TWI223284B - Method and system for high-speed, precise micromachining an array of devices - Google Patents

Abstract

A method and system for high-speed, precise micromachining an array of devices are disclosed wherein improved process throughput and accuracy, such as resistor trimming accuracy, are provided. The number of resistance measurements are limited by using non-measurement cuts, using non-sequential collinear cutting, using spot fan-out parallel cutting, and using a retrograde scanning technique for faster collinear cuts. Non-sequential cutting is also used to manage thermal effects and calibrated cuts are used for improved accuracy. Test voltage is controlled to avoid resistor damage.

Description

1223284 发明 Description of the invention (The description of the invention should state: the technical field to which the invention belongs, the prior art, the content, the embodiments, and a brief description of the drawings) L The technical field to which the invention belongs] 1 Cross-reference of related patents and applications

The title of this application for the application filed on March 28, 2002 is "Laser 5 Based micro-Machining Method And System, and an Application To High Speed Laser Trimming Of Chip Components And Similar Structures". 60 / 368,421 as a claim. This case also claims priority to the following patent cases and its subsequent applications: on March 27, 2002, filed 10 applications with the title "Method for Processing a Device, Method and System for Modeling Same and the Device "U.S. Patent Application No. 10 / 18,101, which is now published as U.S. Patent No. 2002/0162973. The title of" Method and Apparatus for Shaping a Laser_Beam 15 Intensity Profile by Dithering ", which is incorporated herein by reference in its entirety. This application is also related to the US patent case entitled" Pulse Control In Laser Systems "assigned to the assignee of this case No. 6,339,604. This application is also related to the following patents. Application was made on March 27, 2002 The pending application number is 10 / 107,027 and Assignee 20 gives the assignee of the present invention the title "High Speed, Laser Based Method and System for Processing Material of One or More Targets Within a Field" and is now announced as the United States Patent case No. 2002/0170898. Field of invention 6 1223284 (1). Description of the invention The invention relates to a method for high-speed precision micro-cutting array device. The invention also relates to the scope of resistor cutting, especially on thin film block structure Cutting of zigzag resistors. Laser resistor cutting refers to laser-cut cutting in the area of the resistive material between the right conductors. 5 [] BACKGROUND OF THE INVENTION Resistor cutting (and other electronic components and circuits cutting and micro-cutting) Xiao'J) has evolved over the past 20 years and is now used to adjust circuits used in thick films, thin films, and other electronic technologies. Published in 2001, the Laser 10 Material Handling LIA Handbook, No. Chapter 17, 583-588, Cutting, contains revelations that illustrate several points of laser cutting. Figures ^ to 卜 in this case are incorporated from that publication. Figure la illustrates the current lines without cutting resistors, and figure 1 illustrates the effect of laser cutting on the current lines. Figure 1 (: The figure summarizes several results (stability, rate, and tolerance) with various resistor geometries and cutting patterns. The following typical US patents are related to laser cutting systems and methods, patent case number: 6,510,605; 6,322,711 · '5,796,392; 4,901,052, 4,853,671; 4,647,899; 4,511,607 and 4,428,298. U.S. Patent No. 4,429,298 is about many viewpoints of zigzag cutting. 20 Basically, this zigzag resistor is formed by continuous feed cutting, and This final cutting is performed parallel to the edges of the resistor produced by the previous feed cutting. Its description starts from a field that is made on the resistor alternately, continuously, and feeds the knife into consideration. This feed cutting is the largest And the minimum length, for the cutting blade. 〖The cutting resistance of the feed cutting ^ cutoff value for faster cutting of feed cutting 7 1223284 发明, invention description ^, rate 乂 and structured processing with various resistance and cutting length test There are requirements for improved high-speed micro-cutting, such as precision cutting at all operating scales, ranging from thick films Circuit-to-wafer cutting. [Abstract] 5 Summary of the Invention The purpose of the present invention is to provide an improved method and system for high-speed and accurate micro-cutting arrays. In carrying out the above and other purposes of the present invention, Provide a method for high-speed and precise micro-cutting array device. Each device has at least a measurable 10-quantity characteristic. This method includes the following steps: • Selectively micro-cut the device in the array to change the measurable The value of the characteristic; (b) abort this selective micro-cutting step; (c) when this selective micro-cutting step is aborted, at least another device in the series will be selectively micro-cut to change its measurable The value of the characteristic; and (d) resume the discontinued selective microcutting step 15 to change the characteristic that the device can measure until the value is in the desired range. The device can be a resistor. This The resistor can be a thin film resistor. This selective micro-cutting step can be performed by cutting at least one laser ray 20 of the device. This method can further include measuring at least one of these devices. Measure the characteristics to obtain a measurable value. This method may further include: comparing the measured value with a preset threshold to obtain a comparison result, and based on the result, at least one of these other devices 8 玖, invention description micro-cutting This method may further include selectively micro-cutting at least one of these other devices according to the values in the table. This method may further include: deciding not to measure at least one of these other devices based on the values in the table. Measurable value. This method can be used for high-speed and accurate φ through Waseda cutting resistor arrays. One of these measurable characteristics can be resistance. Each of these selective micro-cutting steps can include selective removal of the material. Include at least one of one or more columns and / or multiple rows. At least one of these selective micro-cutting steps can be performed with multiple focusing laser pulses irradiating multiple devices substantially simultaneously. ~ At least one selective micro-cutting step can be performed with multiple focused laser pulses. This method may further include allocating the focused laser pulse. This distribution step may include generating a distribution pattern having a plurality of laser rays and focusing the laser rays. This laser cutting can produce a series of interdigital cuts in the area of resistive material between the conductors of the resistor. The at least one selective micro-cutting step may include the step of positioning the laser light at the position of each of the micro-cutting devices at this time, and selectively irradiating at least one of the micro-cutting devices with at least one laser pulse. Parts. At least one of these selective cutting steps may include a step of: generating laser rays and transmitting them relatively positioned in a first direction in the array range, and selectively illuminating at least one of the ranges with at least one laser pulse. Device description, invention description '' At least one part. The method further includes generating laser rays and transmitting them relative to a second direction substantially opposite to the first direction in the range, and selectively irradiating at least one device in the range with at least a laser pulse. One part five two. At least one of these selective micro-cutting steps may include generating laser light and transmitting it relative to the scan case across these devices, overlapping the second scan pattern with the first scan pattern, And at least one device is irradiated with at least one laser pulse. The second scanning pattern may be a rescanning scan, and at least the laser pulse scanning speed of the at least one device is lower than the corresponding scanning speed of the first scanning pattern. The laser energy can be concentrated on at least the device, and the concentration period is longer than the period related to the first-literacy pattern, thereby increasing the yield. This second cat sweep pattern may include jumping from the first device to the second device. This selective micro-cutting step may be performed with multiple laser pulses, and at least-the laser pulse may have an energy in the range of Joules to 25 Joules. This measured value can be the value of the measured temperature. These devices may be substantially the same. 20

