201221973 六、發明說明： 【發明所屬之技術領域】 本發明係有關於-種萃取材料 法，尤指一種具有可降低介電常數萃取二t裝置及其方 介電常數萃取可用頻寬者。’以及提高 【先前技術】 在某些習知電容法萃取介電常 積之電容器作為驗㈣且，嗲、叙都以固疋面 载/、透過1測電容器電容值，並盘 積做運算取得材料之介電常數，但由於常用的運 i以常數較材料介電常數高。習知電容法介 及：疋藉由量測所得之電容器電容值與面積， 反推材料之介電常數，而實際 、 二感與本體等效電容所組成，如圖一所示， ⑨感的存在使得電容器在頻率提升時，電容值也 曰加’因此造成使用電容法萃取介電常數時，因寄生 較所萃取之介電常數誤差隨著操錢率提升而 跟者k升，讓電容萃取法之可用頻寬降低，而電容器之自 ，頻率正疋判別其寄生電感高低的一項指標。圖二為某種 習知介電常數萃取裝置之立體透視圖，請參考圖二，由於 該結構在量_需要L線12,因此該饋人線12之 =生效應會使得電容器自振頻率降低，成為造成習知電容 萃取介電常數方法之可用頻寬縮減的主要原因。 201221973 在傳統電容法介電當叙w 厂— yArea— “數卒取方式I則是藉由公式 之計算取得材料之介電常數值乃尺，其 中，％為空氣之介電常數 、 ，~為兩電極中之較直技咖咖料的厚度值 兩電極平板之間的電容，作在=面積。由於電容值〇為 電容值還包含電容器邊緣;量測時’所量測之 電容會造成介電常數萃取電容，因此該雜散 高於材料本身之介電常數。5、差’使所萃取之介電常數 因此，如何研發出一種萃 方法’其可較某些習知電 ’I電治數的哀置及其 量’並可提高本方法的 ？取方式具有更低的誤差 討之處。 τ用頻見，將是本發明所欲積極探 【發明内容】 其主種二材料介電常數的裝置及其方法， 量，並有寬方式具有更低的誤差 包含’其 -第-電極’其覆設於該待測頂：3及-：測材料’· 其覆設於該待測材料之底面，及^二電極， 】=形二;面係被該第-電二=二r: 第-介電當數萃取單元，其包含:該待d; 201221973 一電極，其覆設於該待測材料之頂 覆設於該待剩材料之底面，並形成及—第四電極，其 極所形成之覆面係被該第三電極弟-電容’該第四電 中’該第二電極的邊長與該第四電極的=覆面涵蓋’其 之面積不H中，根據該第 的=相同’但形成 電容，值，以推算得該待測材::介電電：值及該第二 其包含下列㈣·· 取H f常數的方法， 提供一第一介電常數萃取 料；-第-電極，豆覆嗖沖主”包含：一待測材 該待測材料之底面，並形成-第f 成之覆面涵蓋；1所开'成之覆面係被該第一電極所形 提供一第二介電常數萃取 料；一第三電極，A覆机於^主丨八包^ :該待測材 四雷亥待測材料之頂面；一第 底面’並形成-第= 成之覆面涵蓋，直中，，第第三電極所形 極的邊長相同，但^^積電^的邊長與該第四電 量測該第一電容的電容值； 量測該第二電容的電容值；以及 根據該第一電容的電容值以及 值，以推算得該待測材料之介電常數值。*的電容 【貫施方式】 201221973 一 為充分瞭解本發明之特徵及功效，茲藉由下述具體之 實施例，並配合所附之圖式，對本發明做一詳細說明广 明如後： 况 圖三A及圖三b為本發明之一具體實施例的立體透視 圖，請同時參考圖三A及圖三b，本發明為一種萃取材料 介電常數㈣置，其包含有：―第-介電常數萃取單元卜 其包含：一待測材料2(例如：片狀材料、基板、薄膜、紙 張型材料或由粉末及複合材料所壓合而成之平面狀材料， 但不限於此）；—第—電極3，其覆設於該待測材料2之頂 面；及一第二電極4,其覆設於該待測材料2之底面，並 ;成一第一電容，該第二電極4所形成之覆面係被該第一 極3所形成之覆面涵蓋；以及—第二介電常數萃取單元 贿^含：該待測材料2…第三電極6，其覆設於該待 測材料2之頂面；及—笛 ^ ^ 第四電極7，其覆設於該待測材料2 底面’並形成一第二雷玄，兮结 .§亥第四電極7所形成之覆面 係被该第三電極6所形成 4的邊具愈〜 A之復面涵盍，其中，該第二電極 4的邊長與該第四電極7的喜 ,，的邊長相同，但形成之面積不同’ 由於製程上的誤差，習知 預个 -π θ ,. 技☆之人士應能理解前述邊長 相同僅疋概略，一般來 可接為1田水 邊長誤差在總邊長的⑽内均 I:::用者可根據該第-電容的電容值以及該第二電 :=，：推算得該待測材料2之介電常數值。根攄 之結構，貫際操作時，使用者只要將該第一電極3及 201221973 該第二電極4連接至儀器或電表之極性相反之訊號端，以 及將該第三電極6及該第四電極7連接至儀器或電表之極 性相反之訊號端，便可測得該第一介電常數萃取單元丨及 該第二介電常數萃取單元5的電容值，再將該第一電容的 電容值C]減去該第二電容的電容值q，乘上該待測材料 2之厚度值7¾油卿，除以該第二電極4與該第四電極7 的面積差為第二電極4的兩邊長；义2、 匕為第四電極7的兩邊長’本實施例係以矩形為例，實際 tr不限於此），再除以空氣之介電常數值％，以㈣ 4測材料2之介電常數值沉，為令使用者更為清楚本 毛月之叶算方式’另將公式(以矩形為例)整理如下： c, ~c2 =201221973 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to an extraction material method, and more particularly to a method for extracting a dielectric constant from a dielectric constant extraction two-t device and a dielectric constant extraction. 'And improve [prior art] In some conventional capacitive methods to extract the dielectric constant capacitance of the capacitor as a test (4) and, 嗲, 都 以 疋 疋 疋 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 The dielectric constant of the material, but because of the usual operation, the constant is higher than the dielectric constant of the material. The conventional capacitance method is: 疋 By measuring the capacitance value and area of the capacitor, the dielectric constant of the material is reversed, and the actual, second sense and the equivalent capacitance of the body are composed, as shown in FIG. When the capacitor is raised in frequency, the capacitance value is also increased. Therefore, when the dielectric constant is extracted by the capacitance method, the dielectric constant error due to the parasitic extraction increases with the operating rate, and the voltage is increased. The available bandwidth of the method is reduced, and the self-frequency of the capacitor is an indicator of the parasitic inductance. 