IES70056B2 - A measurement system - Google Patents
A measurement systemInfo
- Publication number
- IES70056B2 IES70056B2 IES960562A IES70056B2 IE S70056 B2 IES70056 B2 IE S70056B2 IE S960562 A IES960562 A IE S960562A IE S70056 B2 IES70056 B2 IE S70056B2
- Authority
- IE
- Ireland
- Prior art keywords
- component
- optical head
- light source
- computer
- view
- Prior art date
Links
Landscapes
- Supply And Installment Of Electrical Components (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
A measurement system (1) has a robotic system (3) mounted over a base plate (2) which provides a vertical height reference. In the optical head (8), a CCD camera (17) is adjustable to be pre-set in height. The optical head (8) also has an upper light source (18) comprising LEDs and a lower light source (25) comprising a white fluorescent lamp (26) for lateral component illumination. An image analysis computer (10) automatically performs various tests to co-ordinate optical and mechanical operation.
Description
A Measurement System
The invention relates to a measurement system, and more particularly to a system for identifying components on an electronic circuit board and measuring their locations. Such systems generally comprise a support for supporting the circuit board and a robotic system which drives an optical head having a camera connected to an image processor.
Heretofore, many developments have been made separately in the fields of optics and robotic drive control. These have lead to improvements, however, they generally result in increased complexity. An example is the multiple camera arrangement described in EP 341,806.
There is therefore a need for an improved measurement system in which the optical chain signal (from the components to the image processor), and the robotic system are improved in a co-ordinated manner to achieve better results without need for added complexity.
The invention is directed towards providing such a measurement system.
According to the invention, there is provided a measurement system comprising :an optical head driven by a robotic system for capturing images of a circuit board moved into the optical head field of view, the robotic system having an accuracy of at least +/- 2.5 gm over the working area;
an image analysis computer connected to the optical
S70 056 4 'ι
- 2 head and comprising means for capturing circuit board images and using the images to measure component locations;
wherein the optical head comprises :a support block mounted on the robotic system;
a camera of the tele-centric gauging lens type mounted on the support block in a vertically adjustable manner, an upper light source comprising a set of light emitting diodes mounted in a ring arrangement for near-on-axis illumination, and *
a lower white light source comprising an annular fluorescent lamp with higher intensity than the upper light source and being mounted at a sufficiently low level for lateral component illumination to emphasise component sides;
wherein the computer is programming to capture and display colour or monochrome images for analysis; and wherein the image analysis computer is programmed to perform automated repeatability tests comprising the steps of :for each of a plurality of optical head acceleration values performing a plurality of component measurements and recording average time duration and a standard deviation of the measured location;
plotting responses of said standard deviation and time versus acceleration; and determining an optimum optical head acceleration value based on both of said plots analysed in conjunction with each other.
Preferably, vertical position of the upper light source is adjustable and vertical position of the lower light source is also preferably adjustable.
In one embodiment, the lower light source comprises a 10 reflective housing surrounding the lamp and extending to a lower level.
In a further embodiment, the fluorescent lamp of the lower light source is operated at a frequency in the range of 23 to 40 kHz.
In one embodiment, the image analysis computer comprises means for performing automatic scaling by :directing movement of a reference target to the optical head field of view;
directing robotic movement of the target with an 20 accuracy of less than +/- 2.5 microns over the working area and determining the corresponding movement in the image as a number of pixels; and determining a correlating relationship between the robotic and image movements and using said relationship to determine and store a fresh scaling factor.
v
- 4 Preferably, the image analysis computer comprises means for storing in a database a flag indicating components which are larger than the field of view, and for capturing images of pads in rows of such components when they are identified and subsequently merging the associated measurements to generate a composite measurement.
In one embodiment, the image analysis computer comprises means for directing a second pass for components which are identified in a database as potentially having different identification colours, the second pass being initiated according to comparisons with grey scale thresholds on the first pass and being limited to a pre-defined area of the component.
In a further embodiment, the computer performs initial view allocation with a view hung on a first component in a direction, and in a second pass eliminates singlecomponent views of the component in another view.
