After the release of an instructable detailing an Inkjet direct-to-pcb conversion, Dan Fourie and I thought it would be way more awesome to convert a laser printer to be able to print directly onto copper clad.
Inkjet masking is cool because you don't have to deal with the delicate process of smushing plastic dust (toner) onto metal, but it also has several flaws. For one, you need to purchase and replace the ink in an inkjet cartridge by hand. The difficulty here is that cartridges range wildly from printer to printer. Secondly, the process isn't clean - results are often pitted and inaccurate due to etchant eating away at the printed mask.
Laser printer masking is cool because toner is an excellent mask. It doesn't dissolve in ferric chloride or cupric chloride, and you can buy an entire toner cartridge from the store for any laser printer. Additionally, the print time is arguably faster, the possible resolution is higher, and the traces printed are extremely accurate - it's possible to print out 2 mil traces! Also, contrary to popular belief, there's no issue with laser printing on a conductive surface. All of the fancy charge depositing, optical neutralization, and toner patterning is done on a special photoelectric film wrapped around the heating drum, so the copper won't repel or destroy the image. In fact, printing on folded aluminum foil works extremely well! On the other hand, copper clad board... well... see below.
Over the weekend, Dan disassembled a laser printer that he found in the trash. He was able to tear it down to the point where he could fit a board in the (quite linear) print path and gave it a go. To his dismay, the board jammed and nicked the heating drum. I came over and we decided that by removing a metal guide and chamfering the edge of the circuit board, we'd have better luck - and we did! The print quality turned out pretty good, but the toner wasn't fusing to the board. We figured that it was either a heat/pressure issue or a charge issue and called it a day.
The next day, Dan and I went to miters to improve the process. We tried a whole slew of ideas include charge compensation, increasing roller pressure, and pre-heating the boards to get the toner to fuse. The last one did the trick. Although print quality isn't 100% there yet, we're certain that we can hammer out those last pixels and get this process rolling. We etched my first lab in 6.331 - a 4MHz 50-ohm line driver with 2 watts of output power.
More to come next weekend!