Homebrew PCBs

I just made a little batch of boards using a fresh batch of Riston MM540
from the Czech Republic, via eBay:

  http://www.tech-place.com/pyralux/23-photosensitive-film.html
  http://myworld.ebay.com/gaminn/

As I described earlier, I am using a Brother HL-5250DN on laser film
from Canada:

  http://screenprinting.asc365.com/index.asp?ID=PD007003

where the black blocks all but about 1/7th of the light.

This is with a 500W 240V incandescent floodlight with about 40cm from
the filament to the PCB.  The protective glass from the floodlight was
removed, and I used it to hold the phototool to the PCB, which was on
some firm foam.  I used extra weight to hold the glass down more firmly.
 Glass is not as good for pressing the phototool to the PCB as a
flexible plastic vacuum frame would be.

I did two exposure times for this phototool which contained 12 little
PCB images.  The boards which got 2 minutes 20 seconds mainly worked,
but there were a few defects.  Those which got 4 minutes worked very
well.  The boards were single-sided, with SMD components including an
SOIC-16, with 0.010" tracks.

This is quite a range of exposure times.  In the past I had a test
exposure where boards worked out not too bad with exposure times which
were a factor of two different.

I think this Riston is quite a high contrast material, so the ~7:1 ratio
of light and dark I get, which is probably 5:1 or so for the weakest
parts of the "black" image, is perfectly adequate without any need to
make the black parts darker.

I think the key to this is to use a clear film which the laser toner
bonds to evenly - as is the case with the material I am using.  A
diffuse film would reflect some light back and would scatter the light
it passes.

I am using a Lowell laminator modified for cooler operation at around
125C instead of 150C, as I wrote to the list on 19th May.  I cut away
the top housing so I can see the aluminium channels which contain the
rollers.  I also got rid of a pesky metal thing on the output slot.
(This is arguably unsafe, since the metal is not grounded and there are
potentially exposed mains wires - so don't take this as a recommendation
for a safe way of working.)

I find the trick for laminating large areas of Riston without wrinkles
or bubbles is:

1 - Clean the copper with Jif (abrasive white liquid for cleaning sinks
etc.) of similar, to give it a clean and perhaps
slightly rough surface.

2 - Press one edge of the Riston onto the leading edge of the copper
clad board.  All the rest of the Riston should be laying flat, on the
copper *except* that there is a sheet of paper between the Riston and
the copper.

3 - As I feed the board into the laminator, I hang onto the paper so its
edge is always a few mm away from the rollers.  I may lift the Riston up
a little and provide a little tension on it.  The goal is to stop the
Riston sticking to the copper until it is very close to, or exactly at,
the rollers.

As per the Riston documentation I develop with 0.85% sodium carbonate
solution, warmed up somewhat to I guess 30 or 35C.  There is a rinse
step as well, but I haven't yet got suitable mildly hard water.  Maybe
Melbourne tap water is OK.

The MM500 series datasheet is the general processing guide is at:

  http://www2.dupont.com/Imaging_Materials/en_US/tech_info/datasheets
/index.html

I am etching with ferric chloride and citric acid, as I will describe in
another message.  I strip the Riston off the etched board with 3% sodium
hydroxide.  These circuits were with 0.5 oz copper on thin FR4:

 http://www.pcbfx.com/main_site/pages/products/rigid_pc_boards.html

  - Robin   http://www.firstpr.com.au/pcb-diy/