TFE+ Prototype Manufacturing Issues
Posted by retromaster on March 12, 2010
It’s been a while since my last post. Most of this time I’ve been trying to deal with the problems I encountered making the TFE+ prototype PCB.
Since I use a CNC to drill the PCB holes, my process is that I get the holes drilled first and then apply the toner transfer. For this to work, it’s imperative that I get the laser printer printout to match the CNC-drilled holes. Previously, I had tested with some smaller boards, and I managed to obtain some fine results. The TFE+ board is quite a bit larger and this is where the problems started. It seemed that a non-linear error kept creeping into the printout and despite trying many different things, I never managed to get a good match.
The printer I’ve been using is a Samsung ML-1610. It’s a fine entry level printer and it has worked well with toner transfer during time I did manual drilling. Once I started having dimensional accuracy problems after the transition to CNC-drilling, I took a micrometer and measured the printout (on plain paper). It seemed that one side of the printout was smaller by almost a millimeter! The error was consistently there in subsequent printouts I took. I have no idea where this error comes from. I am sure that it was not in the source image. It could be a software/firmware error, or it might be that the some parts of the mechanism have worn-out (it’s more than 3 years old now). Or perhaps the printer was never intended to be very accurate (which is fine for a printer in its class).
Deciding that it is time to get a newer (and better) printer, I did some research and came up with the Kyocera Mita FS-1100. It is significantly more expensive than the ML-1610. But it’s built like a tank, it has a much higher speed and 1200dpi resolution, and it seems it may be more economical in the long run. In many ways it reminds me of the older Laserjet line of printers from HP. Last night I did some tests with laser printer transparencies (just to see). The printout from the FS-1100 still needed some scaling, but there was no non-linear error and I managed to get an almost perfect match. Far better than what I got with the ML-1610.
It still remains to be seen whether the FS-1100 output will work fine with toner transfer (e.g. with PnP blue paper). Over the weekend I’ll do some more tests, and if all goes well, I might even have a prototype PCB in my hands next week!