1331487 玖、發明說明： 【發明所屬之技術領域】 本發明係指對輸送射頻功率之諧波含量的監視，尤指一種對 輸送射頻功率之諧波含量的監視系統與方法。 a ' 【先前技術】 電漿蝕刻及沉積製程在過去20年已成為半導體製造上所用 的主要圖案轉印方式。大部份以電漿為基礎的製程均運用到施加 射頻（RF)功率錢進給氣體分離的基本原理。如同所有的電漿負 載，電漿負載的主要特性之-即為非直線性型。這種負載的非直 線性型會產生普遍㈣波失真，以致料輪送射頻功率的電壓與 電流正弦波H 皮頻率振幅以及電流諧波相對於對應電壓諸波 之相關相位角表示的正確諧波失真量，是產生它們之㈣所特有 者。為求更精密，各種《參數，包括離子與電子密度和能量、 碰撞頻率、中性成分、以及它們各自的密度全都以獨特方式對功 率輸送源施加之基頻的特定諧波成分發揮偏，據以達成意欲的 分離和後續製程的結果。 因此顯射知，若對功率輸送源施加之基頻的譜波成分予以 監視’即可增進電槳沉積與姓刻製程的製程控Μ。結果，現已開 發出幾種在設計上能監視射頻諧波含量而增進製程控制的： 品。遺撼的是’ s為現行技術㈣項基本關，以致這技術尚未 普及。 現行技術最大的限制之—與產品㈣構有關。現行產品通常 都裝有-健f是位在測量點的傳感^封裝件，和-個通常位在 這測量點遠喊的對應分析、控似賴封裝件。由於各傳感器 封裝件均提供-獨特的輸出，所以這二封裝件要經過特別校準後 始可彼此配合。因此，不可能在單獨更換任—封裝件的情況下， 1331487 不對系統重新校準。由於半導體加工線的停機時間十主 以對於該等射頻感測式製程控制解決方案的使用者來說， 在維護及支財面造成重大的問題。 雖然已知有幾種可供監視輸送射頻功率之譜波含 置，但各«置均需精密校準個別的組件。現行解決方案中設^ 成用於電漿製程控制的原始硬體均是以下列任—項為基礎 射頻開關路由式（RF switch routed)帶通濾、波器；b)單向輕合器玨 或0以可程式局部振盪器賦能的外差或數位訊號處理器電口：。， 無論如何，這設計都無法在不重新校準整個射頻感測器裝置（由 傳感器和分析/通訊封裝件二者組成）的情況下更換任—组 裝件。 因此，本技藝需要能支援現場更換策略的方法和裝置，以便 在不減損賊㈣需重難準的情況下，使任—傳Μ封裝件可 跟任一對應的分析及通訊封裝件妥善發揮功能。 用於監視輸送射頻功率諧波含量之現行裝置的另一問題跟 自偏壓電壓有關。電絲刻反應器的射頻功率通常是被輸送給一 個電容性輕合電極。該電容性搞合會使直流電流停止朝著會:電 極表面產生「自偏壓」電壓之射頻功率輸送網路的方向流動。這 種「自偏壓電n終為負性，可使電㈣子加速朝著基板表面 方向流動，以致提供聚合製程揮發所急需的活化能量，所以對蝕 刻製程有助益。 鑒於系統自偏壓電壓具有發揮助益的效果，因而對其以及輸 送f頻功率的各部份宜瞭解。然而，本技藝中雖曾有人提出幾種 電壓取樣方案，以便監視輸送射頻功率的諧波含量，但迄今卻無 -種方案可供監視自偏壓電壓。揭示一種共用電容性耦合射頻電 壓探針的美國第5,867,020號專利（頒給Mo〇re等人）便是個例 子。是以，本技藝需要一種除了監視輸送射頻功率之諧波含量 外’也能監視自偏壓電壓的方法。 用於監視輸送射頻功率諧波含量之現行裝置的再一問題跟 感應式傳感器的屏蔽有關。由於半導體加工所用之許多氣體的壓 力-流量制度與分子量穩定性，往往需要較較高的射頻電壓來引 動及維持製程的電漿。此外，電漿類似二極體的特性能使射頻電 流在引發後變得極高。射頻功率通常是被輸送到一個電容性耦合 電極，於該處阻擋直流電流的流動，以致造成直流「自偏壓」電 壓。結果，不需直流耦合射頻電流傳感器來監視輸送射頻功率的 電流部份。然而，對於依照法拉第（Faraday)定律運作的純感應式 監視裝置來說，卻急需保護其免於接觸雜散場（磁場及電場二 者）’否則會對精確測量射頻電流所用的電勢造成重大影響。 臨界條件分析指出，接地屏蔽必需位在感應性傳感器與射頻 電流載體之間，以便妥善屏蔽感應性傳感器使其不接觸一次射頻 電流載體所放射出的電場，以致避免在電壓與電流之間發生串擾 情形。此外，為使感應式傳感器受屏蔽而不接觸測量所屬特定之 周圍環境中（例如阻抗匹配網路的線圈）可能存有的電場和磁 場，感應性傳感器應圍封在接地屏蔽内。遺憾的是，用於 感器不受周圍雜散場之影響的習用屏蔽也會妨礙意欲—次射頻 電流磁場的測量。 本技藝中曾有人提出種種的屏蔽設計。然而，卻無一種設計 能克服前述的缺點。舉例來說，美國帛5，8〇8,4i 5號（頒給H〇pkins 者）及第6,〇61，0〇6號（頒、給Hopkins者）專利曾教示一種用於龄視 射頻電流的雙迴路天線法。美國第6,5〇i，285 (頒給H〇pkin=人 者）號專利也曾教示-種使用與金屬填充孔穴互連之個別印刷電 路板來組裝感應H，，以便在各層之間提供連接的方法。美國第 1331487 5,834,931號（頒給Moore等人者）專利亦曾教示一種對法拉第定 律的單匝第一原理（single turn fisrt principles)實施法，但遺憾的 是’其卻受限於一次射頻電流載體與感應式迴路屏蔽之間的發弧 傾向。 與射頻功率源之諧波含量監視相關的另一問題跟端點檢測 有關。過去20年來’化學蒸鍍（CVD)及電漿增強型化學蒸鍍 (PECVD)業已成為半導體製造的重要部份^ cVD和pecVD製程 一般疋用於在低溫下澱積介質薄膜，據以作為防钱消耗層或在金 屬層之間作為介質分隔。 與CVD和PECVD二者相關連的一項非加值，但卻重要的 製程步驟是對隔室及相關連的組件進行電漿基清潔，以便去除澱 積製程之後所留下的殘餘薄膜。在澱積製程進行期間，薄膜是被 有計劃地澱積在半導體基板上。等半導體基板從隔室取出後，便 對隔室進行清潔，此舉是澱積製程成功與否的重要事項，但實際 上部非半導體裝置之製造的—部份。隔室清潔的—般方式就是使 澱積薄膜接受電漿基揮發。 大部份以電聚為基礎的製程均運用到施加射頻（RF)功率而 使進給氣體分離的基本原理。它雖不具加值效果，但在減少隔室 清潔時間方面卻甚重要。另外，也有文件記載長時間的清潔可能 會減損隔室組件的品質，因而產生良率有限的粒子。所以，為減 低製造成本同時增進製程步驟的良率，必須瞭解何時該停止清潔 過程。使清潔過程停止的正確時刻稱為端點。 射頻端點檢測Μ輸送射頻功率各部份的監視為基礎。當薄 膜脫離隔室組件時，揮發薄膜的副產品就會在電漿裡發生容積減 低的情形。這種電漿組件的容積變化會產生 網路見到的阻抗變化，因而間接導致射頻電壓、電流、相位= 1331487 自偏壓電壓跟著變化。監視這些訊號中的變化，就可獲得射端點 的正綠判定。值得注意的是，由於訊號分析演算法是個補償因 數’所以各批作業之間的薄膜型式、薄膜厚度、或圓案密度不必 一致，就可讓檢測器妥善發揮功能。 現已有種種裝置被設計成能在半導體加工處理時監視輸送 射頻功率的各部份。該等裝置在，例如，美國第5,770,992號（頒 給 Waters 者），第 5,565,737 號（頒給 Keane 者），第 6,046,594 號（頒 給 Mavretic 者）’第 5,808,415 號（頒給 Hopkins 者），和第 6,061，006 號（頒給Hopkins者）專利中均曾討論過。所有這些裝置都依賴當 作諸波分析之鑑頻檢測電路輸入訊號的輸送射頻功率之交流搞 合電壓與電流測量。然而，該組態對能用於分析寬頻諧波失真射 頻訊號的檢測器電路卻有所限制。再者，這些裝置均需界面電子 裝置先處理取樣的訊號’始可用於任一後續的應用中。另外，各 該裝置的組態係被設成讓傳感器封裝件與相關連的分析或界面 電子裝置封裝件一起校準，以致無法在整體性能不失效或不減損 的情況下分開。是以，本技藝需要一種能克服這些缺點的輸送射 頻功率諧波含量監視裝置。 與射頻功率供應相關的再一問題是跟射頻功率輸送網路的 診斷有關。半導體製造設施的造價及運轉費用十分昂貴。因此， 業界無不全力设法將製造工具的停機時間減至最低程度，而離線 工具的維護及恢復始終受到過多時間的限制。往往在工具因為不 符性能規格而離線取下時，修復工作都因缺乏診斷而受影響。結 果，該等修復工作經常變得十分昂貴。 第十四圖所示者係該系統的一典型組態。系統〗〇〇包括一射 頻產生器101，一阻抗匹配網路13〇,和一負載15〇。產生器1〇〇 經由一已知阻抗120而被耦合到阻抗匹配網路丨3〇。這阻抗通常 疋個公稱特性值，例# 50歐姆。阻抗120可促使產纟器到匹配 網路的功率傳送達到最佳程度。匹配網路130與負載150之間的 阻抗140 —般不詳，且會隨著時間而異。 八β大部份射頻功率產生器都有「内建」輸出測量能力，但它通 吊疋叹在阻抗匹配網路的遠距處。在阻抗匹配網路輸入處測量輸 入功率，以往均是利用測輻射熱計，量熱器，二極體和其它各種 儀表來提供。在同軸環境中測量射頻功率之習用方法的範例可在 美國第4,547,728號（頒給Mecklenburg者），第4,263,653號（頒給1331487 发明, DESCRIPTION OF THE INVENTION: FIELD OF THE INVENTION The present invention relates to the monitoring of the harmonic content of the transmitted radio frequency power, and more particularly to a monitoring system and method for the harmonic content of the transmitted radio frequency power. a ' [Prior Art] Plasma etching and deposition processes have become the main pattern transfer method used in semiconductor manufacturing for the past 20 years. Most of the plasma-based processes are applied to the basic principles of applying radio frequency (RF) power to feed gas separation. Like all plasma loads, the main characteristic of the plasma load is the non-linear type. The non-linear type of this load produces a common (four) wave distortion, so that the voltage and current sinusoidal H-frequency amplitude of the ferroelectric power and the correct harmonics of the current harmonics are related to the phase angles of the corresponding voltages. The amount of distortion is unique to them (4). For the sake of precision, various parameters, including ion and electron density and energy, collision frequency, neutral composition, and their respective densities, all uniquely bias the specific harmonic components of the fundamental frequency applied to the power delivery source. To achieve the desired separation and subsequent process results. Therefore, it is known that if the spectral wave component of the fundamental frequency applied by the power transmission source is monitored, the process control of the electric paddle deposition and the surname process can be improved. As a result, several products have been developed that are designed to monitor RF harmonic content and improve process control. The testament is that 's is the basic technology of the current technology (4), so that this technology has not yet been popularized. The biggest limitation of the current technology - related to the product (four) structure. Current products are usually equipped with -f, which is the sensing package at the measuring point, and a corresponding analysis, control-like package that is usually located at this measuring point. Since each sensor package provides a unique output, the two packages are specifically calibrated to match each other. Therefore, it is not possible to recalibrate the system without the replacement of the package-package. Due to the downtime of semiconductor processing lines, the users of these RF sensing process control solutions pose significant problems in maintenance and financial aspects. Although several spectral wave configurations are known to monitor the delivery of RF power, each component requires precise calibration of individual components. In the current solution, the original hardware used for plasma process control is based on the following items: RF switch routed band pass filter, wave device; b) one-way light combiner玨Or 0 heterodyne or digital signal processor electrical port enabled with a programmable local oscillator: In any case, this design cannot replace any component without recalibrating the entire RF sensor device (consisting of both the sensor and the analysis/communication package). Therefore, the present technology requires a method and apparatus that can support the on-site replacement strategy so that the Ren-Tang package can be properly functioned with any corresponding analysis and communication package without detracting from the thief (4). . Another problem with current devices for monitoring the harmonic content of the delivered RF power is related to the self-bias voltage. The RF power of the wire-wound reactor is typically delivered to a capacitive light-sense electrode. This capacitive engagement causes the DC current to stop moving in the direction of the RF power delivery network where the "self-biased" voltage is generated on the surface of the electrode. This self-biasing voltage n is negative, which allows the electric (four) sub-acceleration to flow toward the surface of the substrate, so as to provide the activation energy that is urgently needed for the evaporation of the polymerization process, so it is beneficial to the etching process. The voltage has a beneficial effect, so it is necessary to understand the various parts of the f-frequency power. However, several voltage sampling schemes have been proposed in the art to monitor the harmonic content of the transmitted RF power, but so far However, there is no such scheme for monitoring the self-bias voltage. A US Patent No. 5,867,020 (issued to Mo〇re et al.), which is a shared capacitively coupled RF voltage probe, is an example. The method of monitoring the harmonic content of the transmitted RF power can also monitor the self-bias voltage. Another problem associated with current devices for monitoring the harmonic content of RF power is related to the shielding of inductive sensors. Gas pressure-flow regimes and molecular weight stability often