TW535473B - System and method for identifying solder joint defects - Google Patents

System and method for identifying solder joint defects Download PDF

Info

Publication number
TW535473B
TW535473B TW91114381A TW91114381A TW535473B TW 535473 B TW535473 B TW 535473B TW 91114381 A TW91114381 A TW 91114381A TW 91114381 A TW91114381 A TW 91114381A TW 535473 B TW535473 B TW 535473B
Authority
TW
Taiwan
Prior art keywords
solder joint
solder
printed circuit
value
circuit board
Prior art date
Application number
TW91114381A
Other languages
Chinese (zh)
Inventor
Tracy Kristine Ragland
Original Assignee
Agilent Technologies Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Agilent Technologies Inc filed Critical Agilent Technologies Inc
Application granted granted Critical
Publication of TW535473B publication Critical patent/TW535473B/en

Links

Abstract

An improved circuit inspection system (100) incorporates an automated measuring technique (1400) that accounts for acceptable Z-axis elevation variance across both a printed-circuit device (410) and a mounting surface (420) when making solder-joint pass/fail decisions. The improved solder-joint inspection system (100) applies a near neighbor solder joint diameter error analysis (1100) for each solder joint on a printed-circuit device (410). A near neighbor solder joint diameter error outlier analysis (1300) is used to identify solder joint defects with improved accuracy.

Description

535473 五、發明說明 本揭示係-般有關於使用測量系統對電路板總成之言 解析度檢查的系統及方法,且特別是有關於為了品質^ 分析而測量電路板總成内之焊料接合的系統。 貝工1 電子裝置之焊接與組裝的快速且精準品質控制檢 變為電子製造業的優先項目Q 一 元件與焊接連接的尺寸變小、電路板總成上元件” 提高的結果、及將焊接連接放在裝置封裝下以免被看= 表面安裝技術(贿)之出現已使裝置間電子装置與電氣連 接之快速與精確檢查在製造環境中非常難以進行。 電子裝置與連接之很多現存的檢查系統使用穿透韓射 以形成影像’其強調該等裝置與連接内部構造的外貌呈 現。這些系統經常運用慣常x光照相術,其中穿透辕射包 含X光。例如胸、臂、腳、脊椎等人體各種部位的醫療χ 先可,是慣常X光照相影像之最熟悉的例子。當物體被X 一束"射t戶斤形成的影像或圖呈現被檢查之物體所定型 之X光影。’亥X光影被如底片或其他適當設施之對X光 敏感的材料記錄。 、X光影或X光圖之形狀不僅被該物體之内部構造特徵 ^射的X光心擊該物體的方向所決定◦所以不管用人 視覺或用電腦數值所實施的X光影像之完整解釋與分析智 :需要有關對物體特性與其針對x光束之排向的某些假 ^ /、、、二#有必要對有關該物體之形狀與内部結構 :: …^ X光對邊物體的方向採取特定的假設。根據這些 光〜像之形貌可被分析以決定該物體對應的結構 4 535473 五、發明說明 特徵,如在焊料連接内之瑕疵,其產生各種影像外貌。這 些假。X經¥創造模糊性,其使影像解釋的可靠度及根據X 光影像分析之決策品質降級。因慣常X光照相分析之假設 的根本模糊性結果為在焊料連接之瑕疵如形狀、密度與大 小的結構特徵之小變異通常被該焊料連接本身以及附近焊 料連接、電子裝置、電路板與及他物體之遮陰塊所遮蔽。 由於遮陰塊與附近物體通常對每種焊料接合為不同的,要 採取足夠假言史以精石崔i也決定各別焊料#合内之焊料的形 狀、尺寸與位置是極端繁瑣且常是幾近不可能的。 為了試圖補償這些缺失,某些系統採用由數個角度觀 ㈣物體之能力。此額外的觀察促成這些系統部分轉決 在X光影投射影像所呈現之模糊性。然而運用多觀察角須 用到複雜機械操縱系統’通常需要多達五個獨立的非正交 運動軸。此機械複雜度導致費用提高、尺寸i重量辦加 檢查時間較長、因機械複雜性所致之不良的定位精:度、 及因非正交運動軸所致之校正與電腦控制複雜度。 與上面討論之慣常乂光照相術有關的問題可藉由產生 :=物體的斷面影像被減輕。如斷U光檢查 輔助X光斷層照相(CT)之X光斷# ^ 啤層妝相技術已在醫療用途 中被使用以產生斷面或斷層影像。在醫療用途中已廣泛地 主要是因大約為-或二公_〇·。… 王又的相當低之解析度是令人滿意的,且 求不像對應的工業要求般地嚴格。 、屋里要 在電子檢查情形中,特別e料 丨如焊料接合之電氣連接 五、發明說明(3 ) 的檢查,例如20微米(大約〇.⑽〇8英 _一的。進而言之,工業== 必須每秒產生多個影像以便在工業生產線之實^用—先 能達成電子檢查所需之速度與精確度要求的 仏查系統在下列專利中被描述:發給等人之: 「用於電子檢查之自動斷層χ光檢杳 ^ ”、、 " ㈣層X先檢查系統」的美國專利第5,097 492號 二Γ二Γ等人之題目為「用於電子檢查之自動: 日 ^乐'..先」的美國專利第5加,㈣號專利、心 之題目為「用於偵測過度/不足谭料瑕 及衣置」的美國專利第5,291,535號專利、發給r〇如等人 之題目為「用於偵測與控制焊料财t之學f :美:專利第5糊"虎專利、發給—人Si 5 561於檢查電氣連接之方法及裝置」的美國專利第 斤产號專利、發給缺·等人之題目為「用於高解 =子t目檢查之方法及裝置」的美國專利第卿54 : 給Baker等人之題目為「斷層χ光檢查系統及 ::引:夕路徑輻射源之方法」的美國專利第5,259,。12 給址㈣等人之題目為「連續線性掃描斷層X α — 2統與方法」的美國專利第5,583,904專利、發給 S寻人之題目為「用於斷層Χ光檢查電路板檢查之自 Γ、曲補償」的美國專利第卿^ …、之母專利的被納於此處作為參考。 535473 五、發明說明(4) 在一斷層X光檢查系統中,其觀察一固定系統且具有 比被檢查之物體小的成像區,其可能有必要繞著移動該物535473 V. Description of the Invention The present disclosure is generally related to a system and method for inspecting the resolution of a circuit board assembly using a measurement system, and in particular to measuring solder joints in a circuit board assembly for quality analysis. system. BeiGong 1 Fast and accurate quality control inspection of welding and assembly of electronic devices has become a priority item in the electronics manufacturing industry. Q. The size of components and soldering connections has become smaller, components on circuit board assemblies have been improved. Put under the device package to avoid being seen = The advent of surface mounting technology (bribery) has made rapid and accurate inspection of electronic devices and electrical connections between devices extremely difficult in manufacturing environments. Many existing inspection systems for electronic devices and connections are used Penetrating the Korean radio to form an image, it emphasizes the appearance of these devices and connecting internal structures. These systems often use conventional x-ray photography, where penetrating radiography includes X-rays. For example, the chest, arms, feet, spine and other human bodies The medical treatment of the part is first possible, and it is the most familiar example of the usual X-ray radiographic images. When an object is formed by an X-ray beam, the image or picture formed by the X-ray beam presents the X-ray shadow shaped by the object being inspected. Light and shadow are recorded by X-ray sensitive materials such as negatives or other suitable facilities. The shape of X-rays or X-ray images is not only reflected by the internal structural features of the object The direction of the X-ray center hitting the object is determined. Therefore, a complete interpretation and analysis of X-ray images implemented by human vision or computer values is required. Some wisdom about the characteristics of the object and its orientation to the x-ray beam is needed ^ / 、、、 二 # It is necessary to make specific assumptions about the shape and internal structure of the object:… ^ The direction of the X-rays on the edge of the object. Based on these shapes of light ~ image, it can be analyzed to determine the corresponding object. Structure 4 535473 5. Inventive features, such as flaws in the solder connection, which produce various image appearances. These fakes. X warp creates blur, which degrades the reliability of image interpretation and the quality of decision-making based on X-ray image analysis The fundamental ambiguity due to the assumptions of conventional X-ray analysis is that small variations in solder connection defects such as structural characteristics of shape, density, and size are usually caused by the solder connection itself and nearby solder connections, electronic devices, circuit boards, and The shading blocks of other objects are obscured. Since the shading blocks and nearby objects are usually different for each type of solder, it is necessary to take a sufficient hypothetical history to fine stone Cui i also Determining the shape, size, and location of the solder within each solder joint is extremely tedious and often almost impossible. In an attempt to compensate for these deficiencies, some systems employ the ability to view objects from several angles. This additional Observation causes these systems to partly depend on the ambiguity presented in the X-ray projection image. However, the use of multiple viewing angles requires the use of complex mechanical manipulation systems. 'Usually requires up to five independent non-orthogonal motion axes. This mechanical complexity results in Increased cost, longer inspection time, size and weight, poor positioning accuracy due to mechanical complexity, and the complexity of calibration and computer control due to non-orthogonal motion axes. As usual with the discussion above 乂The problems related to photography can be alleviated by generating: = cross-sectional images of objects. For example, X-ray fracture assisted by tomography X-ray tomography (CT) # ^ Beer layer makeup technology has been used in medical applications To produce a section or tomographic image. It has been widely used in medical applications mainly due to approximately -or two public _〇 ·. … Wang You's relatively low resolution is satisfactory and is not as stringent as the corresponding industrial requirements. In the case of electronic inspection, special electrical materials such as solder joints are required in the house. V. Inspection of the invention description (3), for example, 20 micrometers (approximately 0.08 in._one. In addition, industrial == Multiple images must be produced per second for practical use in industrial production lines-an inspection system that first achieves the speed and accuracy required for electronic inspection is described in the following patent: issued to others: "Using US Patent No. 5,097,492, Γ 二 Γ, etc. for the automatic tomographic X-ray inspection of electronic inspection ^, "quoting the X-ray inspection system first", etc. The title is "Automation for Electronic Inspection: Japanese ^ Le US patent No. 5 plus ".. first", US patent No. 、, US patent No. 5,291,535 entitled "Used to detect excessive / insufficient material defects and clothing", issued to r〇 The title of "U.S. Patent for Detecting and Controlling Solder Property: U.S. Patent No. 5" & "Tiger Patent, Issued-Method and Apparatus for Checking Electrical Connections by Si 5 561" The patent of Jinjin No., issued to the deficiencies, etc. is entitled "Used for high-resolution = child inspection. Method and Device "U.S. Patent No. 54: U.S. Patent No. 5,259, to Baker et al. Entitled" Tomographic X-ray Examination System and :: Introduction: Method of Radiation Path Radiation Source "12 U.S. Patent No. 5,583,904 entitled "Continuous Linear Scanning Tomography X α-2 System and Method", U.S. Patent entitled "Self-Gamma and Curve Compensation for Tomography X-ray Board Inspection" issued to S Search The mother of the first ^, ..., patent is incorporated herein by reference. 535473 V. Description of the invention (4) In a tomography system, it observes a fixed system and has a smaller imaging area than the object being inspected , It may be necessary to move the object around