2 is a perspective perspective view of a conventional dielectric constant extraction device. Referring to FIG. 2, since the structure requires the L line 12 in quantity _, the feedback effect of the feed line 12 reduces the natural frequency of the capacitor. It has become the main reason for the available bandwidth reduction of the conventional capacitance extraction dielectric constant method. 201221973 In the traditional capacitive method dielectric dynasty w - yArea - "number of strokes I is calculated by the formula to obtain the dielectric constant value of the material, where % is the dielectric constant of air, ~ The thickness of the two electrodes is the capacitance between the two electrode plates, which is made at = area. Since the capacitance value is the capacitance value, the capacitor edge is also included; when measuring, the measured capacitance will cause the capacitance to be measured. The electric constant extracts the capacitance, so the spur is higher than the dielectric constant of the material itself. 5. The difference 'make the extracted dielectric constant. Therefore, how to develop a extraction method' which can be compared with some conventional electric 'I The number of sorrows and their quantity 'can improve the method of the method has a lower error. τ used frequently, will be the initiative of the present invention [invention] The main species of two materials dielectric The constant device and its method, the quantity, and the wide mode have a lower error including 'the -first electrode' which is applied to the top to be tested: 3 and -: the test material'· it is applied to the test The bottom surface of the material, and ^ two electrodes, 】 = shape two; the face is the first - a second-two-r: a first-dielectric-extraction unit, comprising: the to-be; 201221973, an electrode disposed on the top of the material to be tested, which is disposed on the bottom surface of the material to be left, and formed a four-electrode, the surface formed by the pole is covered by the third electrode-capacitor 'the fourth electric medium', the side length of the second electrode and the fourth electrode's = covering area, the area of which is not H, according to The first = the same 'but the capacitance, the value is calculated to calculate the material to be tested: the dielectric: value and the second which contains the following (four) · Take the H f constant method, providing a first dielectric constant extraction The first electrode is covered by the first electrode of the material to be tested, and the surface of the material to be tested is covered by the first surface. The shape provides a second dielectric constant extracting material; a third electrode, A covering machine is in the main body of the eight packs ^: the top surface of the material to be tested, the fourth bottom surface of the material to be tested; a bottom surface 