require higher RF voltages to motivate and maintain the plasma of the process. The characteristics of the plasma-like diode can make the RF current become extremely high after the initiation. The RF power is usually sent to a capacitive coupling electrode, where the DC current is blocked, resulting in a DC self-bias. "Voltage. As a result, no DC-coupled RF current sensor is required to monitor the current portion of the RF power delivered. However, for purely inductive monitoring devices operating in accordance with Faraday's law, there is an urgent need to protect them from contact with stray fields (both magnetic fields and electric fields), which would otherwise have a significant impact on the potential used to accurately measure RF currents. The critical condition analysis indicates that the grounding shield must be located between the inductive sensor and the RF current carrier to properly shield the inductive sensor from the electric field emitted by the primary RF current carrier, so as to avoid crosstalk between voltage and current. situation. In addition, inductive sensors should be enclosed within a grounded shield in order to shield the inductive sensor from exposure to electric fields and magnetic fields that may exist in the particular ambient environment (e.g., the impedance matching network coil). Unfortunately, conventional shielding for the sensor to be unaffected by surrounding stray fields can also hamper the intended measurement of the secondary RF current magnetic field. Various shielding designs have been proposed in the art. However, none of the designs overcome the aforementioned shortcomings. For example, US 帛5,8〇8,4i 5 (issued to H〇pkins) and 6, 〇61,0〇6 (issued to Hopkins) patents have taught a kind of radio frequency for ageing The dual loop antenna method of current. U.S. Patent No. 6,5, i, 285 (issued to H〇pkin = Person) has also taught the use of individual printed circuit boards interconnected with metal-filled holes to assemble the sensing H to provide between layers. The method of connection. US Patent No. 1,331,487, 5,834,931 (issued to Moore et al.) has also taught a single turn fisrt principles implementation of Faraday's law, but unfortunately it is limited to a radio frequency current carrier. The tendency to arc between the inductive loop shield. Another problem associated with the monitoring of harmonic content of RF power sources is related to endpoint detection. In the past 20 years, 'Chemical vapor deposition (CVD) and plasma enhanced chemical vapor deposition (PECVD) have become an important part of semiconductor manufacturing. ^ cVD and pecVD processes are generally used to deposit dielectric films at low temperatures. The money is consumed or separated as a medium between the metal layers. A non-valued, but important, process step associated with both CVD and PECVD is the plasma-based cleaning of the compartments and associated components to remove residual film left after the deposition process. During the deposition process, the film is deposited on the semiconductor substrate in a planned manner. After the semiconductor substrate is taken out of the compartment, the compartment is cleaned, which is an important matter of the success of the deposition process, but the actual upper non-semiconductor device is manufactured. The general way to clean the compartment is to allow the deposited film to undergo evaporation of the plasma. Most electropolymer-based processes use the basic principle of applying radio frequency (RF) power to separate the feed gas. Although it does not have a value-added effect, it is important in reducing the cleaning time of the compartment. In addition, there are documents that long-term cleaning may detract from the quality of the compartment components, resulting in particles of limited yield. Therefore, in order to reduce manufacturing costs while increasing the yield of the process steps, it is necessary to know when to stop the cleaning process. The correct moment to stop the cleaning process is called the endpoint. The RF endpoint detection is based on monitoring the various parts of the RF power delivered. When the film is removed from the compartment assembly, the by-product of the volatilized film is reduced in volume in the plasma. This change in volume of the plasma assembly produces a change in impedance seen by the network, which indirectly causes the RF voltage, current, and phase = 1331487 self-bias voltage to change. By monitoring the changes in these signals, a positive green decision is made at the endpoint. It is worth noting that since the signal analysis algorithm is a compensation factor, the film type, film thickness, or round file density between batches of operations does not have to be consistent, so that the detector can function properly. Various devices have been designed to monitor portions of the RF power delivered during semiconductor processing. Such devices are, for example, U.S. Patent No. 5,770,992 (issued to Waters), No. 5,565,737 (to Keane), No. 6,046,594 (to Mavretic), No. 5,808,415 (to Hopkins), and No. 6,061,006 (issued to Hopkins) has been discussed in the patent. All of these devices rely on the AC and RF measurements of the RF power delivered as input to the frequency detection circuit of the wave analysis. However, this configuration has limitations on detector circuits that can be used to analyze broadband harmonic distortion RF signals. Moreover, these devices require the interface electronics to process the sampled signals first for use in any subsequent application. In addition, the configuration of each device is configured to align the sensor package with associated analysis or interface electronics packages such that the overall performance does not fail or be degraded. Therefore, the present technology requires a transmission frequency power harmonic content monitoring device that overcomes these disadvantages. A further problem associated with RF power supply is related to the diagnosis of the RF power delivery network. The cost and operation of semiconductor manufacturing facilities is very expensive. As a result, the industry has tried to minimize the downtime of manufacturing tools, and the maintenance and recovery of offline tools has been limited by excessive time. Repairs are often affected by a lack of diagnostics when the tool is taken offline because it does not meet performance specifications. As a result, these repairs often become very expensive. The figure shown in Figure 14 is a typical configuration of the system. The system 〇〇 includes an RF generator 101, an impedance matching network 13A, and a load 15〇. The generator 1 is coupled to the impedance matching network 经由3〇 via a known impedance 120. This impedance is usually a nominal characteristic value, for example #50 ohms. Impedance 120 promotes optimal power transfer from the sputum to the matching network. The impedance 140 between the matching network 130 and the load 150 is generally unknown and will vary over time. Most of the beta beta RF power generators have "built-in" output measurement capability, but it hangs over the distance of the impedance matching network. Measuring input power at the impedance matching network input has historically been provided using bolometers, calorimeters, diodes, and various other instruments. An example of a conventional method of measuring RF power in a coaxial environment is available in U.S. Patent No. 4,547,728 (issued to Mecklenburg), No. 4,263,653 (issued to
