Homebrew_PCBs

I'm marching down another path to plot circuit
boards directly on copper.
The .020" pitch TQFP part with the .008" tracks
is pushing the limits of the plotter pen.

There are "fat spots" in the trace every time the
pen touches down. And protel 99SE likes to break
up tracks and touch the pen down a lot.

No problem. I wrote some code to read in the gerber
and combine sub-tracks with common beginning/end points
into a single track that gets plotted without
lifting the pen.

Problem.
When you clean up a protel board, there's a feature
that lets you dynamically reroute a track by just running
a new track segment from any two places on the existing
track. It puts in the new segment(s) and removes the old segment.

But in manipulating the gerber, I discovered
that the end of one segment is not exactly at the same point
as the beginning of the next. There are actually a lot
of gaps in the tracks.
Looks ok on the gerber plot because the aperture covers up
the joint.
But I can't figure out how to determine track continuity from
the gerber...without doing a LOT of extra work to find stuff
that's "near".

Am I just doing the board cleanup wrong? Or maybe some automatic
cleanup pass that I've not yet discovered in Protel 99SE?
Everything automatic I've found insists on rerouting the whole
board and messing up all my hard work.

Suggestions?
mike

I did plotting on copper. It was an amdek plotter with a program
somone wrote in Pascal and it used to be avaialble free on internet.
It all worked fine but first it took a very long time to plot and
second, towards teh end of plot the holder must have gotten tired of
holding a pen and dropped it. Overal I was not happy with quality nd
all kinds of problems. That was a pen plotter.

The second experiment was to use a regular inkjet printer. I rigged
up a printer and it was nicely feeding a PCB. I printed with
standard ink and it printed nicely except for it was a water soluable
ink. I kept changing inks, all that ammonia stuff and the printhead
died. The nice thing is that I chose a printer with a printhead in
the ink cartridge. I dropped the project due to lack of etch
resisting ink.

Then I tried copper milling- too much of everything, noise, dust,
headache.

Then I tried a scratch method in a plotter. It scratched nicely but
the scratches were too thin for the etchant.

Then I started coating my own boards with foil and I saw beautiful
patterns after UV exposure/etch. it was awsome.

Then I tried a TT with a hand iron and that turned out to be a pile
of manure. I could never get the right conditions.

Then I tried TT with HC200 and Staples paper and I stopped making UV
exposure boards because it turned out so nicely and QUICK.

Then I got a commercial laminator and it laminates awsome boards.

Then I tried the www.pulsar.gs TT paper and I said WOW! Never seen
anything like that. That paper flies off the PCB like a flea off a
dog. Mike

--- In Homebrew_PCBs@yahoogroups.com, "Phil" <phil1960us@y...> wrote:

> by the way, in researching refilling my canon inkjet tanks I came
> across waterproof ink for the BCI-3e tanks.
> http://www.alotofthings.com/catalog/bulkinks_canon_bci-3.html
> Look at the top two items.
>
> I have no idea if works as a resist but waterproof is a good first
> step. My printer (pixma 4000) doesn't have a straight through paper
> path so it wouldn't matter to me. These tanks go with a lot of canon
> printers so maybe there is one that has the right paper path.

It's waterproof on paper, don't know about on a slick nonporous
surface. I can tell you from experience that anything not able to
quickly and completely absorb the ink will cause severe puddling of
the ink.

Almost all stock Canon black inks are pigmented, until you get into
the 6 and 7 color photo printers. Many of them have dual black tanks,
the BCI-3 black is pigment, and the 6 is dye. I have several Canon BJC
6000 for my digital photo business, and I've seen what happens on
paper that is not absorbent enough, or if I accidentally stick a sheet
of laser transparency in it.

But don't let me stop you from trying. Just sharing what I know.

Steve Greenfield