--- Brian Ellis <[log in to unmask]> wrote:
> Peter
>
> I disagree with you, sorry! These boards are
> definitely rejectable. A surface defect
> which crosses between two tracks is just cause for
> reject, IMHO, no matter what the
> cause.

Herein lies the problem. You call this thing a surface
defect. I say that its a blob of mask.  The mask is
supposed to be there.  Let me offer a different
scenario, if the fab had double coated the board and
all of the mask was equally dark (as dark as the
puddles) would you reject the board as defective?
This double thick mask might have bubbles etc but
because it is even you will probably accept it.

I won't argue that this isn't an ugly, ugly board.
But there may be a reason this board shipped in this
condition.  (cost of the board comes to mind)  Whether
we like it or not, when the price of the board is low
enough you often get what you pay for.

> Although, on the photo, there is no crossing
> (even though one comes precious
> close to it), some of the bubble sizes are over half
> the spacing. It is therefore
> conceivable to have two bubbles from opposite sides
> joining. Any trapped contamination
> could be a leakage path.
>
Please also note that I was trying to convey the idea
that these are not (might not) be voids but are (may
be) puddling.  Hey, I didn't make the board and I
don't have a sample in front of me.  From the pictures
these things look like puddles, not voids.

> A bubble is always highly suspect and the top skin
> may break away during soldering,
> leaving craters which would be contaminant traps.
>
> If I remember correctly, my mind going back 30 or 40
> years, there was a spec which
> stated something like "an isolated unspecified fault
> between two conductors will be
> deemed acceptable provided that it does not reduce
> the insulation path by more than
> 1/3. Two or more such faults within any inch square
> will be cause for rejection."

6012 Class X specifies this.  But looking at this
board, looking at the other defects shown from this
fabricator I doubt very much that this board was
ordered under any such spec.  And yes, I am a firm
believer that a clear understanding of the
specification requirements must be established between
the customer and supplier.  But that would be another
subject for another thread....

pete

>
> Brian
>
> Peter Menuez wrote:
>
> > First of all the pictures you show look pretty bad
> and
> > I don't believe I would ever ship something like
> > these. However, I want to point out that these may
> not
> > be 'rejectable'.
> >
> > After coating LPI there is a hold time to allow
> the
> > mask to debubble.  This debubble time varies on
> many
> > things but primarily on the thickness of the
> copper
> > trace.
> >
> > Think about the geometry of the trace.  If the
> trace
> > was 1oz (.0014") the mask would cover fairly
> evenly
> > and there would be little or no mask build up at
> the
> > trace.  At the trace thickness increases the trace
> > acts as a dam forcing the mask to build up along
> the
> > trace.  This is exactly what you are showing in
> the
> > pictures. The fact that the mask is very thin
> (most
> > pigment removed) at the top of the trace is a good
> > indicator that there is high copper here.)
> >
> > The debubbling hold time is designed to allow the
> > volatiles (solvents) to escape the mask.  What it
> > looks like in your picture is that the boards were
> not
> > debubbled long enough before they were tack cured.
> > What you have left are small dots where the
> solvent
> > was forced out of the mask leaving behind a puddle
> of
> > mask.
> > I have done a considerable amount of cross
> sectioning
> > on this anomaly and have yet to find a void in the
> > mask.
> >
> > I glanced over several of the emails responding to
> > your  original question and suggest that
> > contamination, curing, moisture etc. are based on
> > peoples inexperience with LPI's. This is a classic
> > debubbling issue.
> >
> > I said in my opening statement that these boards
> may
> > not be rejectable. They are rejectable for the
> simple
> > reason that they are ugly but probably not
> rejectable
> > for performance issues.  If you cross section
> these
> > puddles you will see that there is more mask
> present
> > (rather than less which would be the case with  a
> > void).  If you have more mask you have more
> insulation
> > and in fact your insulation performance would be
> > increased.  Ok, this is pretty weak but...
> >
> > The area I would concentrated on with your
> supplier
> > are: a) control of the plating process and
> >      b) if there is anything you could do to help
> them
> > control the plating (robber buss/thief area)
> >
> > pete menuez
> > [log in to unmask]
> >
> > --- Brian Ellis <[log in to unmask]> wrote:
> > > Steve
> > >
> > > This is incredible. Change your supplier as top
> > > priority :-)  Seriously, as the others
> > > say, the "bubbles" are probably due to
> solvent/water
> > > entrapment in the track corners.
> > > However, I see another problem which may/may not
> be
> > > related. Am I correct in thinking
> > > that it has been hot and humid where you are,
> > > lately? If so, it would seem that there
> > > has been hygroscopic contamination left on the
> > > board. On your first photo, see the
> > > myriad white points between the tracks? This
> looks
> > > like vesication to me. The most
> > > common cause is hygroscopic contamination (ionic
> or
> > > non-ionic) which pumps atmospheric
> > > humidty across the mask layer (all polymers can
> > > allow humidity through). The osmotic
> > > pressure rises until it actually lifts the mask
> off
> > > the surface, locally, forming a
> > > vesicle or, if you like, a microblister (see my
> book
> > > for detailed explanation). This
> > > could possibly be the cause of your bubbles
> along
> > > the track edges, as well, although
> > > these seem rather gross: the photo I use to
> > > illustrate vesication also has bubbles
> > > along some of the track edges, but the
> vesication
> > > between the tracks is on a similar
> > > scale. Whatever, it is a totally unacceptable
> fault.
> > >
> > > Brian
> > >
> > > "Stephen R. Gregory" wrote:
> > >
> > > > Hi all!
> > > >
> > > > We just got some boards in at receiving
> inspection
> > > that have thousands of
> > > > little "bubbles" in the soldermask all over
> the
> > > boards...they seem to
> > > > concentrate all along the edges of the
> features of
> > > the board. You can see
> > > > pictures of it at:
> > > >
> > > >
> > >
> >
>
http://www.driveway.com/share?sid=e25a88c4.8e904&name=Pictures
> > > >
> > > > One picture is bubblemask.jpg, the other is
> > > bubblemask2.jpg...what would
> > > > cause that? It's a LPI...don't know the brand
> > > name. I'm thinking that either
> > > > the boards weren't dry or cleaned well prior
> to
> > > soldermasking them. I'm
> > > > rejecting the boards of course...
> > > >
> > > > My freedrive is filling up fast with all the
> > > pictures of the "pretty boards"
> > > > that we get from our vendors!!!
> > > >
> > > > -Steve Gregory-
> > > >
> > > >
> > >
> >
>
##############################################################
> > > > TechNet Mail List provided as a free service
> by
> > > IPC using LISTSERV 1.8c
> > > >
> > >
> >
>
##############################################################
> > > > To subscribe/unsubscribe, send a message to
> > > [log in to unmask] with following text in
> > > > the body:
> > > > To subscribe:   SUBSCRIBE TECHNET <your full
> name>
>
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.com/

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################