Method and apparatus for in-line solder paste inspection

• US 5,912,984 A
• Filed: 12/19/1996
• Issued: 06/15/1999
• Est. Priority Date: 12/19/1996
• Status: Expired due to Term

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Roberts, Peter, Nichani, Sanjay, Michael, David J., Koljonen, Juha
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Tadayon, Bijan
• Application Number:
  US08/769,739
• Time in Patent Office: 908 Days
• Field of Search: 348/87, 348/130, 438/7, 438/669, 430/311, 430/396, 250/492.1, 382/146, 382/147, 382/151, 382/195, 382/145, 382/141, 382/142, 382/143, 382/144, 382/148, 382/149, 382/150, 382/152, 382/153, 382/154, 382/181, 382/190, 382/209, 382/218 , 382/224, 382/275, 382/276, 382/286, 382/293, 382/302, 382/305, 382/307, 382/312, 382/168, 382/170, 382/173, 382/199, 382/217, 382/225
• US Class Current: 382/149
• CPC Class Codes: B23K 2101/40 Semiconductor devices
• CPC Class Codes: B23K 3/08 Auxiliary devices therefor cleaning pipes or tubes or systems of pipes or tubes, e.g. before soldering, B08B9/02
• CPC Class Codes: G06T 2207/30152 Solder
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: H05K 1/0269 for visual or optical inspection
• CPC Class Codes: H05K 2203/163 Monitoring a manufacturing process
• CPC Class Codes: H05K 3/1216 by screen printing or stencil printing
• CPC Class Codes: H05K 3/3485 Applying solder paste, slurry or powder thick film methods for applying conductive paste or ink patterns H05K3/12

Abstract

Inspection of solder paste on a printed circuit board using a before printing image (pre-image) to normalize an after printing image (post-image) of the printed circuit board. Existing lighting and optics used for alignment of the screen printing stencil to the printed circuit board are used for the solder paste inspection. A stencil in the screen printing process is also inspected using a before printing image (pre) to normalize an after printing image (post) of the stencil.

Claims

1. A method of using a vision system to inspect a material applied in at least one predefined pattern on a portion of a substrate, said method comprising the steps of:

acquiring a model set of characteristics of said portion of said substrate and said at least one predefined pattern from a model substrate;

acquiring a pre-application set of characteristics of said portion of said substrate before said material is applied on said substrate;

performing a first registration to align said pre-application set of characteristics to said model set of characteristics to form an aligned pre-application set of characteristics;

acquiring a post-application set of characteristics of said portion of said substrate and said at least one predefined pattern after said material is applied on said substrate;

performing a second registration to align said post-application set of characteristics to said aligned pre-application set of characteristics to form an aligned post-application set of characteristics;

processing said aligned pre-application set of characteristics and said aligned post-application set of characteristics to provide an image representative of said material applied in said at least one predefined pattern on said portion of said substrate; and

analyzing said image representative of said material applied on said portion of said substrate to determine information about said material applied in said at least one predefined pattern on said substrate relative to said portion of said substrate.

2. The method of claim 1 wherein said material is solder paste and said substrate on which said material is applied is a printed circuit board (PCB).

3. The method of claim 2 wherein said solder paste is applied to said substrate by screen printing.

4. The method of claim 2 wherein said portion of said substrate is a solder pad on said PCB and said at least one predefined pattern is a solder paste blob.

5. The method of claim 4 wherein said step of acquiring said model set of characteristics of said portion of said substrate and said at least one predefined pattern involves training said vision system by constructing at least one data structure including solder pad data corresponding to a width of said solder pad, a height of said solder pad, and a location of said solder pad.

6. The method of claim 5 wherein said solder pad includes curved features and said at least one data structure further includes radial values characterizing said curved features.

7. The method of claim 5 wherein said at least one data structure further includes solder paste blob data corresponding to a width of said solder paste blob, a height of said solder paste blob, and a location of said solder paste blob.

