201208795 六、發明說明： 【發明所屬之技術領域】 本發明係有關可捲繞於載體上之鋸線捲。載體可爲線 軸或芯棒》該線局部或整體以黏著劑處理。鍵線可爲適用 於鬆散磨料（loose abrasive)鋸切程序（鬆散磨料鋸線） ’或者其可具有固定在其表面上的磨料（固定磨料鋸線） ，也可以用於放電加工（EDM線）。 【先前技術】 用來切割諸如矽、石英、鍺等硬而脆的材料之鋸切線 的頭幾件專利可能要歸功於海沃德，其等於上個世紀之 5〇年代（參見GB 771622，GB 717874)獲得頒證。當時 ，矽錠具有1 1/8英寸（28.5mm (毫米））之直徑。寬鬆 硏磨程序被提到，其中使用鎢線來將含磨粉之漿（金剛砂 ’亦即碳化矽）送入切割。鋸切藉由三體硏磨發生（工件 '磨料和線），且所用的線被稱爲「鬆散磨料鋸線」。線 被導經四個具有平行溝槽之線導上方，此等溝槽從而形成 平行配線的平面，該平面稱爲線捲面（wire web)。該線 以前後往復的方式移動，俾在線捲出口不斷地移除線的一 部分線，同時，在線捲入口添加新線。本機器具有目前稱 爲「多線鋸切機」者的所有特點，而其用詞不當之處實際 上在於，只有單一線存在，但該單線被導入多個環圈。摩 托羅拉公司似乎已率先爲半導體產業大規模開發用來切割 矽晶圓之技術（例如參見GB 1 3 97676 )。當時，鋼線被 201208795 用來將具有磨料之漿送入切割。 如今，多線鋸切技術臻於成熟，並已完全超越藉圓內 徑鋸葉片切割矽錠之先前技術。此一內徑鋸因大切口損失 而不再有成本效益（約300//m (微米））。而且，切割 目前技術之直徑300 mm (毫米）矽錠對已幾乎完全放棄 的技術障礙業已造成不能克服之技術障礙。目前用於半導 體用途的所有3 00 mm (毫米）晶圓藉由往復模式之線鋸 操作切割。 在太陽能電池製造領域中，線鋸亦已取得領先地位， 其中’實際上所有目前的晶體太陽能電池均藉線鋸切割， 從單晶或多晶矽錠切割。由於太陽能電池晶圓的幾何要求 較半導體晶圓更不嚴格，因此，太陽能電池並未以往復線 運動模式，而是以單向模式切割。由於線方向逆轉並未浪 費時間（爬坡和緩降），因此，這容許更快的切割。 不僅錠的大小已改變，所用線的直徑和長度亦已演變 。最初使用之180微米鋼鋸線（導致約200 /z m (微米） 的切口損失）業已逐漸減小尺寸，175、160、150、140 至120，甚至lOOyni (微米）。目前低至80/zm (微米 )，其相關切口損失約90 # m (微米）。由於用在鋸線程 序之緊張並未歷時減少，因此，線的拉伸強度須增加，以 維持線的充份強度。雖然最初的抗拉強度就18〇vm (微 米）線而言，約爲3400 N/mm2，目前120 a m (微米）線 的抗拉強度卻超過 4000 N/mm2。目前，單一鋸線的長度 已由原來的30 km (公里）增至約900 km (公里）。這種 ⑧ -6- 201208795 長度必須提供沒有任何捻接或缺陷。 最近，正如火如荼地尋求替代鋸線技術來消除繁瑣的 漿料製備、控制、處理和處置。藉由固定磨料於線，可無 需漿料。僅需冷卻劑來從切割物清除切屑和清潔線。此外 ’切削效率更高，此乃因爲硏磨顆粒不再滾動、黏附在線 和工件之間，而是直接從工件切割材料。該線一般稱爲「 固定磨料鋸線」。 用來切割導電材料之另一線鋸技術是「線放電加工」 或「線EDM」。於該切割技術中，藉由將放電拉入工件 與導電線間的介電媒介物（例如油），進行切割，不斷更 新和定義切割踪跡。這種技術首次進軍來切割如矽之半導 電材料（例如參見WO 200 6/0 27946 A1) » 在下文中，須知「鋸線」係指用於鬆散磨料鋸切，或 固定磨料鋸切或適用於線E D Μ之線。 由以上可知，在鋸切程序中，不管鬆散、固定磨料或 線EDM中，線扮演中心角色；一旦線折斷，切割即立刻 停止。絕對應防止線折斷，此乃因爲，它們不只會導致材 料損失，在許多情況下，工件會因品質問題而損失，亦因 其費幾個小時清潔和重穿線捲（太陽能電池切割可包含超 過1 000個環圈）而耗費運轉時間。 鋸切程序中線折斷的一個來源是線的捲繞。由於鋸線 作成很長’因此，其被捲在線軸或線軸上。替代地，其可 捲在芯棒上，該芯棒可或不可在此後從線捲移除。在任一 種情況下，鋸線捲形成在線軸或芯棒上。於下文將一致提 201208795 到「鋸線捲」，藉此槪念化線捲載體：其與線是或不是某 種載體類型的發明無關。 爲參考，假設繞組軸是水平的，在捲繞期間，線朝觀 看者移動，以及，當自其軸釋放時，線到達軸上方的線軸 ，亦即，線軸朝觀看者旋轉（參見第1圖）。這參考架絕 未限制，此乃因爲，繞線同樣可極佳地完成在垂直配置的 繞線機上，或在線軸站著，而線則透過旋轉飛輪捲繞在線 軸的繞線機上，該飛輪同軸配置於線軸，且軸向上下移動 〇 在線軸捲繞期間，繞組配置成層，此乃因爲，線被以 往復方式從線軸之左側引至線軸的右側和背部。爲此用途 ，一「層」被界定爲在線從線軸一側繞至另一側時形成的 一系列繞組。若線自另一側回繞至第一側，即形成另一層 。爲此用途，常法遵循，奇數層是在捲繞期間由左到右的 層，而偶數層則是在捲繞期間由右至左的層。 一層「繞組」取具有一半徑、一間距長度和間距方向 (右手螺旋，具有peven間距之偶數層之「Z」方向，左手 螺旋，具有Podd間距之奇數層之「S」方向）之形式。半 徑等於業已形成之線捲之直徑的一半，並在捲繞期間，隨 著連續層相互層疊而增加。間距長度係軸向位移’線呈現 僅有一整圈繞線捲軸。線捲之間距對周長的比例等於捲繞 角度的正切。因此，當以恆定間距捲繞時，捲繞角度隨著 線捲直徑增加而減少。 由以上可知，奇數和偶數層繞組在捲繞期間恆在某處 ⑧ -8- 201208795 交叉，且相對於線捲「C」之寬度「W」，交叉數爲： C = W(1/P〇dd + 1/Peven) 線捲密度（WCD )被界定爲線所佔體積除以線捲之總 體積的比例，且可以百分比表示。更高的WCD容許將更 長之線放入相同線軸體積內。 在選擇線繞組的間距方面有許多替代方案： -使用等於線直徑「D」之間距係實現「完善繞組」 的必要條件。於此情況下，每一層封閉，且奇數 和偶數層建構在彼此之頂頂部上。線捲的密度達 到最大，此乃因爲線將成六角形的包裝（除非它 們交叉）。很難以高繞組速度實現完美繞組，惟 其容許如導彈所需極高的解捲速度（參見例如美 國 4950049、美國 5064490)。 -使用等於「D」之倍數的間距將導致並排層，於其 間無間隙。再者，此種繞組很難以高捲繞速度和 大量的層實現。 -藉由使用大間距，可減少線軸寬度上方的交叉數， 但每一層的總長度將減少。這造成較所欲更小的 線捲密度。 在鋸線產業內，經確認可有以下幾種繞組。 -線鋸之解捲站可支配繞組間距。某些線鋸設有起落 滑輪’其以恆定的橫切速度移動。若線捲之間距 不嚴格遵循這個解捲間距，線即以相對於線之軸 向垂直平面之無法接受的大角度解捲。 -9 - 201208795 -在美國2007/0023027中提議用於固定磨料鋸線之 「D」和「2 X D」間之間距》 -目前用於鬆散磨料鋸切之線捲以大於2 X D但小於 200 xD的間距捲繞。這證明是達到夠高的WCD和 解捲品質的折衷。 以下問題可能在鋸線使用期間發生： 每當以往復方式使用線’存在有鋸線之「自行損壞J 結果。由於所用的線在高張力下反覆回繞於線捲，因此’ 彼此相對位移之磨料顆粒，可能會嚴重損壞線。在寬鬆磨 料鋸切情況下，於回繞在新線上時，磨料顆粒黏附在線表 面上之漿料。在固定磨料鋸線的情況下，鋸線的問題更彰 顯，因爲磨料顆粒磨料出現在回繞的捲繞線及新線上。這 種損壞可能會導致過早的線折斷。 「被夾線」也有問題。如前面所述，鋸線已從相對較 厚（1 8 0 m (微米））演變成不折不扣之「細如髮」的 (8 0 // m (微米））。由於線之剛度與直徑之第四次方成 正比，因此，細鋸線不耐彎曲。由於拉伸力量亦已歷年增 加，因此，鋸線極難塑性變形。這兩種傾向已促成容易撓 曲的彈性電線，即使在其自身的重量下，仍很容易撓曲， 但很難永久性彎曲。在線製造期間.，繞組相互鄰接，而後 ，線在彼此互疊。在解捲期間線，線以和其捲繞時相同之 角度離開線軸，不會發生解捲的問題。若因繞組干擾以致 —或更多個稍後放入之繞組到達線捲與較稍後繞組更早放 下之相同或另一層之間，即可能於解捲期間發生線被夾。 ⑧ -10 - 201208795 爲了使此問題可更清楚理解，請參考第2圖。第2圖 的部分1顯示位於線捲210外側之單層212。層212包括 業已依序捲繞於線捲上之不同繞組‘a’、‘b’、‘c’、‘d’及 ‘e’。可見一半繞組以一實線標示，而不可見半繞組則以 —虛線標示。由於干擾，繞組‘e’之環圈216困在先前放 入之繞組‘d’與線捲210之間。在藉由沿方向214拉出線 而解捲時，繞組‘e’和‘d’拉緊，且‘d’抵住線捲210，夾緊 繞組‘ e ’之環圈2 1 6 »在解捲期間出現短暫激烈力量（「障 礙」），這可能會導致線斷裂。 若操作者以環圈216作爲線端部，並從繞組‘e’拉出 自由端，問題即更加嚴重。遍及線捲全寬，線捲解捲困難 ，直至到達線捲端部爲止。 在第2圖的部分2中，繞組‘Γ困在先前放入的繞組 ‘ e ’和‘ d ’下面，圖的解釋仍然相同。 每當稍後置入之繞組於下方與第一道置入的線交叉， 「夾緊線」發生。於第2圖中，這些交叉以虛線圓圈218 標示。 由以上可知，當線軸之配置未改變且線正確解捲時， 可能沒有問題。但是，經常發生以下困擾： -在線軸丟棄期間，繞組之配置可能因在改變線軸之 位向時（藉由使其垂直）繞組掉落或紙包裝，附 接線端（通常藉由打結），而改變。 -在線軸處理和運輸（振動）期間，繞組可能受到干 擾。 -11 - 201208795 -由於若干鋸床具有垂直位向之線軸的軸，因此，有 繞組掉落之真正風險，並因此在安裝線軸時，夾 住電線。 -若在解捲期間，線以相對於軸垂直平面之大「解捲 角度」從線軸拉動，即有頂部繞組在下層繞組上 方移動，並被側拉之虞，從而擾亂下層繞組之順 序。這種風險對在往復線移動之返回循環中使用 原始繞組系統的鋸床特別顯著，該繞組系統不考 慮到最後解捲新線的位置。 -每當失去鋸線端部上的張力，有繞組干擾之虞。因 此，線端必須時常保持張緊》 由以上可知，許多事情可能出錯，它們大多數在鋸線 生產者的控制外。業已尋求各種解決方案來例如將線捲固 定在收縮包裝，外繞組貼以膠帶，或藉清晰可見之端部籤 條，更清楚標示線捲端。然而，這些解決方案均令人不滿 意，其原因在於，收縮包裝在拆封期間對外繞組造成相當 大的困擾。清晰可見的端部籤條只有在鋸機操作員設法於 需要相當技巧和能力之機器運轉期間保持從操作員獲得對 線的指示情況下有用》 因此，本發明人等自行肩起尋找對「夾緊」和「自行 損壞」之充份證明的解決方案。 【發明內容】 因此，本發明之主要目的在於消除任一種鋸線之「被 ⑧ -12- 201208795 夾」和「自行損壞」問題。更特別的是，發明者想要消除 固定磨料鋸線、鬆散磨料鋸線和EDM線的這些問題。發 明人解決兩個層面的問題：鋸線捲層面及鋸線本身上之鋸 線層面。亦提供一種方法來防止線被夾》 根據本發明之第一態樣，請求一種以鋸線成層捲繞之 鋸線捲，該等層之每一者包括複數個繞組。由前段可知「 鋸線捲」所指爲何（[0009]段），「層」所指爲何（ [001 1]段），以及「繞組」所指爲何（[0012]段）。 在本申請案之上下文內（除了已在段[0007]中提到） ，鋸線係金屬線。較佳地，鋸線包括鋼芯或和一或更多塗 層，其內可裝入磨料或無磨料。鋼芯是由普碳鋼（具有 0.70 wt % (重量百分比）碳之最低碳含量。該塗層可爲黃 銅（鬆散磨料鋸線的較佳變種），鋅（在EDM線情況下 )或上面有鎳塗層之銅（在固定磨料鋸線情況下，磨料 顆粒夾持於銅塗層中）。 此外，鋸線之較佳直徑和拉伸強度爲若直徑小於... 然後拉伸強度大於... 2900 N/mm2 3 600 N/mm2 3700 N/mm2 3900 N/mm2 2 5 0 // m (微米） 150ym (微米） 140/zm (微米） 120 // m (微米） 有關鋸線捲之特點在於，於線捲外層之至少一個區域 上有黏著劑（爲清楚起見，參見第3圖）。 -「外」層意指最後捲繞在線軸（並因此率先被解捲 -13- 201208795 的層）。具有可從外面見到之層被認爲是「外層 」（於第3圖中以3 02標示）。 -「區域」意指線釋表面的一部分或全部。「區域」 可分佈於非連續子區，該區域可爲凸或凹，這一 切均未限制本發明。 -須知，「黏著劑」係指有保持鋸線繞組固定之能力 的物質。「黏著劑」之同義詞係「膠水」或「膠 」或「膠黏物質」或「黏性物質」。 -膠水分佈於線上的方式無關緊要：它可藉用刷、海 綿、噴灑頭，帶分配器或本技藝中周知之任何手 段。然而，黏著劑並非意指上面有黏著物質之帶 (如透明膠帶（sellotape，scotch tape)等），因 爲，其已經過證明，未解決這個問題。 較佳地，該區域3 06至少包括鋸線3 04結束的點（參 見第3 a圖）。 黏著劑的功能係僅充份夾持線繞組，使其易移動。該 「夾持力量」可藉由沿垂直於線捲外表面之方向將線剝離 來決定。黏著劑之夾持力量由黏著劑之具體力量和黏著劑 與線之間的接觸區域來決定；接觸區域越大，夾持力越強 。至少夾持力量必須能夠克服至少有一個繞組上的重力， 以防止在運輸和操縱中，繞組掉落。該量約200至300 # N (微牛頓）。較佳地，黏接面積必須能夠夾持一些繞 組，以增加無繞組掉落之確定性。因此，若夾持力量大於 lmN (毫牛頓），即較佳。 ⑧ -14- 201208795 夾持力量不應大於線從線軸拉出的力量。一般，夾持 力量低於25 N。較佳地，夾持力量低於5 N，其約爲發生 於被夾線的瞬間力量。較佳地，這個夾持力量必須低於1 N，俾在黏著區域展開不誘發瞬間力量峰値。一般，黏著 劑是相對弱的黏著劑，其可藉由純粹機械錨固（在鋸線周 圍固化，藉此保持線定位）或弱化學相互作用（范德華力 ，極性相互作用，但不共價接合），或兩者。試驗顯示， 2 0至2 6 0毫牛頓之間的力量足以保持線定位和不會妨礙 線的解捲。 較佳又是，至少某些繞組應藉黏著劑夾持。若在線捲 的外層上，區域3 08從線捲的一端延伸到另一端（參見第 3b圖），外層3 02的所有繞組即可藉黏著劑固定。外層 的每一繞組將因此夾持繞組的至少一部分。以此方式，鋸 線端上之張力可在線鋸之安裝和穿過期間永不喪失；線恆 被夾持且無繞組可能掉落。 替代地，黏著劑可出現在環繞線捲軸之圓形帶區域 3 1 0 (參見第3 c圖）。該帶應夠寬，寬到至少有足夠的繞 組穿過鋸床，此乃因爲，一旦繞組進至帶外，此等繞組即 不再被夾持。 當該區域成螺旋形時，亦可爲兩者的結合，該螺旋自 線捲之一端延伸到另一端，同時環繞線捲軸（圖式中未顯 示之實施例）。 黏著劑也可出現在外層的整個表面3 1 2，其無疑地容 許足夠數量的繞組在穿過鋸床時被夾持（參見第3d圖） -15- 201208795 黏著劑也可在整個線捲上，亦即，所有層覆蓋黏著劑 (圖式所示實施例）。若機器以往復模式運作，其即尤佳 ，此乃因爲，此時，新線捲仍然覆以黏著劑，該黏著劑用 來作爲用在覆以黏著劑之重繞的用過的線之緩衝墊。該實 施例尤佳用於固定磨料鋸線。 鋸線通常以線直徑D之2與2 00倍間的間距捲繞。 尤佳者係介於直徑D之4與200倍間。一般而言，如例 如在光纖線包中所進行情形（美國4 950 049、美國 5 064 4 90 )將鋸線捲繞成「完善繞組」既不佳亦不可行 。如已解釋（第[〇〇1 8]段），鋸線因其極低彎曲剛度和高 伸張強度級別而有「彈性」，且未以鋸線捲繞慣見之捲繞 速度，達到完善繞組。 較佳亦係間距（Ρ。^和Peven )並非線直徑D之整數 倍。若間距爲D之m整數倍，層一般關閉（除了在交叉 發生的點上）。「關閉」意指不同層與具有相同半徑者彼 此相鄰放置。接著，m +1層開始形成於這些m層上。若 間距不是D的倍數，間隙即出現在線之間，從而形成一 個「開口」的線網，經此，黏著劑滲到較小半徑層，藉此 ，提高線捲的固定。因此，開口層繞組較佳。 根據本發明第二態樣，揭示一種本身由本發明第一態 樣之鋸線捲構成之鋸線。 鋸線的特徵在於，其包括在其表面之至少一部分上之 黏著劑。在本申請案之上下文中業已闡明「鋸線」（第 -16- ⑧ 201208795 [0024]段）和「黏著劑」（第[0025]段）的意涵。「在該 鋸線之至少一部分上」意指，在鋸線長度上方的某處經察 可能有一些黏著劑附著於線上。 較佳地，黏著劑可溶解在極性介質中。極性介質的例 子係： -水，一般用來作爲固定磨料線鋸中的冷卻劑。 -聚乙二醇（PEG )或特定化合物二乙二醇（DEG )、三乙二醇（TEG )，四乙二醇。若乙二醇單 體數目大於4，其即一般稱爲PEG。聚乙二醇，乃 至二甘醇特別是用來作爲個鬆散磨料鋸的磨料載 體》 -醇之所有變種，例如甲醇、乙醇、η -丙基醇、異 丙醇。 也可使用上述極性介質的混合物，且其在某些情況下 會於使用期間發展。例如：聚乙二醇已知有吸濕性，且在 使用期間會吸收水。 須知，在鬆散磨料鋸情況下，當使用黏著劑於鋸線時 ，該黏著劑可能會干擾漿液的組成。事實上，線上的黏著 劑會在線捲之第一環圈中被磨掉，並被漿料帶走。爲了防 止配管堵塞或黏著劑殘餘物上磨料之凝塊，較佳係黏著劑 溶解於漿料中。 在固定磨料鋸的情況下，冷卻液必須能夠很容易清掉 黏著劑，使塗層中的磨料顆粒可以很容易地與工件相互作 用。 -17- 201208795 因此，其爲使用可溶解於上述極性介質之較佳實施例 。達到此目的之可能黏著劑係選自包括以下之群組：聚乙 烯吡咯烷酮（噴膠的黏性成份）：聚乙酸乙烯酯、甲基纖 維素、聚乙稀醇、或乙烯/乙烯醇共聚物，聚乙基唑啉或 其混合物。 替代地，可使用其他類型的黏著劑，其未必可溶於極 性媒介物。於此情況下，無論如何，黏著劑殘餘透過摩擦 ，於鋸切程序中移除。於本實施例中，粒子被漿料或冷卻 劑帶走。因此，較佳係當使用這種黏著劑時，藉由例如最 小化塗佈有黏著劑之區域，將進入冷卻劑或漿料的黏著劑 量減至最小。 替代地，當黏著劑不與線交互化學反應且僅機械地錨 固線時，可在進入鋸切捲面之前，提供機械清洗。這可以 如藉由彎曲線，或藉由將其導經刷完成。 可從包括熱塑性黏著劑，熱熔黏著劑，或熱固性黏著 劑如環氧樹脂等之群組選擇未必可溶於極性媒介物的黏著 劑。 替代地，黏著劑可爲從包括酪蛋白、澱粉衍生物、聚 糖或蛋白質系膠之群組中選擇之生物黏著劑。 由於黏著劑與鋸線接觸，因此，它不應該誘發鋸線的 腐蝕。爲了防止這種情況，腐蝕抑制劑可以被添加到黏著 劑。例示類型之腐蝕抑制劑係磷酸鹽、矽酸鹽、矽烷、碳 酸鹽或碳酸酸、硫化物或锍基衍生物、胺或磺酸鹽或其組 合0 ⑧ -18 - 201208795 黏著劑的存在可藉由檢查線端的夾持力量來確定。替 代地，線顯示有手可感覺到的一定「黏性」。黏著劑類型 可從能於線上直接進行之紅外線光譜分析推斷。 在明顯的有益效果出現之前，最小量的黏著劑必須存 在於鋸線上。藉由利用熱裂解判定線表面上之有機炭渣含 量，最佳地決定黏著劑量。在這測試中，只需要一個有限 的樣本（1至2克）。樣品加熱至480°C，直至樣品上的 有機殘留物（但不是在鋼中的碳）分解成一氧化碳和二氧 化碳。於觸媒中，在850°C下，所有一氧化碳轉換成二氧 化碳。由二氧化碳的紅外線吸收計算碳總量。炭渣總量必 須至少大於每克鋸線約400 y g (微克）殘碳。其應大於 每克鋸線約3 000 (微克）殘碳。由於黏著劑不僅由碳 (也由氫和氧）構成，因此，數目低於如雙稱重法之其他 方法所判定者。亦須評估，殘炭的確是因黏著劑（而不是 其他一些有機化合物）而產生，它可很容易藉由紅外線光 譜分析評估。 根據本發明之第三態樣，提供一種防止在鋸線解捲期 間鋸線被夾的方法。該方法包括捲繞鋸線捲繞組的步驟。 在第一較佳實施例中，該線於捲繞期間塗佈有黏著劑。在 第二較佳實施例中’於線捲已完全捲繞後，該鋸線捲之外 表面塗佈有黏著劑。該黏著劑至少塗佈於鋸線捲之外表面 區域上。 【實施方式】 -19- 201208795 實施本發明之模式 在克服線被夾問題的第一個嘗試中，發明人咸認爲， 上油於線，可抑制被夾，此乃因爲，鋸線繞組更容易位移 ，使他們更容易取得平衡。很令人吃驚，線被夾增加。這 導致它們進入完全相反方向，亦即使線黏附黏著劑。這有 違直覺，不期望藉由將鋸線黏合在一起而達到較佳解捲。 進行乙酸乙烯酯（50/50的重量基礎）之水中不同稀 釋度之爾後稱爲「膠水」之水系乳液的一系列試驗。 於第1試驗中，以約3 N的張力及1 .5 mm間距，捲 繞約8 km之長度1之140//m鍍銅鋸線於具有156 mm芯 徑之線軸上。在捲繞期間，1比4體積份之水溶液之膠水 被不斷塗佈在鋸線上。於乾燥後，以500米/分的速度和9 N解捲張力移除約2 km。透過張力計監控捲張力。解捲 在相對於垂直於線捲軸之平面約45°〜85°的嚴格解捲角度 下完成。僅發現小的夾緊力量（小於0· 1 N )，其可歸因 於太強的黏著劑「夾持力量」。在視覺上未發現有位移之 繞組。 準備與第1長度線軸同等狀況惟不供應黏著劑之長度 R之參考線軸。線軸亦在與第1長度線軸相同狀況下解捲 。在視覺上，可以清楚地觀察到位移之線繞組，並發現整 個線軸有重夾緊❶ 在第二個系列的實驗中，測試水中不同稀釋度的膠水 (體積比）。在3N張力下，以1.5 mm (毫米）（12.5xD )間距充塡5 km之120 /z m之鋸線上，寬30 mm (毫米 ⑧ -20- 201208795 )之周帶從線捲端部，並在3 0 mm「控制帶」（未塗佈） 的後續另一邊塗覆1 〇 mm (毫米）黏著劑。藉漆刷’在第 一線軸上以1:20(長度2)之稀釋，在第二線軸上以1 :1〇(長度3)之稀釋，及在第三線軸上以1:5(長度4 )之稀釋塗佈於周帶。 以解捲張力約14 N，在600米/分的速度下，成相同 嚴格角度再次將線軸解捲。獲得以下結果（表Π : 比例 (膠冰） 炭渣 (μδ/g) 解捲等級 1:5 1515(132) 無夾緊 1:10 659(104) 夾緊發生2次 1:20 375(12) 重級夾緊 整體 控制帶 95(24) 重級夾緊 整體 括號內的數字係指標準差。數値結果是至少四個個別 値的平均。「炭渣」藉由裂解方法決定。須知，「控制帶 」顯示一些因無塗層樣品上碳殘餘而產生之炭渣。解捲評 級確認需要最低量的膠水來防止被夾。由此可以得出結論 ，至少1比1 〇體積水稀釋膠水足以達到所欲夾持力量， 以達到所需的夾持力量，而1比4體積水稀釋膠水的解決 方案已顯不有太大的夾持力量。 膠液可用多種方式，例如藉由噴灑，浸塗，塗敷或浸 泡，予以塗佈。膠液可在捲繞期間（例如，藉由使膠液通 .過浸槽或膠液塗佈氈）、間歇地（例如在已捲繞多數層時 -21 - 201208795 ，自動噴槍塗佈一些膠液於中間形成之線捲而不停止捲繞 程序）或最終在捲繞後（例如，當線捲鬆脫時，外表面 塗上一圈膠液，較佳地在解除線的緊張之前塗上），不斷 塗佈。 【圖式簡單說明】 第1圖係一參考架構，其顯示可如何於其上使繞組和 多數層位在線軸上。 第2圖詳細顯示被夾線如何升起。 第3圖本發明可如何實施之四個不同實施例‘a’，‘b’ ，‘ c ’ 和 ‘ d，》 【主要元件符號說明】 2 1 0 :線圈 2 1 2 :單層 214 :方向 216 :環圈 2 1 8 :虛線圓 3〇2 :外層 3〇4 :鋸線 306 ·區域 308 ·區域 312 :表面 -22-201208795 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to a saw wire wound which can be wound around a carrier. The carrier can be a bobbin or a mandrel. The line is treated partially or entirely with an adhesive. The key line can be used for loose abrasive sawing procedures (loose abrasive sawing wire) 'or it can have abrasives fixed on its surface (fixed abrasive sawing wire) or can be used for electrical discharge machining (EDM wire) . [Prior Art] The first few patents used to cut sawing wires such as hard, brittle materials such as enamel, quartz, and enamel may be attributed to Hayward, which is equal to the 5th century of the last century (see GB 771622, GB). 