MeCkIenburg 者）’和第 4,080,566 號（頒給 Mecklenburg 者）專利中 找到，所有這些方法全都依賴一種感生線圈設計來對射頻電壓取 樣。然而，由於執行的測量通常屬於診斷性，且只在維護與故障 排除期間始需要’因而安裝的成本、獲利性及則度均甚受囑目。 用於測量阻抗匹配網路輸出處之功率的習用方法通常依賴 對輸送至負冑之射頻功率的交流(AC)耦合電壓與電流測量。這些 測量會被輸入到鑑頻電路以供執行諧波分析。用於監視半導體加 工處理之射率功率輸送組件的習用方法範例在許多專利中，包 括例如美國第5,770,992號（頒給Waters者），第5,565,737 號（頒給Keane者），第6,046,594號（頒給Mavretic者），第 5,8〇8’415號（頒給Hopkins者），和第6 〇61〇〇6號（頒給出沐⑹ 者）專利中均有所說明。這些系統也包括一傳感器封裝件和相關 連的分析或界面電子裝置封裝件，但因它們是被一起校準，所以 具有别述的缺點（亦即它們無法在不減損整體性能的情況下予以 分開）。 傳統的射頻功測量技術在輸送網路特性阻抗部份或非特性 阻抗方面提供了解決方案，但卻無法整合這兩種測量裝置。雖曾 有人試圖整合昂貴也難以安裝的鑑頻射頻感測器，但卻因價格及 1331487 安裝因素而令人難以接受。是以，本技藝需要—種可讓現場工程 人員迅速、輕易、低成本與精確診斷射頻功率輪送網路各部份， 以及接定系統之故障部份的方式。 採用本文所揭示及如後所述的裝置及方法，就可符合前述需 求。 【發明内容】 本發明一方面是提供一種用於監視輸送到一射頻功率裝置 之射頻訊號職含量的系統。該系統包括⑷—可對射頻訊號電 壓取樣並輸出-個表示該電壓之第—訊號的電壓傳感器，⑸一 可對射頻訊號電流取樣並輸出—個表示該電流之第二訊號的電 流傳感器’和⑹—可與電壓傳感器和電流傳感器其中至少一 個，但宜與二者均可通訊的記憶體裝置，該記憶體裝置裝有該至 個，但且為該二傳感器的專用校準資訊。由於校準資訊係專 用性（locally)儲存，所以可在不減損性能以及不需重新校準的 情況下，於現場單獨更換傳感器封裝件或分析及通訊封裝件，無 需顧及另一封裝件。 本發明的另一方面是提供用於監視射頻功率源諧波含量之 電壓部份的方法與裝置。不同於交流搞合式，因而需避免測量自 偏麼電壓的現行電壓取樣方案，本文所揭示的方法及裝置屬於利 用交流耦合對射頻電壓取樣者。所以，可同時和在同一測量點來 測呈射頻電壓與自偏壓電壓。射頻電壓波形與自偏壓電壓的寬頻 樣本結合對製程控制提供了更完整的數據集。 本發明的再一方面是提供一種用於保護依照法拉第定律運 作之感應式監視裝置的方法與裝置，使其在不妨礙意欲一次射頻 電流磁場的測量了免於接觸雜散場（磁場及電場二者），否則會對 精確測量射頻電流所用的電勢造成重大影響。依這方面所製成的 1331487 裝置宜包括一射頻電流傳感器和一用於容置該傳感器且由一金 屬頂壁及金屬側壁構成的機殼。該機殼的構造宜設成在其置於一 平面基板上時，側壁斜離頂壁和朝向基板為原則。另外，側壁宜 - 與頂壁隔開，且側壁可使傳感器與周圍電場或磁場隔絕。可由第 - 一及第二端壁予以支承且宜接地的頂壁比側壁高。該裝置可跟射 頻電流載體組合使用，以便讓這裝置設在射頻電流載體的磁場近 接處，頂壁則可防止因為該射頻電流載體之電場干擾而發生的串 擾情形。在該等實施例中，側壁的設置方式可為以其不致將射頻 載體相關連之磁場過度衰減為原則。 本發明的又一方面是提供一種非鑑頻射頻澱積室清潔端點 籲 檢測器。該檢測器符合澱積室清潔端點檢測器的價格、安裝、現 場維護性、及功能性等限制。由於射頻檢測器未運用鑑用，所以 各式各樣的檢測器電路均可用來分析寬頻諧波失真射頻訊號。這 些包括但不限於峰值檢測器，平均值檢測器和真實rms檢測器。 此外，射頻檢測器與其它習用射頻感測器的不同處在於它不需要 界面電子裝置’而鑑頻單元卻需要以便先處理取樣訊號，始可用 於任一後續應用。 本發明的另一方面是提供一種在阻抗匹配網路輸入及輸出鲁 處測量射頻功率的方法及裝置。因而可確定和㈣間監視該網路 的效率才康以查明其健康狀況。是以’在一較佳實施例中，提供 一種可在經由-已知阻抗環境而麵合到一產±器之阻抗網路輸 入處測里功率Pi的方法。該方法包括下列各步驟：⑷將—個 與阻抗網路輸入處之電壓成比例的訊號耦合到一電壓檢測器，據 · 、產生RMS電壓訊號（Vrms)，和（b)處理RMS電壓訊號據以 確定阻抗網路輸人處的功率。使用簡單公式pi = v2rms/(Zc)即 可確定輸入功率，其中Zc表示的阻抗特性已知。 12 的方=發Θ的再—方面是提供—種測量阻抗網路輸出處之功率 及雷〜方法包括下列各步驟：（a)確定該輸出處的電壓（V) 確定測^電壓及電流的相位差φ，和⑹從這 計算出°。功率pQ。尤其’功率可從下列公式P。= VIcoscp 本發明的又一方面是接供— 網路效率量;使用p4po之測定值來確定 :…度ε的方法：ε = ρ〇/Ρι。於是，阻抗匹配網路之效率 、平估即提供可用以查明網路健康狀況的診斷資訊。Found in the patents of MeCkIenburg) and No. 4,080,566 (issued to Mecklenburg), all of these methods rely on an inductive coil design to sample the RF voltage. However, since the measurements performed are generally diagnostic and only need to be maintained during the maintenance and troubleshooting period, the cost, profitability and degree of installation are highly appreciated. Conventional methods for measuring the power at the output of an impedance matching network typically rely on alternating current (AC) coupled voltage and current measurements of the RF power delivered to the negative. These measurements are input to the frequency discrimination circuit for performing harmonic analysis. Examples of conventional methods for monitoring the rate power delivery assembly for semiconductor processing are in many patents, including, for example, U.S. Patent No. 5,770,992 (issued to Waters), No. 5,565,737 (issued to Keane), No. 6,046,594 (issued to Mavretic), No. 5,8〇8'415 (for Hopkins), and No. 6〇61〇〇6 (issued to Mu (6)) are described in the patent. These systems also include a sensor package and associated analysis or interface electronics package, but because they are calibrated together, they have the disadvantages described (ie they cannot be separated without detracting from overall performance) . Traditional RF power measurement techniques provide a solution for delivering characteristic impedance or non-characteristic impedance of the network, but cannot integrate the two measurement devices. Although there have been attempts to integrate expensive RF radios that are difficult to install, they are unacceptable due to price and 1331487 installation factors. Therefore, this skill requires a way for field engineers to quickly and easily, inexpensively and accurately diagnose various parts of the RF power transfer network and to locate faulty parts of the system. The foregoing needs are met by the apparatus and method disclosed herein and as described hereinafter. SUMMARY OF THE INVENTION One aspect of the present invention is to provide a system for monitoring the level of radio frequency signals delivered to an RF power device. The system comprises (4) - a voltage sensor capable of sampling the RF signal voltage and outputting - a signal indicating the voltage - (5) a current sensor capable of sampling the RF signal current and outputting a second signal indicating the current 'and (6) - A memory device that can communicate with at least one of the voltage sensor and the current sensor, but preferably both, and the memory device is provided with the specific calibration information for the two sensors. Because the calibration information is stored locally, the sensor package or analysis and communication package can be replaced in the field without degrading performance and without recalibration, regardless of the other package. Another aspect of the present invention is to provide a method and apparatus for monitoring the voltage portion of the harmonic content of a radio frequency power source. Different from AC, it is necessary to avoid the current voltage sampling scheme for measuring the self-bias voltage. The method and device disclosed in this paper belong to those who use AC coupling to sample RF voltage. Therefore, the RF voltage and the self-bias voltage can be measured simultaneously and at the same measurement point. The combination of RF voltage waveforms and wide-band samples of self-bias voltage provides a more complete data set for process control. A further aspect of the present invention is to provide a method and apparatus for protecting an inductive monitoring device operating in accordance with Faraday's law so as not to interfere with the measurement of a primary RF current magnetic field from contact with stray fields (both magnetic fields and electric fields) ), otherwise it will have a significant impact on the potential used to accurately measure the RF current. The 1331487 device made in this respect preferably includes an RF current sensor and a casing for accommodating the sensor and consisting of a metal top wall and a metal sidewall. The housing is preferably constructed such that when it is placed on a planar substrate, the side walls are inclined away from the top wall and toward the substrate. In addition, the side walls should be spaced apart from the top wall and the side walls can isolate the sensor from the surrounding electric or magnetic field. The top wall that can be supported by the first and second end walls and is preferably grounded is taller than the side walls. The device can be used in combination with an RF current carrier to place the device in close proximity to the magnetic field of the RF current carrier, and the top wall prevents crosstalk due to electric field interference from the RF current carrier. In such embodiments, the sidewalls may be arranged in such a way that they do not excessively attenuate the magnetic field associated with the radio frequency carrier. Yet another aspect of the present invention is to provide a non-frequency-frequency radio frequency deposition chamber cleaning endpoint detector. The detector meets the price, installation, field maintainability, and functionality limitations of the deposition chamber cleaning endpoint detector. Since the RF detector is not used for identification, a wide variety of detector circuits can be used to analyze the broadband harmonic distortion RF signal. These include, but are not limited to, peak detectors, average detectors, and true rms detectors. In addition, the difference between a radio frequency detector and other conventional radio frequency sensors is that it does not require an interface electronics' and the frequency discrimination unit needs to process the sampled signals first, and can be used in any subsequent application. Another aspect of the present invention is to provide a method and apparatus for measuring RF power at an input and output of an impedance matching network. It is therefore possible to determine and (iv) monitor the efficiency of the network to determine its health. In a preferred embodiment, a method is provided for measuring the power Pi at an impedance network input that is integrated into a device via a known impedance environment. The method includes the following steps: (4) coupling a signal proportional to the voltage at the input of the impedance network to a voltage detector, generating an RMS voltage signal (Vrms), and (b) processing the RMS voltage signal. Determine the power at the impedance network input. The input power is determined using the simple formula pi = v2rms/(Zc), where the impedance characteristics represented by Zc are known. The side of the ============================================================================================== The phase differences φ, and (6) calculate ° from this. Power pQ. In particular, the power can be derived from the following formula P. = VIcoscp A further aspect of the invention is the supply-network efficiency; the method of determining the degree ε using the measured value of p4po: ε = ρ〇/Ρι. Thus, the efficiency and flattening of the impedance matching network provides diagnostic information that can be used to ascertain the health of the network.