體以在該成像區内定位該物體之不同區域而產生多斷層X 光檢查,其各塊在一起形成整個物體之影像。此經常藉由 在如一 X-Y-Z定位台之機械操縱系統支撐該物體而被達 成。此台被移動以將該物體之所欲的區域帶至該成像區 内。X與Y方向的移動定出被檢查之區、而z方向之移動 將該物體上下移動以選擇要取其斷面影像之物體内的平 面。 數個上面參照之專利揭示裝置與方法用於產生在固定 或可廷擇之斷面影像聚焦平面的測試物體之斷面影像。在 这些系統中,一 x光源系統與一 χ光偵測器系統在z軸方 向以固疋距離被隔離且斷面影像聚焦平面位於Z軸方向之 預=特定位置。此位置針對x光㈣統Μ光偵測器系 、、著車由方向位於中間Q該χ光偵、測器系統收集資料, 、、j式物to位t斷㈣像聚焦平面之外貌的斷面影像可由此 被I成m统主張要被成像之外貌位於沿著z轴方向 之預設特定位置的固定或可選擇之斷面影像聚焦平面内。 :而在這些系統中,重要的是該斷面影像聚焦平面與將被 一像之物體内的平面之位置被組配以與沿著Z軸方向之同 -位置相符。若此條件未符,則測試物體内被選擇之 不會被獲得。代之的是,在包括被選擇之外貌的 π方^方㈣測試物體内之平面的斷面影像將被與 某些配置中,少於燁料接合之最適視域(V一可: 535473 五、發明說明 成像後果為,這些少於最適視域之分析會導致相關焊料 接合之不精準的分析。 目前普遍被用以定位斷面影像聚焦平面内之測試物體 的被選擇外貌之技術實體測量該被選擇外貌之z軸位置。 3測4物體便使用此測量被定位,使得被外貌之外貌與斷 面衫像聚焦平面之z轴位置相符。任何種類之標準方法與 $器可被用以實體測量該被選擇外貌之2軸位置。2轴測 1糸統可在市面取得數種類型,其被用以決定在z上已知 1置/、4表面上或恰在表面下之該測試物體的外貌間之距 離^類系統簡單地以機械式固定該測試物體、一機械探 :=雷射光學三角系統、—光學干涉計與在其間之-超 、系、充這些z軸位置測量系統之一可被用以形成該測 試物體之表面的「z圖」。該「z圖」典型上含有χ_γ位置 之陣列配以在特定^位置之該測試物體的表面之ζ值。 該,位置在與測試物體共用之平面上被點出,其實質上與 肩面〜像^焦平面平行。在電路板上斷面影像外貌系統 2取晋遍錢用m為使用以f射為基礎之測距猫準 測距晦準器已特別被用於斷面χ光影像系統,立被用 =將電子電路板總成形成影像。電路板總成比起元件被安 衣之表面區域典型上是很薄的。某些電路總成用如陶 非常穩定的材料被做成。然而大多數的電: 板士用猶微有彈性或在某些情形為非常有 構成。此彈性讓電路板在垂直於主要表面區域(即含有= 535473 五、發明說明 接腳之表面區域)或z軸的軸上形成纏繞。此外,有些電路 板總成的厚度會有變異。除了電子總成外,很多其他物體 在比例上具有維度之變異,其在斷面X光成像之Z聚焦平 面的視野深度是重大的。藉由測量被纏繞之測試物體的表 面’ Z維度之變異量便可被用以針對該斷面X光成像之z 水焦平面適當地調整該測試物體之位置關係,使得該測試 物體内所論及之外貌的所欲影像可被成像。 、i疋 種此測距目士準糸統被設計用於如在Baker 等人之美國專利第4,926,425號所描述之系統内,此專利 此後簡稱為,452專利。該,452專利揭示一種斷層χ光檢查 系統,其中一Μ χ光為基礎之成像系統具有非常淺之視野 被用以k查如印刷電路板之固態物體。視野之深度線提供 用於檢查焊料接合之設施而不致干擾該焊料接合上方或下 方之元件。焊料接合上方或下方之材料是在焦點外且因此 為大致均勾的背景。為提供所要的選擇性,斷層χ光檢查 成:系統之視野深度為小於大約二百萬之一英綱 不幸的是印刷電路板之表面變異經常超過此容差。 2服此缺點,印刷電路板之表面使用-雷射測晴器 不日、對。然而雷射測距瞄準器The volume is used to locate different areas of the object in the imaging area to generate a multi-slice X-ray examination, and the pieces together form an image of the entire object. This is often achieved by supporting the object with a mechanical manipulation system such as an X-Y-Z positioning table. This stage is moved to bring the desired area of the object into the imaging area. The movement in the X and Y directions determines the area to be inspected, and the movement in the z direction moves the object up and down to select the plane within the object whose cross-sectional image is to be taken. Several of the above-referenced patent disclosure devices and methods are used to generate cross-sectional images of a test object at a fixed or alternative cross-sectional image focus plane. In these systems, a x-light source system and a x-light detector system are isolated at a fixed distance in the z-axis direction and the sectional image focus plane is located at a pre-specified position in the z-axis direction. This position is for the X-ray system, the M-ray detector system, and the direction of the vehicle is in the middle. The x-ray detection and measurement system collects data. The planar image can thus be claimed by I to M to be fixed in the fixed or selectable sectional image focusing plane at a predetermined specific position along the z-axis direction. : And in these systems, it is important that the positions of the focal plane of the cross-sectional image and the plane within the object to be imaged are assembled to coincide with the same -position along the Z-axis direction. If this condition is not met, the selected objects in the test object will not be obtained. Instead, the cross-sectional image of the plane in the test object that includes the selected appearance π square ^ square test object will be in some configurations less than the optimal field of view for the material to join (V-one: 535473 five According to the invention, the imaging consequence is that these analyses that are less than the optimal field of view will lead to inaccurate analysis of related solder joints. At present, it is generally used by technical entities that are used to locate the selected appearance of the test object in the focus plane of the cross-sectional image. The z-axis position of the selected appearance is used. 3 measurements and 4 objects are positioned using this measurement so that the appearance of the appearance matches the z-axis position of the focal plane of the profile shirt. Any standard method and device can be used for the entity Measure the 2-axis position of the selected appearance. 2-axis measurement 1 systems are available in several types on the market, which are used to determine the test object on 1 or 4 surfaces that are known on z or just below the surface The distance between the appearances of the ^ type system simply mechanically fixes the test object, a mechanical probe: = laser optical triangulation system,-optical interferometer, and the z-axis position measurement system One can be used Forms a "z-map" of the surface of the test object. The "z-map" typically contains an array of χ_γ positions with a ζ value of the surface of the test object at a particular position. The position is on a plane shared with the test object. It is spotted on the top, which is substantially parallel to the shoulder surface to the focal plane of the image. The cross-sectional image appearance system 2 on the circuit board takes the money used as the m and uses the f-based range-finding cat. The device has been particularly used in cross-section x-ray imaging systems. It is used to form an electronic circuit board assembly. The circuit board assembly is typically thinner than the surface area where the component is being mounted. Some circuit assemblies It is made of a very stable material such as pottery. However, most of the electricity: Boards are slightly elastic or in some cases very structured. This elasticity allows the circuit board to be perpendicular to the main surface area (i.e. contains = 535473 V. Description of the invention The surface area of the pin) or the z-axis is wound. In addition, the thickness of some circuit board assemblies will vary. In addition to the electronic assembly, many other objects have dimensional variations in proportion. X-ray imaging The depth of field of view of the Z focusing plane is significant. By measuring the surface of the test object being wound, the variation in the Z dimension can be used to appropriately adjust the test object's z-focal plane for X-ray imaging of the section. The positional relationship allows the desired image of the appearance discussed in the test object to be imaged. This rangefinder standard is designed for use as described in US Patent No. 4,926,425 to Baker et al. Within the system, this patent is hereinafter referred to as the 452 patent. The 452 patent discloses a tomographic x-ray inspection system in which an M x-ray-based imaging system has a very shallow field of view and is used to check the solid state of a printed circuit board. Objects. The depth line of the field of view provides facilities for inspecting solder joints without disturbing the components above or below the solder joint. The material above or below the solder joint is out of focus and therefore has a roughly uniform background. To provide the desired selectivity, tomographic X-ray inspection results in: the system's field of view depth is less than about one millionth of an inch. Unfortunately, the surface variation of printed circuit boards often exceeds this tolerance. 2 To serve this disadvantage, the use of the surface of the printed circuit board-laser detector is not good, right. Laser rangefinder

忠士你3 之坪圖被用以針對X 先成像以將印刷電路板定位,使得所論 在 内,甚至在該板由所論及之視野移到另-個時亦Γ 電路板檢查系統之一。 缺點為在决定被測量焊料接合外 方法淪為指示一焊料接合為「可 、 瑕疵的」。本谭料接合檢 二、」或「有 母4科接合外貌施用一 9 535473 五、發明說明(7 預設的門檻值以進行這些連接。 之方法& 此辨4瑕疵」焊料接合 的!右何%陣列封裝之各別焊料接合為「瑕巍」 路電路」_比例特別明Γ。购目關的焊料接合之「開 球:陣:A)之電路封裝例如具有平的 表面,其一般為長方形或正 \文衣 體覆蓋,並在「日片。此面可能被小的球形導 來回?恭:」或積體電路封裝的部位之積體電路 … 。如所習知者,該平的底面形成-基體 /、型上為多層基體)’積體電路模被固定於此。 數個因素組合使得有一陣 料接合)至焊球三種重要的可二:表面上輪(即焊 會纏繞形成該封裝安^面Γ 先’其封裝材料 衣表面與—印刷電路板間相對距離的 y、者Ί 一般而言,BGA „、職使得其邊 喝路板之安裝表面(即該等邊緣向上翻轉)。然而J 1 了角午BGA封裝纒繞主要以各種方式發生,而以咖 封裝之安裝表面與印刷電路板間距離在整個安裝表面上以 很多不同的方式變化。如柱陣列封裝(CGA)、正反器、晶 片尺寸型封裝(csp)與四角形平面封裝(QFp)等但不限於此 之其他封裝亦遭受纒繞之問題。 ”除了封裝纒繞外,印刷電路板材料也會纏繞。其應被 I %此一狀況不是彼此相互斥的。以另一方式述之,印刷 電路板(即安裝表面)與封裝可被纒繞。進而言之,印刷電 路,與(或m褒之任—會遭受「傾斜」。當安裝表面高度 Z維度之高度)在封裝或印刷電路板之任一上由一邊至另 10 五、發明說明(8) 一邊變化時呈現。The loyalty map of you 3 is used to image the X first to position the printed circuit board so that it is within one of the circuit board inspection systems even when the board is moved from the field of view to another. The disadvantage is that in deciding the solder joint to be measured, the method is reduced to indicating that a solder joint is "possible, defective". This Tan material joint inspection II, "or" There are 4 families of joints with a female appearance appearance application 9 535473 V. Description of the invention (7 preset thresholds to make these connections. Method & identify 4 defects "solder joint! Right What percentage of the individual solder joints of the array package are "defective" circuit circuits? The proportion is particularly clear. The circuit package of the "kickoff: matrix: A) of the solder joints of the purchase order has a flat surface, which is generally It is covered by a rectangular or regular body, and it is "Japanese film. This side may be led back and forth by a small ball? Christine:" or the integrated circuit of the integrated circuit package ... As is known, the flat bottom surface is formed-a substrate / a multilayer substrate on the mold) 'integrated circuit mold is fixed here. The combination of several factors makes the bonding of a matrix of materials to the solder ball three important two: the surface of the wheel (that is, the welding will be wound to form the package mounting surface) First, the relative distance between the surface of the packaging material and the printed circuit board y, or Ί Generally speaking, BGA „, makes its side drink the mounting surface of the road board (that is, the edges flip up). However, J 1 angle corner BGA package winding mainly occurs in various ways, and it is packaged with coffee. The distance between the mounting surface and the printed circuit board varies in many different ways across the entire mounting surface. Such as pillar array packages (CGA), flip-flops, chip size packages (csp), and quadrangular planar packages (QFp), but not Other packages limited to this also suffer from entanglement problems. "In addition to package entanglement, printed circuit board materials are also entangled. It should be 1%. This situation is not mutually exclusive. In another way, printed circuits The board (that is, the mounting surface) and the package can be entangled. In addition, the printed circuit and (or any of the m-any-will be subject to "tilt". When the height of the mounting surface Z dimension) on the package or printed circuit board Ren 10 from the upper side to the other five invention is described in (8) exhibits a change when the side.