717874) Obtained the certificate. At that time, the bismuth ingot had a diameter of 1 1/8 inch (28.5 mm (mm)). A loose honing procedure was mentioned in which a tungsten wire was used to feed the milled pulp (corundum, i.e., tantalum carbide) into the cut. Sawing occurs by three-body honing (workpieces 'abrasives and wires') and the line used is called a "loose abrasive sawing wire". The lines are guided over four wire guides having parallel grooves which form a plane of parallel wiring, referred to as a wire web. The line moves back and forth in a reciprocating manner, and the online roll exit continuously removes a portion of the line, and a new line is added to the line roll entry. This machine has all the features of what is currently referred to as a "multi-wire sawing machine", and its misuse is actually that only a single line exists, but the single line is introduced into multiple loops. Motorola seems to have taken the lead in developing large-scale technologies for cutting silicon wafers for the semiconductor industry (see, for example, GB 1 3 97676). At that time, the steel wire was used by 201208795 to feed the abrasive slurry into the cutting. Today, multi-wire sawing technology is mature and has completely surpassed the previous technology of cutting bismuth ingots with round internal diameter saw blades. This inner diameter saw is no longer cost effective (about 300//m (micrometer)) due to large cut loss. Moreover, the cutting of 300 mm (mm) diameter bismuth ingots of the current technology has created an insurmountable technical obstacle to the technical obstacles that have almost completely abandoned. All 300 mm (mm) wafers currently used for semiconductor applications are cut by a wire saw operation in a reciprocating mode. In the field of solar cell manufacturing, wire saws have also taken the lead, in which virtually all current crystalline solar cells are cut by wire saws and cut from single crystal or polycrystalline germanium ingots. Since the geometry requirements of solar cell wafers are less stringent than semiconductor wafers, solar cells are not cut in a reciprocating mode but in a unidirectional mode. This allows for faster cutting because the line direction reversal does not waste time (climbing and slowing down). Not only has the size of the ingot changed, but the diameter and length of the wire used have also evolved. The 180 micron hacksaw wire originally used (resulting in a cut loss of about 200 /z m (micron)) has been gradually reduced in size, 175, 160, 150, 140 to 120, or even 100 yni (micron). Currently as low as 80/zm (micron), the associated kerf loss is about 90 #m (micron). Since the tension applied to the saw thread is not reduced over time, the tensile strength of the wire must be increased to maintain the full strength of the wire. Although the initial tensile strength is about 3400 N/mm2 for the 18 〇vm (micrometer) line, the current 120 am (micrometer) line has a tensile strength of more than 4000 N/mm2. Currently, the length of a single saw wire has increased from the original 30 km (km) to approximately 900 km (km). This 8 -6- 201208795 length must be provided without any splicing or defects. Recently, alternative cutting technology has been sought to eliminate cumbersome slurry preparation, control, handling and disposal. By fixing the abrasive to the wire, no slurry is required. Only coolant is required to remove chips and cleaning lines from the cut. In addition, cutting efficiency is higher because the honing particles no longer roll, stick between the wire and the workpiece, but cut the material directly from the workpiece. This line is generally referred to as a "fixed abrasive sawing wire". Another wire saw technology used to cut conductive materials is "wire discharge machining" or "line EDM." In this cutting technique, cutting is performed by pulling a discharge into a dielectric medium (e.g., oil) between the workpiece and the conductive line, and the cutting trace is continuously updated and defined. This technique was first introduced to cut semi-conducting materials such as enamel (see, for example, WO 200 6/0 27946 A1). » In the following, it is to be understood that "saw line" means for loose abrasive sawing or fixed abrasive sawing or for Line ED Μ line. From the above, in the sawing procedure, the line plays a central role in loose, fixed abrasive or wire EDM; once the line breaks, the cutting stops immediately. Absolute should prevent the wires from breaking, because they not only cause material loss, in many cases the workpiece is lost due to quality problems, but also because it takes several hours to clean and re-thread the coil (solar cell cutting can contain more than 1) 000 loops) and running time. One source of wire breakage in the sawing process is the winding of the wire. Since the saw wire is made very long, it is wound onto a bobbin or bobbin. Alternatively, it can be wound onto a mandrel that may or may not be removed from the coil afterwards. In either case, the wire roll is formed on a spool or mandrel. In the following, we will unanimously mention 201208795 to the "saw line roll", thereby mourning the line roll carrier: it is irrelevant to the invention that the line is or is not a carrier type. For reference, assume that the winding axis is horizontal, the wire moves toward the viewer during winding, and, when released from its axis, the wire reaches