Issue link: https://iconnect007.uberflip.com/i/1288481
SEPTEMBER 2020 I PCB007 MAGAZINE 119 mal shock. But if you did reflow—which is this slide, plus the thermal shock—you would say, "No, I don't want to assemble those four seri- al numbers because they actually went open at reflow, even after thermal shock. They are still connected, and maintaining their connec- tion those boards that have failed could lead to problems out in the field later." There is also the ability to evaluate using IST coupons and current to induce heat in the speci- men to evaluate whether the structure is strong. It's a great system to compare changing one pro- cess to the other, and I highly agree with that. It's a good comparison system to say these are stronger than those, but it doesn't take the cou- pon to reflow temperatures across the board, like we do with an actual board in assembly. On the bottom right in Figure 12, what you see is the thermal image of a coupon, and a current is flowing through the daisy chain. The hottest point in the coupon is in the dead cen- ter, which is going to have the most stress. The connector is soldered on there. It is not as if it had reflowed the coupon and was going to re- flow to have the problem with the connector. The actual coupon itself did not reflow across the whole coupon, just like your board is go- ing to get reflow across the entire board. Then, there's the example with the arrows. You can see the little black dots where the ar- rows are pointing, where every failed microvia was on five different coupons. They are all in the center, where the heat's the strongest. By the way, these all come from different panels and positions on those panels. I do not have the abil- ity to make the weakest microvias in the middle of a coupon specifically over again when they are all the same geometries. It is a good compar- ison test to validate your process. But it does not take your coupon, which should have the same structures as your board and takes the reflow temperature just like your portal experience. IPC-TM-650 2.6.27A comes to the rescue. What we are using is a D-coupon, which is approved by IPC. The coupon can be built by anybody by going to Conductor Analysis Tech- nology and building your own D-coupon that meets the IPC requirements. When you gen- erate your coupon, you have generated it just like your board. You need to know the diam- eter of your microvia or mechanical or buried vias and use the same land size. You specify whether there is a plane layer on layer two or if it is a signal layer. You build the coupon exactly like your board so that it tests a structure just like your board. It is critical that you have the distance between the drill laser and the mechanical drill at the same distance, Figure 12: In IST testing the greatest heat is in the center of the coupon. • In some test methods, coupons are not thermally cycled at actual reflow temperature. The connector is soldered to the coupon and, therefore, cannot be reflowed. • Coupon is a heating coil with five amps of current through a microvia daisychain. The temperature is theoretical, not measured. The greatest heat is at the center of the coupon as seen in the thermal image shown. • OM coupons use the actual recorded temperature readings.