依據本發明的另—方面，係提供—種能將第—射頻感測器所 、供確疋阻抗匹配網路輸入處之射頻功率的輸入功率值傳 、/遠距位置，以便觀察、記錄、或進一步處理。此舉宜以一 種採用TCP/IP的_BUS (數據機匯流排）標準協議或其它適當 協議（例如高速序列）予以達成。同樣地，由第二射頻感測 »算出以供綠定阻抗匹配網路輪出處之射頻功率的輸出功率 值亦，傳達到該遠距位置，以便觀察、記錄、或進一步處理。此 舉也且以-種採用TCP/IP的腦BUS標準協議來達成。接著，便 可在這遠距離位置算出效率。According to another aspect of the present invention, an input power value and/or a distance position of a radio frequency power of a first radio frequency sensor for correcting an impedance matching network input are provided for observation, recording, and Or further processing. This should be done in a _BUS (Data Machine Bus) standard protocol using TCP/IP or other suitable protocol (such as a high speed sequence). Similarly, the output power value calculated by the second radio frequency sensing » for the RF power at the green impedance matching network turn is also communicated to the remote location for observation, recording, or further processing. This is also achieved by a brain BUS standard protocol using TCP/IP. Then, the efficiency can be calculated at this long distance position.
依據本發明的再一方面，係揭示一種經由乙太網路供電 jpower over術而對第一和/或第二感測器供電的方 法。。在這方法的-些實施例中’可將獨糾IIM立址儲存在各感 測器專用的記憶體内，和傳送到遠距觀察位置，以供辨識所觀察 之特定測量的位置。 【實施方式】 各圖式所例舉說明者係本發明的一些較佳實施例。各圖式中 相同的參照號碼則指相同和對應的部件。 A.製程控制用射頻感測器 13 1331487 參閱第一圖的方塊圖’其中顯示出一射頻感測器使用在依本 文教示而製成之射頻控制裝置中的情形。在這系統10中，有個 屬於射頻產生器形式的電源11 (射頻源）係利用傳輸線14而經 由一匹配網路12被耦合到一處理反應器13。反應器13可以是 用於處理包括半導體晶圓在内之各種樣材料的任一反應器，例如 電漿反應器。此外，網熟本技藝者均知有各種處理系統可使用電 或微波能（包括射頻）源，且其中任一種或該等系統的種種組合均 可用於實施本文所述的教示。再者，雖然較宜使用匹配網路12， 但在本文所述的各種感應器應用中未必全然如此。In accordance with still another aspect of the present invention, a method of powering a first and/or second sensor via an Ethernet powered jpower over technique is disclosed. . In some embodiments of the method, the uniquely corrected IIM address can be stored in a memory dedicated to each sensor and transmitted to a remote viewing position for identifying the particular measured position observed. [Embodiment] Each of the drawings is illustrative of some preferred embodiments of the present invention. The same reference numbers in the drawings refer to the same and corresponding parts. A. RF Sensor for Process Control 13 1331487 Referring to the block diagram of the first figure, there is shown a case where a radio frequency sensor is used in an RF control device made in accordance with the teachings herein. In this system 10, a power source 11 (radio frequency source) in the form of a radio frequency generator is coupled to a processing reactor 13 via a matching network 12 using a transmission line 14. Reactor 13 can be any reactor for processing various materials, including semiconductor wafers, such as a plasma reactor. In addition, it is well known to those skilled in the art that various processing systems may utilize electrical or microwave energy (including radio frequency) sources, and any one or combination of such systems may be used to implement the teachings described herein. Again, while it is preferred to use the matching network 12, this is not necessarily the case in the various sensor applications described herein.
如第一圖所示，有個傳感器封裝件15是在一個與反應器13 近接的位置串聯插入傳輸線14 (通常為同軸者），且其宜位在匹 配網路12後方的某―點。傳感器封裝件15宜儘量靠近反應器 13’以便從傳感器封裝件15獲得的測量可指明流進反應器^的 實際v和I值。v和I值二者實質係'在傳輸線14的同—點予以 感測，以便確定流進反應器13的功率，和在某些情況下，還可 確定V和I之間的相位關係。 在傳感器封裝件15中裝入適當的寬頻電壓傳感器16及電洛As shown in the first figure, a sensor package 15 is inserted in series with the transmission line 14 (usually coaxial) in a position proximate to the reactor 13, and is preferably located at a point behind the matching network 12. The sensor package 15 should preferably be as close as possible to the reactor 13' so that measurements taken from the sensor package 15 can indicate the actual v and I values flowing into the reactor. Both the v and I values are substantially sensed at the same point of the transmission line 14 to determine the power flowing into the reactor 13, and in some cases, the phase relationship between V and I. Mounting the appropriate broadband voltage sensor 16 and the pedicle in the sensor package 15
^器Π。該等傳感器係被設計成能分別對輸送之射頻功率合 2壓及電流部餘樣。此傳感器封裝件另設有—高速類比對數七 轉換器（ADC) 63，-數位訊號處理器（Dsp) 65，和一（宜 ===置…傳感器封裝件15是被保持在-_ 射頻傳輸線14專用的測量位置。 第-圖所不的组態另包括一個位在傳感器封裝件1 的分析及通訊封裝件69。傳感器封裝件15盥 ^ 封裝件69之間的通訊_船）㈣4遠距分析及通郭 立葉變換（FFT)、*寅篡^ 匕 达，证用以妥善運算快速傅 葉變換㈣U算法所需的調定（set_up)指令，和接收卿所 14 丄331487 發送的FFT結果。 · 設於傳感器封裝件内的記憶器67是項對現有傳感器封裝件 ， 的顯著改良。這記憶體裝置儲存傳感器封裝件專用的必要校準資 . 訊以供系統存取，另可儲存其它資訊，例如用於追蹤的序號和其 它必要資訊。由於各傳感器的校準資訊係備存於這傳感器封裝件 内，所以在必須更換傳感器封裝件或相關連的通訊及分析封裝件 時，不必重新校準系統。因此，本文所揭示的傳感器封裝件，在 必須更換傳感器封裝件或相關連的通訊及分析封裝件時，解決了 RF感應器裝置所常見的有關保持校準的現場維護問題。現建議 重編第一圖的號碼以便配合第二圖。 參 參閱第二及三圖，所示者係傳感器封裝件15的一可行實施 例。第二圖係傳感器封裝件15各組件的分解圖，而第三圖則是 組裝圖》有個機殼20用以容置傳感器封裝件15的各組件。在這 實施例中，機殼20是個矩形箱盒，一端設有一輸入口，另一端 則設有輸出口。位在機殼20内的一中間導體19為傳輸線14 (參 閱第一圖）的中間（或内）導體提供傳導介質。是以，這中間導體 19變成主傳輸線η在機殼20輸入和輸出端之間的從動部件。 設於機殼各端的輸入/輸出（1/0)連接器21被安裝到各自所屬的 鲁 端板22，各該端板22又被分別安裝到機殼2〇的對應端。輸入/ 輸出連接器21的一端耦合到傳輸線14,另一端則穿入端板22 的一開口而耦合到中間導體19。有個傳導套管24用以使連接器 21的插頭端配接到開口孔徑比這插頭大的中間導體19。另在各 端分別有個具一中間孔口的絕緣墊片23，據以在各連接器21與 端板22被安裝到機殼20上時，將中間導體19支承在機殼2〇内 的中間定位。另外，採用螺絲、螺栓或其它緊固裝置將連接器 21女裝到端板22上，和將端板22安裝到機殼20上。 15 1331487 電壓傳感器16和電流傳感器17實質上係被安裝在機殼2〇 兩端的中間，但彼此卻設在機殼20的對向兩側。如第二及三圖 所示電壓傳感器16和電流傳感器17二者係被安裝在機殼2〇 上，各具有所屬的蓋板28或29以便將電壓傳感器16和電流傳 感器Π總成安裝到機殼20内。電壓傳感器16和電流傳感器17 各有所屬的連接器53或42，其穿過各自的蓋板28或29以便提 供對外連接’如同輸入/輸出連接器21。 第五圖所示者係電流傳感器17 一可行實施例的詳情。這實 7例中的電流傳感Μ有型拾取線圈4Q (或線圏迴路），一 盍板29，一端接器41，和一連接器42。拾取線圈4〇是種同轴 傳輪線，具有-個由一傳導回路（⑽ducting加则）圍繞住的 中間導體。雖然可用配置編織回路的典型同轴電纜，但這實施例 的線圈迴路40卻用-種圍住内導體的實心金屬外殼。將這外殼 的終端耦合到蓋板29’另一端則耦合到蓋板29和連接器42。線 圈40内導體43的-端被輕合到連接器42的内導體，另一端則 耦合到端接器4卜端接器41是個電路組件(例如電阻器），以便 ^線路40末端提供匹配的終端阻抗。這終端阻抗應與線圈利的 特性阻抗(例如5〇歐姆)匹配，且其係應用到 接::2之編封裝件15所給與的阻抗。是以，線^ =兩&係以其特性阻抗終結。線圈4〇的外殼在口型的頂 接^以切Γ (或裂縫）44，以便在這頂點完全斷開外殼的電連 竟的間時’這切°44便位在線圈40的中點。要注 .疋〜、有外殼被切開，線圈4Q的内導體並未如此。 具=妥電^傳感器17並將其置於機殼4〇内時，如第三圖所示， :Sr套 體19的開口，另將u型線圈4〇的頂 16 1331487 二(ΓΓ::44)插入絕緣套管45㈡。絕緣套管45便使線圈外 双（處於大地電位）跟導體電絕緣。切口44位在絕緣套管45内。 緣套管45宜以佩獅品牌的含氟聚合物或其它能提供音欲 介電常數的材料製成，以便在過渡區提供阻抗匹配。 u 在某些實施例中，可不將線圏4〇完全插入電流載體Η的令 曰，屆時便可不用絕緣套管45。就該等實施例而言，法 加咖）定律可能未予完全實施，因而需適當的校準。缺而， 2方式卻有能解決發弧電勢的優點，而那也正是最初會用、到絕 緣套管45的主因。 蓋板29係被安裝到機殼2〇上，並構成傳輸線14回程脖 的一部份。是以，線圈44的外殼_兩端接地，但因存有切口 44:所以不會形成—連續接地路徑。於是，蓋板29就被安裝到 機殼20上。 第六圖所示者係電壓傳感器16的—可竹施例的詳情。該 電壓傳感器包括-⑼合到__同軸線52内導體51的扁平傳導板 50。同軸線52的另一端則耦合到一個安裝在蓋板28之上的連接 器53。第-圖所示的傳感器封裝件15被麵合到連接器犯，據以 測量電壓_16所感測的電壓。安裝妥時，同軸㈣的阻抗 將跟輕合到連接器53的外線相同，而傳感器封裝件15 (第二圖） 也將會阻抗匹配。因此，阻抗從傳感器封裝件15到導體Μ尖端 路均匹配，並於該尖端處與扁平傳導板5〇配對（通常經由焊 接）〇 把盖板28安裝到機殼2〇上，藉以將電壓傳感器組裝到機殼 内時，扁平傳導板5G即位在中間導體19的近接處，但卻不會碰 觸中間導體19,也不會碰觸外導體18。如同電流傳感器17，扁 平傳導板5G約位在導體19的中間點。電壓傳感器16在導體Β 17 1331487 間 上且位在電流傳感器1 γ的對向側。電流及電壓的感測宜在中 19的同一線性位置取得，以便能獲得正確的功率量度（ρ 、相對於嫻熟本技藝者已知的現有裝置與方法，前述裝置及方 法的獨特特點之一就是可將高頻類比訊號轉換成適於測量點專 用之處理的數位協議β若與儲有必要校準資訊的專用儲存器結合 時’此特點便可提供一種校準的測量值數位輸出。 2.射頻感測器電壓傳感器 第七及八圖分別是依本文教示所製作之一種直流耦合射頻 感測器電壓傳感器1 0 1實施例的側視及平面圖。此電壓傳感器 101與第二、三、四和六圖之電壓傳感器16的不同處在於它是 直流麵合，因而可測量直流自偏壓的電壓。如圖所示，這傳感器 包括個與主射頻電流載體（未顯示）直接接觸的錢銅合金彈簧 103 »組態則設成可對一扁平或圓柱狀電流載體保持壓入配合接 觸為原則。直流耦合射頻電壓傳感器1〇1係由第一高壓無感電阻 器107和第二高壓無感電阻器1〇9以及一可使射頻與直流自偏壓 電壓部份分離的偏壓Τ形裝置111所組成。高壓射頻與高壓自偏 壓電壓二者係按照（R2/(R1+R2))的比予以取樣。黏合墊提供對表 面黏著裝置之射頻電壓（VRF) 115,直流電壓（VDC) 117和接地 (GND) 119所用輸出的連接。 直流耦合射頻電壓傳感器101可被包封在一適當的介電盒 (未顯示）’使其不與周圍環境接觸，同時把負載阻抗所受的影響 減至最低，否則就會被射頻功率輸送網路改變特性。如同前述， 電壓傳感器亦可與適當的頻率相關分析電子裝置麵合，以便提供 諧波含量分析’另可跟相對測量所用的檢測電路，以及適當組態 的射頻電流傳感器耦合。 1331487 3.半導體處理用射頻檢測器 第九圖所示者係依本文教示所製成之一種射頻檢測器401 (未顯示）可行實施例的功能架構。射頻檢測器4〇1被耦合到射頻 功率輸送線403上的射頻功率輸送網路405電漿射頻負載407之 間。適當的直流耦合電壓傳感器409與交流耦合電流傳感器411 可供對使用點的射頻電壓與電流訊號取樣。接著視意欲的測量正 確性度而定，可用種種方式來檢測該等樣本。 射頻檢測器401裝有一個可將電壓傳感器409所輸出之訊號 分成直流部份與射頻部份的偏壓T形裝置413。該訊號的直流部 份被輸入到一個直流偏壓檢測器415,再由其輸出直流電壓（VDC) 參數。該訊號的射頻部份則被輸入到一個第一雙向〇。分路器 417。該第一雙向〇。分路器417與一可供輸出射頻電壓（vrf)參數 的射頻電壓檢測器419和一相位檢測器421相通。射頻檢測器 401另裝有一個第二雙向〇。分路器423。該第二雙向〇。分路器423 則與一可供輸出射頻電流（IRF)參數的射頻電流檢測器425和相 位檢測器421相通。根據來自第一雙向〇。分路器417及第二雙向 〇°分路器423的輸入，相位檢測器421輸出射頻電流相對於射頻 電壓之相位角（PHRF)的參數。 對於並無精確規格的一般應用’或對於不想提供作用功率的 應用而$，該等檢測器可以屬於峰值或平均值性質。但對於注重 精確性且能提供作用功率的應用來說，該等檢測器則可以屬於 「真實有效值（true RMS)」的型式。 去除鑒頻能力’就可將端點檢測器的成本及實際尺寸減至最 低，因而易於安裝該裝置和擴大其應用。由於電壓及電流波形的 諧波失真，以峰值或平均值檢測器所輸出的訊號通常只是具有相 對精確性者。然而，如提供在任一分析軟體中所用的校準係數， 19 丄^1487 就能使真實RMS檢測器變得精確。 β美國第5’576,629號（頒給T〇mpkins等人）和第5 939 886 號（頒給Turner等人）專利曾揭示用以控制輸送射頻功率各部份 的方法與系統，並說明它們可用於端點檢測。然而，不同於這些 參考:貝料所述的方法’本文所述的方法並未控制輸送射頻功率的 各邻伤，只疋檢測該等部份，其後可將對應的訊號提供給諸如美 國第5’ 576’ 629號（頒給T〇mpkins等人）和第5, 939, 886號（頒給 Turner等人）專利所述者之類的系統從事實際控制。 4.射頻感測器電流傳感器 第十至十二圖所示者係依本文教示製成的射頻感測器之纟 · 面黏著式傳感器線圈303所用的一機殼總成3〇1。該機殼總成3〇1 是由一頂壁305，第一側壁307和第二側壁309 (參閱第十一及 十二圖），以及第一端壁311和第二端壁313 (參閱第十和十一 圖）組成。端壁的末端是分別利用接觸墊321、323而黏到基板 318上的凸緣315、316。同樣的，側壁3〇7、3〇9也分別經由接 觸墊325、327而被黏著到基板318上。 為求清晰，側壁在第十圖中被省略以便易於瞭解表面黏著式 傳感器線圈303在機殼總成301的安置情形。傳感器線圈3〇3是 鲁 依照法拉第（Faraday)定律運作，並經由第一 317和第二319接 觸墊而被黏著到基板318上。 宜為金屬的頂壁305，因為電場干擾，所以可防止設在傳感 器線圈303磁場近接處的射頻電流載體（未顯示）發生串擾情 形。側壁307，309也宜為金屬。側壁的主要用意是讓傳感器線 圈303跟周圍環境中可能存有的雜散電或磁場隔絕，否則可能會 因感生誤差而導致測量品質的減損。 第十一圖係機殼總成301的俯視圖。如圖所示，該總成在基 20 1331487 板318上係設成讓第一 307與第二3〇9側壁的取向跟基板318保 持一角度，並使側壁斜離頂壁305 ^側壁3〇7, 3〇9取向與基板 318保持的角度，是以側壁不會讓一次射頻電流载體（未顯示）的 磁場過度衰減作為選擇原則。另外，如第十二圖所示，側壁3〇7， 309的咼度且與頂壁305不同。側壁307，309與基板318保持 角度，以及降低側壁307, 309高度的效果，就是可對一次射頻 電流載體重要的磁場線創造一個像是漏斗的東西。 5.射頻功率輸送診斷系統 第十二圖所示者即為依本文教示製成之一種射頻功率輸送 診斷系統的較佳實施例。如圖所示，該系統2〇1包括一射頻產生 器203，一阻抗匹配網路205，和一電漿射頻負載2〇7。電漿射 頻負載207可跟種種射頻從動裝置的任一種對應，包括，例如， 電漿姓刻反應器。 所设的第一射頻感測器2 0 9係用以在阻抗匹配網路的輸入 處測量射頻功率，第二射頻感測器211則用以在阻抗匹配網路的 輸出處測量射頻功率。從感測器209和211所測得的量度則可傳 達到一遠距位置。 射頻功率產生器203係經由一條具有能展現出阻抗ze之阻 抗環境的電源線213而被耦合到阻抗匹配網路205。通常，Ze的 值約為50歐姆，但對本技藝略有涉獵者均瞭解經常會遇到能展 現出其它阻抗的環境’因而本文所揭示的系統並不以任何特定的 阻抗值為限。 