In addition, in implementing the above and other objects of the present invention, a system for a high-speed laser-based precision micro-cutting array device is provided. Each of these devices has at least-measurable characteristics. This system includes a pulsed laser subsystem. The optical subsystem is connected to a pulsed laser system, and a part of the device is selectively illuminated with a laser pulse. This connection is connected to the subsystem. Control the subsystem to: (a) Select the devices in the array selectively.

Micro-cutting to change this measurable characteristic > 佶, the value of the mI injection, (b) abort this selective micro-cutting; (C) when this selective micro-cutting is aborted, at least another device in the array Selective micro-cutting to change the value of its measurable characteristics, and ⑷ Restore this selective micro-cutting to change the measurable characteristics of the device until its value is in the desired range. The optical subsystem may include a polarizer and a polarization controller for controlling the polarizer to scan the laser light along the first cat pattern (which includes each of the micro-cutting devices to be waited for). The system further includes a measurement subsystem to measure one of the 10 measurable characteristics of at least one of these devices. The microcutting can be laser cutting, the array can be a resistor array, and the measurement subsystem can be a detector array. The optical subsystem may include a second polarizer to re-focus the high-speed second scanning pattern onto the first scanning pattern, thereby increasing the output of the system. 15 A controller can be coupled to the subsystem so that the subsystem can be controlled to produce a cutting sequence for at least one of these devices, which reduces the temperature of the device during micro-cutting. The above and other objects, features and advantages of the invention will be apparent from the following detailed description of the best mode for carrying out the invention and with reference to the attached drawings. Brief description of the drawing The first la-lb diagram is a schematic diagram, each of which illustrates the current lines before and after laser cutting; the first lc diagram is a diagram illustrating the various cutting parameters on the number of cutting parameters 11 1223284 玖, the effect of the invention description Figure 2a is a schematic diagram of a chip resistor array arranged in rows and rows, which illustrates the results of using a laser cutting step according to an embodiment of the present invention; Figure 2b is a step-definition cutting step corresponding to Figure 2a Block 5 flowchart; Figure 3 is a block flowchart, which further defines the cutting operation of Figures 2a and 2b in the system of the present invention; Figure 4a is a schematic diagram of a chip resistor array arranged in columns and rows, which Describe the results of using laser cutting steps according to another embodiment of the present invention; 10th diagram is a block flow diagram, which further defines the cutting steps corresponding to FIG. 4a; FIG. 5 is a block flow diagram; The cutting operation of Figures 4a and 4b in the inventive system; Figure 6a is a schematic diagram of a laser cutting 15 system that can be used in at least one embodiment of the present invention; Figure 6b is a schematic diagram of a resistor, which The measurable geometric characteristics, especially the edges of the resistors, can be measured using the data obtained with the system in Figure 63; Figure 7 is a chart showing a laser array scan period of 20 in one embodiment A plot of the position of light versus time (VS), where a fast scan of a solid-state polarizer and an electro-mechanical linear scan are overlapped, and the cuts in Figure 2 or Figure 4 are selectively formed at an increased speed. FIG. 8 is a schematic diagram of the system, which transmits a plurality of focused rays to at least one resistor to increase its cutting speed; and FIG. 9 is a schematic diagram of the system, which provides multiple functions in a laser cutting system. Light to at least one resistor. Embodiment C] Detailed description of the preferred embodiment High-speed zigzag cutting process In the cutting of a resistor, this cutting guide current flows through the resistance material along the resistance path. As illustrated in Figures 1a to 1c, the precise control and adjustment of the cutting size and shape changes the resistance to the desired value. These chip resistors are typically arranged in columns and rows on a substrate. Figure 2a shows a configuration in which the processing resistors R1, R2 ... RN are arranged. Bring the conductor with the detector 200 and the array of resistors and resistor rows illustrated by the arrows in Figure 2a to the contact 202. The matrix switch will be used to address the contacts of the first pair of conductors (for example, the contact of ^ R1), and a series of cuts and measurements will be performed to change the resistance between the conductors # to the desired value. #When this resistor cutting is completed, the matrix switches to the second set of components in the next column (for example, R2) to contact and repeat the cutting process. When the entire column of these resistors (by ... rn) has been cut, the contact between this contact and the detector array is broken. Position this substrate relative to another column, contact the detector array with it, and process the second column in the same way as the previous column. The cutting of the zigzag thin film resistor as illustrated in Fig. 1c, for example, is related to laser processing, and an interdigital cut is generated in the area of the resistance material between the conductors. These inter-digital cutting guide currents flow through the resistive material along this tortuous path around the cutting. Its geometry allows a wide range of resistances to be produced in a single area thin film / conductor layout. This above method will place the zigzag cut sequence in the measurement step at the resistor position 1223284, description of the invention, and then move to the next resistor.