這些與其他因素可—& A ^ ^ 之作用,使得印刷電路總成合含 :數個夕腳裝置(如陣列封裝),具有在 ; 合測量有顯著的變里。你"如 的坏科接 表面導體的實質均勾(在=在被施用至BGA封裝之安裝 . 合和)之球格陣列的情形中,bga 女衣表面與印刷電路板間距離的增加會致使—個以 上的焊料接合伸長,使得复古t 便個以 焊料接合。一般而tn接人在預先再流動狀況的 越靠近陣列中心之輕比Λ其Γ再流動狀況崩潰。 列崩潰較多。m “近陣列外部邊緣的球格陣 人虽ί ί在,列封裝上連聊位置之整個多數被測量的焊料 、廣’印刷電路總成與陣列封裝上相關的 垃或電氣上經由對應的焊料接合適當地被連 。焊料接合之各種測量中的相關差異(在對應的焊料接人 適當地被連接的情形中)為整個數個焊料接合測量之可; 受的變異之代表。 伐 可接受的焊料接合測量之廣大範圍使為精準地反映每 各料料接合之真實電氣與物理狀況的各種焊料接合測量 設定通過/失敗門播值極端地困難。在實務上,為⑽裝 置用丈干料接合檢查系統精準地偵測「開放電路」已是有問 題的。不僅是其耗時於重製每一 BGA封裝被辨識為具有一 <、上的瑕苑焊料接合測試結果、缺乏印刷電路(即完全植 入之印刷電路板)總成之全功能測試、或部件之電路内測 試’其需100%電氣測試之取得,非破壞性認證方法不存 535473 五、發明說明(9) 在以獨立地確認瑕疲結論之精準性。 在回應下’某些製造者已具有使用各種印刷電路裝置 之橢圓形的墊片。該等橢圓形的墊片在該墊片適當地被焊 料黏結時使焊料球形成橢圓形的焊料接合。該等^圓形= 塾片允許偵測「開放」的焊料接合,而相關的焊料影像維 持為環狀。雖然在偵測焊料接合之此與其他改良,其欲具 有改良的系統與方法用於改良焊料接合檢查系統:精; 度,其在於精準地辨識被用以物理上及電氣式連接印刷電 路板上各種印刷電路裝置之瑕疫焊料接合,其考慮到在被 測量之印刷電路裝置中可接受的變異。 在回應於習知技藝之這些與其他缺失下,並一 種焊料接合檢查系統及方法用於採用測試門檻值來^焊 科接合瑕疵。在一配置中,谭料接合檢查系統記錄 電路總成(即完全植人之印刷電路板)之每-印刷電路裝置 上有關數個接腳(即焊料接合介面)的位置資訊以及有關該 印刷電路總成之安裝表面與裝置封裝間之距離變里的資 :敗:谭,料接合檢查系統為每-谭料接合產生-期望通過: =直f考慮到在每印刷電路裝置上為該組痒料接合 =錄之真⑽量值的可接受之變異。簡要地描述,在架 …改良的痒料接合檢查系統用測量焊料接合之m 汁异用之設施;以及一資料儲存設施被實現。… 記錄在-印刷電路總成上每―谭料接合之測量,·為: 心===被視為提供-種方法用於辨識 …一一 έ ’此方法可被囊整為如下的步驟: 焊 12 五、 發明說明 10 料接合估計期望值,其考慮到該印刷電路裝置針對—印厣 電路板之高度的變異;為每一嬋料接合比較被記錄 =為每-焊料接合產生-誤差值;以及辨識界外的測量: 、該系統之其他實施例可被視為提供一種方法用於扩 剛試門檻值。就此而言,此方法可被彙整為如下的步:: 為一印刷電路裝置上的數個接腳獲得位置資訊;獲^:乂 忒印刷電路裝置之整個安裝表面的高度變異之資訊;2錄 與該等數個接腳相關之數個焊料接合的測量;回應於敕個 =裝表面之高度變異估計各別焊料接合點之可㈣的= 範圍;以及設定至少一門檻值回應於該範圍之一界限。里 與採用考慮到在-印刷電路板總成之可接受的焊料接 合直徑變異之電氣連接瑕疲有關之其他系統、方法與特點 對熟習該技藝者在檢視下列的圖與詳細描述下將變得明白 的。其欲所有包括在此描述内之這類系統、方法與特點如 所附申請專利範圍所保護之用於確認瑕庞的系統與方法之 領域内。 用於確認電氣連接瑕疲之系統及方法可參照下列各圖 而較佳地被了解。圖中之元件不見得強調其比例尺而是注 重清楚地說明該測試方法在確認印刷電路板上一個以上之 «間的電氣連接瑕苑之原理。進而言之,在所有圖令相 同的元件編號被指定為各圖中對應的部位。 第1圖為-釋例性電路檢查系統之方塊圖。 第2圖為-釋例性球格陣列之焊料接合直徑值的描 535473 五、發明說明(n 圖。 第3圖為另一釋例性球格陣列之焊料接合直徑值的描 點圖。 弟4圖為在各別焊料接合直徑上之球格陣列·繞的影 響之側面圖。 / 第5圖為球格陣列纒繞與印刷電路板傾斜被耗合及各 別焊料接合直徑上之纒繞的影響之側面圖。 第6圖為-釋例性球格陣列裝置之平面圖,顯示用於 為二=料接合決定自一裝置之中心的距離之方法。 #圖為由圭于I中心之焊料接合直徑值對焊料接合 (接腳)距離的釋例性描點圖。 一圖為由封叙中心之焊料接合直徑值對焊料接合 (接腳)距離的第二釋例性描點圖。 口 …圖為在釋例性球格陣列封裝上焊料接合之第_ 配靖料接合直徑之一誇張表示的平面圖。 一第0圖為在一釋例性球袼陣列封裝上焊料接合之第 二配置的焊料接合直徑之—誇張表示的平面圖。 第圖示意地顯示在-釋例性球格陣列封裝上被辨 1焊料接合與其鄰近焊料接合的相關性,其可被第1圖 之電路檢查系統施用。 f U圖為用封裝型式與誤差大小分類之靠近相鄰焊 的於σ直值誤差的描點圖,其可用第1圖之電路檢查系統 的按曰記錄所創立。 弟13圖為一描點圖,顯示靠近相鄰谭料接合直徑誤差 14 五、發明說明(丨2) ’’外:析’其可用第1圖之電路檢查系統被施用。 、… 圖為流程圖,顯示用於採用測試門檻值之方 法’其可用第1圖之電路檢查系統被實作。 第1 5圖為功能方塊圖,顯示第丨圖之焊料接合分析 應用程式的釋例性元件。 、乐16圖為—流程®顯示用於辨識焊料接合瑕蔽之方 法,其可用第1圖之電路檢查系統被實作。 本發明針對上面列的問題被導向一種系統及方法。特 別重要的是-種改良的焊料接合檢查系統被組配以施用該 方法,用於採用測試門摇值,其可顯著地減少在為每一各 引烊料接口刀析一個以上的測量之際被找出的冑「瑕疵」 數目。依照用於採用測試門檻值之方法被組配的改良焊料 接合檢查系、統在設定通過/失敗準則時考慮到印刷電路裝 置之安裝表面與印刷電路板間距離的可接受之變異。更明 確地况於才木用该測試門橄值之方法在辨識相鄰焊料接 合間高頻率的變異時接受在印刷電路裝置整個安裝表面之 相鄰焊料接合被測量的參數之低頻率的變異。 為促進σ亥糸統與方法之描述,一釋例性系統參照該等 圖被討論。該釋例㈣統與相關的方法僅就說明之目的被 提供。各種修改為可行的而不致於偏離其發明性概念。 例如,釋例性資料、圖與相關的描述之焦點為被固定 於B G Α封裝與焊料接合相關的直#測量。熟習本技藝者將 了解在其他焊料接合型式(即非球形接合)發生的低頻率之 變異至少疋因造成BGA封裝與印刷電路板纒繞的相同原 535473 五、發明說明 因所致。 八—依照較佳實施例,如由加州Palo Alto之Agllem科技 =司可取*之5DX χ光檢查系統的市面上可購得之電路板 檢,系統記錄在-印刷電路總成(即在印刷電路板被植入 之叙置)上有關數個接腳(即焊料接合介面)的位置資訊。节 焊料接合檢查系統使用該位置資訊將每一各別印刷電料 上-組相鄰的接腳與數個接腳的每一個配上關係。在記錄 ㈣觀察之每-焊料接合的特徵之至少—實際測量後,該 焊料接合檢查系統為每_各別印刷電路板相關的每一測量 產生It過/失敗準則。其考慮到被記錄之實際測i的可接 受(即低頻率)之變異。焊料接合之釋例性可為被測量的厚 度、形狀、小丘高度、焊料體積及其他。 數種估計技術可被用以為印刷電路板之每一接腳辨識 各別化的期望值。例如’ 一實施例使用自相鄰焊料接合測 量被導出之統計值以估計目前受到測試之期望測量。統計 值可包括平均值、均數、中位數等。在此第—實施例的變 形中,由整個安裝表面被記錄之測量在估計該期望值時可 被抽樣。換言之,被用以輕合100個接腳裝置之每一焊料 接合㈣計值可由來自整個100個接腳裝置之焊料接合的 部分集合之實際測量被決定。 -第二實施例對測量資料實施二維的多項式配適為受 到測試之焊料接合辨識—期望測量值。—第三實施例對測 量結果利FOU⑽分析以辨識相鄰桿料接合之高頻率的 變異。反Fourler變換可被用以為特定的焊料接合辨識-期 16 535473 發明說明 望測量值。一旦受到測試之一封裝的每一焊料接合之相關 期望測量值被辨識,焊料接合檢查系統為每一所論及之特 疋烨料接合的貫際測量值實施與被相關的期望值之比較以 產生一誤差值。在每一印刷電路裝置上受到測試之各別焊 料接合的相關誤差值被辨識後,界外分析可被實施以為每 一焊料接合導出慣用的通過/失敗準則。These and other factors can—the effect of & A ^ ^, make the printed circuit assembly contain: several devices (such as array packaging), which have significant changes in the; measurement. In the case of a ball grid array where you " such as a badly connected surface conductor is substantially equal to (in the case of installation applied to BGA packages. Hopewell), the increase in the distance between the bga women's clothing surface and the printed circuit board will increase As a result, more than one solder joint is elongated, making the retro t solder joints. In general, the lighter the tn is in the pre-reflow condition, the closer it is to the center of the array, and its reflow condition collapses. Columns crash more often. m "Although the ball grid array near the outer edge of the array, the entire majority of the measured solder on the column package, the wide printed circuit assembly and the array package are related to the electrical or electrical via the corresponding solder. The joints are properly connected. The relevant differences in the various measurements of solder joints (in the case where the corresponding solder joints are properly connected) are acceptable for the entire number of solder joint measurements; representative of the affected variation. The wide range of solder joint measurement makes it extremely difficult to set the pass / fail gate value for various solder joint measurements that accurately reflect the true electrical and physical conditions of each material joint. In practice, it is necessary to use a dry material joint for a concrete device. It is problematic for the inspection system to accurately detect "open circuits". Not only is it time consuming to reproduce each BGA package, it is identified as having a < <, > defective solder joint test results, a full-featured test lacking a printed circuit (i.e., fully embedded printed circuit board) assembly, or In-circuit testing of components' requires 100% electrical testing. Non-destructive certification methods are not available. 535473 5. Description of the invention (9) The accuracy of the conclusion of the defect is independently confirmed. In response ', some manufacturers already have elliptical gaskets using various printed circuit devices. The oval-shaped pads cause the solder balls to form an oval solder joint when the pads are properly bonded by the solder. These ^ circle = cymbals allow detection of "open" solder joints while the associated solder image remains circular. Although there are other improvements in detecting solder joints, it wants to have improved systems and methods for improving solder joint inspection systems: precision; it is based on accurately identifying the PCBs used to physically and electrically connect printed circuit boards. Defective solder joints for various printed circuit devices that take into account acceptable variations in the printed circuit device being measured. In response to these and other deficiencies in the known art, a solder joint inspection system and method is used to apply test thresholds to solder joint defects. In one configuration, the material bonding inspection system records the position information of several pins (ie, solder bonding interface) on each printed circuit device of the circuit assembly (that is, a fully implanted printed circuit board) and the printed circuit. The distance between the mounting surface of the assembly and the device package is changed: failure: Tan, the material joint inspection system is generated for each-Tan material joint-expected to pass: = straight f to allow for the group itching on each printed circuit device Material junction = acceptable variation of recorded true value. Briefly describe, on-the-shelf ... an improved itchy material joint inspection system for measuring m-fluid use for solder joints; and a data storage facility is implemented. … Recorded in the printed circuit assembly for each “tank material connection measurement,” is: heart === is considered to be provided-a method for identification ... one by one 'This method can be encapsulated into the following steps: Welding 12 V. Description of the invention 10 Expected expected value of material bonding, which takes into account the variation in the height of the printed circuit device against the printed circuit board; a comparison is recorded for each material bonding = an error value is generated for each solder joint; As well as identifying out-of-bounds measurements, other embodiments of the system can be seen as providing a method for expanding the rigidity test threshold. In this regard, this method can be summarized into the following