8. The method of claim 7 wherein said solder paste blob includes curved features and said data structure further includes radial values characterizing said curved features.

9. The method of claim 2 wherein said vision system comprises a camera and a memory and said step of acquiring a pre-application set of characteristics of said portion of said substrate before said material is applied involves capturing an image of at least a part of said PCB using said camera and storing a captured image in said memory as a pre-image.

10. The method of claim 2 wherein said step of performing a first registration involves aligning an image of at least a part of said PCB with an image of at least a part of a model PCB comprising said model set of characteristics.

11. The method of claim 2 wherein said vision system comprises a camera and a memory and said step of acquiring a post-application set of characteristics involves capturing an image of at least a part of said PCB using said camera and storing said captured image in said memory as a post-image.

12. The method of claim 9 wherein said vision system comprises a camera and a memory and said step of acquiring a post-application set of characteristics involves capturing an image of at least a part of said PCB using said camera and storing said captured image in said memory as a post-image and wherein said step of performing a second registration involves aligning said post-image to said pre-image in order to facilitate combining said post-image and said pre-image.

13. The method of claim 12 wherein said step of processing said aligned pre-application set of characteristics and said aligned post-application set of characteristics involves combining said post-image and said pre-image to provide a combined image comprising first pixel values representative of solder paste applied on solder pads on said PCB and second pixel values representative of solder paste on said PCB in areas other than on solder pads.

14. The method of claim 13 wherein combining said post-image and said pre-image involves performing an image subtraction process in which a subtraction operation is performed with said post-image and said pre-image to provide said combined image which is a difference image comprising said first pixel values as signed values representative of solder paste applied on solder pads on said PCB and said second pixel values as signed pixel values representative of solder paste on said PCB in areas other than on solder pads, and wherein said difference image is duplicated and processed to provide a solder paste/pad image having information related to solder paste located on solder pads and a non-paste/pad image having information related to solder paste located on said PCB in areas other than on solder pads.

15. The method of claim 14 further comprising steps of,

creating masks to effect a separation of said pre-image into an on-pad area image having information relating to solder pad areas and an off-pad area image having information relating to said PCB in areas other than on solder pads;

combining said on-pad mask with said solder paste/pad image to form an on-pad difference image representative of position of solder paste on said solder pads;

combining said off-pad mask with said non-pad/paste image to form an off-pad difference image representative of position of solder paste on areas of said PCB other than on said solder pads;

converting said on-pad difference image and said off-pad difference image, using thresholding operations, to a first binary image and a second binary image respectively; and

combining said first binary image and said second binary image to form a single binary image representative of solder paste area distribution on said PCB.

16. The method of claim 15 wherein the step of combining said on-pad mask with said solder paste/pad image to form an on-pad difference image is effected by logically ANDing said on-pad mask with said solder paste/pad image.

17. The method of claim 15 wherein the steps of combining said off-pad mask with said non-pad/paste image to form an off-pad difference image is effected by logically ANDing said off-pad mask with said non-pad/paste image.

18. The method of claim 15 wherein the step of converting said on-pad difference image and said off-pad difference image to a first binary image and a second binary image respectively, using thresholding operations, involves using different respective thresholds for each pixel in the respective images which thresholds are locally adjusted based on said pre-image.

19. The method of claim 15 wherein the step of combining said first binary image and said second binary image to form a single binary image representative of solder paste on said PCB is effected by logiacally ORing said first binary image and said second binary image.

20. The method of claim 1 wherein the step of performing a first registration involves using normalized correlation to align said pre-application set of characteristics to said model set of characteristics.

21. The method of claim 1 wherein said step of analyzing said image representative of said material applied on said portion of said substrate includes analyzing said image as a single binary image, using a connected components analysis method to determine area of distribution of said material.

22. The method of claim 1 wherein said step of analyzing said image representative of said material applied on said portion of said substrate includes analyzing said image as a single binary image, to measure at least one of area, center of mass, and orientation of said at least one predefined pattern.