With reference to Figure 2a in March, it illustrates the initial laser position for any cutting as 205, and the light locator directs the light along the linear path through the electrical resistor material. A new model according to the present invention cuts a branch on the first resistor (e.g., cutting cut 204 of R1) and measures its resistance. If its resistance is lower than the preset threshold, similar collinear cutting is performed across the other resistors R2 ~ rn in the stop row. In Fig. 2a, 210 illustrates the collinear cutting performed along the column, and in Fig. 2b, the corresponding square is further defined. In at least one embodiment of the invention, a subset of the resistors can be measured to determine the uniformity of the film across the substrate. If the film has a known uniformity, one measurement is sufficient.

This collinear cutting group along the resistor column is implemented in the same manner as shown in 211 in Fig. 2a, and further defined by 15 in block 221 in Fig. 2b. The resistor RN is first cut. This process is repeated as shown in Figures 2a-212-213, and further defined in Figures 2b, blocks 222-223. If the measurement shows that it exceeds the critical value, the cutting of the columns R1 ... RN is performed with the measurement of each resistor, so that each resistor is cut to the critical value before switching to the next resistor (as described in block 224 as 214) . 20 Limit the number of measurements and maintain collinear cutting trajectories to increase cutting speed. The process in Figure 3 further defines the steps corresponding to Figures 2a-2b, as well as other processing steps used in the cutting system (such as addressing and loading). In at least one embodiment, the cutting step can be implemented according to preset information. As an example, for some resistor types, the _ series element can be cut before measuring the resistance. This sequence is based on the resistor's preset parameters: (such as: geometric shape) and / or known film characteristics ( (Eg 'thin film resistor). Similarly, the number of non-measuring cuts can be determined in the learning mode of the first resistor (for example, including at least _ measurements, or: repeat testing. Implement repeated measurements in the learning mode, and determine the non-measurement based on the measurement and material characteristics Number of times of cutting. In at least one embodiment, the number of times of non-measurement cutting can be calculated.

For example, four cuts can be made without measurement. Please refer to Di Juntu, whose statement 10 states 410 at the beginning. The detector is placed in contact with the column 202 as shown in Figure 2a. Referring to Figure 4b, this initial situation is further defined at block 420. By way of example, Figs. 4a-4b illustrate an embodiment of the cutting process, in which the most four cuts are made without any measurement. As shown in the figure, block 421 defines a predetermined set of cutting times (for example, 4 times) according to at least one value or condition before cutting without measuring 15 $. This is used to complete four

The Sweep Conceal Path is illustrated at 405. The first resistor R1 in the 406 series is then cut and measured to determine whether the target value is reached. If not, as described at 412 and further defined by block 422, then (e.g., not measured) the remaining resistors R2 ... RN are cut. 20 This process is then repeated, starting with RN's cutting 407, and then as shown at 413 and further defined by block 423 from R (N-1) to cutting. Therefore, R1 or RN is cut with a change in each direction, and if the target value is not reached, each of the remaining resistors R2 is cut. The final step occurs after R1 or RN reaches the target value. As in 15 1223284 (1), the description of the invention 414 and further defined by block 424, the resistors are connected and sequentially cut. The flowchart of Figure 5 further defines the steps corresponding to Figures 4a_4b and additional processing steps used in the cutting system (for example, it includes steps of addressing and loading). In one embodiment, iterative measurements are used to obtain pre-set information to provide values before cutting. These values can be specified by the operator or process engineer, or otherwise obtained. The swivel provides the ability to specify or use the target value before cutting so that the applied test voltage 10 and / or current can be controlled. This characteristic is useful for avoiding high voltages that are sufficient to damage this part in the range of resistance changes associated with this zigzag cutting. When a fast resistance H measurement system is used in the embodiment of the present invention, the application of the initial low resistance cut J to the voltage drop of the measurement resistor is used to limit the current flowing through the resistor and the possible damage to it. During subsequent cutting, the electrical resistance increases and the measurement voltage increases. 20

Force to modify the typical cutting and queuing of Figures 2a, 2b, 4a, and 4b to allow changes in material properties and other process parameters and tolerances: for example, in at least An additional step can be used when the measured chipping reaches the target value and its length is within a predetermined edge of the maximum allowable cutting length. Changes in material properties in this edge can make some cutting cuts fail to reach the target value and require additional cuts. In the first mode, at the beginning of the component, cutting and cutting are sequentially performed from ^ thousand of 歹 j, and the component positions that have not reached the target value are captured and saved. And subsequent cutting and cutting 16 1223284 玖, description of the invention The remaining components in the saved position are cut to the target value. In the second mode, according to the value of the cut length of the first component, the cut length is reduced to prevent reaching the target value, and a non-measurement cut is performed to complete the cut of the column. These subsequent cuts will bring all 5 pieces in the column to the target value. In the second mode, at least one pre-cut length on the component is corrected to prevent subsequent cuts from falling into the edge case. In at least one embodiment, an additional step may be used when the measured cutting cut value is in a predetermined edge of the target value. Changes in material properties in this margin 10 can cause some components to exceed the target value of using completely non-measurement cutting. In the first mode, cutting and cutting are performed in sequence among these components, and the locations of components that have not reached the target value are saved. Then, the remaining components in the saved position are cut to the target value by subsequent cutting. 15 In the second mode, the cutting length is reduced according to the value measured in the first component to prevent the target value from being reached, and a non-measurement cutting is performed to complete the cutting of the column. Subsequent cutting cuts bring all components in the column to the target value. τ In the second mode, the length of at least one previous cut on the component is adjusted to prevent subsequent cuts from falling into the edge condition. Different from the traditional single resistor cutting technique, as shown in the experimental data in Figures 2 to 4, the yield is increased by cutting all the resistors in the row. As an example, the approximate results are shown in the following table: 17 1223284 玖, Description of the invention