Ideally, the computer stores a computer aided design circuit board layout, displays the layout, and adjusts component colours according to parameter trends.
In one embodiment, the computer applies a colour coding scheme to each parameter tracked.
The invention will be more clearly understood from the following description of some embodiments thereof, given by way of example only with reference to the accompanying drawings, in which :Fig. 1 is a front view of a measurement system of the invention;
- 5 Fig. 2 is a more detailed and side perspective view of the system;
Fig. 3 is a side view showing an optical head of the system;
Fig. 4 is a perspective view from underneath of the optical head;
Fig. 5 is a schematic representation illustrating the manner in which a surface mount component is illuminated;
Fig. 6 is a flow diagram illustrating the manner in which the system performs automated repeatability tests;
Figs. 7 and 8 are diagrams illustrating operation of the system to perform automated repeatability tests;
Fig. 9 is a flow diagram illustrating the manner in which the system performs automatic scaling; and
Fig. 10 is a flow diagram illustrating the manner in which the system monitors large components.
Referring to the drawings, and initially to Figs. 1 to 4, there is illustrated a measurement system of the invention indicated generally by the numeral 1. The system 1 comprises a base plate 2 having a machined upper flat surface which provides a base reference for the system. A robotic system 3 is mounted over the base plate 2 on a pair of rails 4 mounted on risers 5, in turn mounted on the base plate 2. The height of the risers is carefully selected as this height provides the base height for the robotic system and the optical head which it drives. In more detail, the robotic system 3 comprises a longitudinal drive 6 and a transverse drive 7, both of the linear motor system type.
An optical head 8 is mounted on the robotic system and is connected to display screens 9 via an image analysis computer 10. In addition to performing image analysis, the computer 10 also performs motor control and digital input/output functions.
Conveyor rails 11 are mounted above the base plate 2 and beneath the optical head 8 and these are for conveying printed circuit boards (PCB's) 12 underneath the optical head for measurement of components on the board.
The apparatus also comprises a keyboard 13 on sliding rails, as shown in Fig. 1.
The optical head 8 is mounted on a support block 15 having a pair of vertical bars 16 onto which a camera 17 is mounted. The camera 17 is of the CCD tele-centric gauging lens type and is mounted on the bars 16 in a manner which allows vertical pre-setting.
The optical head also comprises, beneath the camera 17, an upper light source 18 comprising a support 19 supporting a ring of red light emitting diodes (LEDs) 20 around an aperture 21.
The optical head 8 also comprises a lower light source 25 comprising a ring-shaped white fluorescent lamp 26 mounted in a reflective housing 27. The housing 27 has an aperture 28 which is larger than the maximum field of view of the camera 17.
The system 1 is suitable for use before reflow in a surface mount technology (SMT) manufacturing process so that trends in the placement operations can be determined from the measurement results. The system 1 initially measures the position of fiducials on the printed circuit board (PCB) and then measures each component's position relative to the fiducial coordinates and identifies the components. Due to a combination of optical and drive features, the system achieves very high accuracy and excellent repeatability in making measurements at production rates to provide real-time statistical trends of the manufacturing process. This is invaluable to the manufacturer as the trend information allows a preventative maintenance program to be implemented for the placement machines in the process.
Referring now to Fig. 5, the manner in which a component being measured is illuminated is now described. The upper light source 18 generates a near-on-axis red light beam A directed vertically through the aperture 28. The beam A is relatively weak and is useful primarily for identifying fiducials at the beginning of a measurement process.
As is clear from Fig. 5, the fluorescent lamp 26 is close to the level of the PCB, extends in a circle having a larger diameter than the field of view, and is surrounded by a reflective housing which extends downwardly. Because of this arrangement light rays B impinge laterally on the target. In this drawing, an SMT component 41 having leads 42 is within the field of view. An important aspect of the lighting arrangement is that the light rays B impinge on the sides of the component 41, namely at the leads 42 and some of them reflect upwardly within the field of view to be captured by the camera 17. This puts a large degree of emphasis on the sides of the components, and more particularly the leads 42. As these are the most
- 8 important parts of the components for measurement purposes, this optical emphasis reduces the complexity of image analysis which is required and therefore allows a faster response time for real-time measurement to be achieved. Because white light is emitted by the lamp 26, the computer 10 may capture and display colour or monochrome images for analysis. Another important aspect of the lower light source 25 is that high-frequency (the frequency level being 23-40 kHz) fluorescent lighting is used so that image capture beat is avoided.