steps :: Obtaining position information for several pins on a printed circuit device; Obtaining information about the height variation of the entire mounting surface of the printed circuit device; 2 recordings Measurements of several solder joints related to the plurality of pins; response to = = height variation of the mounting surface to estimate the possible range of the respective solder joints = range; and setting at least one threshold value to respond to the range A boundary. Other systems, methods, and features related to the use of electrical connection defects that take into account acceptable solder joint diameter variations in printed circuit board assemblies will become apparent to those skilled in the art by reviewing the following figures and detailed description understandable. It is intended that all such systems, methods, and features included in this description be in the field of systems and methods for identifying defects that are protected by the scope of the attached patent application. The system and method for confirming the electrical connection defect can be better understood with reference to the following drawings. The components in the figure do not necessarily emphasize their scales, but they clearly emphasize the principle of the test method in confirming the electrical connection between more than one of the printed circuit boards. Furthermore, the same component number is designated as the corresponding part in each drawing. FIG. 1 is a block diagram of an exemplary circuit inspection system. Figure 2 is a description of the solder joint diameter value of an exemplary ball grid array. 535473 5. Description of the invention (n figure. Figure 3 is a drawing of the solder joint diameter value of another exemplary ball grid array. Brother Fig. 4 is a side view of the influence of the ball grid array and winding on the respective solder joint diameters. / Fig. 5 is the ball grid array winding and the printed circuit board tilt are consumed and the respective solder joint diameter is wound. Figure 6 is a plan view of an exemplary ball grid array device, showing the method used to determine the distance from the center of a device for two-material bonding. #Figure is solder bonding by the center of I Exemplary tracing diagram of diameter value to solder joint (pin) distance. A picture is the second exemplary tracing diagram of solder junction diameter value to solder joint (pin) distance from the center of mouth. Mouth ... The figure is an exaggerated plan view of the solder joint on the exemplary ball grid array package. One of the second configuration of solder joint on an exemplary ball grid array package. The solder joint diameter is an exaggerated plan view. The intention is to show the correlation between the identified solder joint and its adjacent solder joint on the -exemplary ball grid array package, which can be applied by the circuit inspection system of Fig. 1. f U is the proximity of the package type and error size classification. The tracing point diagram of the σ straight value error of the adjacent welding can be created by using the daily record of the circuit inspection system in Fig. 1. Figure 13 is a tracing point diagram showing the error of the joint diameter near the adjacent material. 14 5 、 Explanation of the invention (丨 2) '' External: Analysis '' can be used in the circuit inspection system shown in Figure 1.… The figure is a flowchart showing the method used to use the test threshold value. '' It can be used in the circuit inspection shown in Figure 1 The system is implemented. Figure 15 is a functional block diagram showing the exemplary components of the solder joint analysis application shown in Figure 丨. Figure 16 is-Process ® shows a method for identifying solder joint flaws, which The circuit inspection system of Fig. 1 can be implemented. The present invention is directed to a system and method for the problems listed above. It is particularly important that an improved solder joint inspection system is configured to apply the method for use in testing. Door swing value, which can significantly reduce the number of "defects" found during the analysis of more than one measurement for each material interface. Improvements that are assembled in accordance with the method used to test threshold values The solder joint inspection system takes into account the acceptable variation in the distance between the mounting surface of the printed circuit device and the printed circuit board when setting the pass / fail criteria. More specifically, Caimu uses the test gate method to identify High-frequency variation between adjacent solder joints accepts low-frequency variation of measured parameters of adjacent solder joints across the entire mounting surface of a printed circuit device. To facilitate the description of the sigma system and methods, an exemplary system reference The diagrams are discussed. The interpretation system and related methods are provided for illustrative purposes only. Various modifications are possible without departing from their inventive concepts. For example, the explanatory materials, diagrams, and related descriptions The focal point is a direct measurement related to solder bonding that is fixed to the BG A package. Those skilled in the art will understand that the low-frequency variation that occurs in other solder joint types (that is, non-spherical joints) is at least the same reason that caused the BGA package to circulate with the printed circuit board. 535473 5. Description of the invention. Eight—According to a preferred embodiment, if a commercially available circuit board inspection is performed by Aglolem Technology of Palo Alto, Calif. = 5DX x-ray inspection system available from the company, the system is recorded in the printed circuit assembly (ie, in printed circuit The board is implanted) on the position of several pins (ie solder joint interface). The solder joint inspection system uses this positional information to match each of the adjacent groups of pins on the printed electronics to each of the plurality of pins. After recording ㈣observed-at least-features of solder joints-actual measurements, the solder joint inspection system generates an IT failure / failure criterion for each measurement associated with each individual printed circuit board. It takes into account the acceptable (i.e. low frequency) variation of the actual measured i recorded. Exemplary solder joints can be measured thickness, shape, hill height, solder volume, and others. Several estimation techniques can be used to identify individualized expectations for each pin of the printed circuit board. For example, an embodiment uses statistics derived from adjacent solder joint measurements to estimate the desired measurement currently being tested. Statistics can include average, mean, median, and so on. In a variation of this first embodiment, measurements recorded from the entire mounting surface may be sampled when estimating the expected value. In other words, the value of each solder joint used to lightly close the 100-pin device can be determined from the actual measurement of the set of solder joints from the entire 100-pin device. -The second embodiment implements a two-dimensional polynomial fit to the measurement data for solder joint identification under test-the desired measurement value. -The third embodiment uses a FOU⑽ analysis of the measurement results to identify the high-frequency variation in the joining of adjacent bars. The inverse Fourler transform can be used to identify specific solder joints-Issue 16 535473 Description of the Invention Expected measurements. Once the relevant expected measurements for each solder joint of one of the packages under test are identified, the solder joint inspection system performs a comparison of the inter-period measurements of each of the special joints in question with the associated expected values to produce a difference. After the associated error values for the individual solder joints tested on each printed circuit device have been identified, out-of-bounds analysis can be performed to derive conventional pass / fail criteria for each solder joint.

