Homebrew PCBs

* 1969 US patent for Riston and the like.
* Explanation of the role of the protective sheets.
* Too narrow a light source may cause uneven exposure.
* Delaying development with the polyester sheet on after exposure.
* My thoughts on storage requirements for DIY PCB quantities of Riston.

(This message is intended to be read in fixed width fonts without Yahoo
Groups trick of removing leading whitespace.  In the archives, use:
Options > Use Fixed Width Font.)


Todd F. Carney mentioned a fascinating DuPont document concerning the
polyester (top) and polyethylene (bottom) cover sheets for Riston and
the like - dry film, heat-applicable, negative photoresists:


http://www2.dupont.com/Imaging_Materials/en_US/assets/downloads/techtalk/TT1207.pdf

    Tech Talk - Fine Lines in High Yield (Part CXLVII = 147)
    Dry Film Photoresist Coating Base Film and Coversheet
    Considerations

    Karl Dietz (for CircuiTree Magazine, December 2007)

      (According to
       http://pcb007.iconnect007.net/pages/zone.cgi?a=75827
       CircuiTree was a magazine established in 1987 and
       which ceased publication in April 2011.)

This refers to a 1969 US patent 3469982 as the basis for these resists:

   http://www.google.com/patents/US3469982

I have printed this but not yet read it.


Narrowness of the light source
==============================

Karl Dietz's piece mentions an interesting limit to how narrowly
collimated the exposure light source should be, though I don't recall
seeing any mention of this or quantification of the limits in the
datasheets and processing guide I have read from DuPont:


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

If the light rays are too narrow, little grains in the polyester layer
can induce uneven exposure of the photoresist.  I never thought The
polyester film I remove from MM540 does had any graininess, diffusion or
the like, because it looks clear enough and such films generally are
perfectly clear.  However, Karl Dietz explains their reasons for
introducing such grains, and on close examination with an eyepiece with
suitable lighting I can see a general low-level irregularity in the film.

I would be surprised if this is a problem with the 500W incandescent
lamp exposure technique I use (as describe in other messages) but it
might be a consideration if using the Sun or perhaps a laser source of UV.


Delaying development?
=====================

In a recent message (sorry, I can't easily find it now) there was a
reference to a website which recommended leaving the exposed PCB with
its polyester top sheet on for a few minutes before development.  I
don't recall reading any such thing in the DuPont documentation.  This
technique would make sense if the delay time was required to complete
polymerization, but I wonder to what extent it would affect sensitivity.


Polyester top sheet excludes oxygen and must be left on during exposure
=======================================================================

I was intrigued by this statement by Karl Dietz on the importance of
retaining the polyester film during exposure (and perhaps, I guess, time
after the exposure leading up to development):

   The polyester has an additional important function.  It
   serves as a barrier against oxygen diffusion into the
   resist.  The oxygen concentration in the resist needs to
   be kept low because oxygen is a potent radical scavenger.
   It reacts with acrylic monomer radicals before they have a
   chance to induce polymerization.

   Thus, the first acrylate radicals formed in the exposure
   process react with the oxygen that is present in the resist
   and a (are) lost as polymerization initiators.  Once the
   oxygen is depleted, and the polyester oxygen barrier
   prevents the oxygen concentration from building up again,
   then the radical induced polymerization can begin.

This prompted the following thoughts on the exposure threshold of the
resist and on storage arrangements which are particular to DIY PCB folk.

In my exposure tests with Riston MM540 from Martin Dusek in the Czech
Republic:

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

I noticed there was a sharp contrast.  Up to a certain level of
exposure, no photoresist would remain.  Marginally above that, the whole
layer of photoresist would remain.  This is a very good thing since it
improves contrast and enables the use of phototools with lower contrast,
such as the direct laser-printer to (slightly matte on one surface)
polyester/Mylar film.

From the above passage I conclude that this threshold is created largely
or entirely by the level of oxygen in the film itself at the time of
exposure.  (I tried various techniques of removing the polyester sheet,
to bond the phototool directly to the photoresist, or with various
adhesives, but these techniques were impossible to use reliably due to
variable and often extreme or total loss of sensitivity.)