Referring again to Figs. 1 to 4 the optical axis of the system is configured so that it is parallel to the Z vertical axis of the robotic system 3, but offset in one direction by a pre-determined amount. Further, the risers
are of a specific height to ensure that the focal plane of the camera 17 is centred around the point of inspection. The vertically adjustable mounting of the optical head allows optimum pre-setting at the start of a measurement session.
Another aspect of the manner in which optical and mechanical drive features are combined to achieve high accuracy and excellent repeatability is the fact that the system 1 carries out repeatability tests, as described in the flow diagram of Fig. 6. In this drawing, a method 50 is illustrated which has the initial step 51 of measuring the location of a particular target on the X axis and monitoring the time taken to make the measurement. As indicated by the decision step 52, step 51 is repeated n times, after which in step 53 the computer 10 computes a standard deviation for the measured location of the target and an average measurement time. In step 54, the computer 10 stores a value for the standard deviation at that particular robotic system optical head acceleration value, and also stores a value for the average measuring time for that acceleration value.
As indicated by the steps 55 and 56, these tests are repeated m times and after each test the computer adjusts the motor control settings to increase the acceleration to a new, incremented value before the next test is carried out. After steps 51 to 54 have been carried out m times, tables of values corresponding to the graphs shown in Figs. 7 and 8 are developed. The former graph plots a vertical axis parameter being six times the standard deviation for the measured location against acceleration on the horizontal axis. As is clear from this diagram, there is a straight-line relationship until an acceleration level between 5 and 6 gm/ms2 is reached. It is of course desirable to use an acceleration which is on the straight-line part of the response and as low as possible. In the latter diagram, there is a gradual reduction in the measurement time until an acceleration level of approximately 4 gm/ms2 is reached. The interrupted line between the two graphs indicates a possible optimum acceleration level to achieve a narrow standard deviation with a low measurement time for optimum results. This acceleration level is determined in steps 57 and 58 by the computer 10.
It will be appreciated that this aspect of operation of the system 1 combines optical and mechanical drive control to achieve excellent accuracy and repeatability.
The system 1 also combines optical and mechanical drive features in an automatic scaling process 70 shown in Fig. 9. In this process, in step 71 a glass calibration plate is mounted beneath the optical head 8 and in step 72, the robotic system 3 moves the plate to cause a precision fiducial reference mark on the plate to be located within the field of view. In step 73, the robotic system 3 moves the fiducial a total of Nmm with an accuracy of +/- 2.50 x 10‘6 mm over the working area on the X axis. This level of mechanical drive accuracy is achieved by use of a linear motor drive system and accurate machining of the base plate, risers and rails. In step 74, the computer 10 determines the number of image pixels (pixel distance”) which the fiducial mark traverses during this physical motion, using sub-pixel algorithms. As indicated by the decision step 55, the steps 73 and 74 are repeated n times and in step 76 the physical distances of the robotic system and the pixel distances are averaged. In step 77, the computer 10 stores an automatic scaling factor correlating these two distances. Then as indicated generally by the steps 78, it repeats steps similar to steps 71 to 77 for the Y axis. This automatic scaling is achieved because the robotic system employed is accurate to one micron repeatability and +/- 2.5m micron accuracy over the working area and use of highly accurate sub-pixel algorithms in conjunction with the optimum lighting head configuration.