This overall cutting rate increases as the number of resistors in the column increases, fewer measurements, and as the final (i.e., precision) cutting time decreases by 5 times. In addition, each resistor must have additional time to recover from the energy generated by the laser. The order of the cuts can be determined to manage temperature changes in the component (for example, reducing the maximum component temperature during cutting). For example, referring to Figure 4a ', the sequence 405 can be reversed, so that a group of cuts is performed starting from the center of the approaching element 10 and proceeding to one end of the element approaching the conductor and the detector. Other suitable sequences can be used (for example, any non-adjacent cutting sequence has the advantage of being used for thermal management). The second element may preferably be cut before the further measuring step. The range of resistance change for zigzag cutting ranges from about 15 orders of magnitude (for example, 10X), typically two orders of magnitude (100x), to about 500X for current materials. Laser Cutting System In at least one embodiment of the present invention, the laser cutting system can be first calibrated using the method described in U.S. Pat. No. 4,918,284, Calibrating Laser Trimming Apparatus, i 20. Therefore, the two patents disclose that the laser cutting device is calibrated by controlling the laser light positioning mechanism. This positioning 18 1223284 玖, the invention description mechanism moves the laser light to the desired rated laser position on the substrate area. Print a mark on the media (for example, cut a line) to establish the actual laser position, scan the printed mark to detect the actual laser position, and compare the actual laser position with the desired rated position . It is preferable to operate the laser light on one wave length, and the detection device operating on different wavelengths scans this mark. The range viewed by this detection device covers a part of the entire substrate area, and determines the position of the mark in this range. Therefore, the 284 patent further discloses determining the position of the laser ray relative to the viewing range of the camera. Other calibration techniques can be used alone or in combination with this 284 method. For example, US Patent No. 6,501,061, Laser Calibration Apparatus and Method, "discloses a method to determine the coordinates of the cat scanner and accurately locate the focused laser light. The laser scanner will therefore The focused laser ray is scanned over the area of interest (eg, aperture) on the work plane. The position of the focused laser ray is measured by light detection 15 at a preset time or space interval or when The focused laser light is detected when it appears through the aperture in the work plane. The detected position using this focused laser light is generated based on the position of the laser scanner when the focused laser light is detected. The data of the position of the scanner vs. the position of the light. You can use the data of the position of the scanner (VS) to determine the center of the aperture or the position of the scanner corresponding to the desired position of the focused laser ray 20 After the system is calibrated (which preferably includes the calibration of various other system components), 'at least one substrate with the device to be cut is loaded into the cutting table. Please refer to Figure 6a, part of which Since the 1284 patent, this improved laser 19 1223284 玖, description of the cutting system may include an infrared laser 602, which typically has a wavelength from 1.047 microns to about 1.32 microns' which outputs laser light 603 along the optical path control 604 Via the laser light positioning mechanism 605 to the substrate area 606. In order to apply to the cutting of the thin film array, the output frequency of the IR laser can be doubled by using various technologies known in the art5 and commercially available A better wavelength of about 0.532 microns can be obtained. The laser light positioning mechanism 605 preferably includes a pair of mirrors and attached galvanometers 607 and 608 (available from the assignee of the present case and available for use). This light positioning mechanism 605 guides the laser light 603 through the lens 609 (which may be telecentric or non-telecentric, and the two wavelengths are preferably colorless) to the substrate region 606 in the range. If it is maintained sufficiently Accuracy, this χ_γ galvanometer mirror system can provide the angle coverage of the entire substrate. Otherwise, various positioning mechanisms can be used to provide the relative between the substrate and the laser light For example, you can use the dual-axis accurate step and repeat 15 repeater illustrated in 617 to position the substrate in the range of the galvanometer-based mirror system 607 and 608 (for example, in the χ_γ plane). (Middle). This laser ray positioning mechanism 605 moves the laser ray 603 along two vertical axes, thus providing two-dimensional positioning of the laser ray 603 across the substrate area 606. Each mirror and the relevant galvanometer 607, 608 Under the control of the computer 610, the laser light is moved along its X or Y axis. The irradiation device 6II can be a halogen lamp or a light emitting diode, which generates visible light to illuminate the substrate area 606. The beam splitter 612 (a kind of part The reflector is located in the optical path 604 to guide the light energy along the path 604, and is reflected from the substrate area 606 back to the detection device 614. This detection device 614 includes a camera 615, which can be a digital 20 1223284, a CCD camera plate (such as a color or black / white) and an associated daylight extractor 616 (or a digital camera provided by this camera) Daytime buffer), this device digitizes the image input from the television camera 615 to obtain a two-dimensional image representing a portion of the substrate area 606. This pixel data is stored in the memory of the frame grabber 5 616, or is directly transmitted to the computer 61 for processing via a high-speed connection, for example. The light positioning subsystem may include other optical components, such as a computer-controlled optical subsystem, for adjusting the laser spot size and / or autofocus of the laser spot at the substrate position. 10 In the application of the present invention to thin film cutting of a resistor array, at least one thin film array is supported by a substrate. This calibration data, as obtained above, is preferably used in combination with the automatic cutting horizon algorithm to position the array element (for example, resistor R1) and measure the position of at least one geometric feature of element 62 in the second figure. This feature can be found, for example, by analyzing pixel data in memory by using one of the 15 available edge detection algorithms: one of horizontal edges 621 (eg, edges parallel to the _ direction), And one of the vertical edges 622 (eg, edges parallel to the γ_ direction). This edge may include multiple edges measured along the entire periphery of the resistor, such as the edge of the original, or the edge of multiple resistors from the array. The resistance-visibility 'can then be determined and can be used as a pre-set percentage of the width to define the cut length. This edge information is preferably obtained automatically and used, for example, with calibration data to control the length of each cut in the row ri " rn. Other suitable measurement algorithms can also be used, such as image correlation algorithms or speckle detection methods. 