IPC International Community magazine an association member publication
Issue link: https://iconnect007.uberflip.com/i/1546209
128 I-CONNECT007 MAGAZINE I AUGUST 2026 A Sticky Situation Here's a recent one I hadn't heard of in several years: a black, tar-like substance that builds up in the sump of a cupric chloride etcher. It coats the walls of the sump and plumbing, leaving an oily sur- face in the etchant sump, and eventually plugging the spray nozzles. Neither acid nor alkaline washes would get rid of this stuff. Many years ago, this was diagnosed as organic residue leaching from the dry film etch resist as it passed through the etcher. However, the problem diminished over the years as dry film manufac- turers responded by improving the product such that leaching rarely occurred. I assumed this was no longer a problem, but apparently it can still happen. Fortunately, the solution is a simple carbon filter that removes unwanted organics and can be retro- fitted as a closed-loop system on most etchers. It's a good example of an old problem and an easy solution that has been forgotten because it rarely happens anymore. The main concern has been that people tend to forget the activated charcoal in the filter needs to be changed occasionally, or the problem will re-emerge. Keeping the organics out of the etch bath is simple. Cleaning it off the walls and plumbing in the etch sump after it forms is not. Fan Fiction One question that continually arises, especially as line and space tolerances get tighter, is "Can I get better etch uniformity and quality if I switch the etch- er from cone nozzles to fan nozzles (or vice versa)?" Sadly, probably not. Nozzle dynamics is a lot more complicated than it appears on the surface. For instance, one common misconception is that fan nozzles have better undercut performance (less undercut) than cone nozzles and therefore should give better etch quality as circuit traces and spaces get smaller because the spray is coming straight down rather than being spread out as with a cone nozzle. However, a massive design of experiment done in conjunction with a large computer company (not IBM) showed that there was no difference under the right conditions. The experiment showed that the undercut performance of fan nozzles improved as the nozzle was moved further from the surface, and that cone nozzles performed better as they were moved closer to the surface. When each nozzle was at its optimum distance from the surface to be etched there was no difference in undercut. The misconception probably comes from putting cone nozzles in an etcher optimized for fan nozzles and observing an increase in undercut. Another misconception is that if you have a smaller space to etch out, a nozzle with a smaller droplet size would do a better job of penetrating it. The truth is that, even with the smallest practical nozzle, the droplet diameters are still in the 500 µm range, which is still huge compared to a 50 µm space. Also, reducing the droplet diameter by half reduces the droplet volume and therefore mass by a factor of eight. Since impact force is mass x acceleration, decreasing the droplet size reduces the impact of the droplet on the surface and results in a decrease in etch performance. Perhaps the next time we have a monthly theme that I know nothing about (something that seems to be happening more frequently as I get older and the PCB industry gets more complicated and sophisti- cated) I'll go a little deeper into nozzle dynamics. By all means, feel free to experiment with spray nozzles, but don't get your hopes up too high. T H E C H E M I CA L C O N N ECT I O N " Fortunately, the solution is a simple carbon filter that removes unwanted organics and can be retrofitted as a closed-loop system on most etchers."