PCB007 Magazine


Issue link: https://iconnect007.uberflip.com/i/1104607

Contents of this Issue


Page 89 of 107

90 PCB007 MAGAZINE I APRIL 2019 shape showed good to average "throw" in the blind slots (Figures 1 and 2). Our results showed that the longer the slot, the easier it is to plate. Shorter and deeper slots tend to exhibit a threshold beyond which they experience a lack of plating. The threshold is defined by the depth and length of the slot. Presently, we target an AR threshold of maxi- mum 4:1 in our designs with a ratio of slot length to slot width at a minimum of 3:1. An interesting development from this initial work is that the AR definition for a blind slot now requires an addition - al dimensional variable. Along with slot width, depth, and the new parameter, length. Ultimately, the objective is to create deep slots up to 2 mm for regular circuit boards and even 3 mm for more ad - vanced products. In addition to the slot depth, NextGIn's experiments concentrated on slot diameters in the region of 0.2–0.5 mm. NextGIn selected this range because larger slot widths are not as useful with respect to BGA component footprints, and smaller width slots—width of interest—are difficult to form reliably for production due to the stability and useful life of the mechanical drill bits. Copper Plating Tests Naturally, plating thicknesses in the slot need to comply with industry standards. In Figure 2, the cross-sections show the results of plating a 0.5 mm 0.7 mm 0.9 mm 1.1 mm 1.3 mm 1.5 mm 1.7 mm Figure 1: Slot dimensions used to conduct initial "throw" testing. Figure 2: Test results, demonstrating that the more a slot resembles a circle, the shorter the plating "throw" capability.

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB007-Apr2019