第一射頻感測器209内設有一分壓網路，其係由二個分別具 有阻抗Z1和Z2的阻抗部件215與217構成。此分壓網路可對一 RMS射頻電壓檢測器221給與一個與阻抗匹配網路205輪入處之 射頻電壓成比例的電壓。分壓網路的阻抗部件可用電阻性和/或 21 1331487 電容性分量實現，因而去除習用技藝以電感性分量實現時所遇到 的困難。 電壓從分壓網路輸入到RMS射頻電壓檢測器221後，RMS射 頻電壓檢測器221即回應而產生一個表示阻抗匹配網路輸入 處之RMS電壓的輸出類比訊號。抑5射頻電壓檢測器221可用市 面販售的積體電路予以實現。 有個處理器228嵌置於第一射頻感測器2〇9内。該處理器 228接收射頻電壓檢測器221的輸出’再利用—類比對數位轉換 器（ADC) 223將其轉換成一數位訊號。處理單元225可用能達成 本文所述之必要功㈣微處理器或其它數位裝置實現。處理 單元225亦被麵合到一含有校準係數的板載記憶體單元挪。該 等校準隸是用來改正因偏離公稱值之分量偏差所造成的誤 差。舉例來說，若以電阻器作為分壓網路之阻抗zi^z2的阻抗 部件=5與217’那麼這些電阻器通常會其公稱錢微偏離一點。 舉例來說’校準係數的選擇可採取能改正該等誤 項式解決方案作為原則； < y = c + bx + ax2 (等式 1) 其中X是從檢測器221發出之RMS訊號的輸人數位表示a， b和c为別是記憶體單元227發出 跪訊號Vrras。接著，依據f的^係數，而y則是校準的 處的m射頻功ΐ 心來物且抗匹配網請輸入 =V2rras/(Zc) (等式 2) 犯的已知2=1生器2〇3輕合到阻抗匹配網㈣之電源線 接著，可將測定輸入功率傳達 傳達到-輸出功率感㈣，以遠距位置’歧為或額外 觀察、g錄、傳輸或進一步處理。 22 1331487 =遠距位置和/或輸出功率感測器與輸人功率感測器之間的通訊 宜以-種採用標準TCP/IP協議的M_s (數據機匯流排）229 予以實現。另外，同—通訊埠可採用經由乙太網路供電（power overEthernet)的技術對一輸入功率感測器供電，以便可在這感 測器上設立單一的數據及功率接達點。 記，體單元227亦可儲存一個用以識別帛-射頻感測器209 特疋測1位置的IP位址。因此，舉例來說，在一個設有若干位 於不同位置之感測H的網路中，便提供了連絡各個位置的通訊， 以致選擇用以執行功率測量的位置和識別所接收之功率量度 的位置。 在阻抗網路205的輸出處，係以一第二射頻感測器211來測 量功率。網路205輸出處的電壓係被直流耦合到電壓傳感器 241，而該傳感器在一較佳實施例中，實質具有分壓器的作用， 以便將感測的電壓降壓。電壓傳感器241的輸出被饋送到一個偏 壓T形裝置243 ’由其將接收電壓的直流及交流部份分開。直流 部份由直流偏壓檢測器245予以檢測，交流部份則被饋送到一個 第一雙向功率分路器247。該第一功率分路器247的一個支路係 對一射頻電壓檢測器259饋送’另一輸出支路則對一相位檢測器 251饋送。 。阻抗匹配網路205輸出處的電流係被交流耦合到電流傳感 器253，其輸出則被饋送給一個第二雙向功率分路器255。該第 一又向功率为路器255的一個支路係對一射頻電流檢測器257饋 送’另一輸出支路則對相位檢測器251饋送。 相位檢測器251用以測量從第一功率分路器247所接收之電 壓訊號與從第二功率分路器255所接收之電流訊號二者間的相 位差。射頻電壓檢測器259提供一個表示阻抗網路2〇5輸出處之 23 1331487 RMS電壓值的訊號。直流偏壓檢測器245，相位檢測器251，射 頻電壓檢測器259 ’和射頻電流檢測器257的輸出則被輸入到一 個設在嵌置式CPU (中央處理器）230内的類比對數位轉換器 (ADC) 26卜再由其將該等訊號轉換成數位形式以供處理單元263 使用 處理單元263可用能達成本文所述之必要功能的微處理器 或其它數位裝置予以實現。處理單元263亦被耦合到一含有校準 係數的板載記憶體單元265。該等校準係數是用來改正因為，例 如’偏離公稱值之分量偏差所造成的誤差。 舉例來說，校準係數可供實現諸如前述用以改正該等誤差之 嫌 等式1的多項式解決方案，其中x是從射頻電壓檢測器257或射 頻電流檢測器259發出之RMS訊號的輸入數位表示，a，b和c 刀別是s己憶體單元265發出的某一檢測器之訊號的校準係數，而 y貝i疋校準的RMS輸出訊號Vrms或Irms。接著，從下列的算法 來確定阻抗匹配網路205輸出處的RMS射頻功率： P〇 = Vrms Irms coscp (等式 3) 其中c〇S(p是電壓與電流之間的相位餘弦。 、接著，可將測定輸出功率傳達到—遠距位置，或改為或額夕卜· 傳達到輸入功率感測器，以便觀察、記錄、傳輸或進一步處理。 該遠距位置和/或輸人功率感測器與輸出功率感測器之間的通訊 =以-種採用標準TCP/IP協議的_哪（數據機匯流排）聊 、實現3外’同-通訊埠可採用經由乙太網路供電（叫· Ethernet)的技術對-輸出功率感測器供電，以便可在第二 ' ’=測g 211上3又立單—的數據及功率接達點。記憶體單元 211 ^個用以朗位置，因而可識別第二射頻感測器 特定測量位置的1ρ位址。是以，舉例來說，在-個設有 24 置之感測器的網路中，便提供了連絡各個位置的 3 b、擇用以執行功率測量的位置和識別所接收之功率 重度的位置。 =輸出功率…已知時，可經由時間算出和觀測 效率值ε⑴’以便追縱阻抗匹配網路2G5的健康情形。效率值 =遠距位置算出，或可在第—射頻感測器簡或第二射頻感測 器211的處理單元内算出。 綱熟本技藝者可從前述内容瞭解此射頻功率輸送珍斷系統 對阻抗匹配網路輸人處的功率和阻抗匹配網路輸出處的功率二 者均能測量，錢能制整個射頻功率輸送，包括阻抗匹配 肩路的6乡斷特性化。此外，本發明也能保持各個測量地點的特定 校準資訊，因而免除習用方法所需的鑑頻電路和諧波分析。 6.各種組合與次組合 本文所述的裝置和方法可彼此或與本技藝已知的組件形成 種種不同的組合與次組合，據以達成各式各樣有用的裝置。舉例 來說，本技藝已知的各種頻率相關的射頻感測器便可跟所述的射 頻電壓與射頻電流傳感器任—個或二者組合，和/或與所述的功 率輸送診斷系統及射頻感測器任一個或二者組合。尤其，本技藝 已知的各種頻率相關的射頻感測器可跟所述的射頻電壓與射頻 電流傳感器二者組合。3夕卜，本技藝已知的各種頻率相關的射頻 感測器可跟所述的射頻電壓與射頻電流傳感器二者組合，再與所 述的功率輸送診斷系統及射頻感測器組合。 同樣地，本文所述的非頻率相關的射頻檢測器可跟所述的射 頻電壓與射頻電流傳感器任一個或二者組合。舉例來說，本文所 述的非頻率相關的射頻檢測器可跟所述的射頻電壓與射頻電流 傳感器二者組合。該等組合又可跟所述的射頻感測器和功率輸送 25 1331487 診斷系統任一個或二者組合。例如，太 本文所述的非頻率相關的射 頻檢測器可跟所述的射頻電流傳感器與射頻電壓傳感器二者⑯ ， 合，再與所述的射頻感測器及功率輪送診斷系統組合。 · 另外’本文所述的射頻感測器可跟所述的射頻頻率與射頻電 壓傳感器任一個或二者組合。舉例來說，本文所述的射頻感測器 可跟所述的射頻頻率與射頻電壓傳感器二者組合。 *以上所舉實施例僅用以說明本發明而已，非用以限制本發明 之範圍。舉凡不違本發明精神所從事的種種修改或變化，俱屬本 發明申請專利範圍。 【圖式簡單說明】 鲁 第一圖所示者係依本發明構成之射頻檢測器的一功能方塊圖。 第二圖所示者係依本發明構成之射頻檢測器的一分解圖。 第三圖係第二圖所示射頻檢測器的一透視圖，其中該裝置的壁面 弄成透明以方便例舉說明。 第四圖係第二圖所示射頻檢測器的一側視圖，其中有部份呈剖 面。 第五圖係可用於本發明之裝置及方法中的一電流傳感器側視圖。 第六圖係可用於本發明之裝置及方法中的一電壓傳感器側視圖。鲁 第七圖所示者係依本發明構成之電壓傳感器的一側視圖。 第八圖所示者係依本發明構成之電壓傳感器的一俯視圖。 第九圖所示者係依本發明構成之端點傳感器的一功能圖。 第十圖所示者係依本發明構成之射頻感測器電流傳感器的一屏 蔽方案側視圖。 第十一圖所示者係依本發明構成之射頻感測器電流傳感器的— 屏蔽方案俯視圖。 第十二圖所示者係依本發明構成之射頻感測器電流傳感器的一 26 1331487 屏蔽方案側視圖。 第十三圖所示者係 一射頻感測器的圖 第十四圖所示者係 【符號說明】 用於測量阻抗匹配網路輸入處之射頻功率的 解圖。 習用射頻功率輸送系統的說明圖。 系統10 射頻源11 匹配網路12 反應器13 傳輸線14 傳感器封裝件15 電壓傳感器16 電流傳感器17 外導體18 中間導體19 南速類比對數位轉換器6 3 數位訊號處理器65 記憶體裝置67 分析及通訊封裝件69 機殼20 連接器21 端板22 絕緣墊片23 蓋板28、29 拾取線圈40 端接器41 連接器42 導體43 切口 44 絕緣套管45 同轴線5 2 内導體51 扁平傳導板50 連接器53 電壓傳感器101 鈹銅合金彈簧103 第一高壓無感電阻器1〇7 第二高壓無感電阻器1〇9 偏壓T形裝置in 射頻電壓115 直流電壓117 接地119 系統201 射頻產生器203 阻抗匹配網路205 電漿射頻負載207 第一射頻感測器209^器Π. The sensors are designed to separately supply the RF power and the current portion of the delivered RF power. The sensor package is additionally provided with a high speed analog logarithmic seven-converter (ADC) 63, a digital signal processor (Dsp) 65, and a (should ===...the sensor package 15 is held at -_ RF The measurement position dedicated to the transmission line 14. The configuration of the first diagram also includes an analysis and communication package 69 located in the sensor package 1. The sensor package 15盥^ the communication between the packages 69_ship) (4) 4 far Distance analysis and Tong Guo Lie transformation (FFT), *寅篡^ 匕达, proof of the set (up_up) instruction required to properly operate the fast Fourier transform (IV) U algorithm, and the FFT result sent by the receiver 14 丄 331487 . • The memory 67 provided in the sensor package is a significant improvement over existing sensor packages. This memory device stores the necessary calibration resources dedicated to the sensor package for system access and other information such as serial numbers for tracking and other necessary information. Since the calibration information for each sensor is stored in this sensor package, it is not necessary to recalibrate the system when the sensor package or associated communication and analysis package must be replaced. Therefore, the sensor package disclosed herein solves the field maintenance problem common to RF calibration devices that is common to RF calibration when the sensor package or associated communication and analysis package must be replaced. It is now recommended to rewrite the number in the first figure to match the second picture. Referring to Figures 2 and 3, a possible embodiment of the sensor package 15 is shown. The second diagram is an exploded view of the components of the sensor package 15, and the third diagram is an assembly diagram. A housing 20 is provided for housing the components of the sensor package 15. In this embodiment, the casing 20 is a rectangular box having an input port at one end and an output port at the other end. An intermediate conductor 19 located within the housing 20 provides a conductive medium for the intermediate (or inner) conductor of the transmission line 14 (see the first figure). Therefore, the intermediate conductor 19 becomes a driven member of the main transmission line η between the input and output ends of the casing 20. The input/output (1/0) connectors 21 provided at the respective ends of the casing are mounted to the respective end plates 22 to which they belong, and each of the end plates 22 is attached to the corresponding end of the casing 2, respectively. One end of the input/output connector 21 is coupled to the transmission line 14, and the other end is inserted into an opening of the end plate 22 to be coupled to the intermediate conductor 19. A conductive sleeve 24 is provided for adapting the plug end of the connector 21 to the intermediate conductor 19 having a larger opening aperture than the plug. Further, at each end, there is an insulating spacer 23 having an intermediate opening, so that the intermediate conductor 19 is supported in the casing 2 when the connectors 21 and the end plates 22 are mounted on the casing 20. Intermediate positioning. In addition, the connector 21 is attached to the end plate 22 by screws, bolts or other fastening means, and the end plate 22 is mounted to the casing 20. 15 1331487 The voltage sensor 16 and the current sensor 17 are substantially mounted in the middle of the two ends of the casing 2, but are disposed on opposite sides of the casing 20. Both voltage sensor 16 and current sensor 17 are mounted on housing 2, as shown in Figures 2 and 3, each having an associated cover 28 or 29 for mounting voltage sensor 16 and current sensor assembly to the machine. Inside the shell 20. The voltage sensor 16 and the current sensor 17 each have an associated connector 53 or 42 that passes through a respective cover 28 or 29 to provide an external connection 'like the input/output connector 21. The figure shown in the fifth figure is a detail of a possible embodiment of the current sensor 17. In the seventh example, the current sensing has a pick-up coil 4Q (or coil loop), a seesaw 29, an end connector 41, and a connector 42. The pickup coil 4 is a coaxial transmission line having an intermediate conductor surrounded by a conduction loop ((10) ducting plus). Although a typical coaxial cable of a braided loop can be used, the coil loop 40 of this embodiment surrounds the solid metal casing of the inner conductor. Coupling the terminal end of the housing to the cover 29' is coupled to the cover plate 29 and the connector 42. The end of the inner conductor 43 of the coil 40 is lightly coupled to the inner conductor of the connector 