gbell@pnet12.cts.com (Greg Bell) (01/17/89)
Wow! I got about 14 E-Mail requests for the article I wrote! I E-Mailed
a few responses before I realized it would be much easier to just post it.
Maybe the word "article" is misleading. What I wrote is simply a document
of what I know about the method.
Personally, I prefer wire-wrapping for prototyping, but this method has
potential. In fact, a few companies are trying to making this method a
commercial product.
Anyway, here's what I've got. Please let me know what you think, any
questions you have, and any experience you gain trying out the method...
###########################################################################
Making PC Boards Using Xerox Transparencies
This method is not perfect, but its simplicity makes it worth
trying. I have successfully made a few simple PC boards using
this method and have tried a more complicated board. Due to a
design error with the complicated board, I never figured out
whether the circuit board was OK (I ended up wire wrapping the
thing).
The beauty of this method is that you can create the artwork for
the board using regular pens and drafting supplies- much as you
would be able to for the standard photographic technique. This
is welcomed by those of use who have tried to use etch resist
pens or lacquer. The cost of making boards this way is much
cheaper than regular photographic techniques.
Make sure all lines on the finished artwork are as black as
possible (use black ink).
The artwork must be Xeroxed onto transparency plastic. Beware!
You can NOT use just any kind of clear plastic sheets. The
temperature inside a Xerox machine will melt most plastics and
ruin the machine. There is a specific type of plastic made just
for Xeroxing things onto transparencies. I got a pack of 25
sheets for $14 at my university's bookstore.
There are different types of this stuff, and some types work
better than others. I was fortunate in that I was able to try
out a few sheets of different types before I bought an entire
pack. The types that don't work too well have an extra layer of
thin plastic laminated onto the thicker sheet of transparency.
This is not apparent until later in the process. You might be
able to tell by looking at light reflecting off the surface of
the transparency- both sides should be shiny. If one side is
shiny, and one side is somewhat dull, then chances are the sheet
is of the laminated type. Perhaps these types of sheets would
work if you Xeroxed onto the shiny side but I haven't bothered to
experiment with this.
Once you have located the right kind of transparency (or have
decided to try whatever transparency plastic you could get),
Xerox your artwork onto the transparency with the machine set at
its darkest contrast. This assures that the most toner gets
deposited onto the plastic. Its a good idea to make more than
one copy as a backup. Another tip: try Xeroxing onto paper
first to see what the image quality is like. If the image looks
less than acceptable, try another machine.
When a Xerox is made onto paper, the porous surface of the paper
absorbs the toner. But, on the transparency plastic, the toner
can be removed...
The next step requires a vise and a clothes iron. Set the iron
to a moderately high temperature (again, there is a lot of room
for experimentation here). Mount the bare circuit board
(cleaned well with steel wool) vertically in the vise so that you
have access to both sides. Cut the circuit image out of the
transparency and line it up on the copper board. Tape the
plastic's edge to the board's edge so that you have a hinge of
sorts (masking tape is best). You should be able to lift the
plastic up like a book cover, and be able to put it back down
against the board without the image shifting.
I have used double sided circuit boards every time I have used
this method, even though I was making a single sided board.
Reason: I'm not sure if the non-metal side of the board will hold
up to the iron's heat (I haven't tried because I didn't want to
ruin the iron). I suppose its possible to heat the board from
the one metal side.
I have not tried making a double sided board. The only
difficulty I anticipate is in lining up the two sides' images so
that the holes will match up.
The copper must be heated to a high enough temperature to re-melt
the toner. I have had success, like I stated above, heating the
board from behind while I press the transparency plastic hard
against the front part of the board. I use a bundled up rag to
press against the copper since it gets VERY hot.. Press the
plastic against the board using a "smoothing" wiping away from
the tape hinge. Continue this for a couple minutes. The time
is a big variable - too short of a time and the toner won't
melt. Next, carefully start to peel the transparency from the
board. Do this SLOWLY. If you see some of the toner lift off,
you can put the plastic back against the board and continue
heating/smoothing. Eventually you should be able to get the
plastic off and a nice sharp black image of your circuit will be
left behind on the copper, most of the toner should come off the
transparency.
If the transparency is laminated, chances are that the layers
will separate at some time during this stage. In my case, the
transparency left the thin layer of plastic on the copper, and it
was a hassle to get it all removed.
Don Lancaster, of CMOS Cookbook, TTL Cookbook, and
Radio-Electronics fame (Hardware Hacker column) suggested to me
that the board be placed in a freezer for a while after the
plastic is removed to "fuse" the toner.
If the image looks pitted in some areas, touching up the lines
may help. An etch-resist pen might work, or lacquer could be
used.
Finally, etch the board using standard etchant. The toner makes
an excellent etch resist (ie. it doesn't dissolve). When the
board's finished, the toner can be removed using many different
solvents.
Let me know what experiences any of you have with this. I'm
especially interested in tips and success/failure stories.
I can be reached on USENET or on the Circuit Cellar BBS at
203-871-1988.
By the way, I first read this idea in a letter in
Radio-Electronics. Don't bother trying to look up the original
letter because it wasn't very long or detailed.
Greg Bell
serene!pnet12!gbell
###############################################
for a better return address, see my sig below...
Incidentally, I need to find out what brand/part number the particular
transparency plastic I use is. But I don't have the package here, so I will
post that information when I get it.
Greg Bell_________________________________________________________
Hardware hacker |
Electronics hobbyist | UUCP: uunet!serene!pnet12!gbell
EE major at UC San Diego |smadi@rlgvax.UUCP (Smadi Paradise) (01/17/89)
In article <308@serene.UUCP> gbell@pnet12.cts.com (Greg Bell) suggest
an ingenious way to fabricate short runs of PCB. He uses a copying
machine to transfer the artwork onto an OHP transparency, then irons it
onto the PC board. The transfered copier toner acts as an etch-resist.
A suggestion:
Use a laser printer to generate transparencies. This works great
with Apple LaserWriters(tm). MAKE SURE YOU USE COPIER-SAFE MATERIAL.
And a query:
Using manual feed, NEC machines can print on rather thick material.
Will they print directly on thin PCB? Unfortunately* I have no access
to one; otherwise I'd try it on thin material and smooth the edges to
save the printer drum from scratches. Has anybody tried this?
* My bad luck may have just saved a $3k printer...
On (just a guest of ...!uunet!rlgvax!smadi) Paradisepostmaster@mailcom.FIDONET.ORG (Bernard Aboba) (01/19/89)
An article on what you describe was in the BMUG Newsletter (available from BMUG, Inc. (415) 549-BMUG in the fall of 1986. They describe in that article how to make PC boards using MacPaint, and an Imagewriter Printer. -- "Don't flame him, don't flame me, flame that fellow behind the tree." Via apple!mailcom, Fido 1:204/444
ritchie@hpldola.HP.COM (Dave Ritchie) (01/19/89)
Running both the PCB and transparency under cold water after getting it hot causes almost 100% transfer of toner from transparency to PCB. Dave
gbell@pnet12.cts.com (Greg Bell) (01/22/89)
Excellent suggestion! I can't wait to try it.
I'm really interested in the limit of the resolution using this method. I
got some E-Mail from a guy who was talking about the fact that Xerox machines
usually fuzz lines over a bit. It would be nice to know if this method would
work for boards with 40 pin (or larger) ICs.
Greg Bell_________________________________________________________
Hardware hacker |
Electronics hobbyist | UUCP: uunet!serene!pnet12!gbell
EE major at UC San Diego |