現在參照第1圖,所顯示者為一功能方塊圖,顯示在 一釋例性焊料接合檢查系統100内之各種元件。一般而 言,第1圖顯示以電腦為基礎之焊料接合檢查系統的各種 功能構建方塊,其可施用各種用於辨識及確認焊料接合瑕 A之方法 般而5,该焊料接合檢查系統1 〇〇可包含廣 泛種類之有線與(或)無線計算裝置的任何之一,如桌上電 腦、攜帶式電腦、專用伺服器電腦、多處理器計算裝置。 太干料接合檢查系統100不管其特定配置為何,例如可包含 處理裝置102、記憶體丨丨〇、一個以上的使用者介面裝置 u〇、顯示器130、及一個以上的輸入/輸出(1/〇)裝置14〇, 其每一個經由一區域介面118被連接。 處理裝置102可包括任何訂製或市面上可購得之處理 為、配於焊料接合檢查系統1〇〇之數個處理器之一輔助處 理為或一中央處理單元(cpu)、以半導體為基礎之微處理 $ (微晶片之形式)、一巨集處理器、一個以上的依用途而 定之和體電路(ASIC)、數個適當地被組配之數位邏輯閘、 及其他相當習知之電子組配,包含各別地與成各種組合之 分散元件,以協調焊料接合檢查系統1〇〇之整體作業。 17 535473 五、發明說明(l5 記憶體110可包括依電性記憶體元件(如隨機存取記 憶體(RAM如DRAM,SRAM等))與非依電性記憶體元件(如 R〇M、硬碟'磁帶、CDROM等)的任一組合。該記憶體1⑺ 典型地包含一 Ο/S 112、一個以上的應用程式,如焊料接 合分析應用程式114。具有本技藝之一般技術人員將了解 記憶體110可(且典型上將)包含其他元件,其為了簡潔之 目的被省略。這些可包括主程式被組配以控制焊料接合檢 查機構之各種層面。 。& 一個以上的使用者使用者介面裝置120包含這些元 件使用者可用與其焊料接合檢查系統1 〇〇互動。例如, 在丈干料接合檢查系統1〇〇包含個人電腦(pc)的情形中,這 些元件可包含—鍵盤與—滑鼠。此處焊料接合檢查系統 ⑽,期待被用於極端的環境(如靠近一焊料流動機器) 中广』兀件可包含功能鍵或按鈕、-觸摸敏感螢幕、一 尖筆等(未畫出),顯示器130可包含一電腦監視器或%用 之電漿螢幕、或者如所欲地之液晶顯示器(LCD)。 進一步參照第!圖’一個以上的1/〇裝置i4〇可被採 ^促進連接焊料接合檢查系統⑽至另一系統與(或)裝 匕括個以上的串列、平行的小電腦系統介面 F. 通用串列匯流排卿)、鹏⑽(如 八::二人):與(或)其他介面元件’其可被用以通訊式地搞 用5 檢查系制。與-個以上的遠端資料儲存裝置 ==測試測量結果。網路介面裝包含各種元件 被用以在-網路(未畫出)上傳輸與⑷接收資料。舉例而 18 535473 五、發明說明(a) 5 ’该網路介面裝置150彳包括與輸入與輸出通訊的裝 置,二如為一調變器/解調器(如數據機)、無線(如射頻(RF)) 收叙态、一電話介面、—播接器、一路由器、一網路卡等。 各種軟體與(或)拿刃體將被用以管理、協調、測量、記 錄估。十及比較期望值與測量值以產生誤差值、以及對所 f生之誤差值實施界外值分析及其他功能。負責使用基準 焊料接合檢查系統100相關的這些與其他功能之相關軟體 與⑷拿刀體可被儲存於任何電腦可讀取的媒體上以便被任 何電腦相關系統或方法中被使用。在此文件之文意中,Φ 腦可讀取的媒體代表電子、磁性、 兒 設施,其可包含或儲存電 二〜4置或 或方法中被使用。這此程式=便被任何電腦相關系統 便在指令執行系統^置;^亥電腦可讀取的媒體以 、相關使用,如以電腦為基礎之 系、洗、3有處理H之m其他能自彳旨令執行 :還指=行該等指令之其他系統。在此文件… ^可項取的媒體」可為儲存、通訊、傳播或輸 送程任何設施以便在指令執行系統、裝置使: 繼可讀取的媒體例如為電子 紅外線或半導體系統、裝置或播放媒體,但:二 腦可讀取的媒體之更特定的例子(非 包: -個以上的配線之電氣連接 0)。括具有 取記憶體(讀)、唯讀記憶體(=的_兹片、隨機存 記憶體(EPROM)、電氣式 W拭m隹讀 阳_)、或快閃記憶體、光纖二讀記憶體 了攜式唯讀記憶體光 19 五、發明說明(17) 碟(CDROM)。注意該電腦可讀取的媒體甚至可為紙或其他 適當的媒體,其上在程式可電子式地經由例如對該紙或其 他媒體光學掃描時該程式被列印,然後在必要時以適當的 方式被編譯、解釋或處理、再被儲存於電腦記憶體内。 第2圖顯示自特定型式之BGA裝置封裝被測量及被記 錄的樣本資料點的描點圖2〇〇。如第2圖之描點圖顯示者, 該貧料包括為350個以上受檢查之焊料接合包括以mn為 單位的直徑。靠近描點圖左邊相關接腳的焊料接合其中之 一已被習知為有瑕疵的,其以元件編號21〇表示。圖示之 知料接合210具有約Μ」mU之直徑且近位於bga封裝被 記錄之直徑的中心範圍。更明確地說,各別焊料接合被記 錄之直彳二在29至31.5 mil範圍内,直徑之總變異為2 5 mi1。在此資料圖中清楚地指出會用主要可接受的焊料接合 之瑕疵隔離與(或)辨識已知的焊料接合來設定高或低門檻 值為不可能的。後果為,在此資料配置下沒有可接受的解 法用於設定通過/失敗門檻值。 旦類似地’第3圖顯示自另一型式之BGA裝置封裝被測 里及被圮錄的樣本資料點的描點圖2〇〇。如第3圖之描點 圖顯示者,該資料包括為35〇個以上受檢查之焊料接^包 括以md |單位的直徑。各別焊料接合被記錄之直徑在 26.75至3 1.5 mil範圍内,直徑之總變異為4·75 _,近乎 第2圖呈現之相關BGA裝置之焊料接合所記錄的直徑範圍 之兩七t要的疋,雖然所記錄的直徑範圍有戲劇性的增 加,在描點圖中並無單—的焊料接合為有職的 535473 發明說明 一步說明唯獨地根據各別焊料接合直徑的通過/失敗門檻 值之直接應用可能會導致不精準的瑕疯判定。Referring now to FIG. 1, shown is a functional block diagram showing various components within an exemplary solder joint inspection system 100. As shown in FIG. Generally speaking, Fig. 1 shows various functional building blocks of a computer-based solder joint inspection system, which can be applied with various methods for identifying and confirming solder joint defects A. The solder joint inspection system 1 〇〇 It can include any of a wide variety of wired and / or wireless computing devices, such as desktop computers, portable computers, dedicated server computers, multiprocessor computing devices. Regardless of its specific configuration, the too dry material joint inspection system 100 may include, for example, a processing device 102, a memory 丨 丨 〇, more than one user interface device u〇, a display 130, and more than one input / output (1 / 〇 ) Devices 14, each of which is connected via a regional interface 118. The processing device 102 may include any custom-made or commercially available processing unit, one of several processors equipped with the solder joint inspection system 100, an auxiliary processing unit, or a central processing unit (cpu), semiconductor-based Microprocessor $ (in the form of a microchip), a macro processor, more than one application-specific sum circuit (ASIC), several appropriately-equipped digital logic gates, and other fairly familiar electronic groups Distribution, including discrete components in various combinations, to coordinate the overall operation of the solder joint inspection system 100. 17 535473 V. Description of the Invention (l5 The memory 110 may include electrical memory elements (such as random access memory (RAM such as DRAM, SRAM, etc.)) and non-electric memory elements (such as ROM, hard Disk, tape, CDROM, etc.). The memory 1 记忆 typically contains 10 / S 112, more than one application, such as solder joint analysis application 114. Those skilled in the art will understand the memory The 110 may (and typically will) include other components, which are omitted for brevity. These may include the main program being configured to control various aspects of the solder joint inspection mechanism. &Amp; More than one user user interface device 120 contains these components that a user can interact with with their solder joint inspection system 100. For example, in the case where the dry material joint inspection system 100 includes a personal computer (pc), these elements may include-a keyboard and a mouse. The solder joint inspection system here is expected to be used in extreme environments (such as near a solder flow machine). The components can include function keys or buttons, -touch sensitive Screen, a stylus, etc. (not shown), the display 130 may include a computer monitor or a plasma screen, or a liquid crystal display (LCD) as desired. Further referring to the first! FIG. / 〇device i4〇 can be adopted to facilitate the connection of the solder joint inspection system to another system and / or install more than one serial, parallel small computer system interface F. Universal serial busbar), Peng如 (such as eight :: two people): and / or other interface components' It can be used to communicate with the 5 inspection system. And-more than one remote data storage device == test measurement results. The network interface contains various components used to transmit and receive data over a network (not shown). For example, 18 535473 V. Description of the invention (a) 5 'The network interface device 150 彳 includes devices that communicate with input and output, such as a modulator / demodulator (such as a modem), and wireless (such as radio frequency). (RF)) receiving status, a telephone interface,-a connector, a router, a network card, etc. Various software and / or blades will be used to manage, coordinate, measure, and record estimates. Ten and compare the expected value with the measured value to generate an error value, and perform out-of-bounds analysis and other functions on the resulting error value. These and other functions related to the use of the reference solder joint inspection system 100 and the gripper body can be stored on any computer-readable medium for use in any computer-related system or method. In the context of this document, Φ brain-readable media represent electronic, magnetic, and child-friendly facilities that can contain or store electricity and are used in methods. This program = will be set by any computer-related system in the command execution system; ^ Hai computer-readable media, related use, such as computer-based systems, washing, and other processing彳 Order execution: also refers to other systems that execute these instructions. In this document ... ^ "optional media" may be any facility that stores, communicates, disseminates, or transports, in order to execute the system, the device: the following readable media such as electronic infrared or semiconductor systems, devices or playback media , But: a more specific example of a brain-readable medium (non-package:-electrical connection of more than one wiring 0). Including memory (read), read-only memory (= _ slices, random access memory (EPROM), electrical W erasure reading _), or flash memory, optical fiber second-read memory The portable read-only memory light 19 V. Description of the invention (17) CDROM. Note that the computer-readable media may even be paper or other suitable media on which the program can be printed electronically via, for example, optical scanning of the paper or other media, and the program can then be printed with appropriate The methods are compiled, interpreted or processed and then stored in computer memory. Figure 2 shows a sample point plot of sample data points measured and recorded from a particular type of BGA device package. As shown in the dot plot in Figure 2, the lean material includes more than 350 solder joints under inspection including the diameter in mn. One of the solder joints near the relevant pin on the left side of the trace is known to be defective and is indicated by the part number 21 °. The illustrated splicing junction 210 has a diameter of about MUm and is located near the center of the recorded diameter of the bga package. More specifically, the respective solder joints are recorded in the range of 29 to 31.5 mil, with a total variation in diameter of 2 5 mi1. In this data sheet it is clearly stated that it would not be possible to set a high or low threshold using the main acceptable solder joint defect isolation and / or identification of known solder joints. As a consequence, there is no acceptable solution for setting pass / fail thresholds under this profile. Similarly, FIG. 3 shows a tracing diagram of a sample data point recorded from another type of BGA device package and recorded in FIG. As shown in the tracing diagram in Figure 3, this data includes more than 350 solder joints inspected, including the diameter in md | units. The recorded diameter of each solder joint is in the range of 26.75 to 3 1.5 mil, and the total variation of the diameter is 4.75 mm, which is nearly two to seven tons of the diameter range recorded by the solder joint of the related BGA device shown in Figure 2. Alas, despite the dramatic increase in the range of recorded diameters, there is no single order in the tracing-solder joints are in place. Application may lead to inaccurate blemish determination.

