Method and apparatus for three dimensional inspection of electronic components

• US 6,072,898 A
• Filed: 01/16/1998
• Issued: 06/06/2000
• Est. Priority Date: 01/16/1998
• Status: Expired due to Fees

• Current Assignee: Scanner Technologies Corporation
• Original Assignee: Elwin M. Beaty
• Inventors: Mork, David P., Beaty, Elwin M.
• Primary Examiner(s): MEHTA, BHAVESH M
• Application Number:
  US09/008,243
• Time in Patent Office: 872 Days
• Field of Search: 382/141, 382/145, 382/146-151, 382/154, 382/144, 356/237.1, 356/376, 356/377, 250/559.23, 250/559.31, 250/559.34, 250/559.4, 348/86-87, 348/94-95, 348/125-126, 702/150-153, 702/166-169
• US Class Current: 382/146
• CPC Class Codes: G06T 2207/30152 Solder
• CPC Class Codes: G06T 2207/30244 Camera pose
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: H01L 2224/13 of an individual bump connector
• CPC Class Codes: H05K 13/0818 Setup of monitoring devices prior to starting mounting operations; Teaching of monitoring devices for specific products; Compensation of drifts during operation, e.g. due to temperature shifts

Abstract

A calibration and part inspection method and apparatus for the inspection of ball grid array, BGA, devices. Two cameras image a precision pattern mask with dot patterns deposited on a transparent reticle. The precision pattern mask is used for calibration of the system. A light source and overhead light reflective diffuser provide illumination. A first camera images the reticle precision pattern mask from directly below. An additional mirror or prism located below the bottom plane of the reticle reflects the reticle pattern mask from a side view, through prisms or reflective surfaces, into a second camera and a second additional mirror or prism located below the bottom plane of the reticle reflects the opposite side view of the reticle pattern mask through prisms or mirrors into a second camera. By imaging more than one dot pattern the missing state values of the system can be resolved using a trigonometric solution. The reticle with the pattern mask is removed after calibration and the BGA to be inspected is placed with the balls facing downward, in such a manner as to be imaged by the two cameras. The scene of the part can thus be triangulated and the dimensions of the BGA are determined.

Claims

1. An apparatus for inspecting a ball grid array, wherein the apparatus is calibrated using a precision pattern mask with dot patterns deposited on a calibration transparent reticle, the apparatus for inspecting a ball grid array comprising:

a) a means for mounting the ball grid array;

b) a means for illuminating the ball grid array to provide an outline of the ball grid array;

c) a first camera positioned to image the ball grid array to provide a first image of the ball grid array;

d) a first means for light reflection positioned to reflect the ball grid array through a second means for light reflection into a second camera, wherein the second camera provides a second image of the ball grid array;

e) a third means for light reflection positioned to reflect an opposite side view of the ball grid array into a fourth means for light reflection and into the second camera as part of the second image of the ball grid array;

f) a means for image processing the first image and second image of the ball grid array to inspect the ball grid array; and

g) a third camera, wherein the first camera enables direct imaging of a bottom view, wherein the first camera is located below a central area of the ball grid array, wherein the third camera is located to receive an image of a single side perspective view and uses fixed optical elements to magnify the single side perspective view in one dimension, wherein the second camera is positioned to image a second side perspective view and uses fixed optical elements to magnify the second side perspective view in one dimension, and wherein the means for image processing calculates a Z position of a ball on the ball grid array from the bottom view, first side perspective view and second side perspective view.

2. The apparatus of claim 1 wherein the means for illuminating the ball grid array further comprises a light source for illuminating the ball grid array, and an overhead light reflective diffuser provides illumination of the ball grid array from the light source.

3. The apparatus of claim 1 wherein the first means for light reflection further comprises a mirror.

4. The apparatus of claim 1 wherein the first means for light reflection further comprises a prism.

5. The apparatus of claim 1 wherein the first means for light reflection further comprises a curved mirror.

6. The apparatus of claim 1 wherein the second means for light reflection further comprises a mirror.

7. The apparatus of claim 1 wherein the second means for light reflection further comprises a prism.

8. The apparatus of claim 1 wherein the second means for light reflection further comprises a curved mirror.

9. The apparatus of claim 1 wherein the ball grid array further comprises balls on a wafer.

10. The apparatus of claim 1 wherein the ball grid array further comprises balls on a die.

11. The apparatus of claim 1 wherein the first camera is connected to a frame grabber board to receive the first image.

12. The apparatus of claim 11 wherein the frame grabber board provides an image data output to a processor to perform a two dimensional inspection.