Storage - excluding oxygen from loosely wound small rolls
=========================================================

This raises questions about storage.  Riston and the like would be
supplied by DuPont in firmly wound rolls, with no appreciable air gap
between the top of the polyester and the bottom of the next layer's
polyethylene.  The tension would probably be quite tight to exclude air
over several years of storage.  The films are made to be smooth enough
that this pressure does not appreciably deform the soft photosensitive
material sandwiched between them.

PCB manufacturers would purchase, store and use full rolls like this and
the material would only come off the roll a few seconds before being
laminated to the PCB (except for that material between the roll and the
PCB in a laminating machine, which might spend hours, days or weeks
(over a production break) exposed to the air on both sides).

In this way, the oxygen content of the photoresist would be unchanged
from the level at which the film left the roll, which I guess would be
unchanged from the level at which it left the factory.

When we DIY PCB folks buy Riston the way I did, in small quantities, for
an agreeably low cost, the film is no longer wound tightly on its
original roll.  Therefore, it would not surprise me if over months of
storage in this form that oxygen would get into the photoresist material
- especially through the thin and presumably not very oxygen-tight
polyethylene layer.

My first batch of MM540 became unusable after a few months.  I initially
attributed this to carelessness on my part in not keeping it
sufficiently protected from light.  However, now I believe the cause was
the loose-wound nature of the small roll.

Potential solutions might include:

  1 - Winding it firmly around a plastic tube, wooden dowel or
      similar and keeping it tightly wound until it is taken off
      for use.

  2 - Keeping it in an oxygen-free environment:

      a - Vacuum - it would need to be a very high vacuum to
          eliminate the small amount of oxygen which might
          contaminate the material.  Also, the vacuum might
          affect volatile substances in the film itself -
          *including* its oxygen.  So the vacuum might
          cause the exposure threshold needed for polymerization
          to *drop*.

      b - Nitrogen, carbon-dioxide or the like.  CO2 might be
          better because I think its density would reduce
          diffusion to the atmosphere if kept in a not completely
          sealed container which resembled a bucket or tub.

          It could be difficult to completely exclude oxygen from
          any such container without pumping in a large quantity
          of oxygen free gas.

          I am not sure where to get such gases on a low-cost
          convenient basis, other than butane which is very
          inexpensive, but also a fire hazard.

          As for vacuum, this could cause the oxygen in a non-
          tightly rolled store of Riston to escape by diffusion
          into the low-oxygen gas it is stored in.

I think the only reliable solution would be tight winding around a
suitably smooth mandrel, with the recently removed end piece, which was
not cut off and used, being firmly applied to the roll again.  However,
I think that if months elapsed after this, any Riston which was not seen
(in safelight) to be closely bonded to the roll layer below should
probably be discarded.

Therefore I think a narrow mandrel would work best, since it makes the
outer turn of Riston smaller than if a larger diameter mandrel was used.
 Several turns of cling-wrap could be wound around the Riston to press
the end-piece down and to reduce contact with the air.

It would be better if the supplier, who presumably removes the Riston
from a larger roll, directly wound it onto a mandrel of some kind,
rather than shipping it in a potentially air-exposed form and requiring
the purchaser to to the tight winding.

The roll I got from Martin Dusek was tightly wound, but had nothing in
the centre.  It is my impression that he wound it onto a mandrel which
he could then split in some way to remove the wound roll.  I intend to
write to him, sending him a copy of this message and any significant
discussion which results.

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

Addenda:

Jeff Heiss in the "photoresist application" thread mentioned the site
which recommended leaving the polyester sheet in place for a while after
exposure:

  "Note that the dry film resist should be exposed with the cover
   sheets on, and then held in a dark place for 15 minutes before
   removing the cover sheets and developing."


http://www.electro-tech-online.com/general-electronics-chat/14427-photoresist-pcb-etching-process.html

This was from 2005, quoting extensive Riston 4713 instructions by Jon Elson:

  http://pico-systems.com/PCBs.html

I recall reading this instruction somewhere else, but not in DuPont
material.

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