現在參照第4圖之側面圖。如第4圖顯示者,印刷電 路總成400可包含一 BGA裝置41〇與一印刷電路板 經由:):干料接合4丨2在實質上被對齊之點電氣式且物理式地 被連接。重要的是要注意該圖示並未按比例顯示。形成印 刷電路裝置400所使用之BGA裝置4丨〇與印刷電路板42〇 的相對高度(即厚度)會常有變化,且由數個焊球形成之各 種焊料接合的尺寸會顯得很小。Reference is now made to the side view of FIG. 4. As shown in FIG. 4, the printed circuit assembly 400 may include a BGA device 410 and a printed circuit board via :): The dry material bonding 4 丨 2 is electrically and physically connected at a point where it is substantially aligned. It is important to note that the icon is not to scale. The relative height (ie, thickness) of the BGA device 4o and the printed circuit board 42o used to form the printed circuit device 400 will often change, and the size of various solder joints formed by several solder balls will appear small.

如第4圖之側面圖進一步顯示者,BGA裝置41〇會纒 繞成向上之曲線使得BGA裝置410的相反端部由印刷電路 板420之安裝(即上層)表面被「拉開」。BGA裝置可能 因封裝内熱澎脹不對齊與(或)如印刷電路板所施加之應力 的其他理由而纒繞。焊料接合412在BGA裝置41〇之材料 月ί) ~部而设立纒繞。因為每一焊球中之材料(焊料)的體積 維持固定,這些連接點以安裝表面(BGA裝置41〇之下側 與印刷電路板420之上層表面)彼此由“z”維度被去除, 而在聚焦平面450被測量時將具有減小的直徑◦如先前解 釋者,BGA裝置纒繞會造成焊料接合412之被測量直徑的 大變異。由於BGA裝置410之向上曲度,靠近BGA裝置 410中心之焊料接合典型上比BGA裝置41〇外層邊緣之焊 料接合具有較大的直徑。 如第5圖之侧面圖顯示者,印刷電路總成5〇〇可包含 BGA衣置5 1 0與一印刷電路板“Ο經由焊料接合4 ^ 2 21 五、發明說明(19 ) 在貝貝上被對背之點電氣式且物理式地被連接。第5圖與 第4圖一樣未按比例被顯示。在第$圖中至少有二因素造 成BGA裝置510之相關焊料接合測量的可接受變異之結 果:印刷電路板纒繞與BGA裝置傾斜。BGA裝置在有一 個以上的角度與其他角度有不同的“z”值時會傾斜。在 第4圖顯示之BGA裝置纒繞的情形中,印刷電路板纒繞可 能為非線性且在X與γ維度二者(即向著該頁紙)變化。後 果為這些連接點以安裝表面(BGA裝置5 1 0之下侧與印 刷電路板520之上層表面)彼此由“z,,維度被去除,而在 聚焦平面450被測量時將具有減小的直徑。如先前解釋 者B G A衣置纏繞會造成焊料接合5 12之被測量直徑的大 變異,而難以使用此資訊獨一地作為焊料接合檢查系統 1〇〇(第1圖)所使用之通過/失敗準則的指標。 現在苓照第6圖之平面圖,其顯示將一釋例性 I置600與位於該裝置安裝表面之數個焊料接合6丨2配出 關係之方法。如第6圖顯示者,焊料接合檢查系統ι〇〇被 用以辨識目前受到測試之特定BGA裝置6〇〇的中心。 在於X-Y表面定出BGA裝置6〇〇之中心61〇後,焊料接 合檢查系統1〇〇可被用以決定及儲存由每一各個焊料接合 612之中心自中心61 〇的距離。 這些距離測量之重要性在觀察第7與8圖之樣本點圖 後將被了解。第7圖顯示描點圖7〇〇為釋例性Bga裝置型 式的各種傳導點用自BGA裝置6〇〇(第6圖)之中心6⑺的 距離來配置焊料接合直徑。與第2圖之樣本點相反的θ該 535473 五、發明說明(20 ) 資料呈現以mil表示之焊料接合直徑與焊料接合自中心 610距離間稍微有線性關係。 類似地,第8圖顯示描點圖8〇〇為釋例性bga裝置型 式的各種傳導點用自BGAt置之中心的距離來配置焊料 接合直徑。與第3圖之樣本點及第7圖之樣本點相反 的是’該貧料呈現以mil表示之焊料接合直徑與焊料接合 自中心距離間麵有線性關係。然而在第8圖被分析及被 描點之BGA封裝型式的情形中,線性關係之斜率針對各別 焊料接合自該裝置中心之距離增加為負的。清楚的是,這 些觀點在4 BGA t置之每一分離的焊料接合組配一個以 上的門檻值時,在該等門健考慮到沿著χ_γ表面被配置 之各種焊料接合可被期待之可接受的變異下是可能的。 例如,知料接合直徑之上門檻值可反映平均直徑對來 自與特定印刷電路裝置相關之焊料接合中心的距離。該上 m 監值可用為特定印刷電路裝置被記錄之平均直徑對距離 白曰勺固疋正差值之線性方式被施用。超過安裝表面間距離不 是相當接近之上Η檻值的實際焊料接合直徑可被視為具有 超過的焊料且可被標示以便進一步調查。 類似地,焊料接合直徑之下門檻值可反映直徑對來自 與=刷電路裝置之焊料接合中心的距離。該下門播值可用 為特定印刷電路裝置被記錄之平均直徑對距離的固定負差 值之線性方式被施用。超過安裝表面間距離不是相當接近 之下門檻值的實際焊料接合直徑可被視為具有開放電路狀 况且可被標示以便進一步調查。 535473 五、發明說明(2ι 第9與1 0圖呈現二個釋例性BGA裝置之平面圖。第 9圖顯示之BGA裝置900呈現在靠近該裝置外部邊緣以行 與列被配置之焊料接合。如第9圖之表面圖顯示者,bga 衣置900之女哀表面貫際上與焊料接合檢查系統1 〇〇(第1 圖)之X與Y維度共表面。為了說明之目的,該誇張的直 徑圖揭示在整個安裝表面之數個焊料接合912。所顯示之 各別焊料接合的直徑實際上是不按比例以顯示在bga裝 置_之各別焊料接合上的整個焊料接合直徑變異。如第 9圖之表面圖顯示者,BGA裝置_揭示每一焊料接合川 之直徑的非線性“z度變異。此處,該焊料接合直徑 變異在整個BGA裝置的安裝表面是多少為隨機的。 第10圖顯示之BGA裝置1000呈現對該裝置之安裝表 中U以行與列被配置之焊料接合1012。裝置1000 女衣表面貝際上與焊料接合檢查系統1(第1圖)之X 維度共表面。為了說明之目的,該誇張的直徑圖揭示 2個安裝表面之數個焊料接合⑻2。如第lG圖之表面 二1不者,BGA裝置1000以典型的方式具有纒繞,靠近 4邊緣之焊料接合具有最小的直徑,而比較靠近BGA裝 置_中心之焊料接合具有最大的直徑。 施改良的焊料接合檢查系、统100(第丨圖)之較佳實 她例與用於採用測試門檻值之 就印刷電路〜 之相關方法,-附近相鄰分析 趣於 衣置(如β G A裝置_)的每—相關焊料接合被 二量㈣電路裝置之整個痒料接合的可接受 文"。该技術假設封裝上之焊料接合測量有變化, 24 五、發明說明(22 ) 但相鄰焊料接合間之測量變化應該很小。BGA裝置的每— 相關焊料接合可如第n圖之部分表面圖顯示地被配以附 近相鄰的焊料接合。當焊料接合如焊料接合1112a之情形 地位於沿著BGA裝置謂的邊緣時,五個焊料接合⑴^ 將形成焊料接合1112a附近相鄰的元件組。當焊料接合如 焊料接合1112b之情形地位於BGA裝置剛的中心時, 二個焊料接合⑴3&將形成焊料接合1U2b附近相鄰的元 :牛組。如進一步顯示者’一個「場」焊料接合(即被八個相 #焊料接合圍繞之-焊料接合)可被配以八個焊料接合。注 意’除了第η圖顯示者外,有數個不同的方法可被用以選 擇鄰近的接腳(即焊料接合)。例如,在特定數目之行或列 ㈣焊料接合可在各種其他方法中被考慮用於以受到測試 之焊料接合來配以相鄰的焊料接合。 種不同的方法可被用以自特定附近相鄰的焊料接合 組的焊料接合同—測量賴來估計適當的焊料接合測量。 :如在一附近相鄰組内每—焊料接合的相關測量值可用統 :來為被辨識之焊料接合設定-期望測量。就此而 d則里之均數、中位數或平均測量值可被用作為該 ?辨識之焊料接合的期望值。在某些配置中,-估計值可 «所測量之特徵被預測。例如一焊料接合之高度可被測 里,則在知道原始焊球之體積下,相_料接合 徑可被決定。 旦金:後一附近相鄰誤差值可藉由比較實際的焊料接合測 里”錢近相鄰組之測量就該谭料接合被導出的期望 五、發明說明(23) 為每一數個焊料接合被計算。先前描述之釋例性BGA裝置 型式的附近相鄰誤差資料在第12圖之描點圖被顯示。就此As further shown in the side view of FIG. 4, the BGA device 41 will be wound into an upward curve so that the opposite end of the BGA device 410 will be "pulled away" from the mounting (ie, upper) surface of the printed circuit board 420. BGA devices can be entangled due to thermal swelling misalignment within the package and / or other reasons such as stress applied by the printed circuit board. Solder bonding 412 is set up around the material of the BGA device 41〇. Because the volume of the material (solder) in each solder ball remains fixed, these connection points are removed from each other by the “z” dimension with the mounting surface (the lower side of the BGA device 41〇 and the upper surface of the printed circuit board 420), and The focus plane 450 will have a reduced diameter when measured. As explained previously, coiling the BGA device will cause large variations in the measured diameter of the solder joint 412. Due to the upward curvature of the BGA device 410, the solder joint near the center of the BGA device 410 typically has a larger diameter than the solder joint on the outer edge of the BGA device 410. As shown in the side view of FIG. 5, the printed circuit assembly 5000 can include a BGA garment 5 1 0 and a printed circuit board "0 bonded by solder 4 ^ 2 21 V. Description of the invention (19) on Babe The back-to-back points are electrically and physically connected. Figure 5 is not shown to scale as in Figure 4. In Figure $ there are at least two factors that cause acceptable variation in the associated solder joint measurement of the BGA device 510 The result: the printed circuit board is coiled and the BGA device is tilted. The BGA device is tilted when there is more than one angle and other angles have different "z" values. In the case of the BGA device coiled shown in Figure 4, printing Circuit board winding may be non-linear and change in both X and γ dimensions (ie toward the page). The consequence is that these connection points are mounted on the surface (the lower side of the BGA device 5 10 and the upper surface of the printed circuit board 520). The dimensions are removed from each other by "z," and will have a reduced diameter when the focal plane 450 is measured. As previously explained, BGA garment winding can cause large variations in the measured diameter of the solder joint 5 12 and it is difficult to use this information exclusively as a pass / fail criterion for the solder joint inspection system 100 (Figure 1). index of. Now, according to the plan view of FIG. 6, it shows an exemplary method for assigning an exemplary I 600 to a number of solder joints 6 and 2 on the mounting surface of the device. As shown in Figure 6, the solder joint inspection system ιο is used to identify the center of the particular BGA device 600 which is currently being tested. After the center 61 of the BGA device 600 has been determined on the X-Y surface, the solder joint inspection system 100 can be used to determine and store the distance from the center 61 to the center of each individual solder joint 612. The importance of these distance measurements will be understood after viewing the sample point plots in Figures 7 and 8. Fig. 7 shows the tracing points. Fig. 7000 illustrates various conductive points of an exemplary Bga device type. The solder joint diameter is arranged at a distance of 6 mm from the center of the BGA device 600 (Fig. 6). Opposite to the sample points in Figure 2 The 535473 V. Description of the Invention (20) The data shows that there is a slightly linear relationship between the solder joint diameter expressed in mil and the solder joint center 610 distance. Similarly, Fig. 8 shows the plotting point. Fig. 800 is an illustration of various conductive points of an exemplary bga device type using a distance from the center of the BGAt to configure the solder joint diameter. In contrast to the sample points in Fig. 3 and the sample points in Fig. 7, 'the lean material exhibits a linear relationship between the solder joint diameter expressed in mil and the distance from the center of the solder joint. However, in the case of the BGA package type analyzed and traced in Figure 8, the slope of the linear relationship is negative for the distance of the respective solder joints from the center of the device. It is clear that when these viewpoints are equipped with more than one threshold value for each separate solder joint set at 4 BGA t, it is considered acceptable for such gates to consider that various solder joints arranged along the χ_γ surface can be expected to be acceptable. Variations are possible. For example, the threshold above the expected bond diameter can reflect the average diameter pair distance from the solder bond center associated with a particular printed circuit device. The upper m-monitoring value can be applied in a linear manner as a positive difference of the recorded average diameter versus distance for a particular printed circuit device. An actual solder joint diameter that exceeds the threshold between mounting surfaces is not quite close to the threshold above, can be considered to have solder that exceeds and can be labeled for further investigation. Similarly, the threshold below the solder joint diameter can reflect the distance of the diameter versus the solder joint center from the brush circuit device. This door-to-door value can be applied in a linear fashion as a fixed negative difference in average diameter versus distance recorded for a particular printed circuit device. Actual solder joint diameters that exceed the threshold below the distance between the mounting surfaces are not quite close, can be considered to have an open circuit condition and can be marked for further investigation. 535473 V. Description of the Invention (2ι Figures 9 and 10 present two plan views of exemplary BGA devices. Figure 9 shows a BGA device 900 showing solder joints arranged in rows and columns near the outer edge of the device. The surface map in Figure 9 shows that the female surface of the bga garment 900 is consistently co-surfaced with the X and Y dimensions of the solder joint inspection system 100 (Figure 1). For illustration purposes, the exaggerated diameter The figure reveals several solder joints 912 over the entire mounting surface. The diameters of the individual solder joints shown are actually not to scale to show the variation in the overall solder joint diameter on the individual solder joints of the bga device. As shown in Section 9 The surface view of the figure shows that the BGA device _ reveals a non-linear "z-degree variation in the diameter of each solder joint. The solder joint diameter variation is random over the entire mounting surface of the BGA device. Figure 10 The BGA device 1000 shown presents solder joints 1012 arranged in rows and columns in the installation table of the device. The surface of the device 1000's women's clothing is coplanar with the X dimension of the solder joint inspection system 1 (Figure 1). in order to For the sake of clarity, this exaggerated diameter diagram reveals several solder joints 2 on the two mounting surfaces. As shown in the second and first surface of Fig. 1G, the BGA device 1000 has a winding in a typical manner, and the solder joints near the 4 edges have the smallest The diameter of the solder joint, which is closer to the center of the BGA device, has the largest diameter. A better example of the improved solder joint inspection system, system 100 (Figure 丨), and the printed circuit used to adopt the test threshold value A related method,-nearby adjacent analysis is interesting for clothing (such as β GA device_). Each of the related solder joints is accepted by the entire volume of the circuit device. It is assumed that the package There is a change in the solder joint measurement, 24 V. Description of the invention (22) But the measurement variation between adjacent solder joints should be small. Each and every relevant solder joint of the BGA device can be matched as shown in the partial surface diagram of n Nearby solder joints. When the solder joint is located along the edge of the BGA device as the solder joint 1112a, five solder joints ⑴ ^ will form adjacent elements near the solder joint 1112a. When the solder joint is located at the center of the BGA device, such as solder joint 1112b, the two solder joints ⑴3 & will form the adjacent element near the solder joint 1U2b: the cow group. As further shown, a 'field' Solder bonding (that is, surrounded by eight phase #solder bonding-solder bonding) can be matched with eight solder bonding. Note that in addition to those shown in Figure n, there are several different methods that can be used to select adjacent pins. (Ie, solder bonding). For example, a particular number of rows or columns of solder bonding may be considered in various other methods for matching adjacent solder bonding with the solder bonding being tested. A different method may be used Based on solder joints from a particular nearby solder joint group—measurements rely on estimates of appropriate solder joint measurements. For example, the relevant measurement values of each solder joint in a nearby adjacent group can be used to set the desired measurement for the identified solder joint. In this regard, the mean, median, or average measurement in d can be used as the expected value of the solder joint identified by this?. In some configurations, the -estimated value can be predicted. For example, the height of a solder joint can be measured, and the phase diameter of the material can be determined by knowing the volume of the original solder ball. Danjin: The value of the adjacent error in the vicinity of the latter can be compared with the actual solder joint measurement. "The measurement of the near-neighboring group of Qian can be derived from the expectation of the material joint. V. Description of the invention (23) For each several solders The joints are calculated. The adjacent adjacent error data of the previously described exemplary BGA device type is shown in the tracing graph of Figure 12.

