SMT007 Magazine

SMT-May2014

Issue link: https://iconnect007.uberflip.com/i/304995

Contents of this Issue

Navigation

Page 21 of 100

22 SMT Magazine • May 2014 C. aXI 3 (Tomosynthesis) AXI 3 machine uses Tomosynthesis technol- ogy which is similar to AXI 2 machine. The AXI 3 has faster testing speed than AXI 1, and AXI 2. We used the same HIP boards tested at four sites, and worked with the vendor to have al- gorithm and threshold settings studies. The al- gorithm threshold settings are different when we tested the boards at different site with the latest software at that time. The main algorithm thresholds for detecting HIP with AXI 3 are listed in Table 5. Because board #495 and #266 were built at different manufacturing facilities with little difference in solder shape and vol- ume, different algorithm thresholds were set to detect HIP effectively at site 3 and site 4. Figure 5 shows detecting HIP pins, escaped pins, and false call pins for board #495, and all pins for board #266 which is from site 2. The AXI 3 flagged pin as HIP when its pad "neighbor outlier" > 2.2 with multi-pass algorithm. AXI3 detected 38 HIPs (escaped 7) with six false calls for board #495, and detected two HIP (no es- capes) with six false calls for board #266. Table 6 listed the test summary from each site with AXI 3 which has better performance than AXI 1 for both HIP detection percentage and false call PPM. The AXI 3 images for detected HIP, escaped HIP and false calls are shown in Figure 6. There is no very clear difference between good and HIP pins similar to AXI 1, and AXI 2. It is not HEaD-IN-PILLoW X-Ray INSPECTIoN continues feATuRe Table 4. Testing summary with AXI 2. Figure 4: axi 2 images of detected hip; escaped hip pins (board #495). Table 5: Algorithm thresholds with AXI 3.

Articles in this issue

Archives of this issue

view archives of SMT007 Magazine - SMT-May2014