the bobbin above the shaft, ie, the bobbin rotates toward the viewer (see Figure 1). ). This reference frame is not limited, because the winding can also be excellently completed on a vertically arranged winding machine, or the spool is standing, and the wire is wound on the winding machine of the bobbin through the rotating flywheel. The flywheel is coaxially disposed on the bobbin and moves axially up and down. During winding of the bobbin, the windings are arranged in layers because the wires are reciprocated from the left side of the bobbin to the right and back sides of the bobbin. For this purpose, a "layer" is defined as a series of windings formed when winding from one side of the spool to the other. If the wire wraps from the other side to the first side, another layer is formed. For this purpose, the usual practice follows that the odd layers are layers from left to right during winding, while the even layers are layers from right to left during winding. A layer of "winding" takes the form of a radius, a pitch length, and a pitch direction (right-handed spiral, "Z" direction with even-numbered layers of peven spacing, left-handed spiral, "S" direction with odd-numbered layers of Podd spacing). The radius is equal to half the diameter of the formed coil and is increased during the winding as the successive layers are stacked one on another. The pitch length is the axial displacement 'line showing only one full turn of the reel. The ratio of the distance between the coils to the circumference is equal to the tangent of the winding angle. Therefore, when wound at a constant pitch, the winding angle decreases as the diameter of the coil increases. It can be seen from the above that the odd and even layer windings are always crossed at some point 8 -8 - 201208795 during winding, and the number of intersections is relative to the width "W" of the coil "C": C = W (1/P 〇 Dd + 1/Peven) The line volume density (WCD) is defined as the ratio of the volume occupied by the line divided by the total volume of the line, and can be expressed as a percentage. A higher WCD allows longer lines to be placed within the same spool volume. There are many alternatives in choosing the spacing of the wire windings: - The requirement to achieve "perfect winding" is achieved using a distance equal to the line diameter "D". In this case, each layer is closed and the odd and even layers are constructed on top of each other. The density of the coils is maximized because the lines are packed into hexagons (unless they intersect). It is difficult to achieve perfect windings at high winding speeds, but it allows for extremely high unwinding speeds such as those required for missiles (see, for example, US 4950049, US 5064490). - Using a spacing equal to a multiple of "D" will result in a side-by-side layer with no gaps between them. Moreover, such windings are difficult to achieve with high winding speeds and a large number of layers. - By using a large pitch, the number of crossings above the spool width can be reduced, but the total length of each layer will be reduced. This results in a smaller wire density than desired. In the sawing industry, the following windings have been identified. - The unwinding station of the wire saw can control the winding pitch. Some wire saws are provided with landing gears' which move at a constant cross-cut speed. If the distance between the coils does not strictly follow this unwinding pitch, the line is unwrapped at an unacceptably large angle with respect to the vertical plane of the line. -9 - 201208795 - The distance between "D" and "2 XD" for fixed abrasive sawing wire proposed in US 2007/0023027 - Current line for loose abrasive sawing is greater than 2 XD but less than 200 xD The pitch is wound. This proves to be a compromise between achieving a high enough WCD and unwinding quality. The following problems may occur during the use of the saw wire: Whenever the wire is used in a reciprocating manner, there is a "self-damaged J result of the saw wire. Since the wire used is repeatedly wound around the coil under high tension, it is displaced relative to each other. Abrasive particles can seriously damage the wire. In the case of loose abrasive sawing, the abrasive particles adhere to the slurry on the wire surface when rewinding on the new wire. In the case of fixed abrasive wire, the problem of the wire is more obvious. Because abrasive grain abrasives appear on the rewinding winding line and the new wire. This damage may cause premature wire breakage. There is also a problem with the "clamped wire". As mentioned earlier, the saw wire has evolved from a relatively thick (180 m (micron)) to an uncompromising "fine as hair" (8 0 // m (micron)). Since the stiffness of the wire is proportional to the fourth power of the diameter, the fine saw wire is not resistant to bending. Since the tensile force has also increased over the years, the saw wire is extremely difficult to plastically deform. These two tendencies have contributed to elastic wires that are easily deflected, and are easily deflected even under their own weight, but are difficult to bend permanently. During the manufacturing process, the windings abut each other, and then the wires overlap each other. During the unwinding of the line, the line leaves the spool at the same angle as it was wound, and the problem of unwinding does not occur. If the windings interfere with each other - or more of the windings that are later placed reach the same or another layer that is placed earlier than the later windings, it is possible that the wire is caught during unwinding. 