而言,該資料描點圖12〇〇顯示該附近相鄰誤差與用BGA 裝置型式及誤差量被分類之接腳(即焊料接合)間的關係。 如a成的資料描點圖1200顯示者,釋例性bgA 346裝置 之誤差符號與釋例性BGA 347裝置之誤差符號顯著地不 同。 現在參照第13圖顯示之附近相鄰誤差界外分析描點 圖。該合成的資料描點圖1300顯示附近相鄰資料描點圖 1200(第12圖)之方盒描點圖分析。方盒描點圖為中央趨勢 與離散之一有效的視覺呈現。其同步地顯示第25、第5〇、 第75百分位數以及最小值與最大值。方盒描點圖之「方盒」 顯示中間或「最典型的」50%之值,而方盒描點圖之「腮 鬚」顯示較極端之值。各別BGA裝置之方盒13〇2與"Μ 之高度為代表性的四分之一位數間距(I〇R)。 其已被發現考慮比各別方盒1302,13〇4頂端後之 三倍以上的任何直徑誤差為BGA裝置之有效的通過/失敗 門捏值如帛13目之合成描點圖的方盒描點圖資料顯示 者,BGA 347裝置型式沒有界外附近相鄰誤差值超過企劃 值(IOR的三倍)。在此㈣中,改良的焊料接合檢查系統 1〇〇已精準地報告釋例性的BGA 347裝置之焊料接合無相 關的瑕疵。如在合成方盒描點圖1 3〇〇隹一 ^ 夕頌不者,就釋 例性B G A 3 4 6的相關焊料接合誤差界外分析不可有相同的 說法。如所顯示I,數個附近相鄰誤差值趨過企劃的誤差 26 535473In other words, the data trace diagram 1200 shows the relationship between the adjacent errors in the vicinity and the pins (ie solder joints) classified by the type and amount of the BGA device. As shown in the data plot 1200 of Figure A, the error sign of the exemplary bgA 346 device is significantly different from the error sign of the exemplary BGA 347 device. Now refer to Figure 13 to analyze the plot of the adjacent error bounds. The synthesized data trace graph 1300 shows a square box trace graph analysis of nearby data trace graphs 1200 (Figure 12). The square box plot is an effective visual representation of central trend and dispersion. It displays the 25th, 50th, and 75th percentiles as well as the minimum and maximum values simultaneously. The "square box" of the square box plot point chart shows the middle or "most typical" value of 50%, while the "branch and whisker" of the square box plot point chart shows the more extreme value. The height of the square box 1302 and the "M" of each BGA device is a representative quarter-digit pitch (OR). It has been found that considering any diameter error more than three times the top of each square box 1302, 1304 is an effective pass / fail gate pinch value of the BGA device, such as a square box trace of a composite plot of 目 13 mesh According to the dot chart data, the BGA 347 device type does not have an adjacent error value outside the boundary that exceeds the planned value (three times the IOR). In this case, the improved solder joint inspection system 100 has accurately reported that there are no defects associated with solder joints of the exemplary BGA 347 device. For example, in the case of drawing a synthetic box with a figure of 1300, it is impossible to say the same thing about the out-of-bounds analysis of the solder joint error related to the exemplary B G A 3 4.6. As shown in I, several nearby neighboring error values have passed the planned error 26 535473

27 53547327 535473

五、發明說明(25) 著X維度之一列焊料接合與(或)沿著γ維度之一行焊料接 合的某些其他特徵可被用以為該被測量直徑所形成之對應 的資料集定出最佳的適配多項式。然後結果所得之最佳適 酉一己多項式可被用以為每-焊料接合產生^期望值。在其他 貝靶例中,每一各別焊料接合之期望值可由在X維度得到 期望最佳適配結果與在γ維度得到期望最佳適配被決定成 為一數學結果。V. Description of the Invention (25) Certain other characteristics of solder joints in one column of the X dimension and / or solder joints in one row of the γ dimension can be used to determine the best for the corresponding data set formed by the measured diameter Adaptive polynomial. The resulting best-fit polynomial can then be used to generate a desired value for each solder joint. In other cases, the expected value of each individual solder joint can be determined as a mathematical result from the expected best fit result in the X dimension and the expected best fit result in the γ dimension.

或者,Fouiier分析可被施用至被測量之資料以為每一 力干料接合產生一期望值。就如已習知者,—變換為 Sm二與—之加權和的二維關係之表示。$進的變換由 連'_的或離散的空間定義域前進為總是為連續的頻率定義 域。因而,F_ier變換可仙以為被測量之直徑的高頻率 變化分析在頻率定義域中之資料。如習知地,高頻率變化 可經由適當地組配之遽波器被去除。然後反F_k變換可 被施用至該資料為每-焊料接合導出該㈣望值。不管被Alternatively, a Fouier analysis can be applied to the data being measured to produce a desired value for each force dry material joint. As already known,-transform into a two-dimensional representation of the weighted sum of Sm two and-. The transformation of $ jin advances from a continuous or discrete spatial domain to a continuous frequency domain. Therefore, the F_ier transform can be used to analyze the data in the frequency domain for high-frequency changes of the measured diameter. As is customary, high frequency variations can be removed via a properly configured chirp. An inverse F_k transform can then be applied to the data to derive the lookout value for each solder joint. Regardless of

