Issue link: https://iconnect007.uberflip.com/i/1313710
DECEMBER 2020 I SMT007 MAGAZINE 79 microvia testing (power-on-sense). The buried vias have an aspect ratio close to eight (300 µm drill diameter in a 2.4-mm core). The num- ber of cycles reached is between 449 and 560, which is above the requirement and in line with expectations. Pre-cycling for the buried vias on the microvia test coupons (SLX3B, 6B, and 9B) shows a similar result of 407 to >500 cycles. The required endurance is reached on all coupons, although the margin on coupon SLX3B is slim. Two out of the three coupons for microvia testing reached the required 400 cycles. Testing of these coupons was stopped after 500 cycles to 210°C. Coupon SLX3B failed after 285 cycles and did not reach the required endurance. As the failure could not be located using thermal imaging, failure analysis was performed by microsectioning the entire coupon row by row. Optical inspection revealed cracks near the target pad of microvia 1-2. These cracks are observed on two microvias in the central zone of the coupon. The other microvia levels in this zone did not show any cracks. In other areas of the coupon, no cracks were detected on any of the microvia levels. Microvia 1-2 in the SLX coupon is located partially above the buried via (0.8-mm pitch fanout configuration). It is known that this leads to higher stress. If the microvias in coupon SLX3B are slightly mis- aligned and thus overlap more with the buried via as compared to the other panels, this could explain the early failure. Three BGA coupons were submitted to the assembly and life test flow. The test samples were subjected to reflow simulation (two times vapor phase reflow at 215°C). In addition, rework simulation (ten times manual solder- ing) was performed on plated through-holes of the high-density connectors on the BGA cou- pon (two times four PTHs). All coupons were baked for eight hours at 120°C before reflow and rework simulation. BGA6A and 9B were subjected to 500 cycles from -55°C to +100°C, while BGA6B was subjected to 200 cycles from -60°C to +140°C. Thermocouples are attached to all BGA coupons to monitor the board tem- perature during testing. The interconnection resistance of the daisy chains in the 0.8-mm pitch fanout was monitored during testing (four-point resistance measurement every five seconds using Keithley 3706A Systems Switch/ Multimeter). The final evaluation was per- formed by microsectioning. Figure 4 shows the details of the daisy chains in the 0.8-mm pitch fanout used for electrical monitoring. DC1 connects all the vias in the outer row of the component fanout, DC2 con- nects all the vias in the second row, and so on. Only the buried vias are included in the daisy chains, and the number of buried vias per chain is indicated in the table on the right. Electrical monitoring during thermal cycling (Table 6 and Table 7) showed only a minimal increase in resistance for both tem- perature ranges. This confirms the positive IST results for the buried vias. Microsection- ing did not reveal any cracks initiating in the barrel of the buried via nor any anomalies in the microvias. Figure 4: Layout and number of vias of the daisy chains in the 0.8-mm pitch fanout.