The Development Of Inspection System Of PCB

With the constant development of technology and continuous progress ofindustry, electronic products tend to become lighter, thinner, shorter and smaller.In the production process of electronic components, it always occurs rosin joint,missed welding, open circuit, short circuit, welding components errors and otherproblems. At present, the inspections of welding quality are mostly completed byvisual observation, because of human factors, missed and misuse of elements intest often happen, which is time-consuming and inefficient. So how to takewelding quality tests of mass electronic production has become an urgentrequirement in electronics industry development.The thesis describes a design of an on power PCB automatic testing systemusing microcomputer, optical and infrared photography, digital image processingand information management database.By in-depth study on the present situation and trend of development of onpower PCB test system, associated with the requirements of high-resolutionpicture, the thesis offers a method combines with machine vision technology,automatic control technology and framing image acquisition technology tocollect the PCB image which solves the conflict between the high imageresolution ratio and small field. In the process of image acquisition, the deviationbetween images is mainly determined by the positioning accuracy of theoperating platform. The kinetic control system of the operating platform choosesa high-precision code as position sensor. The motor control algorithm uses theincremental PID control algorithm, so that the platform can meet the requirementsof the system. A DSP chip TMS320F240 is chosen as the main machine controlerusing bridge driving circuit to drive 2D workstations for horizontal movementwith the features of smooth movement, high integration, low cost and convenientcontrol.The system chooses VC++6.0 as software environment, using median filteringalgorithm and sharpening gradient optimization of image quality imageprocessing algorithm to improve the effectiveness of the detection algorithm. Thesystem uses the database by ADO technology which is developed to manage batch detection circuit board products information, and solves the welding productsquality control issues.