8 -10 - 201208795 For a clearer understanding of this issue, please refer to Figure 2. Part 1 of Fig. 2 shows a single layer 212 located outside of the coil 210. Layer 212 includes the different windings 'a', 'b', 'c', 'd', and 'e' that have been wound on the coil in sequence. It can be seen that half of the windings are indicated by a solid line, while the invisible half windings are indicated by a dashed line. Due to the disturbance, the loop 216 of the winding 'e' is trapped between the previously placed winding 'd' and the coil 210. When unwinding by pulling the wire in direction 214, windings 'e' and 'd' are tightened, and 'd' is against wire coil 210, clamping ring 'e' ring 2 1 6 » in solution A brief intense force ("barrier") occurs during the volume, which can cause the line to break. If the operator takes the loop 216 as the end of the wire and pulls the free end from the winding 'e', the problem becomes more serious. The entire width of the line is wide, and it is difficult to unwind the line until it reaches the end of the line. In part 2 of Fig. 2, the windings are 'trapped' below the previously placed windings 'e' and 'd', and the explanation of the figure remains the same. Whenever the winding placed later is crossed below the line placed in the first pass, the "clamp line" occurs. In Figure 2, these intersections are indicated by dashed circles 218. As can be seen from the above, there may be no problem when the configuration of the bobbin is not changed and the line is correctly unwound. However, the following problems often occur: - During the wire axis discarding, the winding configuration may be due to the falling of the windings or the paper wrap when changing the position of the bobbin (by making it vertical), attaching the terminal (usually by knotting), And change. - Winding may be disturbed during spool processing and transport (vibration). -11 - 201208795 - Since several sawing machines have axes that are oriented vertically to the spool, there is a real risk of the windings falling, and therefore the wires are clamped when the spool is installed. - If during unwinding, the line is pulled from the bobbin by a large "unwinding angle" with respect to the vertical plane of the axis, that is, the top winding moves above the lower winding and is pulled sideways, thereby disturbing the order of the lower winding. This risk is particularly pronounced for sawing machines that use the original winding system in the return cycle of the reciprocating line movement, which does not take into account the position of the final unwinding of the new line. -Whenever the tension on the end of the wire is lost, there is a wind disturbance. Therefore, the wire ends must be kept tight at all times. As can be seen from the above, many things can go wrong, most of them are outside the control of the wire producer. Various solutions have been sought, such as fixing the coil in a shrink wrap, tape the outer winding, or clearly marking the end of the coil with a clearly visible end label. However, these solutions are unsatisfactory because the shrink wrap is quite confusing to the outer winding during unpacking. The clearly visible end tag is only useful if the saw operator is trying to keep an indication of the line from the operator during the operation of the machine that requires considerable skill and ability. Therefore, the inventor has to shoulder to find the pair of clips. A fully documented solution for "tight" and "self-destructive". SUMMARY OF THE INVENTION Therefore, the main object of the present invention is to eliminate the problem of "8-12-201208795 clip" and "self-damage" of any saw wire. More specifically, the inventors wanted to eliminate these problems with fixed abrasive saw wires, loose abrasive saw wires, and EDM wires. The inventor solves two levels of problems: the sawing layer level and the sawing level on the sawing line itself. A method is also provided to prevent the wire from being pinched. According to a first aspect of the present invention, a wire wound wound in layers by a saw wire is claimed, each of the layers comprising a plurality of windings. From the preceding paragraph, we can see what the "saw line roll" refers to (paragraph [0009]), what the "layer" refers to ([001 1]), and what the "winding" refers to ([0012]). Within the context of this application (except as already mentioned in paragraph [0007]), the wire is a metal wire. Preferably, the wire comprises a steel core or one or more coatings which may be filled with or without abrasive. The steel core is made of plain carbon steel (with a minimum carbon content of 0.70 wt% carbon). The coating can be brass (a preferred variant of a loose abrasive saw wire), zinc (in the case of an EDM line) or above. Nickel coated copper (in the case of a fixed abrasive sawing wire, the abrasive particles are held in the copper coating). In addition, the preferred diameter and tensile strength of the saw wire is less than... and then the tensile strength is greater than ... 