Issue link: https://iconnect007.uberflip.com/i/857644
August 2017 • The PCB Magazine 15 speed immediately after centering to get the panel out. That deals with the speed changes. We can deal with speed changes at that lev- el as well so we don't burn out on the mate- rial. We're entering the load/unload station, at the speed of the preceding line, stopping, cen- tering, and then exiting at next line speed. The gaps are managed accordingly. The speed of the etcher is almost one-to-one with the speed of the plater; we've matched that. If the plater runs at one meter a minute, the etcher runs at one meter a minute. If the plater runs at half a meter a meter a minute, the etcher runs at half a meter a minute. Remember, we're panel plat- ing, so if the plater is running at half a meter a minute, it's putting on twice as much copper. Then the etcher runs slower and it matches. It's all matched and balanced. Holden: This is an unusual line in that it's pan- el-plated plus patterned tin. It's not a conven- tional panel plate line; it's kind of a panel/pat- tern hybrid. Stepinski: That was the lowest-cost solution I could come up with. Goldman: Meaning lowest cost to do the panel. Stepinski: Yes. Holden: I'm surprised that the companies that sell inkjet printing for solder mask haven't come out with a plating or etch resist machine or mask. Yours are adapted with this wax-based ink. Where did you start on that inkjet printing for resist and for pattern plating? They're not really even available now, so you made an obvi- ous choice to custom engineer that level of ink- jet printing. You're not ink formulators, so... Stepinski: No, Dow formulated the ink. We tested it here and validated that it worked be- fore we bought the equipment. When the ma- chine showed up, we optimized the preced- ing and follow-up processes to handle it. There were a few uncertainties in designing the pro- cess. I planned it very quickly, so I had to com- pensate for the uncertainties. We have a pre- clean in front of the line, and it turns out that WHELEN ENGINEERING, TWO YEARS LATER and then wouldn't all the other machines have the wrong immersion time? Stepinski: But that's the slowest line so it doesn't really matter. Holden: If you had a spec that called for twice as much copper, that machine would have to be even slower. Do you have the flexibility of changing the current density or the pulse rate to keep the conveyor the same speed and dou- ble the amount of copper? Or, do you have to actually reduce the conveyor speed, changing the immersion time? Stepinski: It depends on how much copper. Holden: Everything tied together means every- thing has to be balanced. One reason why pro- cesses are separated is because you might say for maximum throughput, the conveyors run at different speeds, and so you have isolated con- veyorized processes, with manual loads and un- loads. Stepinski: Relative to that, the individual load- ers/unloaders, when a panel enters then, and they all automatically center the panels, be- cause you get drift going through horizontal equipment. Every time we go through a load- er/unloader, we center and then we change the Figure 3: Strip-etch-strip line.