The manner in which the system 1 deals with components which are larger than the field of view is illustrated by the process 90 in Fig. 10. In this process, in step 91 the computer 10 recognises a component by virtue of its measured location and by cross-reference to a database indicating such large components by a data record flag. These latter steps are indicated by the step 92 and as indicated by the decision step 93 are repeated for each component identified. Where a large component is identified, in step 94 the computer breaks the component into four sub-components or rows of pads and then inspects each pad in isolation. A row naming scheme is used by the computer 10. For example, the component may be called u200 and the system would break it into four rows of pads, one from each side. The component, is now named u500nel2. u200enl0. u200swl2 and u200wsl0. Each of these rows of pads is measured in isolation and the results are then merged back together to yield a measurement for the whole component. An important aspect of this process is the fact that the step 94 of capturing the pad images involves tracking of offsets for each measurement, the offset levels being stored in the database. Once the offsets have been correctly retrieved by the computer 10, it is a relatively simple, operation to capture the pad images in step 94 and merge them in step 95. Thus, while the process 90 may at first appear complex, it may be carried out in real-time with a fast response time.
Another aspect of operation of the system 1 is that the computer 10 differentiates capacitor components, the colour of which may change without notification. This is achieved by using two grey scale thresholds, a high one for the yellow examples and a low one for the black examples. If the image is above the high threshold, a full measurement result set is generated. If it fails the high threshold test, the central area only is tested against the low threshold and if it is below this, the component is considered to be present, but no X and Y information is generated. This effectively means that multiple image analysis algorithms are employed to cater for a situation where there are two potential solutions.
In controlling the fields of view, the computer performs initial view allocation in which the components are sorted in top to bottom and left to right order. Starting at the top left, a view is hung on the first components, and all components that fit within that view are allocated to it in a non-optimal fashion. These components are flagged as unavailable. The search continues with the next component, hanging” another view and allocating as many
A components to it as possible and continuing until all of the views are allocated. To eliminate anomalous results arising from single components allocated to views, the computer 10 uses a compactor algorithm in the second pass. This looks at all single component views and checks if the component lies within another view. If it does, it is allocated to that view and the initial view is deleted. In this way, about 5% of the views can be deleted with a corresponding increase in performance.
For display of capture images, the computer 10 generates a graphical display for ease of viewing of measurement information on a production line. This caters for the fact that there may be as many as 2000 X and 2000 Y measurement data sets for each pcb 12 being measured. The system computes mean X and Y offsets from nominal position and standard deviation of X and Y offsets for each component. Using the mean and standard deviation, and also using known process widths, the system calculates Cp (potential process capability index), K (correction factor), and CpK (process capability index) for each component in real-time. CAD data for the circuit board layout which was previously stored is employed to display a map of the board. By selecting one parameter, the CAD layout map for the circuit board is colour-coded so that the characteristics of individual components become explicit. There is one colour code for each parameter. This display technique gives the operator instant feedback on the manufacturing process.
The invention is not limited to the embodiments hereinbefore described, but may be varied in construction and detail.
Claims (3)
1. A measurement system comprising :an optical head driven by a robotic system for capturing images of a circuit board moved into the 5 optical head field of view, the robotic system having an accuracy of at least +/- 2.5 gm over the working area; an image analysis computer connected to the optical head and comprising means for capturing circuit board 10 images and using the images to measure component locations; wherein the optical head comprises :a support block mounted on the robotic system; a camera of the tele-centric gauging lens type 15 mounted on the support block in a vertically adjustable manner, an upper light source comprising a set of light emitting diodes mounted in a ring arrangement for near-on-axis illumination, and 20 a lower white light source comprising an annular fluorescent lamp with higher intensity then the • upper light source and being mounted at a sufficiently low level for lateral component illumination to emphasise component sides; 25 wherein the computer is programmed to capture and display colour or monochrome images for analysis; and wherein the image analysis computer is programmed to perform automated repeatability tests comprising the steps of :for each of a plurality of optical head acceleration values, performing a plurality of component measurements and recording average time duration and a standard deviation of the measured location; plotting responses of said standard deviation and time versus acceleration; and determining an optimum optical head acceleration value based on both of said plots analysed in conjunction with each other. A measurement system as claimed in claim 1, wherein vertical position of the upper light source is adjustable wherein vertical position of the lower light source is adjustable, and wherein the lower light source comprises a reflective housing surrounding the lamp and extending to a lower level, and wherein the fluorescent lamp of the lower light source is operated at a frequency in the range of 23 to 40 kHz. A measurement system as claimed in any preceding claim, wherein the image analysis computer comprises means for performing automatic scaling by :directing movement of a reference target to the optical head field of view; directing robotic movement of the target with an accuracy of less than +/- 2.5 microns over the working area and determining the corresponding movement in the image as a number of pixels; and determining a correlating relationship between the robotic and image movements and using said 5 relationship to determine and store a fresh scaling factor, and wherein the image analysis computer comprises means for storing in a database a flag indicating components which are larger than the field of view, and for capturing 10 images of pads in rows of such components when they are identified and subsequently merging the associated measurements to generate a composite measurement, and wherein the image analysis computer comprises means for directing a second 15 pass for components which are identified in a database as potentially having different identification colours, the second pass being initiated according to comparisons with grey scale thresholds on the first pass and being 20 limited to a pre-defined area of the component.