'and formed - the first = The covered surface covers, in the middle, the third electrode has the same side length, but the side length of the ^^ electric power and the fourth electric quantity measure the first capacitance Capacitance values; measuring the capacitance of the second capacitor; and according to the capacitance value of the capacitor and the first value to obtain the estimated value of the dielectric constant of the material under test. * Capacitance [Common Application] 201221973 In order to fully understand the features and effects of the present invention, the present invention will be described in detail by the following specific embodiments and with the accompanying drawings. FIG. 3A and FIG. 3b are perspective perspective views of an embodiment of the present invention. Referring to FIG. 3A and FIG. 3b simultaneously, the present invention is a dielectric constant (four) of an extraction material, which includes: “— The dielectric constant extraction unit comprises: a material to be tested 2 (for example, a sheet material, a substrate, a film, a paper-type material or a planar material which is pressed by a powder and a composite material, but is not limited thereto); - a first electrode 3, which is disposed on the top surface of the material to be tested 2; and a second electrode 4 which is disposed on the bottom surface of the material to be tested 2, and is formed into a first capacitor, the second electrode 4 The formed cladding is covered by the cladding formed by the first pole 3; and the second dielectric constant extraction unit includes: the material to be tested 2, the third electrode 6, which is coated on the material to be tested 2 The top surface; and the flute ^ ^ the fourth electrode 7, which is placed on the bottom of the material to be tested 2 The surface 'and forms a second Lei Xuan, the knot is formed. The cladding formed by the fourth electrode 7 is formed by the third electrode 6 and the side of the layer 4 has a complex surface, wherein the second The side length of the electrode 4 is the same as the side length of the fourth electrode 7, but the area formed is different. 'Because of the error in the process, it is conventional to pre--π θ. The length is the same as the outline only, generally it can be connected to the 1 field water side length error within the total side length (10) I::: the user can calculate the capacitance value of the first capacitance and the second electricity: =,: The dielectric constant value of the material 2 to be tested is obtained. The structure of the root can be used, the user only needs to connect the first electrode 3 and the second electrode 4 of the 201221973 to the signal terminal of opposite polarity of the instrument or the electric meter, and the third electrode 6 and the fourth electrode 7 connected to the opposite end of the signal of the instrument or the meter, the capacitance value of the first dielectric constant extraction unit 丨 and the second dielectric constant extraction unit 5 can be measured, and then the capacitance value of the first capacitor is C. Subtracting the capacitance value q of the second capacitor, multiplying the thickness