Hi Xiaofan Thank you.... tell your friend to mail me if he has any questions. I also have a hobby mill too that I use to make my mechanical protos and when I move in my new building I am going to get a more "manly" one. I have hobbies but rolling my own PCBs was never a hobby for me. In the late eighties early nineties when we started the costs of prototype PCBs were orders of magnitude more than they are now. There was wire wrap and perf-board to make protos but we needed controlled impedances and surface mount components to make our circuits work in the GPS (L-band) and S-band radar frequencies. We spent many nights in a garage in Silicon Valley learning through trial and error. If we knew how much work it was going to be we surely would have never started....it was sort of like playing with fire it is fun but you get can get burned if you are not careful....and burned out. I'm glad I learned/did it but even more glad I don't have to do it all over again ...at least I hope not. Phillip Things should be as simple as possible but no simpler Phillip Coiner CTO, GPS Source, Inc. Your source for quality GNSS Networking Solutions and Design Services, Now! -----Original Message----- From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of Xiaofan Chen Sent: Friday, September 15, 2006 5:34 PM To: Microcontroller discussion list - Public. Subject: Re: [PIC] PIC18F97J60 & prototypes using over head transparencies On 9/14/06, Phillip wrote: > Hi Xiaofan/All > I used to do all my circuit boards by hand. (I still whip out prototypes > occasionally with the transparency and ferric chloride method for simple > circuits). > I have sold thousands of homemade PCBs with very low failure rates and great > performance. > > I use photosensitive copper clad laminate from Kepro and expose them using > my personal overhead UV light source (exposure takes a little longer on > cloudy days) and overhead transparencies as my negatives. > > There are few tricks that with a little practice you can do six to eight mil > trace widths and 20 mil vias and yes most tiny packages. > > Transparency negatives: > The first and most important is to mirror the image of the negative. > The reason for this is that this will allow you to flip the transparency > over and place the side with the toner directly against the light sensitive > plastic coating on the board. That way the negative is not held away from > the board the thickness of the transparency. > If it is light will be able to diffuse around the image and make the edges > of your traces/pads fuzzy. > The second trick is to use two transparencies one over top of the other to > get the darkness required so that the light sensitive plastic coating > doesn't receive any exposure so when you develop the board to expose the > copper the edges of the remaining plastic are sharp. > > Vias: > On your program that you use to create your negatives leave a small spot > dark in the center of your vias (the bigger the via the bigger the hole you > want) so there will be a tiny hole of copper removed so when you drill the > vias the tiny hole in the center will prevent the drill from walking. > Then you use bare tined copper wire the thickness/diameter of your drill. > The wicking solder through holes advice is bad jue jue. > You will not have good grounds (RF wise) and you will have mechanical > failures when you use solder only. > Unless you are an expert, don't use an electric drill/drill press to drill > out your vias. > Just use a pin vise and drill by hand or you will spin the annular rings on > your vias or break the tiny drill bits required. > This takes some practice and it is not worth ruining a perfectly etched > prototype board for the little time it will take to drill the vias by hand. > > Stripping: > Don't use the stripper from Kepro it is an expensive and a harsh chemical > and it can dissolve the glue on the copper that holds it to the > board/fiberglass. > To remove the exposed plastic from the traces after etching just make a > batch of developer about five times stronger and it will remove the plastic. > After you strip the etched boards you safely can pour it down the drain with > no environmental headaches. > The developer is sodium carbonate available at you local pool supply store a > few bucks will purchase an endless supply. > You should sprinkle it into your solution as it will turn to lumps if you > just dump it in but not to worry if you do it will dissolve eventually it > just takes a while. > As a developer and stripper sodium carbonate solution works better lukewarm. > > Etching: > When etching the boards use a Pyrex/glass baking pan on a hot plate. > Heat the ferric chloride but only to about 140 0r 150 F. > A little heat makes the stuff etch faster and more evenly more heat only > makes more fumes it does not make the process work any better. > Don't submerge the boards in the ferric chloride they will float!!! > So just drop them gently onto the solution with the side you want to etch > down. > This lets bonded atoms of copper fall away to the bottom of the solution. > So the bare copper is always exposed to fresh solution. > If you are doing two sides leave the side with the least detail up and use > a plastic or glass implement to wash/drizzle the solution over the top. > > Lastly don't even think about pouring the ferric chloride down the drain > Besides the environmental hazard if there are iron pipes anywhere between > you and the ocean they will quickly be nothing but rust. > > Happy protos! > Thanks for the detail information. Really impressive! Today I know a guy who is using iron and laserjet for PCB making. He has two workshops at home, one for mechanical and one for electronics. Seems people here are really into their hobbies! -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist