝碼部1 2b及Y軸方向用砝 碼部1 2 a的扭轉梁1 1 a，1 1 b或凹部1 1 g也同樣地被形成 〇 然後，如第9(a)圖所示地，有SOI基板的活性層1 If 覆蓋表示於第7(b)圖的構造的玻璃基板13的凹部13c般 地，積層表示於第8(b)圖的矽製基板1 1，進行接合兩基 板1 1，13。此時，藉由陽極接合進行接合較佳。之後， -20- 200839242 如第9(b)圖所示地，藉由微影成像及鈾刻來除去SOI基板 的基底層及絕緣層11 d的所定部分而形成z軸方向用砝碼 部12c。然後，如第9(c)圖所示地，將玻璃基板15接合 於SOI基板的基底層1 le上。此時藉由陽極接合進行接合 較佳。之後，在露出於玻璃基板1 3的主面的導電構件 1 6 c，1 6 d，1 6 e上分別被覆電極材料，藉由微影成像及蝕 刻，分別形成抽出電極17c，17d，17e。 如此所得到的靜電容量型加速度感測器，是獨立地可 檢測出X軸方向、Y軸方向及Z軸方向的加速度，亦即 ，他軸感度較低之故，因而在沒有他軸方向的加速度成分 的干擾的狀態下可檢測出各軸方向的加速度。藉由此，用 以加速度檢測的演算較簡單，而可減小訊號電路的負載。 又，各軸方向的加速度，是藉由差動方式被算出之故，因 而精度優異地可求出。 本發明是並不被限定於上述實施形態，經各種變更可 進行實施。在上述實施形態中，說明針對於使用玻璃基板 與矽製基板的情形，惟在本發明中，也可使用玻璃基板與 矽製基板以外的基板。又，針對於感測器的電極或各層的 厚度或材質，在未超越本發明的效果的範圍下可適當地設 定。又，針對於在上述實施形態所說明·的處理並不被限定 於此，也可變更工程間的適當順序加以實施也可以，例如 ，藉由玻璃基板1 3的触刻進行間隙的形成，惟進行相對 的SOI基板11側也可以。其他，在未超越本發明的目的 的範圍內可適當變更。 -21 - 200839242 【圖式簡單說明】 第1圖是表示本發明的實施形態的靜電容量型加速度 感測器的俯視圖。 第2(a)圖是表示沿著第1圖的HA-HA線的斷面圖。 第2(b)圖是表示沿著第1圖的IIB-IIB線的斷面圖。 第3(a)圖及第3(b)圖是用以說明本發明的實施形態的 靜電容量型加速度感測器的各軸的確碼部的圖式。 第4(a)圖及第4(b)圖是用以說明本發明的實施形態的 靜電容量型加速度感測器的各軸的確碼部的圖式。 第5(a)圖至第5(c)圖是用以說明本發明的實施形態的 靜電容量型加速度感測器的各軸的砝碼部的圖式。 第6(a)圖至第6(c)圖是用以說明本發明的靜電容量型 加速度感測器製造方法的圖式。 第7(a)圖及第7(b)圖是用以說明本發明的靜電容量型 加速度感測器製造方法的圖式。 第8(a)圖及第8(b)圖是用以說明本發明的靜電容量型 加速度感測器製造方法的圖式。 第9(a)圖至弟9(c)圖是用以說明本發明的靜電容量型 加速度感測器製造方法的圖式。 【主要元件符號說明】 1 1 :矽製基板 1 la，1 lb :扭轉粱 -22- 200839242 11c :彎曲 梁 1 Id :絕緣層 lie :基底層 1 If :活性層 llg ，13c :凹部 12a ，12b ，12c: 石去碼部 13， 1 5 :玻璃基板 14a ，14b ，14c， 14d ， 14e ， 1 4 f :固定電極 16a ，1 6b ，16c， 16d ， 16e ： 導電構件 17a ，17b ，17c， 17d ， 17e ： 抽出電極 18a ，18b ，18c· 空腔 19 : 電極 -23-BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electrostatic capacitance type acceleration sensor that detects an acceleration using an electrostatic capacity. [Prior Art] As the sensor for detecting acceleration, for example, there is a capacitance type acceleration sensor. The electrostatic capacitance type accelerometer is composed of a fixed electrode and a movable electrode (weight) that is shaken by applying an acceleration (G), and detects a change in electrostatic capacitance between the fixed electrode and the movable electrode. The acceleration can be found. As such a capacitance type accelerometer, there is a complex weight portion in which a movable electrode is formed by processing a substrate, and a physical quantity change corresponding to an acceleration received by the weight portions is calculated, and separated into three mutually positive sides. A three-axis electrostatic capacitance type acceleration sensor having a structure of acceleration in the direction of the coordinate axis (Patent Document 1). [Patent Document 1] Japanese Patent Laid-Open Publication No. Hei. No. PCT-A No. Hei. Since the components are calculated, the method of calculating the acceleration in the coordinate axis direction orthogonal to each other becomes complicated, so that the signal processing circuit of the sensor becomes large. On the one hand, it is also possible to consider three three-axis electrostatic capacitance type accelerometers that have sensitivity in the direction of each coordinate axis. Only the sensor is large, and the sensor structure becomes complicated. Question-5-200839242 . The present invention has been made in view of such matters, and an object of the present invention is to provide a three-axis electrostatic capacitance type accelerometer having a simple sensor structure and a small size, and a small signal load of a signal processing circuit. The electrostatic capacity type acceleration sensor of the present invention belongs to accelerations in the respective X-axis direction, Y-axis direction, and Z-axis direction, and is detected from a change in electrostatic capacitance between a movable electrode and a fixed electrode functioning as a weight. The capacitance type acceleration sensor includes: a first substrate having three movable electrodes; and at least one of the movable electrodes is opposed to each other with a predetermined interval therebetween, and the capacitance is detected as a differential capacity Each of the pair of detecting electrodes used is a second substrate joined to one main surface of the first substrate as the fixed electrode, and a third substrate bonded to the other main surface of the first substrate. The movable electrode has a pair of surfaces facing the thickness direction of the first substrate, and is sealed in each of the cavities formed by the first substrate, the second substrate, and the third substrate, and is used in the X-axis direction and Y. The movable electrode for the axial direction is supported by the torsion beam so as to be rockable to the first substrate, and the movable electrode for the Z-axis direction is supported by the curved beam for the first substrate. Elevating the torsion beam and said beam bending examination is performed in accordance with the configuration, independently X axis, Y-axis direction and the acceleration along the Z-axis direction is formed along one surface of each of the movable electrode. That is, his axial sensitivity is low, and thus the acceleration in each axial direction can be detected without interference of the acceleration component in his axial direction. By this, the calculation used for acceleration detection is simpler, and -6 - 200839242 can reduce the load of the signal circuit. Further, since the acceleration in each axial direction is calculated in a differential manner, it can be obtained with high accuracy. In the electrostatic capacitance type accelerometer of the present invention, the internal pressure between the cavities is set to be approximately the same, and the amount of change in electrostatic capacitance between the movable electrode and the detecting electrode is approximately the same in each axis. In the range of the range, it is preferable to set the area of the detecting electrode for the Z-axis direction and the amount of displacement of the movable electrode. Thereby, when the acceleration is applied, the same damping characteristics as desired can be obtained, and the acceleration can be detected with approximately the same sensitivity in each axial direction. In the electrostatic capacitance type accelerometer of the present invention, the maximum displacement amount of the movable electrode for the X-axis direction and the Y-axis direction is within ±20% of the predetermined interval, and the maximum of the movable electrode for the Z-axis direction is The amount of displacement is preferably within ±10% of the above-mentioned predetermined interval. According to this configuration, the acceleration can be detected within a range in which the linearity is excellent. In the capacitance type acceleration sensor of the present invention, it is preferable that the detection electrodes are formed on the same surface of the second substrate. According to this configuration, the electrode for differential capacitance detection can be formed by one project, and the process can be simplified. In the electrostatic capacitance type accelerometer of the present invention, the movable electrodes for the respective shafts are arranged on the first substrate, and both have approximately the same shape. According to this configuration, the process can be simplified. In the capacitance type accelerometer of the present invention, one of the detecting electrodes for the z-axis direction is independent of the cavity of each of the movable electrodes in the X-axis, the γ-axis, and the z-axis direction. In the cavity, a total of 4 200839242 cavities are preferably arranged in a plan view of the second substrate to be approximately square. According to this configuration, the fields in which the electrodes are formed on the second substrate are uniform, and the influence on the thermal stress is equal. The electrostatic capacitance type acceleration sensor 'in accordance with the acceleration in the independent X-axis direction, the Y-axis direction, and the Z-axis direction, respectively, is detected from the change in electrostatic capacitance between the movable electrode and the fixed electrode functioning as a weight. The electrostatic capacitance type acceleration sensor includes: a first substrate having three movable electrodes; and at least one of the movable electrodes is opposed to each other with a predetermined interval therebetween, and the capacitance is detected as a differential capacity Each of the pair of detecting electrodes used is a second substrate joined to one main surface of the first substrate as the fixed electrode, and a third substrate bonded to the other main surface of the first substrate. The movable electrode has a pair of surfaces facing the thickness direction of the first substrate, and is sealed in each of the cavities formed by the first substrate, the second substrate, and the third substrate, and is used in the X-axis direction and Y. The movable electrode for the axial direction is supported by the torsion beam so as to be rockable to the first substrate, and the movable electrode for the Z-axis direction is supported by the first substrate by bending The support beam is movable up and down, and the torsion beam and the curved beam are formed along one surface of each of the movable electrodes. Therefore, it is possible to realize a three-axis electrostatic capacity type acceleration sensor in which the sensor structure is simple and the size can be reduced and the load of the signal processing circuit is small. [Embodiment] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. -8 - 200839242 Fig. 1 is a plan view showing a capacitance type acceleration sensor according to an embodiment of the present invention. Further, Fig. 2(a) is a cross-sectional view taken along line IIA-IIA of Fig. 1, and Fig. 2(b) is a cross-sectional view taken along line ΠΒ-ΙΙΒ of Fig. 1. The capacitance type acceleration sensor shown in Fig. 1 is a weight portion 12a, 12b that is joined to three movable electrodes that are detected