42, and the other end is coupled to the terminator 4. The terminator 41 is a circuit component (e.g., a resistor) to provide matching at the end of the line 40. Terminal impedance. This termination impedance should match the characteristic impedance of the coil (e.g., 5 ohms) and is applied to the impedance imparted by the package: Therefore, the line ^ = two & is terminated with its characteristic impedance. The outer casing of the coil 4 is crimped (or cracked) 44 in the shape of the mouth so that the cut 44 is at the midpoint of the coil 40 when the apex completely breaks the electrical connection of the casing. Note: 疋~, the outer casing is cut, and the inner conductor of the coil 4Q is not. When the sensor 17 is properly placed and placed in the casing 4, as shown in the third figure, the opening of the Sr sleeve 19, and the top 16 1331487 of the u-shaped coil 4 二 (ΓΓ:: 44) Insert the insulating sleeve 45 (2). The insulating sleeve 45 electrically insulates the outer (double earth potential) of the coil from the conductor. The slit 44 is located within the insulating sleeve 45. The edge sleeve 45 is preferably made of a lion polymer brand fluoropolymer or other material that provides an acoustical dielectric constant to provide impedance matching in the transition zone. u In some embodiments, the turns 4圏 may not be fully inserted into the current carrier, and the insulating sleeve 45 may not be used. For these embodiments, the law of Fagara may not be fully implemented and therefore requires proper calibration. In short, the 2 methods have the advantage of solving the arc potential, and that is the main reason for the initial use of the bushing 45. The cover plate 29 is attached to the casing 2 and forms part of the return line of the transmission line 14. Therefore, the outer casing _ of both ends of the coil 44 is grounded, but since the slit 44 is present, a continuous ground path is not formed. Thus, the cover plate 29 is mounted to the casing 20. The figure shown in the sixth figure is the details of the method of the voltage sensor 16. The voltage sensor includes - (9) a flat conductive plate 50 that is bonded to the inner conductor 51 of the coaxial line 52. The other end of the coaxial line 52 is coupled to a connector 53 mounted on the cover plate 28. The sensor package 15 shown in the figure is bonded to the connector to measure the voltage sensed by the voltage _16. When installed, the impedance of the coaxial (4) will be the same as the outer line that is lightly coupled to the connector 53, and the sensor package 15 (Fig. 2) will also be impedance matched. Thus, the impedance is matched from the sensor package 15 to the conductor turns tip and is mated with the flat conductive plate 5 at the tip (usually via soldering) to mount the cover 28 to the housing 2 to provide a voltage sensor When assembled into the casing, the flat conductive plate 5G is positioned at the vicinity of the intermediate conductor 19, but does not touch the intermediate conductor 19 nor touch the outer conductor 18. Like the current sensor 17, the flat conductive plate 5G is located at an intermediate point of the conductor 19. The voltage sensor 16 is between the conductors 13 17 1331487 and is located on the opposite side of the current sensor 1 γ. The sensing of current and voltage is preferably taken at the same linear position of the middle 19 so that the correct power measurement can be obtained (p, one of the unique features of the aforementioned apparatus and method relative to prior art devices and methods known to those skilled in the art) The high-frequency analog signal can be converted into a digital protocol suitable for measurement-point-specific processing. When combined with a dedicated memory that stores the necessary calibration information, this feature provides a calibrated digital output of the measured value. The seventh and eighth diagrams of the detector voltage sensor are respectively a side view and a plan view of an embodiment of a DC-coupled RF sensor voltage sensor 101 made in accordance with the teachings herein. The voltage sensor 101 and the second, third, fourth and sixth The voltage sensor 16 of the figure differs in that it is a DC face and thus can measure the voltage of the DC self-bias. As shown, the sensor includes a copper alloy spring that is in direct contact with the main RF current carrier (not shown). 103 »The configuration is designed to maintain a press-fit contact for a flat or cylindrical current carrier. The DC-coupled RF voltage sensor 1〇1 is A high voltage non-inductive resistor 107 and a second high voltage non-inductive resistor 1〇9 and a biasing device 111 for separating the RF and DC self-bias voltage portions. High voltage RF and high voltage self-bias voltage The two are sampled at a ratio of (R2/(R1+R2)). The bond pad provides a connection to the RF voltage (VRF) 115, DC voltage (VDC) 117 and ground (GND) 119 output of the surface mount device. The DC-coupled RF voltage sensor 101 can be enclosed in a suitable dielectric box (not shown) to keep it out of contact with the surrounding environment while minimizing the effects of load impedance, otherwise it will be used by the RF power transmission network. The circuit changes characteristics. As mentioned above, the voltage sensor can also be combined with appropriate frequency-dependent analysis electronics to provide harmonic content analysis. The detection circuit used for relative measurement and the appropriately configured RF current sensor can be coupled. 3. Radio Frequency Detector for Semiconductor Processing Figure 9 is a functional architecture of a possible embodiment of a radio frequency detector 401 (not shown) made in accordance with the teachings herein. 4〇1 is coupled between the RF power transfer network 405 and the RF RF load 407 on the RF power transfer line 403. Appropriate DC-coupled voltage sensor 409 and AC-coupled current sensor 411 are available for RF voltage and current at the point of use. The signal is sampled. The sample is then detected in various ways depending on the degree of accuracy of the intended measurement. The RF detector 401 is provided with a bias voltage for dividing the signal output by the voltage sensor 409 into a DC portion and a RF portion. The T-shaped device 413. The DC portion of the signal is input to a DC bias detector 415, which then outputs a DC voltage (VDC) parameter. The RF portion of the signal is input to a first bidirectional 〇. The first multiplexer 417 is coupled to a radio frequency voltage detector 419 and a phase detector 421 for outputting a radio frequency voltage (vrf) parameter. The RF detector 401 is additionally equipped with a second bidirectional cymbal. Splitter 423. The second two-way 〇. The splitter 423 is in communication with an RF current detector 425 and a phase detector 421 that are capable of outputting RF current (IRF) parameters. According to the first two-way 〇. The input of the splitter 417 and the second bidirectional 分° splitter 423, the phase detector 421 outputs a parameter of the phase angle (PHRF) of the radio frequency current with respect to the radio frequency voltage. For general applications that do not have precise specifications' or $ for applications that do not want to provide operating power, such detectors may be of peak or average nature. However, for applications that focus on accuracy and provide power, these detectors can be of the "true RMS" type. The ability to remove the frequency discriminating capability minimizes the cost and actual size of the endpoint detector, making it easy to install the device and expand its application. Due to the harmonic distortion of the voltage and current waveforms, the signal output by the peak or average detector is usually only relatively accurate. However, by providing the calibration coefficients used in any of the analysis software, 19 丄^1487 can make the real RMS detector accurate. U.S. Patent Nos. 5,576,629 (issued to T.mpkins et al.) and 5,939,886 (issued to Turner et al.) disclose methods and systems for controlling the delivery of various portions of RF power and indicate that they are available. Detection at the endpoint. However, unlike these references: the method described in the shellfish' method described in this paper does not control the neighboring injuries that transmit RF power, only the parts are detected, and the corresponding signals can be provided to the US, for example. Systems such as those described in 5' 576' 629 (to T〇mpkins et al.) and 5, 939, 886 (issued to Turner et al.) are subject to actual control. 