[sdiy] DIY PCBs

[The Old Crow]

On Mon, 23 Jul 2001, Grant Richter wrote:

> Can you talk more about this? I did not know you could make acetates
> at home (which is what I think you mean?) some kind of direct transfer
> process? What do you do about chemistry for development and fixing?

>> Actually, I've had good results laserprinting to paper, then

  Well, this isn't exactly homebrew--it is stuff I used (and still use on
occasion) at a local electronics shop that has a small "all-manual" PCB
lab.  It is close enough to DIY to talk about in that sense, however, as
the materials are easily obtainable.  Apologies if this gets long-winded:

  The lab uses Kepro PCB chemicals, which all come pre-packaged.  I
typically use 1.5mm (0.062") board stock with a "2-ounce" copper clad and
photoresist emulsion.  This is a standard Kepro part number, as is the
double-sided board stock.  I'll make a list of the numbers next time I'm
there.

  I take a laserprint of the artwork on white copier paper and place it
face-down on a suitable section of UV film (the 'black' film, which is
*very* finicky and thus requires a bit of fudge-factoring to get right)
and expose the film in a UV lightbox for about 5 minutes.  Developing the
film is the trickiest part: one has to rinse the film in developer
solution and *very lightly* rub away the residue with cotton puffs.  Too
much rubbing or too much pressure will ruin the film.

  Once the film negative is developed it is a much easier task to expose
the copper clad.  I usually remove the plastic covering so as to reduce
blurring caused by the slight distance the plastic would otherwise hold
the film off the emulsion.  A board is exposed for 2 minutes under UV.

  Next comes the "dry film developer": this is a powder that is measured
into a 13x9 glass cake baking dish and dissolved in the warm water used to
fill the pan to a depth of about 1cm or 1/2 inch.  The board(s) are placed
in this solution for a few minutes, then while in solution a clean, soft
brush (like a varnishing brush) is used to brush the unfixed emulsion away
from the board.  Sometimes the boards have to be rinsed under hot running
water and placed back in the developer tray for the proper result.

  Once a shiny copper board with the hardened phororesist circuit is
ready, the board is rinsed and placed into a holder that stands the
board(s) up vertically.  The boards are then immersed in the etching tank
and a timer set for 30 minutes.  Etching tanks aren't aren't exactly
homebrew fare (and the ones I use are under a huge vent-hood/forced-air
system), but I suppose decent results can be obtained using another
glass baking dish setting on a hotplate with the hotplate at its
lowest setting (no hotter than about 40C--the etchant works MUCH better
when warm but don't boil it!)  I would set such an etching dish out on my
porch with a box over it to keep stuff in the air from getting in.

  One note: the ferric chloride *WILL* ruin your clothes if they get
soiled.  I use an apron (which amuses the wife to no end) and rubber
gloves when at the lab, but then I like to keep my pants longer than a
week. ;)

  Anyway, after 30 minutes it is time to raise the boards out of the etch,
take each one and rinse it under hot water, invert the boards to even out
the etching ofthe surface area and re-immerse in the tank.  depending on
the age of the etchant and board size (it terms of the number of boards
processed in a given etchant batch) it can take as little as 40 minutes or
as long as 2 hours to etch a board to completion.  Of course, a board
can't be left in the tank too long or the etchant will eat right under the
photoresist from the sides and your circuit will disappear.  After the
initial 30 minutes elapse, I will check them every 15 minutes.  Each time
I rinse the board and inspect it for signs of unetched, exposed copper.
As the etching time varies as noted, one tends to develop a 'feel' for the
number of times a given board has to be checked.

  When the board etching is complete, the board(s) are rinsed a final
time, then they are immersed in a container of water for about 5 minutes.
This is to dilute out the etchant that has worked under the photoresist
slightly as well as the etchant that has soaked into the edges of the
board.  After the soak, the boards are dried and taken to the drill
presses for drilling, which is the least-exciting part of the whole
ordeal.  I use two Dremels, one chucked with a #70 bit and one with a #60.
The #70 is for transistors, ICs, and the like whereas #60 is for
resistors, caps, diodes, and so on.  For larger holes such as for mounting
screws, connectors, etc. I use the big drill press and appropriate bits.
I always drill the large holes before shearing the board to size and I
drill all holes before stripping the photoresist.  The photoresist
actually helps in piloting the Dremel bits into the pad holes.

  Once drilling is completed, the boards are sheared to size.  On the PCB
artwork, I will indicate the shear-marks using the inside edges of 50mil
track segments to 'frame' the board size.  The board shears are also
Kepro-made; they are large, flat platforms with a 'punch arm' that is
pulled down to shear the board at the blade edge.  Since the board will
move due to the cutting pressure, the board has to be anchored so as to
remain immobile.  The Kepro stand has the anchor-bar built-in.

  After shearing, the boards are immersed in 'dry film stripper', which is
a solution of Kepro-packaged Sodium Hydroxide (aka Draino) measured into a
glass tray of hot water similar to the dry film developer.  Caution: the
stripper is very reactive--particularly to water! By all means measure the
desired amount into a dry tray and put away the jar before turning on the
water.  Once the stripper is dissolved, the board is immersed for about 30
seconds, after which time the blue photoresist should visibly start to
'curdle' and peel up from the copper.  Once this happens, a second brush
can be used to loosen any resist that is stuck to hole edges, etc.  Do not
leave the board in the stripper solution any longer than necessary or the
copper will start to tarnish.  Of course, wear gloves while working with
this and any other chemical mentioned above.  Once the resist is stripped,
remoce the boards and rinse under running water.

  One more optional step: I use 'immersion tin-plate' in a glass tray to
plate the copper traces with tin.  The boards are immersed for about 30
seconds, then removed and rinsed a final time.  The copper must be shiny
prior to immersion or the tin plate will not plate well.  If the copper
got tarnished, use steel wool to burnish it.  Also, do not touch the bare
copper with your fingers unless you are wearing gloves.  Oils from one's
fingers cause the tin plate to occur unevenly.

  At this point you have a finished board.  Dry it off thoroughly, and
start soldering!

Crow

p.s. and totally different topic: does anyone have a Yamaha GS1 voicebook
(a wallet of magnetic voice cards) that I can borrow?  It would of course
be returned.

/**/