value of the material to be tested 2 by 73⁄4 oil, divided by the area difference between the second electrode 4 and the fourth electrode 7 being the length of both sides of the second electrode 4 2, 匕 is the length of both sides of the fourth electrode 7 'This embodiment is based on a rectangle, the actual tr is not limited to this), and then divided by the dielectric constant value of air %, (4) 4 measured material 2 dielectric The constant value is Shen, in order to make the user more aware of the leaf calculation method of the month of the month, and the formula (in the case of a rectangle) is organized as follows: c, ~c2 =
Thinkness ，透過上述之裝置即可消除邊 緣雜散電容對介電常數萃取所造成的影響，使本發明較習 去電奋法卒取方式具有更低的誤差量。 圖四A及圖四B為本發明之另—具體實施 該第:ί=置成T頻寬’根據本實施例，較佳係於 經由节笛第一透孔8’以供一第一饋入單元9 連接^主孔8穿過該待測材料2與該第二電極4電性 ί:入該第三電極6形成-第二透孔1◦，以二 第四電由该第二透孔1Q穿過該待測材料2與該 電極7電性連接，如此只要提供該第-饋入單元9及 201221973 該第一電極3至儀器或電表之極性相反之訊號端，以及提 供該第二饋入單元11及該第三電極6至儀器或電表之極性 相反之訊號端，再根據前述實施例的處理方式，便可獲得 該待測材料2之介電常數值DA：，同時本實施例的結構較 之前一實施例，由於其訊號饋入方式是採用中心穿孔，藉 由中心穿孔使號饋入時具有最短路徑，使饋入路徑之寄 生電感降低，因此可更提高自振頻率，進而提升本發明裝 置之可用頻寬。 圖五為本發明之一具體實施方式的步驟流程圖，為令 使用者更為了解本發明之使时式，請參相五並配合圖 ^含及下本發明之—種萃取㈣介電常數的方法， m—介電常數萃取單元1，形成有-·第-電 ^胺、^··-待測材料2(例如：片狀材料、基板、 之平：:材:=複=合而成 設於該待測材料2之頂面此），第一弟二電極3,其覆 於該待測材料2之底二=4’其覆設 面係被該第-電極3所^；*W4所形成之覆 提供-第二介電常數萃d 容，其包含:該待測材料2早：：有-第二電 於該待測材料2之7頁.★第二電極6,其覆設 該待測材料2 、面，—第四電極7，其覆設於 係被該第三電極6开^亥第四電極7所形成之覆面 二電極4的邊C面涵蓋’其中，該第 ?、 电極7的邊長相同，但形 201221973 成之面積不同； 量測該第一電容的電容值； 量測該第二電容的電容值；以及 根據該第一電容的電容值以及該第二電容的電容 值’以推算得該待測材料2之介電常數值。 f用者〆、要依循上述步驟便可得到欲量測之介電常數 m 上的誤差’ f知此技藝之人士應能理解前述 内均疋概略’一般來說’邊長誤差在總邊長的10% 二雷三又，使用者可根據該第—電容的電容值以及該第 值’以推算得該待測材料2之介電常數值。 連二義只要將該第一電極3及該第二電極4 極6及性減，以聽該第三電 迪^ 按主儀裔或電表之極性相反之訊號 數萃取;ΓΠίϊ:介電常數萃取單元1及該第二介電常 一電容的電容值ς減去該第二電容 者了，3亥第 待測材料2之厚度伽― ς 2後’乘上该 四電極7的面積差㈣—、 “第 邊長；尤2、y Α雷μ 7 1 '、 一電極4的兩 例’貫際使用時不限於此)，再除 月：以矩為 以寐;}日外/士 、 工氧之；丨電常數值， 又㈣待測材料2之介電常數值服 述，在此不另加贅述。 。十A式如如 發明t同圖四A及圖四8之實施例’當使用者欲提古太 乂成一第一透孔8,以供一第一 電極3 饋入早疋9經由該第一透 201221973 =穿，待測材料2與該第二電極4電性連接，同時於 “第一電極6形成一第二透孔1〇，以供一第二饋入 =該第二透孔iG穿過該待測材料2與該第四電極7電性 連接，如此只要提供該第一饋入單元9及該第一 儀^或電表之極性相反之訊號端，以及提供該第二饋2 及5亥第二電極6至儀器或電表之極性相反之訊㈣ ==施例的處理方式便可獲得該待測材料二介 二孔’：由中心穿孔使訊號饋入時具有最短路徑，使饋 =之寄生電感降低，因此可更提高自振頻率，進而提 升本毛明方法之裝置的可用頻寬。 圖六為本發明之另—具體實_的 六圖=明之介電常數萃取的裝置亦可以多層方=圖 為各個不同之待測材料，L|'“為待測 =間之金屬線路，當⑽材料（1與]為h盆中任一 ^，且為待測材料時，則介電常 :=果於，與。上’如此達到測量的萃取= ’在此提供* (ΨίΌ\η FI a » 較之 > 考結構可參照圖二Thinkness, through the above device, can eliminate the influence of the edge stray capacitance on the dielectric constant extraction, so that the present invention has a lower error amount than the conventional method. 4A and FIG. 4B are another embodiment of the present invention: the ί=set to T bandwidth ′ according to the embodiment, preferably via the first through hole 8 ′ of the flute for a first feed The second unit 8 is connected to the second electrode 4 to form a second through hole 1 , and the second through The hole 1Q is electrically connected to the electrode 7 through the material to be tested 2, so as long as the first electrode 3 of the first feeding unit 9 and the 201221973 are provided to the opposite end of the instrument or the meter, and the second is provided. The feeding unit 11 and the third electrode 6 are connected to the signal terminals of the opposite polarity of the