基于芯片引线键合的视觉定位技术设计与研究

发布时间：2018-10-10 15:38

【摘要】：随着集成电路的发展,应对各种新材料与半导体的需要,先进的封装技术也随之发展和改变,为了使芯片能跟外设平滑输入\输出,需要建立电连接,这个连接是在外部管脚和内部芯片之间,还有芯片相互之间,并且是整个厚后封装过程中极重要的步骤,90%以上的所有封装管脚使用引线键合焊接,实现了简单、成本低、适合于各种中的主导地位的连接方式。引线键合之前,首先从引线框架材料(外引线)上选取金属带切断,通过热压方法加压在引线框架上的高纯度Si或Ge的半导体元件,和一个导电树脂如银浆料覆有金或通过在引线框架的表面上少部分电镀一层很薄的金;用特殊焊接方式使电路和金属丝键合,贴合后的电路是带有保护性的树脂封装。到20世纪90年代,机器视觉系统的研究和集成电路芯片制造设备的集成电路芯片视觉检测如雨后春笋般出现了。从印刷电路板(PCB)的检测,芯片封装过程以及向焊点,焊丝质量检测,以实时检测和控制,越来越广泛的使用范畴。视觉系统被用来从简单的存在性验查和分类进行检测,涉及到尺寸检查和表面质量检查,然后对集成电路芯片精度定位,这将很大程度上提高视觉伺服控制和视觉系统的实时性和复杂性。由于芯片的集成度日益变高,芯片封装速度和精度需要更先进的封装设备。近年来先后陆续出现焊接精度低于2.5um,焊线速度超过12线/秒的并且精度很高的自动金丝球焊机,这些机型的精度和速度的控制系统的改进主要是由于线性电机驱动的使用、软触摸控制和许多其他许多新技术的三维工作台的采用;以及视觉系统使用多倍放大的显微视觉系统,高清CCD摄像机和一个独特的视觉定位方法。本文的主要目的是考察机器人精密平台在3D空间进行芯片引线键合等操作,因此主要研究内容将围绕在显微环境下的视觉控制器的设计和摄像机参数的自标定,同时考察芯片图像的精密定位方法及视觉控制的效果、并探讨视觉系统与运动控制系统集成的坐标变换关系,及反馈控制实现方式。在现有的实验条件下,通过对现实生产环境的模拟实验表明,通过本文提出的芯片封装显微视觉系统的标定算法定位所得的坐标更为精确,误差相对更小。
[Abstract]:With the development of integrated circuits, to meet the needs of various new materials and semiconductors, advanced packaging technology has also developed and changed. In order to make the chip and peripheral devices smooth input and output, we need to establish electrical connection. This connection is between the external pin and the internal chip, and also between the chips, and is an extremely important step in the whole thick post-packaging process. More than 90% of all packaging pins are welded by wire bonding, which is simple and low cost. Suitable for all kinds of dominant connections. Before the lead bonding, the semiconductor elements of high purity Si or Ge on the lead frame are first selected from the lead frame material (outer lead) to be cut off by hot pressing method. And a conductive resin such as silver paste coated with gold or by electroplating a thin layer of gold on the surface of the lead frame; the circuit is bonded to the wire by special welding, and the circuit after bonding is a protective resin package. By the 1990s, the research of machine vision system and the detection of integrated circuit chip for IC chip manufacturing equipment appeared. From PCB (PCB) detection chip encapsulation process to solder joint welding wire quality detection to real-time detection and control more and more widely used fields. Visual systems are used to detect from simple existential checks and classifications, involving size checks and surface quality checks, and then positioning the accuracy of integrated circuit chips. This will greatly improve the real-time and complexity of visual servo control and visual system. Due to the increasing integration of the chip, the speed and precision of the chip packaging need more advanced packaging equipment. In recent years, automatic gold-wire ball welding machines with welding accuracy of less than 2.5 umum, wire speed of more than 12 lines per second and high precision have been successively appeared. The improvement of the precision and speed control system of these machines is mainly due to the use of linear motor drive. Soft touch control and the adoption of many other new technologies for 3D workbench; and vision systems using multiple magnification microvision systems, high-definition CCD cameras and a unique visual positioning method. The main purpose of this paper is to investigate the robot precision platform in 3D space chip lead bonding operations, so the main research content will focus on the design of the vision controller and camera parameters self-calibration in the micro-environment. At the same time, the precision positioning method of chip image and the effect of visual control are investigated, and the coordinate transformation relationship between visual system and motion control system is discussed, and the realization of feedback control is also discussed. Under the existing experimental conditions, the simulation experiments on the real production environment show that the calibration algorithm proposed in this paper is more accurate and the error is relatively small.
【学位授予单位】：武汉工程大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN405;TP391.41

【参考文献】