用以產生該等期望值之方法為何,整個陣列封裝之被測量 值的低頻率(可接受的)欒显A ~ 又7 )又呉在邊寻期望值及因而修改之通 過/失敗門根值中被反映。 統100可被規劃程式以為界外物儲存及分析誤差。這些界 外物為在焊料接合測量未被期待(如高頻率)之變異,其已 接著如步驟⑷0顯示者’被改良的焊料接合檢查系統 ⑽可被規劃程式以比較在步彻被產生之估計值或期 望值與在步驟1406被記錄之每一焊料接合的各別被測量 值。如上述及在步驟1412顯示者,改良的焊㈣合檢查系 28 535473 五、發明說明(26) 被蚊㈣BGA裝置之瑕疲焊料接合的精準指標。其應被 了解第14圖之流程圖中顯示的處理步驟可被重複以完成 在印刷電路總成上被安裝之其他印刷電路裝置的焊料接合 分析。 第15圖為一功能方塊圖顯示第丨圖之焊料接合分析應 用程式114的各種釋例性元件。就此而言,焊料接合分析 應用程< U4T包括記錄邏輯1500被組配以接收被焊料接 合檢查系統丨00產生及分析之各種影像中被觀察之一個以 上的特徵值。該焊料接合分析應用程式丨14亦可包括邏輯 15 10被組配為受到測試之每一焊料接合決定一期望值。如 先前描述者,邏輯1510可使用任何數個方法用於為每一各 別的焊料接合產生一期望測量值,其考慮到相關裝置之安 裝表面與該印刷電路板間的變異。 如第15圖顯示者,記錄邏輯1500與邏輯151〇被可組 配以提供有關每一各別焊料接合之資訊至比較邏輯 1 520。其次,比較邏輯可被組配為受到觀察之每一各別焊 料接合決定藉由測量(即被記錄之測量值)與相關期望值間 里的差異。如在此方塊圖進一步被顯示者,該比較邏輯 1520可被組配以遞送每一各別誤差資料值至一誤差值資 料陣列1530。 接著,誤差值界外分析邏輯1 540可被組配為接收數個 各別5吳差值並對该資料貫施界外分析。在已於界外分析邏 輯1540辨識誤差值界外物後,焊料接合分析應用程式U4 可遞送與超過一門檻值之界外資料值相關的辨識資訊至界 29 535473 五、發明說明(27 =報告邏輯,550。該界外誤差報告邏輯⑽ :制接合提供位置與(或)其他辨識元至與該焊料接合檢 一糸統1GG相之-個以上的周邊裝置。 第16圖為流程圖’顯示用於辨識焊料接合瑕疲之方法 〇,其可被第1圖之電路檢查系統實施。就此而言,改 =料接合嶋統⑽可藉由測量及記錄與特 ^相關的每-焊料接合之實體特徵㈣始實施。在 置母一焊料接合後’改良的焊料接合檢查系統·可在步 驟1604使用有關陣列封裝纒繞與(或)印刷電路板纒繞的資 Λ來估计母一被測量之焊料接合的期望值。 接者,如步驟16〇6顯示者,改良的焊料接合檢查系统 ΤΙ可被組配以比較該期望值與各別的被測她^ 硯祭之母一焊料接合產生-誤差值。如步驟刪顯示者, ^U606被產生之數個誤差值可在一誤差資料值界外 分析中被施用以辨識有瑕疵的接合。 二隹…、改良的;C干料接合檢查系統1〇〇之特定實施例已在 則面描述與圖中就舉例之目的詳細地被揭示,其將被熟習 本技藝者了解到其變化與修改可不偏離本發明如下列申請 2利範圍所設立的精神地被作成。例如,與非陣列封裝式 置相關的其他焊料接合測量可被期望反映因各種理由之 低頻率變異’包括上述安裝表面間因“Z” |度變異所致 =果。因此,其有-個以上的相關技術用於為非陣列封 I焊料接合採用通過/失敗門檻值。本揭示欲包括這些技術 之應用。 30 535473 五、發明說明(28) 元件標號對照表 元件編號 譯 名 元件編號 譯 名 100 焊料接合檢查系統 550 聚焦平面 102 處理裝置 600 BGA裝置 110 記憶體 610 中心 112 作業系統(0/S) 612 焊料接合 114 焊料接合分析應程式 612a 焊料接合 118 區域介面 612b 焊料接合 120 使用者介面裝置 700 描點圖 130 顯示器 800 描點圖 140 輸入/輸出(I/O)裝置 900 BGA裝置 150 網路介面裝置 910 焊料接合 200 描點圖 912 焊料接合 210 焊料接合 1000 BGA裝置 300 描點圖 1012 焊料接合 400 誤差總成 1100 BGA裝置 410 BGA裝置 1112a 焊料接合 412 焊料接合 1112b 焊料接合 420 印刷電路板 1113a 焊料接合 450 聚焦平面 1113b 焊料接合 500 印刷電路總成 1200 描點圖 510 印刷電路裝置 1300 描點圖 512 焊料接合 1302 方盒 520 印刷電路板 1304 方盒 31 535473 五、發明說明(29) 元件標 號對照表 元件編號 譯 名 元件編號 譯 名 1400 方法 1402 步驟. 1404 步驟 1406 步驟 1408 步驟 1410 步驟 1412 步驟 1500 邏輯 1510 邏輯 1520 邏輯 1530 誤差值資料陣列 1540 邏輯 1550 邏輯 1600 方法 1602 步驟 1604 步驟 1606 步驟 1608 步驟 32What is the method used to generate these expected values? The low frequency (acceptable) of the measured values of the entire array package (Luan Xian A ~ 7) is again being found in the search for the expected value and the modified pass / fail gate root value. reflect. The system 100 can be programmed to store and analyze errors for foreign objects. These foreign objects are unanticipated (such as high-frequency) variations in solder joint measurements, which have been followed as shown in step ⑷0. The 'improved solder joint inspection system' can be programmed to compare the estimates generated at each step Or the respective measured value of each expected solder joint with each solder joint recorded in step 1406. As mentioned above and shown in step 1412, the improved solder joint inspection system is 28 535473. V. Description of the invention (26) Accurate index of solder joints caused by mosquito bug BGA device. It should be understood that the process steps shown in the flowchart of Figure 14 can be repeated to complete solder joint analysis of other printed circuit devices mounted on the printed circuit assembly. FIG. 15 is a functional block diagram showing various exemplary components of the solder joint analysis application 114 of FIG. In this regard, the solder joint analysis application < U4T includes recording logic 1500 configured to receive one or more of the characteristic values observed in various images generated and analyzed by the solder joint inspection system 00. The solder joint analysis application 14 may also include logic 15 10 each solder joint configured to be tested determines a desired value. As previously described, the logic 1510 may use any number of methods to generate a desired measurement for each individual solder joint, which takes into account variations between the mounting surface of the relevant device and the printed circuit board. As shown in Figure 15, record logic 1500 and logic 1510 can be configured to provide information about each individual solder joint to comparison logic 1520. Second, the comparison logic can be configured to determine the difference between the measured (ie recorded measurement) and the relevant expected value for each individual solder joint being observed. As further shown in this block diagram, the comparison logic 1520 can be configured to deliver each individual error data value to an error value data array 1530. Next, the error value out-of-bounds analysis logic 1 540 can be configured to receive a number of 5 Wu differences and perform out-of-bounds analysis on the data. After the error value foreign object has been identified in the out-of-bounds analysis logic 1540, the solder joint analysis application U4 can deliver identification information related to out-of-bounds data values that exceed a threshold value to the bound 29 535473 V. Description of the invention (27 = report logic, 550 The logic of the out-of-bounds error report is as follows: the system provides a position and / or other identification elements to one or more peripheral devices that are in line with the solder joint inspection system 1GG. Figure 16 is a flowchart showing the identification of solder joints. Defective method 0, which can be implemented by the circuit inspection system of Fig. 1. In this regard, the modification of the material bonding system can be implemented by measuring and recording the physical characteristics of each solder joint that is related to the specific characteristics. • Improved solder joint inspection system after mother-to-solder bonding. The information about array package winding and / or printed circuit board winding can be used at step 1604 to estimate the expected solder-to-measuring solder bonding. For example, as shown in step 1606, the improved solder joint inspection system TI can be configured to compare the expected value with that of each of the tested subjects. ^ The mother of a sacrifice-solder joint generation-error value If the display is deleted in steps, ^ U606 generated several error values can be applied in an error data value out-of-bounds analysis to identify defective joints. Second, ..., improved; C dry material joint inspection system 100. The specific embodiment has been disclosed in detail in the description and the figure for the purpose of example. It will be understood by those skilled in the art that changes and modifications can be made without departing from the spirit of the invention as set forth in the following application. Made. For example, other solder joint measurements related to non-array packages can be expected to reflect low frequency variations for various reasons, including the above-mentioned mounting surface due to "Z" | degree variations = results. Therefore, it has- More than two related technologies are used to adopt pass / fail thresholds for non-array sealing I solder joints. This disclosure intends to include the application of these technologies. 30 535473 V. Description of the Invention (28) Component Labeling Table Component Number Translation Component Number Translation 100 Solder bonding inspection system 550 Focusing plane 102 Processing device 600 BGA device 110 Memory 610 Center 112 Working system (0 / S) 612 Solder bonding 114 Material bonding analysis program 612a solder bonding 118 area interface 612b solder bonding 120 user interface device 700 trace point 130 display 800 trace point 140 input / output (I / O) device 900 BGA device 150 network interface device 910 solder joint 200 traces 912 solder joints 210 solder joints 1000 BGA device 300 traces 1012 solder joint 400 error assembly 1100 BGA device 410 BGA device 1112a solder joint 412 solder joint 1112b solder joint 420 printed circuit board 1113a solder joint 450 focus plane 1113b Solder Bonding 500 Printed Circuit Assembly 1200 Plotted Drawing 510 Printed Circuit Device 1300 Plotted Drawing 512 Solder Bonded 1302 Square Box 520 Printed Circuit Board 1304 Square Box 31 535473 V. Description of the Invention (29) Component Numbering Table Component Number Translation Name Component Number 1400 method 1402 step. 1404 step 1406 step 1408 step 1410 step 1412 step 1500 logic 1510 logic 1520 logic 1530 error value data array 1540 logic 1550 logic 1600 method 1602 steps 1604 step 1606 step 1608 step 32

Claims (1)

535473535473 1 ·種用於採用測試門捏值之方法,包含下列的步驟· 為一印刷電路裝置(410)上之數個焊料接合(412)獲 取(H02)位置資訊; 又 獲得(1404)表示該印刷電路裝置(41〇)之安裝表面與 一印刷電路板(420)間之距離變異;記錄(i4〇6)被用以耦 合該印刷電路裝置(410)至該印刷電路板(42〇)之數個焊 料接合(412)的實體性質之測量; 為回應於該印刷電路裝置(41〇)之安裝表面與該印 刷電路板(420)間之距離變異的各別焊料接合估計(i4〇8) 可接受的測量之範圍;以及設定(1412)回應於該範圍之 至少一門檻值。 2. 如申請專利範圍第丨項所述之方法,其中該估計步驟 (1408)包含對辨識之相鄰焊料接合⑴13)組的被記錄測 量實施統計分析。 3. 如申請專利範圍第i項所述之方法,其中該設定步驟進一 步包含: 比較(1606)該門檻值與被記錄的測量以為該印刷電 路裝置(4U))上之數個焊料接合⑷·生—誤差值;以及 對該等數個誤差值實施一界外分析(16〇8)以建立至 少一門檻值。 4. 如申請專利範圍第3項所述之方法,其中比較該門檻值與 被記錄之測4的步驟(⑽6)包含該«值與職記錄之 測量的數學組合。 5. -種用於辨識焊料接合之方法,包含的步驟為: 33 5354731. A method for adopting a test gate pin value, including the following steps: Obtaining (H02) position information for several solder joints (412) on a printed circuit device (410); and obtaining (1404) indicating the printing Variation in the distance between the mounting surface of the circuit device (41) and a printed circuit board (420); the record (i406) is used to couple the number of the printed circuit device (410) to the printed circuit board (42) Measurement of the physical properties of each solder joint (412); the individual solder joint estimates (i408) in response to the variation in the distance between the mounting surface of the printed circuit device (41) and the printed circuit board (420) may be The range of accepted measurements; and setting (1412) at least one threshold value in response to the range. 2. The method as described in item 丨 of the patent application scope, wherein the estimation step (1408) includes performing statistical analysis on the recorded measurements of the identified adjacent solder joints (13). 3. The method as described in item i of the scope of patent application, wherein the setting step further comprises: comparing (1606) the threshold value with the recorded measurement as a number of solder joints on the printed circuit device (4U)) Generation-error value; and performing an out-of-bounds analysis (160) on the plurality of error values to establish at least one threshold value. 4. The method as described in item 3 of the scope of patent application, wherein the step (⑽6) of comparing the threshold value to the recorded test 4 includes a mathematical combination of the measurement of the value and the job record. 5.-A method for identifying solder joints, including the steps: 33 535473 535473 六、申請專利範圍 9.如申請專利範圍第8項所述之系統, 1進一步包含: 設施(114)用於分析該等數個誤差值以辨識焊料接合 瑕疵。 10.如申請專利範圍第9項所述之系統, •其中用於分析之設施 包含一方盒描點圖(1304)。 35535473 6. Scope of patent application 9. The system described in item 8 of the scope of patent application, 1 further includes: a facility (114) for analyzing these several error values to identify solder joint defects. 10. The system described in item 9 of the scope of patent application, wherein the facility for analysis includes a box plot (1304). 35
TW91114381A 2001-12-17 2002-06-28 System and method for identifying solder joint defects TW535473B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US10024101A 2001-12-17 2001-12-17

Publications (1)

Publication Number Publication Date
TW535473B true TW535473B (en) 2003-06-01

Family

ID=29214395

Family Applications (1)

Application Number Title Priority Date Filing Date
TW91114381A TW535473B (en) 2001-12-17 2002-06-28 System and method for identifying solder joint defects

Country Status (1)

Country Link
TW (1) TW535473B (en)

Similar Documents

Publication Publication Date Title
US6853744B2 (en) System and method for confirming electrical connection defects
US6847900B2 (en) System and method for identifying solder joint defects
JP4227959B2 (en) Inspection method using vertical slice imaging
US6665433B2 (en) Automatic X-ray determination of solder joint and view Delta Z values from a laser mapped reference surface for circuit board inspection using X-ray laminography
US6539107B1 (en) Machine vision method using search models to find features in three-dimensional images
US6490368B2 (en) Automatic X-ray determination of solder joint and view Delta Z values from a laser mapped reference surface for circuit board inspection using X-ray laminography
Moore et al. Three-dimensional X-ray laminography as a tool for detection and characterization of BGA package defects
Rooks et al. Development of an inspection process for ball-grid-array technology using scanned-beam X-ray laminography
JP2011191085A (en) X-ray inspection device, x-ray inspection method, x-ray inspection program, and x-ray inspection system
US7327870B2 (en) Method for inspecting a region of interest
JP4610590B2 (en) X-ray inspection apparatus, X-ray inspection method, and X-ray inspection program
JP3800394B2 (en) Method for optimizing probe card analysis and scrub mark analysis data
JP4580266B2 (en) X-ray inspection apparatus, X-ray inspection method, and X-ray inspection program
JP6025768B2 (en) Noble metal amount calculation device and noble metal amount calculation method
TW535473B (en) System and method for identifying solder joint defects
US20190227001A1 (en) Method for inspecting ball grid array-type semiconductor chip package
JP3643722B2 (en) X-ray inspection method and apparatus
JP7450272B2 (en) Measuring the loop height of overlapping bond wires
Roth et al. Non-destructive inspection of through mould vias in stacked embedded packages by micro-CT
Mahon et al. Automated visual inspection of solder paste deposition on surface mount technology PCBs
KR20150121024A (en) X-ray nondestructive testing device
JP2002076071A (en) Reliability evaluation method and reliability evaluation equipment of component mounting part
TW200409914A (en) Method and device for the detection of defective printed circuit board blanks
Zeng et al. A full-field warpage characterization measurement method coupled with infrared information
JP3332001B2 (en) Recognition method of the tip of the contact probe

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees