Issue link: https://iconnect007.uberflip.com/i/963057
62 PCB007 MAGAZINE I APRIL 2018 balance, thin tin deposits are often the result. Excessive turbulence in the plating solution leads to excessive suppression of the deposi- tion of metal at the knee. At the same time, additive imbalance in the electrolytic copper plating solution leads to thin copper plating at the corner (Figure 2). It has been theorized that levelling components, carrying a positive charge, migrate preferential- ly to high-current density areas, just like metal ions, and inhibit metal deposition. If too much leveller is present, either because the bulk con- centration is too high or the replenishment rate of leveler to the through-hole corner is higher than its consumption rate, then metal deposi- tion at the corner will be suppressed leading to the thin knee condition. (I will present more detail on plating processes and the functioning of addition agents in a future column.) As the micro-section on Figure 3 shows, it is also possible to get micro-voids in the via due to the action of the pre-plate microetch. When the microetch removes a portion of the electro- less copper, there will be sporadic voids in the via. Since these voids create a discontinuity in the via, the electrolytic copper deposition will face an extreme resistance through the via. It will be nearly impossible for the plated copper to bridge these voids. It is not that one type of chemical microetch is more beneficial than another. What the engineer must ensure here is that the microetch chemistry is maintained within critical operating limits. Finally, some voids can be attributed to the photoresist process. Although this is a rarity, one should not discount the possibility that re- sist can flow into the holes. When this situa- tion occurs, rim voids are possible. The resist flows in such a way that the developing solu- tion cannot remove the resist. The resist thus acts as a barrier, preventing the deposition of electroplated copper (Figure 4). How did this situation occur? One plausible cause is moisture entrapment in the via. Moisture re- maining in the via creates a vac- uum during the resist lamination process. Resist flows into the via (remember moisture makes things flow more), creating a situation where the developer is not effec- tive at removing the unexposed re- sist. This resist then prevents sub- sequent electroplating. Howev- er, the electroless copper is pres- ent. And that is the clue that the void was not present after electro- less copper. Thus, with the resist lodged into the via, electroplat- ed copper will not deposit in that Figure 2: Thin knee in copper plating. Figure 3: Pre-electrolytic plating-microetch issue.