2900 N/mm2 3 600 N/mm2 3700 N/mm2 3900 N/mm2 2 5 0 // m (micron) 150ym (micron) 140/zm (micron) 120 // m (micron) It is characterized in that there is an adhesive on at least one of the outer layers of the coil (see Figure 3 for clarity) - "outer" layer means the last winding of the spool (and thus the first to be unwound - 13- The layer of 201208795. The layer that can be seen from the outside is considered to be the "outer layer" (marked by 3 02 in Figure 3) - "Zone" means part or all of the surface of the line. Distributed in a non-continuous sub-region, which may be convex or concave, all of which do not limit the invention. - It should be noted that "adhesive" means A substance that maintains the ability of the wire winding to be fixed. The term "adhesive" is synonymous with "glue" or "glue" or "adhesive material" or "sticky material". - The way the glue is distributed on the line does not matter: it can be borrowed Brush, sponge, sprinkler head, dispenser, or any means known in the art. However, the adhesive does not mean a strip of adhesive material (such as scotch tape, etc.) because it It has been proven that this problem has not been solved. Preferably, the region 306 includes at least the point at which the saw wire 3 04 ends (see Figure 3a). The function of the adhesive is to only fully clamp the wire winding, making it easy to move. The "clamping force" can be determined by peeling the wire in a direction perpendicular to the outer surface of the coil. The adhesive force of the adhesive is determined by the specific force of the adhesive and the contact area between the adhesive and the wire; The larger the contact area, the stronger the clamping force. At least the clamping force must be able to overcome the gravity on at least one of the windings to prevent the winding from falling during transportation and handling. The amount is about 200 to 300 # N (micro Newton) . Preferably, the bonding area must be able to hold some windings to increase the certainty of no winding drop. Therefore, if the clamping force is greater than lmN (millitons), it is better. 8 -14- 201208795 The clamping force should not be Greater than the force the wire pulls from the spool. Typically, the clamping force is less than 25 N. Preferably, the clamping force is less than 5 N, which is about the momentary force that occurs when the wire is clamped. Preferably, this grip The force must be below 1 N, and the deployment of the 俾 in the adhesive area does not induce an instantaneous force peak. Generally, the adhesive is a relatively weak adhesive that can be anchored by purely mechanical (curing around the wire to maintain line positioning) Or weak chemical interactions (van der Waals forces, polar interactions, but not covalent bonding), or both. Tests have shown that a force between 20 and 260 millinewtons is sufficient to maintain line alignment and not interfere with line unwinding. Preferably, at least some of the windings are held by an adhesive. If the area 3 08 extends from one end of the coil to the other end on the outer layer of the coil (see Figure 3b), all of the windings of the outer layer 302 can be secured by an adhesive. Each winding of the outer layer will thus clamp at least a portion of the winding. In this way, the tension on the wire end can never be lost during installation and passage of the wire saw; the wire is clamped and no windings can fall. Alternatively, an adhesive may be present in the circular band region 3 1 0 of the wraparound spool (see Figure 3c). The strip should be wide enough to have at least enough windings through the sawing machine because the windings are no longer clamped once the windings are brought into the belt. When the region is spiraled, it may also be a combination of the two, the spiral extending from one end of the coil to the other end while surrounding the wire spool (embodiment not shown in the drawings). Adhesives can also be present on the entire surface of the outer layer 3 1 2, which undoubtedly allows a sufficient number of windings to be clamped as they pass through the sawing machine (see Figure 3d). -15- 201208795 Adhesives can also be applied throughout the coil. That is, all layers are covered with an adhesive (embodiment shown in the drawings). If the machine is operated in a reciprocating mode, it is particularly preferred because, at this time, the new coil is still covered with an adhesive, which is used as a buffer for the used wire used for rewinding the adhesive. pad. This embodiment is particularly preferred for use in fixed abrasive sawing wires. The saw wire is usually wound at a pitch of 2 to 200 times the wire diameter D. The better ones are between 4 and 200 times the diameter D. In general, it is neither good nor feasible to wind a saw wire into a "perfect winding", as in the case of a fiber optic cable package (US 4 950 049, US 5 064 4 90). As explained (paragraph [〇〇18]), the saw wire is "elastic" due to its extremely low bending stiffness and high tensile strength level, and the winding speed is not conventionally wound by a saw wire to achieve perfect winding . Preferably, the spacing (Ρ.