[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Homemade PCBs!



Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net> 

Having been through both approaches in a variety of environments, I have the
following observations:
1) If you have the room to make the boards (including exposures, etc.) and
do it often enough that your process skills remain fresh, there is a certain
appeal to being able to zap out a board "on impulse".  Draw the circuit,
make the board, solder the parts, and be testing in a couple hours.  20
years ago I bought a big stat camera which can do 1:4 reductions just for
this sort of thing.  Too bad I don't make enough boards to keep the
chemicals and film fresh, much less my photographic skills.  The camera
sits, unused, in my garage.

2) The commercial products will almost always turn out better (unless you're
a business where prototyping is part of the day to day work).  While making
good quality two layer boards is fairly straightforward, drilling by hand is
a pain, decent robotic drills are expensive (more than a kilobuck, although
I'd love to hear about a cheaper one), and doing plated through holes is a
real pain.  Then there's the solder mask and silkscreen..

3) On a cost basis, they'll probably turn out about the same, within the
significance of the measurement.  More dollars invested up front (or time
scrounging equipment and materials) for home fab, but fairly low piece
costs.

4) Multilayer (4 or more) are not real feasible for home fabrication. It can
be done, but process controls are tough. There's also the problem of minimum
order quantities for things like prepreg.

5) A very attractive alternative for a lot of prototyping is the computer
controlled router approach.  It will do the traces, and do the holes.  Won't
do the plated through, but it's definitely a hot ticket for RF prototyping.
A tad pricey (again, if someone knows of a decent (assembled) system for
under $1000 (exclusive of PC), I'd love to hear about it).

6) There are some interesting approaches I've seen commercially, but I
haven't seen in hobby applictions, although I think they'd be appropriate.
Things like the "multiwire" where it essentially "plots" a special insulated
wire onto the substrate, or various stereolithographic schemes (like the
computer controlled glue gun things).  I've also seen systems (based on the
same scheme they use for making custom vinyl signs) that cut copper foil on
a backing, which is then "glued" to a substrate.  Peeling the unused copper
away doesn't strike me as a lot of fun (like peeling rubylith!) but might
give good results.


----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Saturday, April 24, 2004 6:55 PM
Subject: Re: Homemade PCBs!


 > Original poster: "Jim Mitchell" <Electrontube-at-sbcglobal-dot-net>
 >
 > Or for about the same price it costs to get all the materials and things
to
 > do photo etching,  you can have 10, 3X5 soldermask and silkscreened boards
 > for 65$....
 >
 > Regards - Jim Mitchell
 > ----- Original Message -----
 > From: "Tesla list" <tesla-at-pupman-dot-com>
 > To: <tesla-at-pupman-dot-com>
 > Sent: Saturday, April 24, 2004 8:53 PM
 > Subject: Re: Homemade PCBs!
 >
 >
 >  > Original poster: Matthew Smith <matt-at-kbc-dot-net.au>
 >  >
 >  > Eric ("Hydrogen18") writes:
 >  > >I'm looking to make some homemade PCB's for a voltage multiplier,
whats
 >  > >the best way to do this? I've seen those kits where you draw
everything,
 >  > >but that seems pretty tedious. Is their a better way to do it at home?
 >  > >Also, what kind of circuit board do I want to buy? Thanks.
 >  >
 >  > Matthew replies:
 >  >
 >  > I use PCBs for all prototypes - they are NOT hard to make once you get
the
 >  > hang of it. I use a simple photolithographic process with all home-made
 >  > equipment.
 >  >
 >  > Note that a PCB will be more reliable and easier to trouble-shoot than
 >  > prototypes using solderless breadboard or ratsnesting.
 >  >
 >  > Now that I've got my technique down to pat and have got the optimum
 >  > exposure times, etc., I can throw a board together in very little
 >  > time.  I'm even starting to use some SMD components now.
 >  >
 >  > Abbreviated steps are:
 >  > 1) Produce artwork on a transparency.  This could be hand-drawn or
 > produced
 >  > by CAD software such as Eagle <http://www.cadsoft.de>.
 >  > 2) Acquire PCB stock coated with positive photoresist.
 >  > 3) Cut stock to size if required.
 >  > 4) Contact print artwork onto PCB stock with UV source
 >  > 5) Develop board with sodium metasilicate solution (easier to use than
 > NaOH).
 >  > 6) Wash board
 >  > 7) Etch with ammonium persulphate solution.  This is far nicer to work
 > with
 >  > than ferric chloride.
 >  > 8) Wash board
 >  > 9) Drill (unless SMD ;-)
 >  > 10) When ready to solder, remove remaining photoresist with acetone.
(It
 >  > protects the board in the meantime)
 >  > 11) - optional - flux and tin traces or use humungously expensive
tinning
 >  > compound ;-)
 >  > 12) Assemble and solder
 >  > 13) Clean off flux residues
 >  > 14) Step back and admire your work :)
 >  >
 >  > Exposure, developing and etching can easily be done in under half an
hour.
 >  >
 >  > One very important tip that I received from a member of this list
(think
 > it
 >  > was Dan Mccauley) is: remember to put plenty of test points on your
 >  > artwork, especially if this is a prototype!
 >  >
 >  > Contact me off-list if you want to know any specifics, like how to make
a
 >  > UV exposure unit, etc.
 >  >
 >  > Cheers
 >  >
 >  > M
 >  >
 >  > --
 >  > Matthew Smith
 >  > Kadina Business Consultancy
 >  > South Australia
 >  > http://www.kbc-dot-net.au
 >  >
 >  >
 >  >
 >
 >