by acceleration in the respective X-axis direction, Y-axis direction, and Z-axis direction. One of the main surfaces of the tantalum substrate 1 1 of the first substrate of 12c, and each of the pair of weight portions 12a, 12b, and 12c having a change in electrostatic capacitance as a differential capacity is formed at a predetermined interval. The glass substrate 13 of the second substrate of the pair of detection electrodes 14a, 14b, 14c, 14d, 14e, and 14f is formed so that the glass substrate 15 of the third substrate is bonded to the other main surface of the substrate 1 1 . The field (cavity) 16 constituting the weight portion 12a, 12b is shaken and the weight portion 12c is raised and lowered. In the capacitance type acceleration sensor shown in Fig. 1, the movable electrode having the sensitivity in the acceleration in the Y-axis direction is the weight portion 12a, and the movable electrode having the sensitivity in the acceleration in the X-axis direction is the weight portion 1 2b, the movable electrode having the sensitivity in the acceleration in the Z-axis direction is the weight portion 12c. The pair of detecting electrode pairs for the Y-axis direction weight portion 1 2a are the fixed electrodes 14a and 14b, and the pair of detecting electrode pairs for the X-axis direction using the weight portion 12b are the fixed electrodes 14c and 14d. The pair of detecting electrode pairs in the Z-axis direction using weight portion 12c are fixed electrodes 14e and 14f. The Y-axis direction is made of a weight portion 1 2 a which has an approximately rectangular shape on the opposite side in plan view, and is supported by the twisted 粱 1 1 a for the 基板 substrate 1 1 to be supported by -9-200839242, and the X-axis The direction using weight portion 1 2b has a substantially rectangular shape in plan view, and is supported on the opposite side by the torsion beam 1 1 b to be supported by the twisted substrate 11 to be rockable. Each of the torsion beams 11a, lib is disposed in the vicinity of the center of each of the opposing sides of the weight portions 12a, 12b in plan view. On one side, the weight portion 1 2c in the Z-axis direction has a substantially rectangular shape in plan view, and the periphery thereof is supported by the curved beam 1 1 c so as to be movable up and down. On the glass substrate 13, three pairs of detecting electrode pairs are formed. The fixed electrodes 14a and 14b of the Y-axis direction weight portion 12a have approximately the same area. As can be seen from the first figure, the fixed electrodes 14a and 14b in the Y-axis direction are located below the weight portion 1 2a in the Y-axis direction, and pass through the torsion beam 1 . The central part of 1 a is formed as a boundary of the boundary (in the first figure, it is divided vertically). The area of the two fixed electrodes 14a, 14b is approximately equal to the area of the Y-axis direction weight portion 12a. The fixed electrodes 14c and 14d of the X-axis direction weight portion 12b have approximately the same area. As can be seen from the first figure, the fixed electrodes 14c and 14d in the X-axis direction are located below the weight portion 1 2b in the X-axis direction, and pass through the torsion beam 1 The central portion of 1 b is formed as a boundary of the boundary (in the first figure, it is divided into left and right). The area of the two fixed electrodes 14c, 14d is approximately equal to the area of the X-axis direction weight portion 12b. The fixed electrodes 1 4 e, 1 4 f of the z-axis direction weight portion 1 2 c have approximately the same area as the weight portion 1 2c in the Z-axis direction, and one fixed electrode 14e is formed in the Z-axis direction. The lower side of the weight portion 12c is used, and the other fixed electrode 14f is formed in other fields. -10-200839242 In this way, by using three pairs of detecting electrode pairs on the same surface of the glass substrate 13, the electrode (fixed electrode) can be used in one project, which simplifies the process. Preferably. As shown in Fig. 1, it is preferable to arrange the weight portions 12a, 1 2c for the respective axial directions on the tantalum substrate 1 to form a simplified process having approximately the same shape. Further, the fixed electrode pairs 14c and 14d for the X-axis direction, and the fixed electrode pairs 14a and 14b for the direction of use, and the four electrodes of the differential capacitance detecting electrodes 14e and 14f for the Z-axis direction are on the plane of the glass 3. Depending on the arrangement, it is arranged in a square shape. As described above, in the field in which the electrodes are formed in the glass substrate 13 by arranging the electrodes, the influence of the thermal stress is equal, which is preferable. Fig. 2(a) is a view showing a configuration in which the weight of the Y-axis direction weight portion 12a and the X-axis direction are used along the line IIA-IIA of Fig. 1, and Fig. 2(b) shows The IIB-IIB diagram of Fig. 1 shows the configuration of the weight portion 12c for the Z-axis direction. In the figure, the concave portion is formed on one main surface of the glass substrate 13, and the poles 14a, 14b, 14c, 14d are formed on the bottom surface of the concave portion. Further, in the figure (a), the solid 14c and 14d for the X-axis direction weight portion 12b are shown, but the poles 14a and 14b for the Y-axis direction weight portion 12a are not shown on the glass substrate 13. There are members 16a and 16b which are formed to be exposed on both main faces, and one of the exposed faces is electrically connected to the movable electrode 14d. Further, the conductive members 16a, 16b are formed to be detected again, for example, 12b, and the two S-plates 13 of the γ-axis are fixed, and the opposite view, the section of the portion 12b is fixed second (a), in the first The other -11 - 200839242 exposed surface of the fixed electrode fixed electrically conductive structure 14c is formed with extraction electrodes 17a, 17b, and the conductive members 16a, 16b and the extraction electrodes 17a, 17b are electrically connected, respectively. Further, in the second (a) diagram, the conductive members and the extraction electrodes are provided in the same configuration for the fixed electrodes 14a and 14b of the weight portion 12a for the Y-axis direction. A tantalum substrate 11 is bonded to the glass substrate 13. Here, in order to facilitate the formation of the beam portion, an SOI (Silicon On Insulator) substrate is used as the tantalum substrate. Further, a glass substrate 15 is bonded to the sand substrate 11. Thereby, the cavity 18a in which the Y-axis direction weight portion 12a and the fixed electrodes 14a, 14b corresponding thereto are disposed, and the X-axis direction weight portion 12b and the fixed electrodes 14c, 14d corresponding thereto are formed. The cavity 18b is configured. Further, in order to improve the airtightness between the glass substrate 13 and the tantalum substrate 1 or between the glass substrate 15 and the tantalum substrate 1 1 in order to improve the cavities 18a and 18b formed between the substrates. It is preferred to perform anodic bonding. Further, in the cavity 18a, the active layer 11f of the SOI substrate serves as the torsion beam 11a, and the Y-axis direction weight portion 12a is slidably supported, and in the cavity 18b, the active layer 1 1f of the S 01 substrate becomes The torsion beam 1 1 b is slidably supported by the weight portion 12b for the X-axis direction. In the second (b) diagram, a concave portion is formed on one main surface of the glass substrate 13, and fixed electrodes 14e, 14f are formed on the bottom surface of the concave portion. The glass substrate 13 is provided with conductive members 16c, 16d, and 16e which are exposed to be exposed on both main surfaces, and one of the exposed surfaces is electrically connected to the movable electrodes 14e and 14f, respectively. Further, the other exposed surfaces of the conductive members 16c, 16d, and 16e are formed with extraction electrodes 17c, 17d, and 17e, and the conductive members 16e, 16d, 16e and the extraction electrodes 17c, 17d, and 17e are electrically connected, respectively. Ground connection. The sand substrate 1 is bonded to the glass substrate 13 and the glass substrate 15 is bonded to the sand substrate 11. Thereby, the cavity 18c in which the Z-axis direction stone removing portion 12c and the corresponding fixed electrode 14e are disposed, and the cavity in which the Z-axis direction fixing electrode 14f of the weight portion 12c is disposed are formed. Thereby, one of the detecting electrodes for the Z-axis direction is sealed in an independent cavity which is different from the cavity of each of the movable electrodes in the X-axis, Y-axis, and Z-axis directions. Further, bonding between the glass substrate 13 and the tantalum substrate 1 or between the glass substrate 15 and the tantalum substrate 1 is performed in order to improve the airtightness of the cavity 18c formed between the substrates. Anodic bonding is preferred. Further, in the cavity 18a, the active layer 11f of the SOI substrate serves as the curved beam 11c, and the Z-axis direction weight portion 1 2c is supported so as to be movable up and down. The torsion beam 1 1 a, 1 1 b and the curved beam 1 1 c are formed on the bottom surface side of the weight portions 12a, 12b, 12c. That is, each of the weight portions 12a, 12b, and 12c has a pair of faces opposed to each other in the thickness direction of the tantalum substrate 1, and the torsion beam 1 1 a, 1 1 b and the curved beam 1 1 c are along One surface of each of the weight portions 12a, 12b, and 12c is formed. Further, as is apparent from Fig. 1, the torsion beam 1 1 a, 1 1 b and the curved beam 1 1 c are respectively passed through the center of gravity of the