PCB007 Magazine


Issue link: http://iconnect007.uberflip.com/i/663184

Contents of this Issue


Page 17 of 87

18 The PCB Magazine • April 2016 path as you attempt to tech out of the problem. In Figure 1, examine the cross-section close- ly. Carefully determine if what one is seeing is a shrinking or recessing of the resin, or is the plated copper blistering or pulling away? Some would jump to the conclusion that the defect is HWPA. This would then trigger an exercise to brainstorm on a number of processes includ- ing desmear and electroless copper (including a deeper dive into drilling, electroless copper catalyzation and copper plating rates). However embarking on this path would be disastrous. Clearly the issue shown in not HWPA. The resin has recessed, leading to the proper conclusion that this is resin recession. According to IPC-600 H, section 3.1.9, res- in recession is acceptable for all classes of PCB boards after thermal stress. Now, in the case of the defect shown in Fig- ure 2, the plated copper clearly has lost adhe- sion to the hole wall. When a condition such as HWPA is discov- ered, there are several potential causes of the defect: 1. Overactive electroless copper process: Es- sentially, the deposit is being laid down much too fast, leading to a stressed condition. The in- herent stress in the deposit causes the copper to pull away from the substrate. Check the deposi- tion rate on the electroless copper rate panels- was an increase noted possibly due to over tem- perature condition in the electroless copper so- lution? Is the chemistry out of balance? Check the formaldehyde, caustic and copper concen- tration. Is the process being controlled within the operating window of these additives? A higher than normal caustic concentration will lead to a higher deposition rate. Examine the grain structure of the electroless copper deposit. Compare to previous results when HWPA did not exist. 2. Desmear operation leaving a less than desirable surface topography on the resin: Typically, this is an issue related to the per- manganate process. Higher Tg resins are more resistant to desmear chemistry and thus the surface topography and extent of resin remov- al are less than optimum. Weight loss mea- surements utilizing a small coupon made of the same resin system and processed through the desmear process under the current condi- tions will yield some evidence. An SEM taken of the PTH will give a good picture of the to- pography on the resin after desmearing. Do not rely on weight loss data alone. A sufficiently micro-roughened resin surface provides a sur- face area to promote catalyst adsorption as Figure 1: Defect clearly shows resin recession, not HWPA. Figure 2: Severe case of HWPA. a proCess engineer's guiDe to effeCtively troubleshooting pWb DefeCts

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB-Apr2016