I have done some various experiments with engraving (isolation milling)
or plotting of resist lines using CNC machines. I do not feel this is a
good approach, I have much better end products using toner transfer
with a clothes iron and etching in ferric chloride.
Isolation milling has several issues going against it even with a
perfect machine:
1) Milling copper is not good process, even under ideal conditions
(outside the pCB industry). Copper does not machine well.
2) An engraving cutter has to cut with near zero surface speed because
you are cutting with the point of the tool. Again, not a good
condition.
3) Spindle run-out + cutter tip size controls the smallest distance
between traces. Dremels and other similar tools can easily have .010"
of spindle run out which results in a minimum isolation groove of .020"
plus tool bit tip diameter.
Plotting of the resist lines also is not very ideal. Making a pen that
is rigid enough to NOT flex while plotting is a challenge. Then making
a holder to hold the pen which also does not flex adds to the issues.
Trace width is limitted to the minimum width of the pen which is often
thicker than what is easily attainable with toner transfer.
While I did not "beat the issue to death" with extensive testing, it
didn't take long to realize everything I was trying to do with CNC was
in fact a step back from the reliability and precision I could achieve
with toner transfer and chemical machining (etching).
If lasers could be adapted for this "cutting", then I think I would try
it again. However, I don't know nearly enough about electronics to even
attempt a laser at this time.
Chris
--- In Homebrew_PCBs@yahoogroups.com, "Henry Liu" <henryjliu@...> wrote:
>
> Has anyone tried attaching a very fine pen to a cnc then etch the pcb
> board? Right now I'm using pcb-gcode but it's a pain because it takes
> so long to remove all the material when 95% of the board is empty
> space and 5% traces. Also until I get my vacuum holder, it is hard to
> be quick with the setup. I'm thinking a pen mounted in the holder
> will be able to write the traces better perhaps then I can chemically
> etch them. Has anyone tried this? Thanks.
> Henry
>