4. Radio Frequency Sensor Current Sensors The tenth to twelfth figures are shown as a radio frequency sensor made in accordance with the teachings herein. A housing assembly 3 〇1 used for the surface-adhesive sensor coil 303. The housing assembly 3〇1 is composed of a top wall 305, a first side wall 307 and a second side wall 309 (see FIGS. 11 and 12), and a first end wall 311 and a second end wall 313 (see Ten and eleven figures) composition. The ends of the end walls are flanges 315, 316 that are adhered to the substrate 318 by contact pads 321, 323, respectively. Similarly, the side walls 3〇7, 3〇9 are also adhered to the substrate 318 via the contact pads 325, 327, respectively. For clarity, the side walls are omitted in the tenth view to facilitate an understanding of the placement of the surface mount sensor coil 303 in the housing assembly 301. The sensor coil 3〇3 operates in accordance with Faraday's law and is adhered to the substrate 318 via the first 317 and second 319 contact pads. Preferably, the top wall 305 of the metal prevents crosstalk of the RF current carrier (not shown) disposed at the proximity of the magnetic field of the sensor coil 303 due to electric field interference. The side walls 307, 309 are also preferably metal. The main purpose of the side wall is to isolate the sensor coil 303 from stray electric or magnetic fields that may be present in the surrounding environment, otherwise the measurement quality may be degraded due to induced errors. The eleventh figure is a plan view of the casing assembly 301. As shown, the assembly is mounted on the base 20 1331487 plate 318 such that the orientation of the first 307 and second 3 〇 9 sidewalls is at an angle to the substrate 318 and the sidewall is inclined away from the top wall 305 ^ sidewall 3 〇 The orientation of the 7, 3 〇 9 orientation with the substrate 318 is based on the principle that the sidewall does not excessively attenuate the magnetic field of the primary RF current carrier (not shown). Further, as shown in Fig. 12, the width of the side walls 3, 7, 309 is different from that of the top wall 305. The effect of the sidewalls 307, 309 being at an angle to the substrate 318 and reducing the height of the sidewalls 307, 309 is to create a funnel-like object for the magnetic field lines important to the primary RF current carrier. 5. RF Power Delivery Diagnostic System The figure shown in Fig. 12 is a preferred embodiment of a radio frequency power delivery diagnostic system made in accordance with the teachings herein. As shown, the system 2〇1 includes a radio frequency generator 203, an impedance matching network 205, and a plasma RF load 2〇7. The plasma RF load 207 can correspond to any of a variety of RF slave devices, including, for example, a plasma surname reactor. The first RF sensor is used to measure the RF power at the input of the impedance matching network, and the second RF sensor 211 is used to measure the RF power at the output of the impedance matching network. The metrics measured from sensors 209 and 211 can then be transmitted to a remote location. The RF power generator 203 is coupled to the impedance matching network 205 via a power line 213 having an impedance environment that exhibits an impedance ze. Typically, the value of Ze is about 50 ohms, but those skilled in the art are aware that environments that exhibit other impedances are often encountered. Thus the systems disclosed herein are not limited to any particular impedance value. The first RF sensor 209 is provided with a voltage dividing network which is composed of two impedance members 215 and 217 having impedances Z1 and Z2, respectively. The voltage divider network can provide an RMS RF voltage detector 221 with a voltage proportional to the RF voltage at the turn-on of the impedance matching network 205. The impedance components of the voltage divider network can be implemented with resistive and / or 21 1331487 capacitive components, thus eliminating the difficulties encountered by conventional techniques in achieving inductive components. After the voltage is input from the voltage divider network to the RMS RF voltage detector 221, the RMS RF voltage detector 221 responds to produce an output analog signal indicative of the RMS voltage at the impedance matching network input. The RF voltage detector 221 can be implemented by a commercially available integrated circuit. A processor 228 is embedded in the first RF sensor 2〇9. The processor 228 receives the output 'reuse' of the RF voltage detector 221 and converts it to a digital signal by an analog-to-digital converter (ADC) 223. Processing unit 225 can be implemented with a microprocessor or other digital device capable of achieving the necessary functions described herein. Processing unit 225 is also surfaced to an onboard memory unit containing calibration coefficients. These calibrations are used to correct errors caused by component deviations from the nominal value. For example, if the resistor is used as the impedance component of the voltage divider zi^z2 of the voltage divider network = 5 and 217', then these resistors usually deviate slightly from the nominal cost. For example, the selection of the calibration coefficients can be based on the principle that the corrective solution can be corrected; < y = c + bx + ax2 (Equation 1) where X is the input of the RMS signal from detector 221 The digits indicate that a, b, and c are the memory cells 227 that emit the signal Vrras. Then, according to the ^ coefficient of f, and y is the m-RF function of the calibration, and the anti-matching network, please input =V2rras/(Zc) (Equation 2). Known 2=12 〇3 is lightly coupled to the power line of the impedance matching network (4). Next, the measured input power can be conveyed to the sense of output power (4), and the remote location is 'discriminated or additionally observed, recorded, transmitted or further processed. 22 1331487 = Communication between the remote position and / or output power sensor and the input power sensor should be implemented in M_s (Data Machine Bus) 229 using the standard TCP/IP protocol. In addition, the same communication can be used to power an input power sensor via a power over Ethernet technology so that a single data and power access point can be set up on the sensor. The body unit 227 can also store an IP address for identifying the location of the 帛-RF sensor 209. Thus, for example, in a network having a plurality of sensing Hs located at different locations, communication is provided to contact the various locations such that the location for performing power measurements and the location identifying the received power metrics are selected. . At the output of impedance network 205, a second RF sensor 211 is used to measure power. The voltage at the output of network 205 is DC coupled to voltage sensor 241, which in a preferred embodiment substantially functions as a voltage divider to step down the sensed voltage. The output of voltage sensor 241 is fed to a biased T-shaped device 243' which separates the DC and AC portions of the received voltage. The DC portion is detected by the DC bias detector 245 and the AC portion is fed to a first bidirectional power splitter 247. One branch of the first power splitter 247 feeds a radio frequency voltage detector 259 and the other output branch feeds a phase detector 251. . The current at the output of impedance matching network 205 is AC coupled to current sensor 253, the output of which is fed to a second bidirectional power splitter 255. The first and the other branch of the power path 255 feeds an RF current detector 257, and the other output branch feeds the phase detector 251. The phase detector 251 is for measuring the phase difference between the voltage signal received from the first power splitter 247 and the current signal received from the second power splitter 255. The RF voltage detector 259 provides a signal indicative of the 23 1331487 RMS voltage at the output of the impedance network 2〇5. The outputs of the DC bias detector 245, the phase detector 251, the RF voltage detector 259' and the RF current detector 257 are input to an analog-to-digital converter provided in the embedded CPU (Central Processing Unit) 230 ( The ADCs are then converted by these signals into digital form for processing unit 263 using processing unit 263 to be implemented with a microprocessor or other digital device capable of achieving the necessary functions described herein. Processing unit 263 is also coupled to an onboard memory unit 265 that contains calibration coefficients. These calibration coefficients are used to correct errors caused by, for example, deviations from the component deviation from the nominal value. For example, the calibration coefficients may be implemented