Prashant:

I hope that you don't mind that I copied this to the piclist.
I thought it may be of help to someone else as well...


> From prashb@rta.nsw.gov.au Wed Dec 18 18:02:49 1996
>
> I have just begun to make my own PCBs for the first time and am facing a few
> problems. Wonder if you can help me out...
>
> At 12:49 PM 18/11/96 EST, you wrote:
> >I use precoated positive photo board.  Saves worrying about the zillions
> Where do you get it from? I've tried RS and Dick Smith. The Dick Smith
> resist just gets dissolved in the developer and most of the tracks are half
> gone. The resist on RS PCBs does not dissolve completely - no copper is
> exposed. I am using the Riston developer from Dick Smith.

I got mine from a small business in Canberra called Electronic Components.
Unfortunately, they seem to have disappeared from the face of the earth.
Most pre-coated boards that use positive transparencies (i.e. opaque where
copper is required) will use the same developer.  The photoresist is
green, and uses an alkaline developer.  Not sure about Riston, though.

I don't think Riston developer is the right stuff, unless you use it
with Riston boards.  In any case there's no real advantage to using
commercial developers if you can make your own from household chemicals.

If you are having trouble with over- or under-development you will need
to systematically explore the parameters.  This only needs to be done
once and you can use the results forever (or at least until you need to
use different supplies).  Consistency is the key!

The following may seem like a PITA but will save you from wailing and
gnashing your teeth down to the gums (like I used to).

1.  Make a small test transparency with a mix of tracks (10 - 50 mil)
running in vertical and horizontal directions.  Make a 1/2" x 1/2" solid
area.  Add some pads with 13 mil guide hole size.  The complete pattern
should fit in about 1" x 1".  Duplicate this pattern so there are 4
copies spread in a line over about 5".  This is your test pattern (1" x 5").

2.  Cut a strip of PCB 1.5" x 5.5".  This is the sacrificial test piece.

3.  Now we expose each of the 4 patterns for a different time.  The UV
light should be allowed to warm up for 5 min, since the tubes take a
while to develop full output.  The patterns are exposed for 2min, 4min,
8min and 16min respectively, at the same (take a note!) distance from
the lamp.  This is best done using an old photographer's trick: place
an opaque mask over the three rightmost patterns, leaving the leftmost
uncovered.  Expose this for 8min.  Move the mask to the right, exposing
the next pattern as well as the first.  Expose for 4min.  Move mask
again and expose leftmost 3 patterns for 2min.  Finally remove the mask
altogether and expose for another 2min.  Now the total exposure times
will be 16, 8, 4 and 2 mins from left to right.

4.  Now create various concentrations of developer to use.  Make 4
solutions, each in 100ml of water.  Dissolve various quantities of
developer depending on the manufacturer's recommendation.  The minimum
amount, 'D' grams, would be about 5g for commercial powdered developer
or washing soda (see below).  If using NaOH, try 2g.  The 4 solutions
should contain D, 2D, 3D, 6D grams of developer.  Allow all solutions
to reach room temperature (20-25 deg C).  Since these are rather small
amounts of developer (less than a teaspoon) you might like to use larger
amounts of water and nominate your standard 'developer measuring
teaspoon'.

5.  Develop the test board.  In a container which could contain at
least 500ml, add the first solution.  Agitate for 2min and observe the
results.  If this is only just starting to develop the 16min exposure,
then the developer is too weak.  In this case, remove the board, mix in
the second solution.  The mix will have a concentration 1.5 times
higher (3D in 200ml).  The board must be removed to prevent exposure to
temporarily higher concentrations.  Replace the board and agitate for
further 2min.  Continue increasing the concentration in this manner, if
necessary, until one of the exposed patterns is completely developed.
Remove the board, rinse for 30sec and examine closely.  The development
is successful if all traces of resist have been removed from exposed
areas.  The copper should be uniformly very bright, and the remaining
traces should be sharply defined.  Examine the guide holes carefully to
ensure that all resist has been removed (there's nothing more annoying
than missing guide holes - the drill skips everywhere but the centre of
the pad!)  If traces of resist remain in unwanted areas, replace in the
same developer and agitate for another 2min.  If this still doesn't
work, then go to the next higher concentration of developer.

6.  Etch then remove the remaining resist with methylated spirits.
Examine the copper traces for clarity and definition.  Preferably test
each trace for continuity with a multimeter.  Check for short-circuits
caused by thin film of resist.  Note:  when etching with ammonium
persulphate, small bubbles appear over the resist area.  The bubbles
indicate the presence of pinholes which seem to catalyse the formation
of oxygen from the etchant.  If the bubbles are not disturbed they help
to protect the pinholes as much as possible.  Of course, this is not
desirable for the pad guide hole 'pinholes', so on balance it is
probably better to brush the bubbles off occasionally.

6.  Summarise:  note down the exposure time and equivalent
concentration for the developer, which will be D, 1.5D, 2D or 3D grams
per 100ml depending on how many additions were made.  Make note of the
finest traces and separations which may be reproduced reliably.

>
> The RS PCBs are pretty good - after UV exposure I can see the pattern in
> crisp detail. The are a little dearer. They are glass epoxy as opposed to
> DS which is paper epoxy. This may be a disadvantage as glass epoxy is a
> pain to drill.

I am surprised.  I can never see any pattern after exposure, at least when
using fluoro lights, but perhaps the RS boards use a different resist.