21 1223284 Rose, invention description Calibration can be applied to one or more points along the cut. In at least one embodiment, the starting data of at least one cut can be corrected by the weight data. Fortunately, the length and starting point of multiple cuts in Figures 2 and 4 can be corrected. It is preferable that the lengths 5 and the starting points of all the cuts in Figs. 2 & 4a can be corrected. In an embodiment, the first resistor (for example: ... or Guan Jiaozheng) can be corrected correspondingly to all resistors in the column (for example: R1, ~, RN). Fully automated. However, a semi-automatic algorithm with the intervention of 10 operators can be used, for example, when positioning the galvanometer so that the array elements are in this range, and then sequentially positioning the light along the elements in an interactive manner, then by operation A person observes the intensity contour (or a derivative of intensity) on the display 630. This use is only necessary to adjust the coordinates in the array area to improve the accuracy of the 15 laser light positioning without causing damage. Yield reduction. The measurement and alignment data of the resistor width is useful for the following situations: • For controlling the length of ㈣, and for correcting the deviation from linearity, and for correcting the X and γ coordinates of the sweeper Non-orthogonality of the system. The use of this calibration data to correct the shape is particularly suitable for laser cutting systems with one or more linear conversion tables. Geometric correction There is no need to replace other useful system design features, including: linearity of the f-theta lens, fan-out light compensation, etc. Usually this system tolerance can be used < Total based on the expected position error for a few months at the cutting calibration position When the light is fanned out (fan 〇ut), especially 22 1223284, the invention description is for a large gap across many resistors, only one is calibrated and aligned. For example, when the resistor When the interval is very large, a single cut can be calibrated and aligned. The position deviation of these components can be expected, and part of it is due to the system design, f_theta linearity, fan-out expansion and 5 dispersion compensation. Mitigation. Closely spaced horizontal fans are considered to have smaller errors than axial fans. Yield is further improved. Optical technology In at least one embodiment of the present invention, one can use a Or more of the following technologies to increase the effective scanning rate to further increase the yield. 10 can jump across the gap between the resistors in this column to achieve a faster The collinear cutting processing rate is further increased. This gap is shown in Figure 2a. Please refer to Figure 7, in at least one embodiment of the present invention, when the galvanometer is scanned at a constant speed of 702 across the column, this single The axio-optic light deflector (AOBD) overlaps the 踞 -tooth linear scanning pattern 701. During the cutting period, this AOBD scans in the reverse movement 703 and provides a rapid jump 704 to the next between cuttings. One cut. This allows the galvanometer to evade at a constant speed and minimizes the proportion of jumps in the overall processing time. This uses acousto-optic deflectors in combination with galvanometers for speed improvement is known in this technology For example, 'U.S. Patent No. 5,837,962 discloses 20 improved devices for heating, melting, evaporating, or cutting the crop. This two-dimensional acousto-optic deflector provides about a five-fold improvement in the marking rate. U.S. Patent No. 6,341, G29 takes the entirety as a reference, and Figure 5 shows that its embodiment has several elements that can be used in the entire system when the invention is implemented in reverse mode to increase the rate . In '23. Description of the invention, this' 029 U.S. patent shows an acousto-optic deflector and a galvanometer with associated controllers for pulsating CW light to form a laser pattern. For additional details of the system structure, please also refer to this, line 3, line 47 and column 4 of the 029 patent specification. By using the applicable technology, the configuration of the 〇29 patent case can be easily adjusted in order to provide the correction of the optical element and the scanning control contour. It is preferable to implement the reverse scanning of the present invention with an additional hardware calibration procedure. technology. In another embodiment of the present invention, collinear cutting on the meandering resistor can be achieved in parallel with a plurality of light spots along the column. A fan-out grating or other multi-light generating device is used to generate an array of light spots, so that two or more light spots are formed and aligned according to the resistor pitch along this column. For example, U.S. Patent No. ^^ 0, 5,521,628 discloses the use of diffractive optical devices to mark several portions simultaneously. These multiple rays can be lower power rays from a more powerful laser source, or a combination of rays from multiple sources. This scanning system scans multiple light rays and forms a light spot through a common scanning lens across multiple resistors simultaneously. This cutting process is similar to a single light spot method with two or more parallel cuts during a non-measurement cutting step. When the critical value is reached, the system transforms into a single spot pattern and sequentially cuts each resistor to the desired value. Similarly, this collinear cutting on the zigzag resistor can be achieved by forming multiple light spots on the target and performing parallel cutting in a parallel manner. A fan-out grid or other multi-light generating device is used to generate an array of light spots to form two or more light spots, and align this light spot with the component with a gap set in advance between cuts. If a predetermined number of cuts are performed (for example, 4 times shown in Fig. 4a of the invention description), the number of passes can be reduced by 50% in one embodiment (for example, a single pass in each direction) by). This embodiment is most useful if the resistance "smell resistance and tolerance are well established. This grating can be in the light pre-switching path so that multiple light spots or a single light spot can be selectively formed. This published United States Patent application No. 2002/0162973 describes a system and method for generating multiple light spots to handle semiconductor connections used for memory maintenance. Various modifications can be used in the lens system and polarizer system to produce the use in the present invention 10. In a consistent embodiment, a single laser pulse is used to cut up to two resistors at the same time (for example: no, one or two cuts). Please refer to Figure 8 by The aligned laser light 803 is spatially decomposed into two divergent aligned rays 804 805, and two focused light spots, 802 are formed on the two cuts. This difference frequency is precisely adjusted to control the separation of light spots. This uses sound The b-light device