to implement a polynomial solution such as the aforementioned Equation 1 to correct the errors, where x is the input digit representation of the RMS signal from the RF voltage detector 257 or the RF current detector 259. , a, b, and c are the calibration coefficients of the signals of a certain detector issued by the unit 265, and the RMS output signals Vrms or Irms are calibrated. Next, the RMS RF power at the output of the impedance matching network 205 is determined from the following algorithm: P 〇 = Vrms Irms coscp (Equation 3) where c 〇 S (p is the phase cosine between voltage and current. The measured output power can be communicated to a remote location, or converted to an input power sensor for observation, recording, transmission or further processing. The remote location and/or input power sensing Communication between the device and the output power sensor = _ which uses the standard TCP / IP protocol _ which (data machine bus) chat, to achieve 3 external 'same-communication 埠 can be powered via Ethernet (called · Ethernet) technology - the output power sensor is powered so that the data and power access points can be set up on the second ''=g g211'. The memory unit 211^ is used for the Lang position. Therefore, the 1ρ address of the specific measurement position of the second RF sensor can be identified. For example, in a network with 24 sensors, 3b is provided for contacting each location. Selecting the location to perform the power measurement and identifying the received power severity = Output power... When known, the efficiency value ε(1)' can be calculated and observed via time to track the health of the impedance matching network 2G5. Efficiency value = distance position calculation, or can be used in the first - RF sensor Or the processing unit of the second RF sensor 211 is calculated. The skilled artisan can understand the power of the RF power transmission and the power of the impedance matching network at the output of the impedance matching network. Both can measure, the money can make the whole RF power transmission, including the impedance matching shoulder path. In addition, the invention can also maintain the specific calibration information of each measurement location, thus eliminating the frequency discrimination required by the conventional method. Circuit and Harmonic Analysis 6. Various Combinations and Sub-Combinations The devices and methods described herein can be formed in various combinations and sub-combinations with one another or with components known in the art to achieve a wide variety of useful devices. In general, various frequency dependent radio frequency sensors known in the art can be combined with the RF voltage and RF current sensor, or both, and/or The power delivery diagnostic system and the radio frequency sensor are either or both. In particular, various frequency dependent radio frequency sensors known in the art can be combined with both the radio frequency voltage and the radio frequency current sensor. Various frequency dependent radio frequency sensors known in the art can be combined with the RF voltage and RF current sensors described above, and combined with the power delivery diagnostic system and the RF sensor. The non-frequency dependent radio frequency detector can be combined with either or both of the RF voltage and RF current sensors. For example, the non-frequency dependent radio frequency detector described herein can be associated with the RF voltage. The RF current sensor is combined. These combinations can in turn be combined with either or both of the RF sensor and power delivery 25 1331487 diagnostic system. For example, the non-frequency dependent radio frequency detector described herein can be combined with the RF current sensor and RF voltage sensor described above, in combination with the RF sensor and power wheel diagnostic system. • Further, the RF sensor described herein can be combined with either or both of the RF frequency and RF voltage sensors. For example, the radio frequency sensors described herein can be combined with both the RF frequency and RF voltage sensors described. The above embodiments are merely illustrative of the invention and are not intended to limit the scope of the invention. All modifications or variations made without departing from the spirit of the invention are within the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS The first figure shows a functional block diagram of a radio frequency detector constructed in accordance with the present invention. The second figure shows an exploded view of a radio frequency detector constructed in accordance with the present invention. The third figure is a perspective view of the RF detector shown in the second figure, wherein the wall of the device is made transparent for ease of illustration. The fourth figure is a side view of the RF detector shown in the second figure, some of which are partially cut away. The fifth figure is a side view of a current sensor that can be used in the apparatus and method of the present invention. The sixth drawing is a side view of a voltage sensor that can be used in the apparatus and method of the present invention. Lu. Figure 7 is a side view of a voltage sensor constructed in accordance with the present invention. The figure shown in the eighth figure is a top view of a voltage sensor constructed in accordance with the present invention. The figure shown in the ninth figure is a functional diagram of the end point sensor constructed in accordance with the present invention. The figure shown in the tenth is a side view of a shielding scheme of the radio frequency sensor current sensor constructed in accordance with the present invention. The eleventh figure shows a top view of the shielding scheme of the RF sensor current sensor constructed in accordance with the present invention. Figure 12 is a side view of a 26 1331487 shielding scheme for a radio frequency sensor current sensor constructed in accordance with the present invention. Figure 13 is a diagram of a radio frequency sensor Figure 14 Figure [Symbol Description] Used to measure the RF power at the input of the impedance matching network. An illustration of a conventional RF power delivery system. System 10 RF source 11 Matching network 12 Reactor 13 Transmission line 14 Sensor package 15 Voltage sensor 16 Current sensor 17 Outer conductor 18 Intermediate conductor 19 South speed analog-to-digital converter 6 3 Digital signal processor 65 Memory device 67 Analysis and Communication package 69 Enclosure 20 Connector 21 End plate 22 Insulation pad 23 Cover plate 28, 29 Pickup coil 40 Terminator 41 Connector 42 Conductor 43 Notch 44 Insulation sleeve 45 Coaxial axis 5 2 Inner conductor 51 Flat conduction Plate 50 Connector 53 Voltage Sensor 101 Copper Alloy Spring 103 First High Voltage Non-Inductive Resistor 1〇7 Second High Voltage Non-Inductive Resistor 1〇9 Bias T-In Device RF Voltage 115 DC Voltage 117 Ground 119 System 201 RF Generator 203 impedance matching network 205 plasma RF load 207 first RF sensor 209
27 1331487 第二射頻感測器211 電源線213 阻抗部件215、217 RMS射頻電壓檢測器221 類比對數位轉換器223 處理單元225 板載記憶體單元227 處理器228 數據機匯流排229 嵌置式CPU230 電壓傳感器241 偏壓T形裝置243 直流偏壓檢測器245 第一雙向功率分路器247 相位檢測器251 電流傳感器253 第二雙向功率分路器255 射頻電流檢測器257 射頻電壓檢測器259 類比對數位轉換器261 處理單元263 板載記憶體單元265 數據機匯流排267 機殼總成301 表面黏著式傳感器線圈303 頂壁305 第一側壁307 第二側壁309 第一端壁311 第二端壁313 凸緣 315、316 基板318 第一接觸墊317 第二接觸墊319 接觸墊 321、323、325、327 射頻檢測器401 射頻功率輸送線403 射頻功率輸送網路405 電漿射頻負載407 電壓傳感器409 電流傳感器411 偏壓T形裝置413 直流偏壓檢測器415 第一雙向0°分路器417 射頻電壓檢測器419 相位檢測器421 第二雙向〇°分路器423 射頻電流檢測器425 2827 1331487 Second RF Sensor 211 Power Line 213 Impedance Component 215, 217 RMS RF Voltage Detector 221 Analog Digital Converter 223 Processing Unit 225 Onboard Memory Unit 227 Processor 228 Data Engine Bus 229 Embedded CPU230 Voltage Sensor 241 Bias T-shaped device 243 DC bias detector 245 First bidirectional power splitter 247 Phase detector 251 Current sensor 253 Second bidirectional power splitter 255 RF current detector 257 RF voltage detector 259 Analog to digital Converter 261 Processing Unit 263 Onboard Memory Unit 265 Data Machine Bus 267 Chassis Assembly 301 Surface Adhesive Sensor Coil 303 Top Wall 305 First Side Wall 307 Second Side Wall 309 First End Wall 311 Second End Wall 313 Convex Edge 315, 316 substrate 318 first contact pad 317 second contact pad 319 contact pad 321, 323, 325, 327 RF detector 401 RF power transfer line 403 RF power transfer network 405 Plasma RF load 407 Voltage sensor 409 Current sensor 411 bias T-shaped device 413 DC bias detector 415 First bidirectional 0° splitter 417 RF voltage detector 419 Phase detector 421 Second Bidirectional 〇° splitter 423 RF current detector 425 28