CN101058164A - Image calibrating method used for chip dicing machine - Google Patents

Image calibrating method used for chip dicing machine Download PDF

Info

Publication number
CN101058164A
CN101058164A CN 200610025803 CN200610025803A CN101058164A CN 101058164 A CN101058164 A CN 101058164A CN 200610025803 CN200610025803 CN 200610025803 CN 200610025803 A CN200610025803 A CN 200610025803A CN 101058164 A CN101058164 A CN 101058164A
Authority
CN
China
Prior art keywords
dicing machine
chip dicing
chip
angle error
machine
Prior art date
Legal status (The legal status 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 status listed.)
Granted
Application number
CN 200610025803
Other languages
Chinese (zh)
Other versions
CN100439039C (en
Inventor
张昕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Electric Machine Tools & Engineering Co., Ltd.
Original Assignee
SHANGHAI FUAN FACTORY AUTOMATION CO Ltd
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 SHANGHAI FUAN FACTORY AUTOMATION CO Ltd filed Critical SHANGHAI FUAN FACTORY AUTOMATION CO Ltd
Priority to CNB2006100258038A priority Critical patent/CN100439039C/en
Publication of CN101058164A publication Critical patent/CN101058164A/en
Application granted granted Critical
Publication of CN100439039C publication Critical patent/CN100439039C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Dicing (AREA)

Abstract

The invention relates to the image calibration method for a chip scribing system that uses single CCD camera with graphic data collecting system, serial circulation execution adding coordinate values X and Y on one processing baseline, computing the included angle tolerance theta 1 of the working piece processing baseline and the machine processing baseline, instructing the return of the chip scraper on the working platform, rectifying the included angle error, selecting the X and Y values of two points at both ends of the longest processing baseline, computing the included angle error theta 2 of the whole processing baseline and the machine processing baseline, instructing the chip scraper working platform returning to rectify the included angle error, and finally centering alignment for the tool and work piece cutting. Through second computation and rectification angle theta, it can speed up the adjustment of the single crystalline cutting position with improved cutting precision.