23. The method of claim 1 wherein said step of analyzing said image representative of said material applied on said portion of said substrate includes analyzing said image to determine at least one of:

indications of bridging between depositions of said material; and

offset between centers of mass of respective ones of said at least one predefined pattern and respective ones of said portion of said substrate.

24. The method of claim 1 wherein said material is applied in said at least one predefined pattern on said portion of said substrate using a stencil having at least one aperture corresponding to said at least one predefined pattern, and wherein said stencil is periodically inspected according to a stencil inspection method comprising the steps of:

acquiring a stencil model set of characteristics of said at least one predefined aperture from a model stencil;

acquiring a stencil pre-application set of characteristics of said at least one predefined aperture before said material is applied on said substrate;

performing a first stencil image registration to align said stencil pre-application set of characteristics to said stencil model set of characteristics to form an aligned stencil pre-application set of characteristics;

acquiring a stencil post-application set of characteristics of said at least one predefined aperture after said material is applied on said substrate;

performing a second stencil image registration to align said stencil post-application set of characteristics to said aligned stencil pre-application set of characteristics to form an aligned stencil post-application set of characteristics;

processing said aligned stencil pre-application set of characteristics and said aligned stencil post-application set of characteristics to provide a stencil image representative of said material in and proximate to said at least one aperture; and

analyzing said stencil image representative of said material in and proximate to said at least one aperture to determine information about location of said material.

25. The method of claim 24 wherein said step of acquiring a stencil model set of characteristics of said at least one predefined aperture from a model stencil further includes defining at least one blockage region corresponding to said at least one aperture and where inspection is effected for blockage of said at least one aperture using said vision system.

26. The method of claim 24 wherein said step of acquiring a stencil model set of characteristics of said at least one predefined aperture from a model stencil further includes defining at least one smear region corresponding to said at least one aperture and where inspection is effected for smear of said material proximate said at least one aperture using said vision system.

27. A system configured for inspecting a material applied in at least one predefined pattern on a portion of a substrate, said system comprising:

a camera configured to acquire a model set of characteristics of said portion of said substrate and said at least one predefined pattern from a model substrate, said camera being further configured to acquire a pre-application set of characteristics of said portion of said substrate before said material is applied on said substrate;

a processor receiving from said camera said model set of characteristics of said portion of said substrate and said at least one predefined pattern, and said pre-application set of characteristics of said portion of said substrate, said processor performing a first registration to align said pre-application set of characteristics to said model set of characteristics to form an aligned pre-application set of characteristics; and

a memory accessible to said processor, receiving therefrom and storing said aligned pre-application set of characteristics,

wherein said camera further acquires a post-application set of characteristics of said portion of said substrate and said at least one predefined pattern after said material is applied on said substrate, and said processor performs a second registration to align said post-application set of characteristics to said aligned pre-application set of characteristics to form an aligned post-application set of characteristics and processes said aligned pre-application set of characteristics and said aligned post-application set of characteristics to provide an image representative of said material applied in said at least one predefined pattern on said portion of said substrate.

28. The system of claim 27 wherein said processor is configured to analyze said image representative of said material applied on said portion of said substrate to determine information about said material applied in said at least one predefined pattern on said substrate relative to said portion of said substrate.

29. The system of claim 27 wherein said material is solder paste and said substrate on which said material is applied is a printed circuit board (PCB).

30. The system of claim 29 wherein said solder paste is applied to said substrate by screen printing.

31. The system of claim 29 wherein said portion of said substrate is a solder pad on said PCB and said at least one predefined pattern is a solder paste blob.

32. The system of claim 31 wherein said model set of characteristics of said portion of said substrate and said at least one predefined pattern is stored in said memory by constructing at least one data structure including solder pad data corresponding to a width of said solder pad, a height of said solder pad, and a location of said solder pad.

33. The system of claim 32 wherein said solder pad includes curved features and said at least one data structure further includes radial values characterizing said curved features.