2. 4. A measurement system as claimed in any preceding claim, wherein the computer performs initial view allocation with a view hung on a first component in a direction, and in a second pass eliminates single25 component views of the component in another view, and wherein the computer stores a computer aided design circuit board layout, displays the layout, and adjusts component colours according to parameter trends, and wherein the computer applies a colour coding scheme to each parameter tracked.
3. 5. A measurement system substantially as hereinbefore described, with reference to and as illustrated in the accompanying drawings.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
IES960562 IES960562A2 (en) | 1996-08-02 | 1996-08-02 | A measurement system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
IES960562 IES960562A2 (en) | 1996-08-02 | 1996-08-02 | A measurement system |
Publications (2)
Publication Number | Publication Date |
---|---|
IES70056B2 true IES70056B2 (en) | 1996-10-16 |
IES960562A2 IES960562A2 (en) | 1996-10-16 |
Family
ID=11041231
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
IES960562 IES960562A2 (en) | 1996-08-02 | 1996-08-02 | A measurement system |
Country Status (1)
Country | Link |
---|---|
IE (1) | IES960562A2 (en) |
-
1996
- 1996-08-02 IE IES960562 patent/IES960562A2/en not_active IP Right Cessation
Also Published As
Publication number | Publication date |
---|---|
IES960562A2 (en) | 1996-10-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6151407A (en) | Measurement system | |
CN103185549B (en) | Circuit board checker and inspection method | |
US7545512B2 (en) | Method for automated measurement of three-dimensional shape of circuit boards | |
US6161941A (en) | Light array system and method for illumination of objects imaged by imaging systems | |
JP4412914B2 (en) | Controller that automatically adapts illumination light | |
US6246788B1 (en) | System and method of optically inspecting manufactured devices | |
US6084663A (en) | Method and an apparatus for inspection of a printed circuit board assembly | |
US7899239B2 (en) | Inspection method of bonded status of ball in wire bonding | |
KR20100083743A (en) | System and method for inspecting a wafer | |
KR102362929B1 (en) | Method of adjusting needle tip position of probe needle and inspection device | |
US5392360A (en) | Method and apparatus for inspection of matched substrate heatsink and hat assemblies | |
TW201337243A (en) | High speed autofocus system | |
KR101079686B1 (en) | Image recognition apparatus and image recognition method | |
EP0871027A2 (en) | Inspection of print circuit board assembly | |
GB2316168A (en) | Optical measurement system for components | |
JP2021032598A (en) | Wafer appearance inspection device and method | |
IES70056B2 (en) | A measurement system | |
JP3868917B2 (en) | Three-dimensional measuring device and inspection device | |
KR100340013B1 (en) | Computer vision system using rotating mirrors | |
JPH05296745A (en) | Form measuring device | |
JP3725093B2 (en) | In-rib phosphor embedding amount inspection method and inspection apparatus therefor | |
JP6723391B2 (en) | Component mounter and production job creation method | |
JP2017223451A (en) | Measurement device | |
JP2021124374A (en) | Measurement device and method for measurement | |
JPH08285548A (en) | Bump appearance inspection device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FD4E | Short term patents deemed void under section 64 |