for spatially decomposing light in material processing applications is known in this technology. For example, Japanese Patent Abstract JP 53152662 shows a configuration for using Multi-frequency polarizer to drill micro-shaped holes. Laser 806 in Figure 8 pulsates at a preset repetition rate. This 20-ray laser light passes through the transmission optical device 807, which forms the middle image of the laser light waist and enters In the aperture of the acousto-optic modulator (AOM), it is better to use the AOM 808 operated in the Brag ^^ condition to controllably produce two slightly divergent and aligned brothers, a special-level diffraction laser light 'and Control the energy in each light. This AOM is driven by two frequencies fi, f2, and fi = f〇 + df and f2: = f〇_df, and 25 1223284 发明, invention description df is the original RF signal frequency fO A small percentage. The angle between the two rays is approximately equal to the Bragg angle used for fO times 2 (df / f0). This AOM modulates the two frequency components f 1 and f 2 in the RF signal f 12 The signal amplitude and the interactive coupling of the adjustment light control the energy in each laser light. 5 After leaving the AOM 808, this light rotates the light 90 degrees through the selective light rotation control module 809, so that the light is at Or ah direction. In one embodiment, 稜鏡 is used for this rotation, although as in Many of the rotation techniques described in the related U.S. Patent Publication No. 2002/0170898 are well known. 10 Second, the light is positioned at the waist of the light via a set of optical devices, and the size of the light is set to be suitable for adjusting the focal length Optical device and objective lens 810. The focus-adjusting optical device also corrects the angle between the two rays, so depending on the adjustment of the focus adjustment, the angle between the two rays leaving the AOM 808 must be adjusted. In order to cause the desired light at the focal plane to be separated by 15 points. Second, this laser light enters the objective lens 81o, which provides a pair of focused light spots 801, 802 on two resistors. The distance between these two light spots is approximately equal to the focal length of the lens 810 times the angle between the two rays. This inverse and parallel method can be combined for collinear tangent on a tortuous resistor.] For example, this ray Scan by AOBD, then break down into a pair and scan across 20 perimeters. Cut two adjacent resistors at the same time, and this jump is from resistor N to resistor N + 2 and to the next pair of resistors. Alternatively, or with a two-dimensional polarizer, a spot-point can be created in a direction perpendicular to the direction of the zigzag scan. For example, with the very early control and stylization of the eight factors, you can use a polarizer (with appropriate output power 26 1223284 玖, controlled by the description of the invention) to simultaneously produce four implementations as shown in Figure 4a. Cut at least two of the four rays used. Therefore, the scanning time for cutting can be reduced by 50%. Because of its programmable deflection, this AOBD is preferred over the fan-out grating. These multiple light spots can also be generated as required during rough 5 or fine cutting. Fig. 9 outlines an exemplary embodiment of an improved laser cutting system having a module 901 from Fig. 8 for reverse scanning, parallel processing, or a combination thereof. For example, a signal 902 from a computer 61 can be used to control AOBD or other solid-state polarizer 808 in one or more axes, rotating the module 809 with 10 and (if provided) light. The module 901 may include a transmission optical device 807 and other light shaping elements. It is preferred to use at least one AOBD to provide considerable flexibility and ease of use. For example, a digital ruler ^^ generator is used to provide a control signal 812 from a computer 61 °. In addition, this technique for forming extended or elliptical light spots can be used with the present invention to further increase processing speed and quality. The improvement of the cutting rate related to the formation of light spots is described in US Patent Application No. 2002/0170898, which has been published and pending. In at least one embodiment of the present invention, a variety of other design alternatives may be used to enhance system performance and ease of use. For example, these 20 generation methods include but are not limited to the following: 1. This system can provide computer-controlled light spot size and / or focus adjustment. In the US Patent No. M83, G7m, which is assigned to the assignee of the present invention, a kind of optical m which provides the control of the spot size and the dynamic focus for laser-based maintenance of redundant memory. 27 1223284 发明, description of the invention 2 · Another alternative is to control the energy of light with a variable light attenuator. This attenuation state can be an acousto-optic deflector (or modulator). You can use either a manual or automatic adjustment of a neutral density filter or an attenuator based on polarization. As an example, U.S. Patent No. 6,518,54 5 shows a suitable variable attenuator with a rotating half-wave plate and a beam splitter sensitive to polarization. 3. The pulse width can be changed using methods known to those skilled in the art, with the understanding that the energy of this q-switched laser changes with the repetition rate (especially at high repetition rates). For dynamic cutting, where measurements are performed between 10 pulses, it is preferred to maintain a substantially constant pulse energy. This method for pulse energy control is disclosed in the 6,339,104 patent 'It corresponds to the precise measurement period when the resistance value approaches a preset target value, and decreases with the cutting speed (for example: a larger pulse Time interval) while reducing changes in target energy. 15 (In at least the embodiment, a kind of YAG laser with diode as the pump frequency doubled is used as the rubidium resistance material. #Compared with other wavelengths, this output wavelength of 532nm results in low movement and no micro-fracture And can ignore the area affected by heat. It is better to use a pulse width of approximately ⑸, but typically a width less than 30ns. This preferred maximum laser repetition rate is at least 20 ΙΟΚΗζ. This pulse width is more typical. Thick-film systems are much smaller and provide a fairly high repetition rate for removing thin film materials. This reduces the pulse width and high repetition rate to the maximum available pulse energy, allowing and diffractive optical devices (eg, Gratings or AOBD), so multiple light spots can be provided. 28 玖, description of the invention 5. The laser can be focused to a large size limited by the diffraction limit about the light spot Γ type is less than large ⑽ microns or more It is small, and the preferred light spot is 15: half 20 micrometers' in size, and the most preferable light spot size is in the range of about 6 to ~ meters, for example, 10-15 micrometers. 5-In the illustrated embodiment of the present invention In the description of zigzag cutting The listed parallel numbers are cut. However, it should be understood that the application of the present invention is not limited to the formation of parallel cuts. Such cutting or micro-cutting is used to produce

