Head in Pillow X-ray Inspection at Flextronics



Head in Pillow X-ray Inspection at Flextronics
For the study we used two boards exhibiting HIP defects with four types of AXI machines at four sites in Flextronics manufacturing, or vendor laboratory.
Analysis Lab

DOWNLOAD

Authored By:


Alejandro Castellanos, Adalberto Gutierrez, Gilberto Martin, Matthew Vandiver, Ranga Dematampitiya, Hung Le, Elliott Le, Phuong Chau, Hao Cui, An Qi Zhao, Wei Bing Qian, Fuqing Li, Jacky Yao, Jiyang Zhang, Leonard Brisan, Cristian Gurka, Shane Young, Johann Bruenner, Martin Novak, Nadarajan M Singaram, Zhen (Jane) Feng, Ph. D., David Geiger, Murad Kurwa
FLEXTRONICS International Inc.
Guadalajara, Jalisco 45640, Mexico

Summary


Manufacturing technology faces challenges with new packages/process when confronting the need for high yields. Identifying product defects associated with the manufacturing process is a critical part of electronics manufacturing. In this project, we focus on how to use AXI to identify BGA Head-in-Pillow (HIP), which is challenging for AXI testing.

Our goal is to help us understand the capabilities of current AXI machines. For the study we used two boards exhibiting HIP defects with four types of AXI machines at four sites in Flextronics manufacturing, or vendor laboratory. The AXI machines used have different X-ray technologies: Laminography and Tomosynthesis. We collected three sets of data with AXI 1 machine (Laminography), and AXI 4 machines (Tomosynthesis); one set of data with AXI2 (Tomosynthesis); and 4 sets data for AXI3 (Tomosynthesis).

We studied AXI measurement data with the different AXI Algorithm Threshold settings. The data indicated clearly that the Algorithm Threshold settings are very critical for detecting HIP, including open. The defective HIP pins are validated by using 2DX and CT scan.

The test data consist of Defects Escaped %, False call PPM and also Gage R & R. The AXI images for HIP pins, false call pins and defects escaped pins are presented in the paper. The 2DX and CT images are provided for identifying HIP type (shape and size).

Conclusions


•Current AXI has capabilities to detect some HIP with right Algorithm threshold settings. The Tomosynthesis technology looks better than Laminography technology for detecting HIP as the Tomosynthesis technology are utilizing digital technology as compared to the analog Laminography technology.
•The Algorithm threshold settings are very critical for detecting HIP. The false call number has to be reasonable for the production line. The AXI program optimization is based on its measurement data analysis.
•There are no clear image differences between good solder joint and HIP with current AXI machine, especially for HIP type defect AA15 shown as in Fugue 11. Therefore we have to have good balance between HIP defect escaped and false
call.
•We are looking forward to see AXI machine with more accurate and repeatable measurement data, and better image
separation between good solder joint and HIP pins.
•2DX and Large Board CT are very important techniques used to verify the AXI results and fine tune the algorithms.


Initially Published in the IPC Proceedings

Comments

No comments have been submitted to date.

Submit A Comment


Comments are reviewed prior to posting. You must include your full name to have your comments posted. We will not post your email address.

Your Name


Your Company
Your E-mail


Your Country
Your Comments