13. The apparatus of claim 1 further comprising a nonlinear optical element to magnify the second image in one dimension.

14. The apparatus of claim 1 wherein an optical path length of a second side perspective view is equal to an optical path length of a first side perspective view.

15. The apparatus of claim 1 wherein a maximum depth of focus of a side perspective view allows for a fixed focus system to inspect larger ball grid arrays, with one perspective view imaging one portion of the ball grid array and a second perspective view imaging a second portion of the ball grid array.

16. The apparatus of claim 15 wherein one portion of the ball grid array is at least a first half of the ball grid array and the second portion of the ball grid array is at least a second half of the ball grid array.

17. The apparatus of claim 1 wherein a maximum depth of focus of a side perspective view includes an area of the ball grid array including a center row of balls.

18. The apparatus of claim 1 wherein all of the balls on the ball grid array are in focus from both side perspective views resulting in two perspective views for each ball.

19. The apparatus of claim 1 further comprising a means for inspecting gullwing and J lead devices.

20. A method for three dimensional inspection of a lead on a part mounted on a reticle, the method comprising the steps of:

a) locating a first camera to receive an image of the lead;

b) transmitting an image of the lead to a first frame grabber;

c) providing fixed optical elements to obtain two side perspective views of the lead;

d) locating a second camera to receive an image of the two side perspective views of the lead;

e) transmitting the two side perspective views of the lead to a second frame grabber;

f) operating a processor to send a command to the first frame grabber and second frame grabber to acquire images of pixel values from the first camera and the second camera;

g) processing the pixel values with the processor to obtain three dimensional data about the lead, including a rotation, an X placement value and a Y placement value of the part relative to world X and Y coordinates, by finding points on four sides of the part;

h) using a part definition file that contains measurement values for an ideal part;

i) calculating an expected position for each lead of the part for a bottom view using the measurement values from the part definition file and the X placement value and Y placement value;

j) converting the pixel values into world locations by using pixel values and parameters determined during calibration wherein the world locations represent physical locations of the lead with respect to world coordinates defined during calibration; and

k) calculating a Z height of each lead in world coordinates in pixel values by combining a location of a center of a lead from a bottom view with a reference point of the same lead from a side perspective view.

21. The method of claim 20 wherein the lead is a curved surface lead.

22. The method of claim 20 wherein the lead is a ball.

23. The method of claim 20 wherein the part is a ball grid array.

24. The method of claim 20 wherein the step of processing the pixel values with the processor further comprises the step of sending a signal to a part handler to allow the part handler to move the part out of an inspection area and allow a next part to be moved into the inspection area.

25. The method of claim 20 further comprising the step of using a search procedure on the image to locate the lead.

26. The method of claim 20 further comprising the step of determining a lead center location and a lead diameter in pixels and storing the lead center location and lead diameter in memory.

27. The method of claim 26 further comprising the step calculating an expected position of a center of each lead in both side perspective views in the image using a known position of each side view from calibration.

28. The method of claim 20 further comprising the step using a subpixel edge detection method to locate a reference point on each lead.

29. The method of claim 20, further comprising the step of converting the world values to part values using the rotation, the X placement value and the Y placement value to define part coordinates for the ideal part where the part values represent physical dimensions of the lead including lead diameter, lead center location in X part and Y part coordinates and lead height in Z world coordinates.

30. The method of claim 29 further comprising the step of comparing ideal values defined in part value to calculate deviation values that represent a deviation of the center of the lead from its ideal location.

31. The method of claim 30 wherein the deviation values may include lead diameter in several orientations with respect to the X placement value and Y placement value, lead center in the X direction, Y direction and radial direction, lead pitch in the X direction and Y direction and missing and deformed leads, further comprising the step of calculating the Z dimension of the lead with respect to a seating plane based on the Z world data.

32. The method of claim 30 further comprising the step of comparing the deviation values to predetermined tolerance values with respect to an ideal part as defined in the part definition file to provide a lead inspection result.

33. The method of claim 32 wherein the predetermined tolerance values include pass tolerance values and fail tolerance values from industry standards.

34. The method of claim 33 further comprising the steps of:

(a) determining if the deviation values are less than or equal to the pass tolerance values, wherein the processor assigns a pass result for the part;

(b) determining if the deviation values exceed the fail tolerance values, wherein the processor assigns a fail result for the part; and

(c) determining if the deviation values are greater than the pass tolerance values, but less than or equal to the fail tolerance values, wherein the processor designates the part to be reworked.

35. The method of claim 33 further comprising the step of reporting a part inspection result for the part.

36. A method to inspect a ball grid array device comprising the steps of:

(a) locating a point on a world plane determined by a bottom view ray passing through a center of a ball on the ball grid array device;

(b) locating a side perspective view point on the world plane determined by a side perspective view ray intersecting a ball reference point on the ball and intersecting the bottom view ray at a virtual point where the side perspective view ray intersects the world plane at an angle determined by a reflection of the side perspective view ray off of a back surface of a prism where a value of the angle was determined during a calibration procedure;

(c) calculating a distance L₁ as a difference between a first world point, defined by an intersection of the bottom view ray with a Z=0 world plane, and a second world point, defined by the intersection of the side perspective view ray and the Z=0 world plane, where a value Z is defined as a distance between a third world point and is related to Li as follows;

##EQU8## wherein Z is computed based on the angle;

(d) computing an offset E as the difference between the virtual point defined by the intersection of the bottom view ray and the side perspective view ray and a crown of a ball at a crown point that is defined by the intersection of the bottom view ray with the crown of the ball, and can be calculated from a knowledge of the angle and ideal dimensions of the ball where a final value of Z for the ball is;

space="preserve" listing-type="equation">Z.sub.Final =Z-E.

37. A method for finding a location and dimensions of a ball in a ball grid array from a bottom image, the method comprising the steps of:

(a) defining a region of interest in the bottom image of an expected position of a ball where a width and a height of the region of interest are large enough to allow for positioning tolerances of the ball grid array for inspection;

(b) imaging the ball, wherein the ball is illuminated to allow a spherical shape of the ball to present a donut shaped image, wherein the region of interest includes a perimeter of the ball wherein the bottom image comprises camera pixels of higher grayscale values and where a center of the bottom image comprises camera pixels of lower grayscale value and wherein a remainder of the region of interest comprises camera pixels of lower grayscale values;

(c) finding an approximate center of the ball by finding an average position of pixels having pixel values that are greater than a predetermined threshold value;

(d) converting the region of lower grayscale pixel values to higher grayscale values using coordinates of the approximate center of the ball; and

(e) finding the center of the ball.

38. The method of claim 37 further comprising the step of determining a diameter of the ball.

39. The method of claim 37 wherein the diameter of the ball is equal to (2 * sqrt(Area/3.14)) wherein the Area is equal to a sum of pixels in the region.

40. A method for finding a reference point on a ball in an image of a side perspective view of a ball grid array, the method comprising the steps of:

a) defining a region of interest in the image from an expected position of a ball wherein a width and a height of the region of interest are large enough to allow for positioning tolerances of the ball grid array;

b) imaging the ball, wherein the ball is illuminated to allow a spherical shape of the ball to present a crescent shaped image having camera pixels of higher grayscale values, and wherein a remainder of the region of interest comprises camera pixels of lower grayscale values;

c) computing an approximate center of the crescent shaped image by finding an average position of pixels that are greater than a predetermined threshold value;

d) using coordinates of the approximate center of the crescent to determine a camera pixel as a seed pixel representing a highest edge on a top of the crescent shaped image; and

e) determining a subpixel location of the reference point based on the camera pixel coordinates of the seed pixel that define coordinates of a region of interest for the seed pixel.

41. The method of claim 40 wherein in a side perspective image of a reticle calibration pattern, space between dot images is magnified, increasing a number of higher value grayscale pixels when compared to a nonmagnified image.