Description

A kind of image calibrating method that is used for chip dicing machine
Technical field
The present invention relates to IC chip system integrated technology application, be specifically related to a kind of image calibrating method that is used for chip dicing machine.
Background technology
In the manufacture craft of IC chip, get off at present with the IC chip cutting that scribing machine will be printed on the single-chip.Single chip fitly is arranged in (as Fig. 1) on the single-chip, requires the cutting track of scribing machine emery wheel can only be confined in the two cutting belt zones of going between the IC chips during cutting.The cutting belt width is generally 0.04mm~0.05mm, and the groove width after the emery wheel cutting is generally 0.03mm~0.035mm.This just require on the single-chip cutting belt and scribing machine X to movement locus parallel; The cutting belt center overlaps with the grinding wheel mid-depth.
Vision positioning system on the scribing machine is to solve this technological problems.Prefabricated cross datum line on the scribing machine display screen, and as the background of CCD photographed images.Can preset CCD camera lens with mechanical means, the horizontal line that makes the cross datum line and scribing machine X be to parallel, and overlap (cross datum line modulated become machine tooling datum line) with the grinding wheel mid-depth.Single-chip is placed on the scribing machine workbench, and the image that the CCD camera system is adopted is thrown on the display screen, as shown in Figure 2.Manually control scribing machine work table rotation makes cutting belt and planche cross line parallel; Manually control scribing machine Y makes the cutting belt center overlap (as shown in Figure 3) with planche cross line center to motion.
Because the CCD multiplication factor requires to be not less than 100 times, (multiplication factor is little generally to get 100~200 times, the center misalignment), to amplify 100 times is example, display image is of a size of 150mm * 165mm, the cutting belt width amplifies the back and be of a size of 4mm (the cross hairs live width is generally 0.4mm by the decision of kinescope pixel) on screen, and the single-chip actual area that CCD captures has only 1.5mm * 1.65mm.Just screen is gone up 165mm cutting belt length, and actual size has only 1.65mm.Proofreading and correct and the planche cross line parallel with the range estimation of the image of 1.65mm length, if its error has only cross hairs live width size, also be very normal, and actual Cutting Length is very big, and 6 " (inch) sheet X stroke is 165mm, and so omnidistance parallel misalignment reaches 40mm.So must mobile repeatedly X-axis, θ axle and Y-axis, progressively adjust, the planche cross line is included in the omnidistance cutting belt zone, and placed in the middle as far as possible.In fact use the intensified image of 1.65mm length, proofread and correct the cutting belt of 165mm length, operation is that very cumbersome, final precision also is not high.
Advanced both at home and abroad for this reason scribing machine has all adopted twin-lens CCD camera system.(establish apart a), two camera lenses are gathered one group of image to two camera lenses separately, render to two zones (as shown in Figure 4) about display screen respectively at a distance of certain distance.Two cutting belt not necessarily link to each other on screen about on the display screen, if moreover about two cutting belt also not necessarily same cutting belt that link to each other.Manually control the scribing machine work table rotation, two cutting belt are parallel to the planche cross line about making, and are connected to a cutting belt, manually control scribing machine Y to moving, and make the planche cross line overlap (as shown in Figure 5) with the cutting belt center.
Because two cutting belt are actual in being a (supposing a=55mm) on the display screen, estimating and adjusting error also is cross hairs live width size, cuts omnidistance length 165mm, aims at the omnidistance parallelism error in back and reduces to 1.2mm.Omnidistance back and forth more mobile X-axis and fine motion θ axle, Y-axis have just been finished the work of workpiece location for several times.Find out that therefrom a distance is big more, easy more adjusting, but a big require scribing machine X stroke also should strengthen.In addition because the single-chip size has different size from 4 "~12 ", and the excessive twin-lens of a can not capture small dimension wafer epigraph simultaneously, also just can't the calibration of workpieces position; The too small big specification list wafer initial adjustment error of a increases, and regulates efficient and precise decreasing.
Summary of the invention
The technical problem to be solved in the present invention provides a kind of image calibrating method that is used for chip dicing machine, adopts single CCD camera, regulates efficient and precision but can improve.
For solving the problems of the technologies described above, the inventive method chips scribing machine uses single CCD camera, the graph data acquisition system is housed on this chip dicing machine, sequential loop is carried out following steps: two point coordinates value X, Y on the same machining benchmark line in the workpiece that step 1, acquisition chip scribing machine are handled calculate angle error θ between workpiece machining benchmark and machine tooling benchmark by setting equation 1, and instruct this chip dicing machine work table rotation, with finishing angle error θ 1Step 2, gather coordinate figure X, Y at 2, and calculate omnidistance machining benchmark line of workpiece and machine tooling datum line angle error θ by setting equation at the two ends of the longest machining benchmark line of handling of workpiece 2, and instruct this chip dicing machine work table rotation, to repair omnidistance angle error θ 2Step 3, to the cutter and the work piece cut band center of chip dicing machine.
The inventive method is by the process secondary calculating and repair the θ angle, can accurately adjust the single-chip cutting position fast, has improved the work cutting accuracy.
Description of drawings
Fig. 1 is the single-chip schematic diagram;
Fig. 2 is the CCD camera system is adopted in the traditional handicraft an image situation when throwing into traditional die scribing machine display screen;
Fig. 3 manually controls the scribing machine work table rotation in the traditional handicraft, make cutting belt and planche cross line parallel; Manually control scribing machine Y makes the cutting belt center overlap situation with planche cross line center to motion;
Fig. 4 is the display screen that traditional handicraft chips scribing machine adopts twin-lens CCD camera system;
When Fig. 5 is traditional handicraft chips scribing machine employing twin-lens CCD camera system, manually adjust cutting belt, make the situation that overlaps with the cutting belt center.
Fig. 6 is in the inventive method, any 2 coordinate figure situation on the cutting belt in the system acquisition image;
Fig. 7 is the signal of the inventive method flow chart.
The specific embodiment
The present invention is further detailed explanation below in conjunction with accompanying drawing.
Adopting the chip dicing machine of the inventive method to use single CCD camera, and the graph data acquisition system is housed, increased the image data acquiring processing capacity, can the cross hairs central point be the datum mark of data acquisition, can gather the coordinate dimension of arbitrfary point on the workpiece.After the image on the single-chip is rendered on the display screen, as Fig. 6, shown in Figure 7, at first to gather in the image coordinate figure of any 2 points (A, B) on the cutting belt respectively, system can calculate the angle theta of cutting belt and planche cross line, i.e. θ=tg automatically then -1((Y 2-Y 1)/(X 2-X 1)), system command scribing machine workbench rotates θ automatically, cutting belt and planche cross line is transferred to automatically is parallel to each other; Because system has increased segmentation scale (is radix with display screen matrix unit) on vertical cross hairs, therefore can manually control mobile Y-axis,, regulate that the planche cross line is overlapped with the cutting belt center by the segmentation scale on the vertical cross hairs.And then manually control mobile X-axis, cutting belt two ends at maximum Cutting Length, gather the two-end-point coordinate respectively, system still calculates omnidistance angular errors again by above-mentioned equation, and instruction works platform automatic dressing, utilize vertical cross hairs segmentation scale, regulate that accurately cutting belt center whole process is overlapped with the planche cross line.Revise through secondary θ angle,, just finish single-chip location work on the scribing machine with vertical cross hairs center adjustment.
In sum, the inventive method has overcome the deficiency in the traditional handicraft, solved the problem that needs accurately to adjust fast the single-chip cutting position in the actual process, no matter and workpiece size, can both revise workpiece angular errors (satisfy the positioning requirements of the specification list wafer that varies in size simultaneously, and adjust omnidistance error) with the maximum Cutting Length of workpiece two ends coordinate figure with the workpiece maximum spacing.And the thin scale function of vertical cross hairs, improved the accuracy of alignment of emery wheel cutting groove and work piece cut band.