instrument or the electric meter, and according to the processing manner of the foregoing embodiment, the dielectric constant value DA of the material to be tested 2 can be obtained, and the embodiment is Compared with the previous embodiment, since the signal feeding method adopts the center hole, the center piercing has the shortest path when the number is fed, so that the parasitic inductance of the feeding path is lowered, so that the natural frequency can be further improved, and thus the natural frequency can be further increased. The available bandwidth of the device of the invention is increased. Figure 5 is a flow chart of the steps of one embodiment of the present invention. In order to make the user more aware of the timing pattern of the present invention, please refer to the figure and the extraction (four) dielectric constant of the present invention. The method, the m-dielectric constant extraction unit 1, is formed with -·the first electro-amine, ^··-the material to be tested 2 (for example: sheet material, substrate, flat: material: = complex = combined The first electrode 2 is disposed on the top surface of the material to be tested 2, and the second electrode 3 is applied to the bottom of the material to be tested 2=4', and the covering surface is covered by the first electrode 3; The cover formed by W4 provides a second dielectric constant extraction capacity, which comprises: the material to be tested 2 is early:: there is - the second electricity is 7 pages of the material to be tested 2. The second electrode 6 is covered The material to be tested 2, the surface, and the fourth electrode 7 are disposed on the side C of the cladding electrode 4 formed by the fourth electrode 7 of the third electrode 6 to cover 'the middle, the first ?, the side length of the electrode 7 is the same, but the shape of the shape of 201221973 is different; measuring the capacitance value of the first capacitor; measuring the capacitance value of the second capacitor; and according to the capacitance value of the first capacitor and the first The capacitance value of the two capacitors is used to derive the dielectric constant value of the material 2 to be tested. f User 〆, follow the above steps to get the error in the dielectric constant m to be measured' f. Those who know the skill should be able to understand the above-mentioned internal uniformity. Generally speaking, the side length error is on the total side length. 10% of the second and third, the user can calculate the dielectric constant value of the material to be tested 2 according to the capacitance value of the first capacitor and the first value. As long as the first electrode 3 and the second electrode 4 are reduced in polarity, the second electrode is extracted by the number of signals of the opposite polarity of the main instrument or the meter; ΓΠίϊ: dielectric constant extraction The capacitance value of the unit 1 and the second dielectric constant-capacitor is subtracted from the second capacitor, and the thickness of the material 2 to be tested is 伽 ς 2 and then multiplied by the area difference of the four electrodes 7 (four) - "The length of the first