^ and Peven) is not an integer multiple of the line diameter D. If the spacing is an integer multiple of D, the layer is generally closed (except at the point where the intersection occurs). "Off" means that different layers are placed next to each other with the same radius. Next, m+1 layers are formed on these m layers. If the pitch is not a multiple of D, the gap appears between the lines, thereby forming an "opening" wire mesh, whereby the adhesive seeps into the smaller radius layer, thereby increasing the fixing of the wire coil. Therefore, the open layer winding is preferred. According to a second aspect of the present invention, a saw wire which itself constitutes a saw wire roll of the first aspect of the invention is disclosed. A saw wire is characterized in that it includes an adhesive on at least a portion of its surface. The meaning of "saw line" (paragraphs -16-8 201208795 [0024]) and "adhesive" (paragraph [0025]) has been clarified in the context of this application. "On at least a portion of the wire" means that somewhere above the length of the wire is observed that some adhesive may be attached to the wire. Preferably, the adhesive is soluble in the polar medium. Examples of polar media: - Water, generally used as a coolant in fixed abrasive wire saws. - polyethylene glycol (PEG) or the specific compound diethylene glycol (DEG), triethylene glycol (TEG), tetraethylene glycol. If the number of ethylene glycol monomers is greater than 4, it is generally referred to as PEG. Polyethylene glycol, or even diethylene glycol, especially as an abrasive carrier for a loose abrasive saw, - all variants of alcohols, such as methanol, ethanol, η-propyl alcohol, isopropanol. Mixtures of the above polar media can also be used, and in some cases will develop during use. For example, polyethylene glycol is known to be hygroscopic and absorbs water during use. It should be noted that in the case of a loose abrasive saw, the adhesive may interfere with the composition of the slurry when an adhesive is applied to the wire. In fact, the adhesive on the line is worn away in the first loop of the coil and taken away by the slurry. In order to prevent clogging of the pipe or clinking of the abrasive on the residue of the adhesive, it is preferred that the adhesive be dissolved in the slurry. In the case of a fixed abrasive saw, the coolant must be able to easily remove the adhesive so that the abrasive particles in the coating can easily interact with the workpiece. -17-201208795 Therefore, it is a preferred embodiment in which a solvent which is soluble in the above-mentioned polar medium is used. Possible adhesives for this purpose are selected from the group consisting of polyvinylpyrrolidone (adhesive component of spray glue): polyvinyl acetate, methyl cellulose, polyethylene glycol, or ethylene/vinyl alcohol copolymer. , polyethyloxazoline or a mixture thereof. Alternatively, other types of adhesives may be used which are not necessarily soluble in the polar vehicle. In this case, in any case, the adhesive residue is removed by friction and removed during the sawing process. In this embodiment, the particles are carried away by the slurry or coolant. Therefore, it is preferred to minimize the amount of the adhesive entering the coolant or slurry by, for example, minimizing the area coated with the adhesive when such an adhesive is used. Alternatively, when the adhesive does not chemically interact with the wire and only mechanically anchors the wire, a mechanical cleaning can be provided prior to entering the sawing roll. This can be done, for example, by bending the wire or by guiding it through the brush. Adhesives which are not necessarily soluble in polar vehicles may be selected from the group consisting of thermoplastic adhesives, hot melt adhesives, or thermosetting adhesives such as epoxy resins. Alternatively, the adhesive may be a bioadhesive selected from the group consisting of casein, starch derivatives, polysaccharides or protein gums. Since the adhesive is in contact with the wire, it should not induce corrosion of the wire. To prevent this, a corrosion inhibitor can be added to the adhesive. Exemplary types of corrosion inhibitors are phosphates, citrates, decanes, carbonates or carbonates, sulfides or sulfhydryl derivatives, amines or sulfonates or combinations thereof. 0 8 -18 - 201208795 The presence of an adhesive can be borrowed It is determined by the clamping force at the inspection line end. Instead, the line shows a certain "stickiness" that can be felt by the hand. The type of adhesive can be inferred from infrared spectral analysis that can be performed directly on the line. The minimum amount of adhesive must be present on the wire before significant benefits occur. The adhesion dose is optimally determined by determining the amount of organic carbon residue on the surface of the wire by thermal cracking. In this test, only a limited sample (1 to 2 grams) is required. The sample was heated to 480 ° C until the organic residue on the sample (but not the carbon in the steel) decomposed into carbon monoxide and carbon dioxide. In the catalyst, all carbon monoxide was converted to carbon dioxide at 850 °C. The total amount of carbon is calculated from the infrared absorption of carbon dioxide. The total amount of carbon residue must be at least greater than about 400 y g (micrograms) of residual carbon per gram of saw wire. It should be greater than approximately 3,000 (micrograms) of residual