Claims (3)

1, a kind of image calibrating method that is used for chip dicing machine, this chip dicing machine uses single CCD camera, it is characterized in that, the graph data acquisition system is housed on this chip dicing machine, following method is carried out in described systematic order circulation, comprise: step 1, gather in the workpiece that this chip dicing machine handles two point coordinates value X, Y on the same machining benchmark line, calculate angle error θ between described workpiece machining benchmark and machine tooling benchmark by the setting equation of angle error 1, and instruct this chip dicing machine work table rotation, with finishing angle error θ 1The two ends of step 2, the longest machining benchmark line of workpiece handled at described chip dicing machine are gathered coordinate figure X, Y at 2, and calculate omnidistance machining benchmark line of described workpiece and machine tooling datum line angle error θ by setting equation 2, and instruct this chip dicing machine work table rotation, to repair omnidistance angle error θ 2Step 3, to the cutter and the described work piece cut band center of described chip dicing machine.
2, the image calibrating method that is used for chip dicing machine according to claim 1, it is characterized in that, described machine tooling datum line is the cross datum line, the width of this cross datum line line is got minimum pixel unit, i.e. 1 pixel, and on vertical cross datum line, be that unit makes thin scale with the pixel.
3, the image calibrating method that is used for chip dicing machine according to claim 1 and 2 is characterized in that, the setting equation of described angle error is: θ=tg -1((Y 2-Y 1)/(X 2-X 1)).
CNB2006100258038A 2006-04-18 2006-04-18 Image calibrating method used for chip dicing machine Expired - Fee Related CN100439039C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2006100258038A CN100439039C (en) 2006-04-18 2006-04-18 Image calibrating method used for chip dicing machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2006100258038A CN100439039C (en) 2006-04-18 2006-04-18 Image calibrating method used for chip dicing machine

Publications (2)

Publication Number Publication Date
CN101058164A true CN101058164A (en) 2007-10-24
CN100439039C CN100439039C (en) 2008-12-03

Family

ID=38864548

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2006100258038A Expired - Fee Related CN100439039C (en) 2006-04-18 2006-04-18 Image calibrating method used for chip dicing machine

Country Status (1)

Country Link
CN (1) CN100439039C (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101777610B (en) * 2010-01-21 2011-07-20 广东志成华科光电设备有限公司 Method for fast tuning angle of LED chip
CN102248309A (en) * 2010-05-17 2011-11-23 苏州天弘激光股份有限公司 Wafer laser dicing method and wafer laser dicing equipment with charge coupled device (CCD) assisting in positioning
CN104567675A (en) * 2014-12-29 2015-04-29 福建华冠光电有限公司 Conveying precision monitoring method for double small liquid crystal panels
CN105304540A (en) * 2015-09-22 2016-02-03 常州星海电子有限公司 Glass passivation chip reverse cutting method without positioning lines
CN106145658A (en) * 2016-06-29 2016-11-23 昆山国显光电有限公司 Glass cutting method
CN106449900A (en) * 2016-08-31 2017-02-22 导装光电科技(深圳)有限公司 Cutting technology and device for LED white-light chips
WO2017107534A1 (en) * 2015-12-21 2017-06-29 广州视源电子科技股份有限公司 Method and device for measuring angle, and method and device for adjusting angle
CN107205318A (en) * 2016-03-17 2017-09-26 塔工程有限公司 Scribing equipment and dicing method
CN107579028A (en) * 2017-09-12 2018-01-12 北京中电科电子装备有限公司 A kind of edge of incomplete wafer determines method, apparatus and dicing device
CN109581288A (en) * 2018-11-16 2019-04-05 广州杰赛科技股份有限公司 Indoor orientation method, equipment and storage medium

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS63104348A (en) * 1986-10-21 1988-05-09 Toko Inc Semiconductor device
JP2652015B2 (en) * 1987-04-07 1997-09-10 セイコーエプソン株式会社 Semiconductor device
JP2651199B2 (en) * 1988-06-06 1997-09-10 富士通株式会社 Automatic pattern verification system for masks or reticles
JPH08166534A (en) * 1994-12-14 1996-06-25 Toshiba Corp Automatic focusing method for image processor
JPH1064981A (en) * 1996-08-13 1998-03-06 Fujitsu Ltd Method for aligning wafer
TWI240965B (en) * 2003-02-28 2005-10-01 Toshiba Corp Semiconductor wafer dividing method and apparatus
CN1529347A (en) * 2003-10-21 2004-09-15 中国科学院上海光学精密机械研究所 Saphire base nitride chip scribing method

