Issue link: https://iconnect007.uberflip.com/i/1529411
38 PCB007 MAGAZINE I NOVEMBER 2024 a specific gravity of 32°Be (1.283 sg) or about 24% cupric chloride. e freer the acid, the faster the etch rate, so a lot of people then ran their cupric at 32°Be, with free acid levels of up to 2N and some as high as 3N. By etch quality, I am talking about etch fac- tors or straightness of sidewalls. Cupric chlo- ride generally etches 1 unit sideways for every three units down. is means that if you are attempting to etch a 3-mil (75 µm) line in 1-ounce (35 µm) foil, you would measure about 2.5 mils (64 µm) across the top of the line when the bottom of the line reaches its finished etch width of 3 mils (75 µm). Reducing the amount of sideways etch will give a wider measurement across the top of the line and a better etch qual- ity. Perfection would be measuring 3 mils (75 µm) across the top of the line when you mea- sure 3 mils (75 µm) across the bottom. ose carefree days are long gone (although they didn't seem so carefree at the time). Lines and spaces, and specifications have become much tighter, and improving etch quality by reducing sideways etch has become paramount. We did one of our best projects in the mid-'90s with a manufacturer of lead frames who was pushing the limits of finger spacing and finger widths in 6-mil (150 µm) copper etched from both sides. We considered every variable in the chemistry and etching equipment that might affect the sideways etch, and we incorporated them into a complicated design of experiment (DOE) with 32 separate tests. e advantage of the DOE method is that it allows the simulta- neous testing of several variables. en, using statistical methods, it is possible to determine which variables have the most effect on the process being studied. e chemical variables we studied included etchant specific gravity, etchant temperature, free acid level, and oxi- dation/reduction potential (ORP). Equipment variables included spray pressure, type of noz- zles (fan or cone), nozzle height, nozzle flow rate, and oscillation (yes or no). Finally, we included a test with different panel widths. It took a few months to run all 32 tests twice, but the results were conclusive: e variables that had the largest effect on sideways etch were the free acid level followed closely by the specific gravity. e rest of the variables tested had only minor effects and were not worth the effort to test further. It may sur- prise some to learn that fan and cone nozzles did not have a statistical difference in their effect on sideways etch. Tests were then run on specific gravities of between 28–40°Be (1.24-1.38 sg). As the spe- cific gravity increased, the etch rate slowly decreased until reaching 36°Be (1.33 sg), aer which the etch rate began to drop significantly. e rate of sideways etch also slowly decreased (which is what we want), until, again, the spe- cific gravity reached 36°Be, when the rate of decrease also began to drop significantly. Once the specific gravity reaches 36°Be, the reduc- tion in sideways etch becomes insignificant compared to the loss in productivity caused by the significant decrease in etch rate, even in low-productivity environments. erefore, I usually recommend a specific gravity of 36°Be for cupric chloride for the best etch rate and quality. e free acid level had the largest effect on sideways etch, with less sideways etch as the " ose carefree days are long gone (although they didn't seem so carefree at the time). "