side; especially 2, y Α雷μ 7 1 ', two cases of one electrode 4' is not limited to this when used continuously, and then the month: the moment is the moment;} outside the day / Shi, work Oxygen; 丨 electric constant value, and (4) dielectric constant value of the material to be tested 2, which will not be described here. 10A is as invented as the embodiment of Figure 4A and Figure 4 The user wants to introduce the ancient through hole into a first through hole 8 for feeding a first electrode 3 into the early stage 9 through the first through hole 201221973, and the material to be tested 2 is electrically connected to the second electrode 4, Forming a second through hole 1 〇 in the first electrode 6 for a second feed = the second through hole iG is electrically connected to the fourth electrode 7 through the material to be tested 2, so as long as the the first Into the unit 9 and the first instrument or the opposite polarity of the signal of the meter, and the second feed 2 and 5 second electrode 6 to the opposite polarity of the instrument or the meter (4) == the treatment of the example The second material of the material to be tested can be obtained: the shortest path is adopted when the signal is fed by the central perforation, so that the parasitic inductance of the feed is reduced, so that the natural frequency can be further improved, thereby improving the device of the present method. bandwidth. Figure 6 is a further embodiment of the present invention - the six figures = the dielectric constant extraction device of the Ming can also be multi-layered = the figure is the different materials to be tested, L | '" is the metal line to be tested = when (10) The material (1 and ) is any of the h pots, and when it is the material to be tested, the dielectric is often: = fruit, and . 'up to the measured extraction = 'provided here * (ΨίΌ\η FI a » Compared to the > test structure can refer to Figure 2
Si : β(本發明)’其中第二電極4之兩邊 長句為4〇mil ’第四電極7之兩邊長分別為 二’而習知之介電常數萃取裝置之電極 40 nul再加上長度為12ιηϋ宮 咬瓦j兩 ψ lmH=9r Λ 寬又為4mil的饋入線12，其 -5如，圖七為三種介電常數萃取裝置之電容值 201221973 比較圖’圖八為習知介電常數萃取裝置與本發明之 較Γ圖九為f知介電常數萃取㈣本： 之介”‘萃方t圖二至圖九可發現本發明 置及方法有明顯的改善，表-為比較結果= 表，由比較結果可得知本發明不但可降低傳 取錯誤率’更可提升電容器萃取介電常數之可； 頻率範圍。 双 <』用 从」以上所述可以清楚地明瞭，本發明係提供一種萃取 |及，、方法，其可較某些習知電容法萃 式八有更低的誤差量，並可提高本方法的可用頻寬。 述發明專利申請案做一詳細說明’惟以上所 述者’僅為本發日轉利申請案之較佳實 限定本發明專利申請荦竇 田不月匕 請案申請範圍所作之二::二即凡依本發明專利* 專利申請案之專利涵與修飾等，皆應仍屬本發明 表一為比較結果整理表。 201221973 傳統萃取方式 改良式萃取方式 表面積(mil2) (4〇x40)+(4xl2) (4〇x40)-(5〇x30) 電容值(pF) (0.05〜3 GHz) 0.4559-Ό.475 (0.4658^.4765)-(0.4449^.4547) 介電常數 (0.05-3 GHz) 4.4-4.593 3.332-3.477 誤差率 (0〜3 GHz) 33%^39.2% l°/cr-5.4%Si: β (present invention) wherein the two sides of the second electrode 4 are 4 mils long. The length of the two sides of the fourth electrode 7 is respectively two', and the electrode of the conventional dielectric constant extraction device 40 nul plus the length is 12ιηϋ宫咬子jj two ψ lmH=9r Λ Width and 4mil feed line 12, its -5, Figure 7 is the capacitance value of three dielectric constant extraction devices 201221973 Comparison Figure 'Figure 8 is the conventional dielectric constant extraction Figure 9 is a comparison between the device and the present invention. The fourth embodiment