On Wed, 16 Nov 2005 20:47:47 +0100, Alan King <alan@...> wrote: > > > > "Many will find that TT combined with CNC drilling is a bad > > > combination. (Assuming the CNC is of decent accuracy<G>)" > > > Ballendo > > > > Of course, this was more than a bit off. Takes 3 minutes of 10th grade > geometry to fix axis scaling errors. Anyone not able to use the > relative accuracy of both and scale for absolute accuracy in 15 minutes > should really turn their high school diploma back in, they > should've flunked the math requirements. I found you must first adjust your printer properly, then make a trial transfer which you measure, and then simply compensate. The thing is, if the printer isn't aligned horizontally and vertically you usually will have some amount of skew or trapezoid distortions, which is hard to compensate. Once it's aligned properly (see service manual, on the web for many printers) you have only linear error in x and y which is easily compensated for. This way i got it down to no measurable deviation easily. Also, while inkjets are often more accurate, you still need a true print just the same with photoprocess. If you have any shrinkage while transferring from the heat simply run the paper through the laminator before printing, to dry it properly, then it won't shrink any more. I readily admit i didn't master the photoprocess, but i also now make great boards (with legend by the way), much faster, cheaper, and easier than i ever did with photoprocess. Even if i mastered it i would need considerably more gear, effort, and cost than i do now for TT. For photoprocess you _need_ sensitized boards (or you have to add the step of sensitizing which is again not that easy). They are much more expensive than the cheap blank offcuts i get. You _need_ a lightbox to do it properly, not necessarily expensive, but surely comparable in building effort to a fuser. You _need_ either a inkjet or special (expensive) laser paper, i don't have an inkjet and don't want to keep one ever again. You need to develop them, which means another step making mess and effort and chemicals. When i make TT i don't need all that. I plug in the fuser to heat up, meanwhile go and cutoff and clean a suitable PCB, and print the layout. 3 minutes max.. Then i cut the layout out, put it on the pcb, feed it through the fuser - another minute max.. When it's out i scratch the paper with a wirebrush i keep in a desk drawer, and submerge in water (the brush rips open the paper and allows water in quicker. Do not brush too much or it'll go through and damage the transfer). After a moment in water i take it out, hold it over the dustbin, and "roll" the paper off with my thumb starting at the center. submerge again to wet it and rub off the rest. Submerge again to flush off any paper remains and put in the etcher. two minutes max. That's 6 minutes, and i have have estimated very generously - i'll time it next time i make a board. Did you notice there's NOTHING to prepare or put away? All waste lands in the dustbin already. I don't have a sink in the shop so the container with the water is filled and emptied when i go by one anyway. I don't think you can make a photoprocess in the same time, and if it surely requires more space to leave all the gear prepared. But what am i doing, wasting my time here. Let them believe what they want i have work to do. ST
Message
Re: [Homebrew_PCBs] Re: A $500.00 "UV" non-trivial exposure box.....
2005-11-16 by Stefan Trethan
Attachments
- No local attachments were found for this message.