weight portions 1 2 a, 1 2 b, and 1 2 c. By forming such a beam, the sensitivity of the other shaft is reduced, and the acceleration in each axial direction is detected, and the size, shape, and arrangement of the weight portions 1 2 a, 1 2 b, and 1 2 C in the respective axial directions can be detected. The shape of the torsion beam 1 1 a, 1 1 b, and 1 1 c is preferably determined by the state reviewed in the following-13-200839242. In this case. In order to form the crucibles from the same substrate in the same process, it is preferable to make the thickness (T) of the beams equal. Further, the measurement range of the acceleration is made to match the capacity measurement range, and the torsion beam 1 1 a, 1 1 b corresponding to the X-axis direction weight portion 1 2b and the Y-axis direction weight portion 1 2a is formed. Further, in this case, the initial capacity (capacity detection 加速度 with zero acceleration), the minimum sensitivity (minimum capacity detection 値), and the maximum sensitivity (maximum capacity detection 値) are determined according to the sensor specifications. At this time, consider the following parameters. - area of the capacity electrode (S) • distance of the gap forming the capacity (D) - medium coefficient of the gap (ε) - displacement in the gap of the measurement range (d) - mass of the weight (Μ) - weight Plane size (L) • Weight of the weight (Η) - Specific gravity of the weight material (ρ ) \ - Destructive strength of the beam material - Destructive strength required for the product - Resonance frequency from these parameters, the initial capacity is by ε x (S/D) is obtained, and the weight of the code is a square, and the height of the center of gravity of the beam and the weight is obtained by p xL2xH, which is the shape of the weight. The cube is found at approximately H/2. Further, the breaking strength of the beam exceeds the breaking strength necessary for the product, and the width, thickness, and length of the beam are determined. -14- 200839242 The torsion 11 1 b of the torsion beam 1 la and the X-axis code portion 1 2b of the weight portion 12a in the Y-axis direction is preferable only for each of the measured accelerations. For this purpose, the torsion beam 1 1 a, 1 1 b is the main surface side (the bottom surface side in Fig. 2) of one of the positions 12a and 12b. The center of gravity of the weight portions 12a and 12b is preferably. Thereby, the beams 1 1 a, 1 1 b are hardly displaced with respect to the acceleration applied to the width direction of the beam. On the other hand, the beam for the Z-axis direction is a curved beam which is different from the torsion beam which is different from the Y-axis direction by the weight portion 12b and the Y-axis direction. Therefore, it is necessary to prevent the displacement of the variable weight portion 12c. (d) It becomes an excessive situation. In order to correctly detect the acceleration in each axis direction, the linearity between the displacement and the electrostatic capacity must be considered. That is, the acceleration can be detected more accurately by setting the field of linearity between the exact displacement and the electrostatic capacity to the measurement. In the electrostatic capacitance type accelerometer of the present invention, each axial speed is obtained by a differential capacity method. According to the X-axis direction, the displacement of the weight portion 1 2a in the y-axis direction and the Y-axis direction is the patterned static capacitance as shown in the third (a) diagram. For example, the γ-axis direction is oscillated in the vertical direction of the first drawing by the weight π-beam 1 1 a as the axis (the front side and the inner side of the first side are oscillated). At this time, the stone in the Y-axis direction is changed from the distance between one of the fixed electrodes 14a and 14b to the other. In other words, in the third (a) diagram, when the direction weight portion 12a is close to the fixed electrode 14a, the direction is fixed from the fixed side, and the direction in the Y-axis direction using the weight portion 12a close to the fixed electrode direction is in the joint code. For each of the twisted thickness direction code portions 1 2 c : the code portion 1 2 a is more static capacitance, the code portion is protected by the code portion, and the amount of the coded portion 12b is y, which is twisted 2 (a) In the figure, the code portion 1 2 a is large, and when the Y-axis electrode 14b 14b, the Bay ij -15-200839242 is away from the fixed electrode 14a. The relationship between the displacement amount and the capacitance change (◊ mark, □ mark) between the weight portion 12a and the fixed electrodes 14a and 14b in the Y-axis direction is shown in Fig. 4(a). In Fig. 4(a), the horizontal axis is a number 値 of increasing or decreasing the distance between the movable electrode and the fixed electrode by the initial gap interval, and