of the present invention is characterized by a significant improvement in the method of the present invention, and the comparison result is shown in Table IX. From the comparison results, it can be seen that the present invention can not only reduce the transmission error rate, but also improve the dielectric constant of the capacitor extraction; the frequency range. Double <> An extraction method, and method, which can have a lower error amount than some conventional capacitance methods, and can increase the available bandwidth of the method. The invention patent application is described in detail as 'the above-mentioned one' is only the preferred embodiment of the present invention. The patent application of the invention is the second application of the application scope of Dou Tian's application. That is, the patent embossing and modification of the patent application according to the present invention should still be listed in Table 1 of the present invention as a comparison result. 201221973 Traditional extraction method Improved extraction surface area (mil2) (4〇x40)+(4xl2) (4〇x40)-(5〇x30) Capacitance value (pF) (0.05~3 GHz) 0.4559-Ό.475 (0.4658 ^.4765)-(0.4449^.4547) Dielectric constant (0.05-3 GHz) 4.4-4.593 3.332-3.477 Error rate (0~3 GHz) 33%^39.2% l°/cr-5.4%
12 201221973 【圖式簡單說明】 圖一為一習知介電常數萃取裝置之電容器等效電路圖。 圖二為一習知介電常數萃取裝置之立體透視圖。 圖三A及圖三B為本發明之一具體實施例的立體透視圖。 * 圖四A及圖四B為本發明之另一具體實施例的立體透視圖。 - 圖五為本發明之一具體實施方式的步驟流程圖。 圖六為本發明之另一具體實施例的剖視圖 • 圖七為三種介電常數萃取裝置之電容值比較圖。 圖八為一習知介電常數萃取裝置與本發明之萃取介電常數結 果比較圖。 圖九為習知介電常數萃取裝置與本發明之介電常數誤差率比 較圖。 【主要元件符號說明】 1第一介電常數萃取單元 2待測材料 3第一電極 4第二電極 5第二介電常數萃取單元 6第三電極 7第四電極 13 201221973 8第一透孔 9第一饋入單元 10第二透孔 11第二饋入單元 12饋入線 Μι〜Mn待測材料 L丨〜Ln+i金屬線路 (1H5)本發明之一具體實施方式的實施步驟 1412 201221973 [Simple description of the diagram] Figure 1 is a capacitor equivalent circuit diagram of a conventional dielectric constant extraction device. Figure 2 is a perspective perspective view of a conventional dielectric constant extraction apparatus. 3A and 3B are perspective perspective views of one embodiment of the present invention. * Figure 4A and Figure 4B are perspective perspective views of another embodiment of the present invention. - Figure 5 is a flow chart of the steps of one embodiment of the present invention. Figure 6 is a cross-sectional view showing another embodiment of the present invention. Figure 7 is a comparison of capacitance values of three dielectric constant extraction devices. Figure 8 is a comparison of a conventional dielectric constant extraction apparatus with the extracted dielectric constant of the present invention. Figure 9 is a comparison of the dielectric constant error rate of the conventional dielectric constant extraction apparatus and the present invention. [Main component symbol description] 1 first dielectric constant extraction unit 2 material to be tested 3 first electrode 4 second electrode 5 second dielectric constant extraction unit 6 third electrode 7 fourth electrode 13 201221973 8 first through hole 9 First feeding unit 10, second through hole 11, second feeding unit 12, feeding line 〜1 to Mn, material to be tested L丨~Ln+i, metal line (1H5), step 14 of an embodiment of the present invention