carbon per gram of saw wire. Since the adhesive is composed not only of carbon (also composed of hydrogen and oxygen), the number is lower than that determined by other methods such as the double weighing method. It is also necessary to assess that charcoal is indeed produced by adhesives (rather than some other organic compounds), which can be easily assessed by infrared spectroscopy. According to a third aspect of the present invention, there is provided a method of preventing a wire from being caught during unwinding of a saw wire. The method includes the step of winding a wire winding set. In a first preferred embodiment, the wire is coated with an adhesive during winding. In the second preferred embodiment, the outer surface of the wire wound roll is coated with an adhesive after the wire roll has been completely wound. The adhesive is applied to at least the outer surface area of the saw roll. [Embodiment] -19-201208795 Mode for Carrying Out the Invention In the first attempt to overcome the problem of wire clamping, the inventor believes that oiling on the wire can suppress the pinching, because the wire winding is more Easy to displace, making them easier to balance. Very surprising, the line was clipped. This causes them to go in exactly the opposite direction, even if the line sticks to the adhesive. This is counterintuitive and it is not desirable to achieve better unwinding by bonding the saw wires together. A series of tests for aqueous emulsions known as "glue" after the different dilutions of vinyl acetate (50/50 by weight basis) were carried out. In the first test, a 140//m copper-plated saw wire of length 1 of about 8 km was wound on a bobbin having a core diameter of 156 mm with a tension of about 3 N and a pitch of 1.5 mm. During the winding, 1 to 4 parts by volume of the aqueous solution of the glue was continuously applied to the saw wire. After drying, about 2 km was removed at a speed of 500 m/min and a 9 N unwinding tension. The tension of the roll is monitored by a tensiometer. Unwinding is accomplished at a strict unwinding angle of approximately 45° to 85° with respect to the plane perpendicular to the line reel. Only a small clamping force (less than 0·1 N) was found, which can be attributed to the too strong adhesive "clamping force". No displacement windings were visually found. Prepare the reference spool with the same length as the first length spool but only the length of the adhesive R. The bobbin is also unwound under the same conditions as the first length bobbin. Visually, the line winding of the displacement can be clearly observed and the entire spool is found to be heavily clamped. In the second series of experiments, different dilutions of glue (volume ratio) were tested in the water. Under the tension of 3N, fill the 5 km / 120m saw wire with a pitch of 1.5 mm (mm) (12.5xD), and the width of 30 mm (mm 8 -20 - 201208795) is taken from the end of the coil, and The next side of the 30 mm "control tape" (uncoated) is coated with 1 〇mm (mm) of adhesive. By paint brush 'diluted 1:20 (length 2) on the first spool, 1:1 (length 3) on the second spool, and 1:5 (length on the third spool) 4) The dilution is applied to the peripheral zone. With a unwinding tension of about 14 N, the bobbin is unwound again at the same strict angle at a speed of 600 m/min. The following results were obtained (Table Π: Proportion (gel ice) Cinder (μδ/g) Unwinding grade 1:5 1515 (132) No clamping 1:10 659 (104) Clamping occurred 2 times 1:20 375 (12 Heavy-duty clamping overall control belt 95 (24) Heavy-duty clamping The number in the overall brackets refers to the standard deviation. The number of results is the average of at least four individual enthalpies. The "carbon residue" is determined by the cracking method. The “Control Strip” shows some of the carbon residue from the carbon residue on the uncoated sample. The unwinding rating confirms that a minimum amount of glue is needed to prevent it from being pinched. It can be concluded that at least 1 to 1 volume of water dilutes the glue. Sufficient to achieve the desired clamping force to achieve the required clamping force, and the solution of 1 to 4 volumes of water to dilute the glue has not shown much clamping force. The glue can be applied in a variety of ways, for example by spraying, Dip coating, coating or soaking, coating. The glue can be applied during winding (for example, by passing the glue through the dip tank or the glue coating felt), intermittently (for example, in the majority layer Pm-21 - 201208795, automatic spray gun coating some glue to form a line in the middle without stopping Wrap around) or finally after winding (for example, when the coil is loose, the outer surface is coated with a ring of glue, preferably before the tension of the line is removed), and the coating is continuously applied. Figure 1 is a reference architecture showing how the windings and most of the horizons can be placed on the bobbin. Figure 2 shows in detail how the clamped wire rises. Figure 3 shows how the invention can be implemented in four different ways. Embodiments 'a', 'b', 'c' and 'd," [Description of main component symbols] 2 1 0 : Coil 2 1 2 : Single layer 214: Direction 216: Loop 2 1 8 : Dotted circle 3〇 2: outer layer 3〇4: saw line 306 · area 308 · area 312: surface-22-