the vertical axis is the fixed electrode for forming the electrode pair by the initial 値 normalization and The capacity between the movable electrodes and the number of phase differences (differential capacity) between the respective capacities. Further, the fourth embodiment (a) also shows the differential capacity (Δ symbol) for the Y-axis direction weight portion 12a. Further, the relationship between the fixed-electrode portions 14c and 14d in the X-axis direction is the same as in the third (a) and fourth (a) drawings. Further, in the following 4(b), 5(a), (b), and (c), the horizontal axis and the vertical axis are the same as in the case of Fig. 4(a). The electrostatic capacitance of the displacement by the weight portion 1 2c in accordance with the Z-axis direction is patterned to be shown in Fig. 3(b). For example, the Z-axis direction weight portion 12c is moved up and down in the vertical direction of the second (b) diagram by the curved beam 11c. At this time, the weight portion is used in the Z-axis direction, and 12c is the distance between the fixed electrode 14e and the fixed electrode 14e. On the other hand, the fixed electrode 14f and the tantalum substrate 11 (the distance between the active layers 1 1 Ό of the SOI substrate is constant, and thus a fixed capacity is formed. The fixed capacity is used as the reference capacity, and the differential detection is changed. The change in electrostatic capacitance between the weight portion 12c and the fixed electrode 14e in the Z-axis direction changes the amount of displacement and the capacitance between the weight portion 12c and the fixed electrode 14e in the Z-axis direction (◊ The relationship of the □ mark is shown in Fig. 4(b). The figure 4(b) also shows the differential capacity (Δ mark) of the stone-decoded portion 12c in the Z-axis direction. -16- 200839242 A part of the characteristic diagram shown in Fig. 4(a) is shown in Fig. 5(a). As can be seen from Fig. 4(a), the differential capacity represents a bilaterally symmetric change, and the differential range is linear. In this way, according to the principle of the differential capacity described above, the acceleration detection in the X-axis direction and the acceleration detection in the Y-axis direction are considered from the viewpoint of linearity, and the differential capacity is for the initial gap interval. Initial position of the weight part 1 2a, 1 2b (not applied The maximum displacement of the position of the acceleration state is preferably within ±20%. At this time, when considering the displacement of the ideal parallel plate electrode, the linearity of more than 95% is ensured, but the differential capacity of the torsion beam is not The parallel plate electrode has an average displacement of 1/2 of the maximum displacement, so that the average operation is within ±10%, and the differential capacity is preferably a desired detection capacity. The range is detectable for acceleration. On the one hand, there is acceleration detection in the Z-axis direction. As shown in Fig. 4(b), the differential capacity does not indicate a bilaterally symmetric change, and the range of the linear approximation is narrow. In the acceleration detection in the Z-axis direction, as described above, in order to use one of the differential capacities as the fixed capacity, it is possible to ensure that the 90% linear displacement range is ±1 〇% with respect to the initial gap interval. The acceleration detection and the acceleration in the Y-axis direction are detected to the same degree of sensitivity (linearity), and the displacement range is preferably suppressed to ±5 % within the initial gap interval. In order to design and add to the X-axis direction. Degree detection and acceleration detection in the Y-axis direction have the same degree of sensitivity. It is preferable to appropriately set the areas of the fixed electrodes 14e and 14f. For example, as shown in Fig. 5(b), the area of the fixed electrodes 14e and 14f is adjusted to Double the ' or as shown in Fig. 5(c), the area of the fixed electrodes 14e, 14f is adjusted to four times. -17- 200839242 For the curved beam 1 1 C of the weight portion 12c in the Z-axis direction, In the z-axis direction, the displacement of the weight portion 1 2c is set to the above range, and the acceleration applied to the direction other than the Z-axis direction is set so as not to be displaced. To achieve this, the symmetry is made in the X-axis direction and the Y-axis direction. The shape 'and the width of the beam is preferred. Further, the adjustment sensitivity is considered to be the center 间隔 of the interval between the weight portion 1 2 c and the fixed electrode 14 e , and the mass of the weight portion 1 2 c and the length of the curved beam 1 1 c are optimized. The thickness (H) of the weight portion 12a in the X-axis direction and the weight portion 12b in the Y-axis direction is thicker and the sensitivity is more excellent, but the weight portion 1 2 c is used for the Z-axis direction, and the axial direction sensitivity is considered. The smaller the thickness of the weight portion 1 2 c is, the better. Further, the size ' of the weight portion 1 2c in the Z-axis direction is for the purpose of facilitating the manufacture of the sensor, and the relative deviation of the acceleration detection in the respective axial directions is reduced, and the de-coded portion 12b and the Y-axis direction are formed in the X-axis direction. It is preferable to use the joint portion 12a to be the same. It is preferable that the internal pressures of the cavities 18a, 198b, and 18c shown in Figs. 2(a) and (b) are set to be approximately equal. By the "when acceleration is applied", the same damping characteristic as desired can be obtained, and the acceleration can be detected with the same sensitivity in each axial direction. Further, in order to limit the frequency characteristic of the sensor to the resonance frequency, it is appropriately set to an appropriate number in accordance with the use. In the electrostatic capacitance type acceleration sensor having such a configuration, 'when the acceleration in the X-axis direction is applied', the torsion beam 1 1 b is caused to oscillate in the X-axis direction by the & code portion 1 2 b. Thus, the displacement is changed by the weight of the weight portion 1 2 b such that the distance between the opposing fixed electrodes 1 4 c ' 1 4 d is changed, and -18-200839242 will vary according to the electrostatic capacitance of the distance. The change is detected as a differential capacity, and the acceleration can be measured by the change in the electrostatic capacity. Further, when the acceleration in the γ-axis direction is applied, the torsion beam 1 1 a is caused to swing in the Y-axis direction by the weight portion 1 2 a. In this manner, the displacement of the opposing fixed electrodes 14a, 14b is changed by the shaking of the weight portion 1 2a, and the change in electrostatic capacitance according to the distance can be detected as the differential capacity. This change in electrostatic capacity measures the acceleration. Further, when the acceleration in the Z-axis direction is applied, the Z-axis direction is raised and lowered by the weight portion 1 2c by bending the beam 1 1 c. In this manner, the weight portion 1 2c is moved up and down to be displaced, so that the distance from the opposing fixed electrode 14 e is changed, and the change in electrostatic capacitance according to the distance can be detected as the fixed electrode 1 The differential capacity of 4f can be measured by the change in electrostatic capacitance. Hereinafter, an example of a method of manufacturing a capacitance type acceleration sensor having the above configuration will be described. Figures 6(a) to 6(c), 7(a), (b), 8(a), (b), 9(a) to 9(c), A diagram for explaining a method of manufacturing the electrostatic capacitance type acceleration sensor of the present invention. As shown in Fig. 6(a), projections as the conductive members 16c, 16d, and 16e are formed on one main surface of the ruthenium substrate 16 by lithography and dry etching. Then, the glass substrate 13 is placed on the protruding portion of the ruthenium substrate 16 and, as shown in Fig. 6(b), the substrate is pressed while being heated, and the projections are bonded to the two substrates in the glass substrate 13. Thereafter, as shown in Fig. 6(c), the two principal faces ' of the obtained composite are honed and the conductive members 16c, 16d, 16e are exposed on both principal faces. Further, Fig. 6 is shown in accordance with the configuration of Fig. -19-200839242 corresponding to Fig. 2(b), but the configuration corresponding to Fig. 2(a) is also formed at the same time. In other words, the conductive members corresponding to the fixed portions 14a, 14b, 14c, and 14d of the weight portion 1 2b in the X-axis direction and the weight portion 12a in the Y-axis direction are formed in the same manner. Then, as shown in Fig. 7(a), for the structure shown in Fig. 6(c), the concave portion 1 3 c is formed on one main surface by lithography and dry etching. The depth of the recess 1 3 c corresponds to a gap between the weight portion and the fixed electrode. Thereafter, as shown in FIG. 7(b), the electrode materials are coated on the exposed conductive members 16c, 16d, and 16e by sputtering, and the fixed electrodes 14e and 14f are formed by lithography and etching, respectively. Electrode 19. Thereafter, as shown in FIG. 8(a), the active layer Uf, the active layer 1 If of the SOI substrate (tanned substrate 11) having the insulating layer lid and the underlying layer lie is formed into a curved beam by lithography imaging and etching. 1 lc. Further, the thickness of the active layer 1 1 f of the SOI substrate corresponds to the thickness of the beam. Then, as shown in Fig. 8(b), the concave portion 11g is formed by lithography imaging and etching in the field of forming the weight portion of the base layer 1 1 e. Further, Fig. 8 is shown in accordance with the configuration corresponding to Fig. 2(b), but the configuration corresponding to Fig. 2(a) is also formed at the same time. In other words, the torsion beam 1 1 a, 1 1 b or the recess 1 1 g corresponding to the weight portion 1 2b in the X-axis direction and the weight portion 1 2 a in the Y-axis direction are similarly formed, and then As shown in Fig. 9(a), the active layer 1 If having an SOI substrate covers the concave portion 13c of the glass substrate 13 having the structure shown in Fig. 7(b), and the laminate is shown in Fig. 8(b). The substrate 1 1 is bonded to the two substrates 1 1,13. At this time, bonding by anodic bonding is preferred. Thereafter, -20-200839242, as shown in Fig. 9(b), the basal layer of the SOI substrate and the predetermined portion of the insulating layer 11d are removed by lithography and uranium engraving to form the z-axis direction weight portion 12c. . Then, as shown in Fig. 9(c), the glass substrate 15 is bonded to the underlayer 1 le of the SOI substrate. Bonding by anodic bonding is preferred at this time. Thereafter, the electrode members are respectively coated on the conductive members 16c, 16d, and 16e exposed on the main surface of the glass substrate 13, and the extraction electrodes 17c, 17d, and 17e are formed by lithography and etching, respectively. The electrostatic capacitance type acceleration sensor thus obtained independently detects accelerations in the X-axis direction, the Y-axis direction, and the Z-axis direction, that is, the axis sensitivity is low, and thus the direction of the axis is not present. The acceleration in each axial direction can be detected in the state of interference of the acceleration component. By this, the calculation for acceleration detection is simpler, and the load of the signal circuit can be reduced. Further, the acceleration in each axial direction is calculated by the differential method, and thus the accuracy can be obtained. The present invention is not limited to the above embodiment, and can be implemented with various modifications. In the above embodiment, the case of using a glass substrate and a tantalum substrate will be described. However, in the present invention, a substrate other than the glass substrate and the tantalum substrate may be used. Further, the thickness or material of the electrode or each layer of the sensor can be appropriately set without departing from the effects of the present invention. Further, the processing described in the above embodiment is not limited thereto, and the appropriate order between the items may be changed. For example, the gap may be formed by the contact of the glass substrate 13 . It is also possible to perform the opposite SOI substrate 11 side. Others can be appropriately changed without departing from the scope of the invention. -21 - 200839242 [Brief Description of the Drawings] Fig. 1 is a plan view showing a capacitance type acceleration sensor according to an embodiment of the present invention. Fig. 2(a) is a cross-sectional view taken along line HA-HA of Fig. 1. Fig. 2(b) is a cross-sectional view taken along line IIB-IIB of Fig. 1. Figs. 3(a) and 3(b) are diagrams for explaining the correcting portions of the respective axes of the capacitance type acceleration sensor according to the embodiment of the present invention. 4(a) and 4(b) are diagrams for explaining the identification portions of the respective axes of the capacitance type acceleration sensor according to the embodiment of the present invention. Figs. 5(a) to 5(c) are diagrams for explaining the weight portions of the respective axes of the capacitance type acceleration sensor according to the embodiment of the present invention. 6(a) to 6(c) are diagrams for explaining a method of manufacturing the electrostatic capacity type acceleration sensor of the present invention. Fig. 7(a) and Fig. 7(b) are diagrams for explaining the method of manufacturing the electrostatic capacitance type accelerometer of the present invention. Figs. 8(a) and 8(b) are views for explaining the method of manufacturing the electrostatic capacity type acceleration sensor of the present invention. Fig. 9(a) to Fig. 9(c) are diagrams for explaining the manufacturing method of the electrostatic capacity type acceleration sensor of the present invention. [Description of main component symbols] 1 1 : Twisted substrate 1 la, 1 lb : Torsional 粱-22- 200839242 11c : Bending beam 1 Id : Insulation layer lie : Base layer 1 If : Active layer 11g , 13c : Concave portion 12a , 12b , 12c: stone removing portion 13, 15: glass substrates 14a, 14b, 14c, 14d, 14e, 1 4 f : fixed electrodes 16a, 16b, 16c, 16d, 16e: conductive members 17a, 17b, 17c, 17d , 17e : Extracting electrodes 18a, 18b, 18c · Cavity 19 : Electrode-23-