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101777610B (en) * 2010-01-21 2011-07-20 广东志成华科光电设备有限公司 Method for fast tuning angle of LED chip
CN102248309A (en) * 2010-05-17 2011-11-23 苏州天弘激光股份有限公司 Wafer laser dicing method and wafer laser dicing equipment with charge coupled device (CCD) assisting in positioning
CN104567675A (en) * 2014-12-29 2015-04-29 福建华冠光电有限公司 Conveying precision monitoring method for double small liquid crystal panels
CN104567675B (en) * 2014-12-29 2017-08-15 福建华冠光电有限公司 Biplate small size liquid crystal panel transports accuracy monitoring method
CN105304540A (en) * 2015-09-22 2016-02-03 常州星海电子有限公司 Glass passivation chip reverse cutting method without positioning lines
WO2017107534A1 (en) * 2015-12-21 2017-06-29 广州视源电子科技股份有限公司 Method and device for measuring angle, and method and device for adjusting angle
CN107205318A (en) * 2016-03-17 2017-09-26 塔工程有限公司 Scribing equipment and dicing method
CN107205318B (en) * 2016-03-17 2021-01-29 塔工程有限公司 Scribing apparatus and scribing method
CN106145658A (en) * 2016-06-29 2016-11-23 昆山国显光电有限公司 Glass cutting method
CN106449900A (en) * 2016-08-31 2017-02-22 导装光电科技(深圳)有限公司 Cutting technology and device for LED white-light chips
CN107579028A (en) * 2017-09-12 2018-01-12 北京中电科电子装备有限公司 A kind of edge of incomplete wafer determines method, apparatus and dicing device
CN107579028B (en) * 2017-09-12 2020-03-17 北京中电科电子装备有限公司 Method and device for determining edge of incomplete wafer and scribing device
CN109581288A (en) * 2018-11-16 2019-04-05 广州杰赛科技股份有限公司 Indoor orientation method, equipment and storage medium

Also Published As

Publication number Publication date
CN100439039C (en) 2008-12-03

Similar Documents

Publication Publication Date Title
CN100439039C (en) Image calibrating method used for chip dicing machine
CN110828344B (en) Automatic alignment control method for cutting channel of semiconductor device
CN109702290B (en) Steel plate groove cutting method based on visual identification
CN103107121B (en) Wafer angular deviation automatic method based on vision
CN103354770B (en) Laser processing and laser processing device
CN112013787B (en) Blade three-dimensional contour reconstruction method based on blade self-characteristics
CN106041296B (en) A kind of online dynamic vision laser precision machining method
CN104181861B (en) Biaxial numerical control machine tool correction positioning implementation method and system based on counterpoint platform
CN104526743A (en) PCB-V groove cutting device
CN110293559B (en) Installation method for automatically identifying, positioning and aligning
CN102179726A (en) Instrument and method for measuring secondary clamping deviation during numerical control machining based on image technology
CN101298116B (en) Non-coaxial positioning method for processing laser material
CN103676976A (en) Correction method for three-dimensional worktable repositioning error
CN110814398A (en) Machine vision auxiliary curved surface machining device and method
KR20150020217A (en) Scribing method and scribing device
CN112296513B (en) Laser cutting machine leads positive system based on machine vision
JP5709593B2 (en) Processing equipment
CN201223996Y (en) Non-coaxial locating device for processing laser material
CN102615665B (en) Automatic positioning system and method
CN112331560A (en) Automatic alignment Y, T bisection algorithm of dicing saw
CN208147154U (en) A kind of vertical wide cut multifunction laser machine
CN109859273A (en) A kind of caliberating device and method of online dynamic vision system of processing
CN211588658U (en) Machine vision auxiliary curved surface machining device
CN111221298B (en) Machining system with visual positioning function and machining method
CN103419081A (en) Bench synchronous positioning compensation system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C56 Change in the name or address of the patentee

Owner name: SHANGHAI ELECTRIC MACHINE TOOLS + ENGINEERING CO.,

Free format text: FORMER NAME: SHANGHAI FUAN FACTORY AUTOMATION CO., LTD.

CP01 Change in the name or title of a patent holder

Address after: 200041 Huaian Road, Shanghai, No. 681, No.

Patentee after: Shanghai Electric Machine Tools & Engineering Co., Ltd.

Address before: 200041 Huaian Road, Shanghai, No. 681, No.

Patentee before: Shanghai Fuan Factory Automation Co., Ltd.

CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20081203

Termination date: 20150418

EXPY Termination of patent right or utility model