Texas Instruments Corporation

Dennis Lee Doane

Charles K Harris

Rajiv Roy

Thomas J Doty

George L Streckmann

Paul Gerald Barker

Joe Douglas Woodall

Donald E Russell

Anthony S Leonard

David V Dudasko

Methods for measuring the low spatial reflectivity uniformity of a DMD spatial light modulator. These methods are unique since they compensate for the non-uniformities introduced by the tilt angle of the DMD mirrors in addition to the normal system non-uniformities introduced by the illumination source and optics. These methods flatten the image and remove all but the low spatial non-uniformities from the DMD mirrors.

Methods for measuring DMD low frequency spatial uniformity

A visual inspection system (10) and method for inspecting the operation of micromechanical devices. The system has a vision system (16), a vision system controller (18), a system controller (20), an illumination system (12), a digital signal processor and a mounting stage (22). The objects undergoing inspection, either wafers or devices, are mounted under the visions system. A probe card is engaged to the object and power is provided. The vision system (16) records which elements or devices on the object begin operating at that voltage. The voltage is then incremented and the process continues until a determination is made as to the objects operability. The system can also be adapted to inspect micromechanical spatial light modulators by generating images with the modulator and capturing them with the image system. The digital signal processor then analyzes the images while the vision system moves to the next object to be inspected.

Inspection system for micromechanical devices

A device (10) identifies a defective object (12) from an image of the object (12) stored as pixel data in a pixel-data memory (52, 54, 56, 58). An inspection-point memory (72, 74, 76, 78) stores inspection-point data for the object (12). The inspection-point data represents inspection points (122, 124, . . . ) that are arranged in circuital (112, 136) and transverse (110) groups, where member inspection points of the circuital groups (112, 136) are also members of different ones of the transverse groups (110). A program memory (84) stores a program, which is executed by a processor (64, 66, 68, 70). The processor (64, 66, 68, 70) identifies which of the pixel data correspond to the inspection-point data, calculates differences between values of the corresponding pixel data within the circuital groups (112, 136), identifies a deviant pixel (116) if one of the differences exceeds a first threshold, calculates gradients of values of the corresponding pixel data within the transverse group (110a) including the deviant pixel and within at least one other transverse group (110b, 110c), and determines whether the gradients of the transverse group (110a) differs by more than a second threshold from the gradient of the other transverse group (10b, 10c).

Apparatus and method for identifying defective objects

A device 12 acquires substantially centered images of moving items. A strobe (36) illuminates an item (16) while a camera 28 having a view (29) with a view center (39) acquires an image of that item (16). A position sensor (38) generates an indication signal that indicates when an item (16) is substantially a predetermined distance (D) from the view (29), and an image-data memory (31) stores data representing each image. A processor (24) executes a program for waiting a delay from each indication signal and then instructing the strobe (36) to fire, the camera (28) to acquire an image, and the image-data memory (31) to store the image. The processor (24) then determines an offset value representing the offset of the image from the view center (39), updates an average offset value using the offset value, and updates the delay using the updated average offset value. A program memory (20) stores the program, and a working memory (20) stores the offset value, the delay, and the average offset value.

Growing community of inventors

Dallas, TX, United States of America

Dennis Lee Doane