>
> >1.  Draw the board on white bond paper.  Use a laser printer or use
> >a 0.5mm drafting pen.  Print/draw mirror image so that the pattern will be
> I use the TurboCAD CAD package. This lets me fuss over every pad and every
> hole :-) I am using a HP 5MP 600 dpi laser printer to print on overhead
> transparencies. When I put the transparency toner side on the PCB I get
> a mirror image. So I draw looking from the top of the PCB. Any problems at
> this stage are too subtle to detect yet.

The trouble with printing on transparencies with a typical laser
printer is that the toner doesn't stick as well as it does on bond
paper.  I have tried this and it tends to leave scratches and pinholes
all over the place.  Pinholes are OK, but scratches can leave really
tiny breaks in the tracks.  It is best to closely examine the
transparency under a bright backlight (photographer's light table is
ideal) and touch up any defects.  Only try this with a printer in
tip-top condition.

>
> >2.  Sandwich board and pattern between 2 sheets of glass with ink side
> >against the board.  Expose to white fluorescent lights (4 x 20W tubes
> *>about 200mm from board) for about 1 hour.  Yes, this is slow but it gives
> >time for the finer things in life such as preparing the developer and
> >etchant.
> I am using a 16W UV-C tube by Phillips in a wooden box. It does(?) Dick Smith
> PCBs in 4 mins and RS PCBs in 7-8 mins. I don't know if I am messing this
> stage up.
>
> >3.  Develop in commercial developer (or NaOH in a pinch).  Etch in hot
> What developer are you using? This is the biggest problem I have. I use
> the Riston developer from Jaycar/Dick Smith(?). Overdoes things on Dick
> Smith PCBs and does not create much of an impression on RS PCBs(resist
> gets slightly discoloured on the UV exposed parts but bare copper is not
> exposed). I am using it at room temprature. Where do you get NaOH(I am
> in a pinch)?

NaOH is common caustic soda which you should be able to get in
supermarkets etc.  A better alternative (which I haven't tried) is
Na2CO3 (sodium carbonate or 'washing soda') which I think is the major
constituent of the commercial developer and has the advantage of being
safer than NaOH since it is a much milder alkali.

>
> >ammonium persulphate, preferred for its transparency and non-staining
> That is what I am using. Worked kinda OK with a badly developed DS PCB.
>
> >4.  Drill then solder.  I find that removal of the resist is not
> WOW! You actually reach a stage where you get to drill? And solder?
>
> >If cutting the board to size using an angle grinder, leave a 10mm zone
> >around the edge where no copper will be required.  The heat of grinding
> >ruins the resist for about 6mm either side of the cut.
>
> I usually try to fit multiple copies or some other useful PCB so that
> the whole board is almost used up.

I do this too - saves wastage.  If the board is too large it may be
difficult to get even exposure.

>
> >water when making it.  Don't use at over 80 deg C otherwise the resist
> >can get damaged.
> I'll keep that in mind.
>
> >Many a good DS board has been damaged by using blunt drills.  When the
> >drill breaks through the other side it lifts the pad there.  Tungsten
> >carbide drill bits last more than 10 times as long as HSS but must be
> >used in a drill press.  They also cost ten times as much.  It's a pity,
> >but I've never seen nitride coated HSS in PCB sizes.
>
> Where do you get the bits? I've seen HSS sold by Jaycar, DS, et al. RS
> sells 5 bits at 3 bucks a bit or a set at 20 bucks. I'll buy the set if
> I ever get that far. What hole sizes are needed? I'm thinking of 0.8mm
> for IC holes, 1mm for most components and 1.3 for big resistors or 0.2"
> terminal strips.

RS sells at about $3.00 each however if you look around for tungsten
carbide (WC) suppliers you should be able to find a cheaper source.
The advice to use a PCB drill press should be taken seriously, because
WC is extremely brittle and won't take very much side force at all.
For one-off boards you can get away with 0.8mm and 1.2mm.  Anything
bigger just use HSS twist drills.  0.8mm gets used the most - all those
vias and IC holes.  If doing a lot of ICs, I think a 0.6 or 0.7mm is
about optimum.  The smaller the hole, the smaller the pad, hence you
can fit more traces between IC pads.

>
> Someone else said:
> >> Pre-etched PCBs: Been experimenting with these lately. Shows promise. I'm
> >> talking about the Rat Shack boards that are etched to look like
 breadboards.
> >> Simply solder in the sockets, components, and wire.
>
> That's what I've been using all my life.
>
> >Single sided with min. 20mil traces is easy.  Double sided and/or 12mil
> >is a pain.  However, by dint of much practise my boards are turning
> >out so well that I wouldn't consider wire wrapping or solder tags (remember
> >them in the days when components were big enough to see?).
>
> Single sided is good enough for me. And if it is just a matter a perserving
> I'll give that a go as well. Problem is I've got a deadline of mid-Feb...
>
>
> Thanks in advance
>
> Prashant

One project I have in the pipeline is to use an X-Y table in
conjunction with a nitrogen laser to directly expose the board without
the intermediate transparency step.  The X-Y table is driven by HPGL
plotter commands, with 'pen up/down' replaced with 'laser off/on'.  N2
lasers are easy to build, but a bit more difficult when a high
repetition output is required.  Another problem is the fact that the
laser is guided over the copper which will be _removed_, hence a
'negative' of the board will need to be plotted.  An alternative, of
course, would be to use negative resist.


Regards,
SJH
Canberra, Australia