There are multiple disjoint cuts that reduce the number of measurements, and this is considered to be within the scope of the present invention. X 7 · In addition, the embodiments of the present invention are not limited to the measurement of thin film resistors, but can be used in other micro-cutting applications where the physical characteristics are measurable. This measurement is not limited to electrical measurement, but may be (for example, an infrared sensor) temperature monitoring, pressure, vibration, or other characteristic measurements. 0 15 Although the embodiments of the present invention have been described and illustrated above, It is not intended to describe and illustrate all possible forms of the invention for these embodiments. Instead, the words used in the present invention are explanatory words rather than restrictive words, and it should be understood that various changes can be made to the subject without departing from the spirit and scope of the invention. 20 [Schematic description] Figures la-lb are schematic diagrams, each of which illustrates current lines before and after laser cutting; Figure lc is a diagram, which illustrates the effect of various cutting formations on several cutting parameters; 29 1223284发明 Description of the invention Figure 2a is a schematic diagram of a chip resistor array arranged in rows and rows, which illustrates the results of using a laser cutting step according to an embodiment of the invention; Figure 2b is a further definition of the cutting corresponding to Figure 2a Block diagram of steps; 5 FIG. 3 is a block diagram that further defines the cutting operation of FIGS. 2a and 2b in the system of the present invention; FIG. 4a is a schematic diagram of a chip resistor array arranged in columns and rows, It illustrates the result of using a laser cutting step according to another embodiment of the present invention; FIG. 4b is a block flow chart, which further defines the cutting step corresponding to FIG. 10; FIG. 5 is a block flow chart; The cutting operation of Figures 4a and 4b in the invention system; Figure 6a is a schematic diagram of a laser cutting system that can be used in at least one embodiment of the present invention; 15 Figure 6b is a summary of a resistor Figure, which has measurable geometric characteristics, especially the edge of the resistor, which can be measured using the data obtained with the graph system; Figure 7 is a graph showing the resistor array scanning period in one embodiment The plot of the position of the light rays versus time (VS), in which the fast scanning of the solid-state polarizer 20 and the electromechanical linear scanning are superimposed, and the cutting of Figure 2 or Figure 4 is selectively formed at an increased speed. Fig. 8 is a schematic diagram of the system, which transmits a plurality of focused rays to at least one resistor to increase its cutting speed; and Fig. 9 is a schematic diagram of the system 'which provides more 30 1223284 in a laser cutting system. Explain a light to at least one resistor. [Representative symbol table of main components of the figure] 617 ... transponder 620 ... element 200 ... detector 202 ... contact 204,205 ... cut 210 ... collinear cut 211,212,213,214—cut R1, R2 ~ RN ... resistor 405 ... Sweep path 406,407,408 ... Cut 410 ... Cut most cases 411,412,413,414 ... Cut 602 ... Infrared 603 ... Ray path 604 ... Optical path 605 ... Laser beam positioning mechanism 606 ... Substrate area 607, 608 ... Galvanometer 609 ... Lens 610 ... Computer 611 ... Illuminating device 614 ... Detection device 615 ... Camera 616 ... Writer 62 ... Horizontal edge 622 ... Vertical edge 630 ... Monitor 702 ... Constant speed 703 ... Reverse movement 704 ... Jump 801,802 ... Light spot 803 ... Laser light 804,805 ... Light 806 ... Laser 807 ... Transmission optics 808 ... Acoustooptic modulator 809 ... · Light rotation control module 810 ... Objective lens 812 ... RF signal 822 ... Folding lens 901 ... Module 902 ... Signal 220, 221, 222, 223, 224, 420, 421, 422, 423, 424 ~ block

31

Claims (1)

Scope of patent application 1. A method for high-speed precision micro-cutting array devices, each of which has at least one measurable characteristic, which is characterized by the following steps: Selectively micro-cutting the devices in the array to change The value of its measurable characteristic; discontinuing the selective microcutting step; and when the selective microcutting step is aborted, selectively microcutting at least another device in the array to change the value of the measurable characteristic; and Resume this aborted selective microcutting step to change the measurable characteristics of the device until its value is within the desired range. For example, the method of claim 1 in which the device is a resistor. For example, the method of claim 2 of the patent scope is declared, wherein the resistor is a thin film resistor. 4 · If the scope of patent application is the first! The method of claim 4, wherein the step of selective micro-cutting is performed by at least one laser light of a device such as a cutting ground. 5 · If the scope of patent application is the first! The method further includes measuring at least the measurable characteristics of these devices to obtain a measured value. 6. The method of claim 5, further comprising comparing the measured value with a preset threshold to obtain a comparison result, and micro-cutting at least one of these other devices based on the comparison result. 7. The method of claim 5, further comprising selectively micro-cutting at least these devices according to the measured value '. 8. The method of item 5 of the scope of patent application, further comprising deciding not to measure at least the measurable characteristics of these devices based on the measured values. 32 fe. Application for patent: The method of item No. 17 of the scope of patent application. Precision laser cutting array resistor. 0. As described in the scope of patent application, each of these selective micro-cutting steps includes: a step of selectively removing material. u • If the scope of the patent application item 丨, method, where the array includes ... these—at least one of the plurality of columns, the "at least one of these one or more rows. 2 The method of item 其中, in which at least one of these selective micro-cutting steps, is carried out by using a plurality of 10 v ′ fine-cutting pulses to substantially simultaneously shoot multiple devices. … 13. The method according to item 1 of the scope of patent application, wherein at least-these selective micro-cutting steps are performed with a plurality of focused laser pulses, and wherein the method further includes allocating the focused laser pulses. The method of claim 15 is patent application method, wherein the distribution step includes a step of generating a distribution pattern having a plurality of laser rays, and focusing the laser rays. 15. The method of claim 9 in the scope of patent application, wherein the laser cutting produces a series of digital cuts in the area of the resistive material between the conductors of these resistors. 20 9. Where this method is used for high speed and one of the measurable characteristics is 16. The method of claim 1 in patent application, wherein at least one of these selective U-cutting steps includes the step of positioning laser light at each of the positions waiting for the micro-cutting device, and selectively using at least one laser pulse The ground is irradiated with at least a part of each of the waiting micro-cutting devices. 17. If the method of applying patent scope β, at least one of which is selective 33, the patent application Fan ^ cutting step includes these steps: generating laser light and relatively 疋 located in this array range φ 筮. "The target girl's younger brother transmits in one direction. I selectively irradiate at least a part of at least one device in this range with at least one laser pulse. 18. The method of claim 17 in the scope of patent application, comprising generating laser light and its relative positioning in this range is substantially opposite to the first direction, transmitted in the first direction ', and at least _ laser pulse selectivity The ground radiates at least-at least-the second part of the device in this range. 10 10 If you are applying for a method specifically for lJS, at least-these selective micro-cutting steps include these steps: Generate field shots and position them relative to the first across these devices: cat pattern transmission , The second scan pattern and the first scan pattern are repeated, and at least one device is illuminated with at least one laser pulse. 15. The method of claim 19 in the patent application range, wherein the second concealment pattern is a reverse concealment pattern, and wherein the concealment rate of at least one laser pulse irradiating the at least one device is lower than that corresponding to the-concealment pattern The "blind rate", in which the laser energy is concentrated in at least one device for a period of time, which is longer than the period related only to the first scanning pattern, and thus the increase is increased. 20 • The method of item 19 in the scope of claim 4, wherein the second scanning pattern includes a jump from the first device to the second device. 22. The method of claim 1, wherein the selective micro-cutting steps are performed with a plurality of laser pulses, and at least one of these pulses has an energy in the range of ❻ · 1 microjoule to 25 millijoules In the range. 23. The method of item 7 in the scope of patent application, wherein the measured value is the temperature value measured in the scope of patent application and scope of patent application. 24. The method of claim 1 in the patent scope, wherein these devices are substantially the same. 25. —A system for high-speed laser-based precision micro-cutting array devices' Each of these devices has at least a measurable feature, which is characterized by: a pulsed laser subsystem; coupled to a pulsed laser One of the subsystems is an optical subsystem that selectively irradiates a part of the device with laser pulses; and a controller coupled to these subsystems to control these subsystems: Means to change the value of the measurable characteristic; to suspend the selective micro-cutting; to selectively micro-cut at least one other device in the array to change the value of the measurable characteristic when the selective micro-cutting is stopped; and The selective microcutting is resumed to change the measurable characteristics of the device until its value is in the desired range. 26. The system of claim 25, wherein the optical subsystem includes a polarizer and a polarizer controller, which are used to control the polarizer to scan the laser light along the first scanning pattern, and the pattern includes Each of these devices for micro-cutting. 27. The system of claim 25, further comprising a measurement subsystem to measure these measurable characteristics of at least one of these devices. 28. The system of item 27 in the scope of patent application, wherein the micro-cutting is laser 1223284, the scope of patent-application cutting ', and the array is an array of resistors, and wherein the measurement subsystem is a detector array. 29. The system of claim 26, wherein the optical subsystem includes a second polarizer that superimposes the high-speed second scanning pattern on the first scanning pattern, thereby increasing the system output. 30. The system of claim 25, wherein the controller is coupled to these subsystems so as to control the subsystems to produce a cutting sequence for at least one of these devices, which is reduced during micro-cutting Device temperature. 36