TECHNET Archives

October 1998

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Douglas Pauls <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Wed, 21 Oct 1998 19:30:09 EDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (133 lines)
Good day all,

I've been restraining from commenting on John's posting, but right now I'm
watching my instruments do SIR readings, which is about as exciting as
watching paint dry, soooooooooo:

Be aware that John and I have had a (amiable) running argument on this issue
for some time.  I think we are at a point where we agree to disagree.  Each of
us have good data to support our positions.  I have a great deal of respect
for John's opinions.

>
>  Ion chromatography provides a quantitative assessment of those ionic
species
>  in an extract.  One key issue is the extraction methodology - time,
>  temperature, and extraction medium, as well as the part of the
>  board/assembly that is being extracted.

No arguments here.  The high temperature, long term extraction method of
2.3.28 DOES pull ionic materials from the substrate, specifically bromide,
which would not be detrimental to electrical field performance.  This
extraction method was essentially designed to pull ionic materials into
solution such that you did not have residues left on a bare board or finished
assembly.  You want to pull ALL the residues off to examine.  On one end of
the extraction spectrum you have the room temperature ROSE test.  Most agree
that this is really benign and only pulls a fraction of the residues off the
assembly for examination.  We showed this as part of a Georgia Tech study
(ETC96) in 1996, handled by Dr. Laura Turbini.  Considering the wide range of
residues, you would be hard pressed to find a single extraction solution
composition, time, and temperature that would get all residues off the surface
of the assembly, yet not extract anything from the substrate.  Especially so
when considering the wide variability in the porosity of laminates and masks.

The key to accurate analysis is running blanks and developing a good
historical database of laminate X generally produces Y micrograms per square
inch of bromide.  For example, using 2.3.28 (as we do daily), we know that
almost all FR-4 laminate produces roughly 5-7 micrograms per square inch of
bromide from the fire retardant.  We know this from testing boards prior to
flux exposure (e.g. HASL).  There is no other place for the bromide to come
from.  Thermal excursions may increase this to 10-12 micrograms per square
inch.  Blanks and heated controls tell us this.  Therefore, when we see an
assembly with 30 micrograms per square inch, barring a special laminate
formulation, we know a brominated flux was used and about 20 micrograms per
square inch is flux related.

The issue all boils down to: "How much bromide (or any residue) is too much
bromide"?  It is not an easy question to answer.  The IPC committees have been
trying for a long time to define "how clean is clean".  The jury is still out.

>  Simply using 2.3.28 can lead one
>  astray - see the presentation I gave at the Fall 1997 IPC Bare Board
>  Cleanliness Task Group Meeting (detailed in the Fall 97 minutes, page 86).
>  This case study showed that the results from method 2.3.28 were NOT
>  predictive of field performance.  It is important to use electrical testing
>  in combination with extractive analyses.

Don't know that I would have used the word astray in this case, but John does
have a good point.  Relying on one evaluation point (ion chromatography in
Johns Fall 97 presentation) alone is risky.  It is important to verify whether
the risk will have impact on a reliability assessment, using other tools like
electrical testing.  No arguments here on that point.  I am familiar with the
background data from the presentation John cites.  In that case, there were
high bromide levels on hardware as determined using method 2.3.28.  John's
data indicated that there were no field performance issues with these levels
of bromide.  I have no issue with John's position here.  He has good data to
support his position.  We are in the process of doing some internal "data
mining" on this issue to make our position more statisical and less anecdotal.

On the other hand, I have had clients who have had bromide levels far lower
than those seen in John's presentation and experienced extensive corrosion,
electrical leakage, and metal migration.  Once the bromide residues were
cleaned to lower levels, the problem went away.  On the third hand, we have
also had clients who have had bromide levels higher than seen in John's
presentation, with no field failures.  So, yes, ion chromatography will not
always be predictive of field performance. But then, I don't know of any other
single test that is either.

The way that I look at it, the amount of halide present, e.g. chloride or
bromide, represents an electrochemical risk.  Low levels of halide represent a
lower level of risk.  The higher the halide level, the higher the risk.  Many
factors go into whether or not that risk translates to field failure.  Is it
coated, is it not?  Is it indoor computer office environment or automotive
under the hood?  Can water drip on it or can it not?   The list goes on.  In
addition, some designs or combination of materials seem to be "bulletproof"
against halides.  There are lots of factors to consider.  Until you know
better, by alternative test batteries, it is prudent to treat high halides as
a danger, as recommended in my earlier post.

>
>  No-clean fluxes have been used successfully for ~ 10 years.  There are
>  specifications in place (e.g. Bellcore) that allow for production of
>  assemblies that can meet the reliability requirements of the
>  telecommunications industry.
>

Yes they have.  I have seen the same laminate/mask/paste/flux used beautifully
in one facility to make assemblies that will last 50 years, and in another
facility it is used to create scrap.  I have lost track of the number of no-
clean assemblers who have been nailed with field failures due to high halides
on the incoming bare boards.  High bromide from a brominated HASL flux is one
such source, which is why I responded to the earlier post.  Assemblers need to
be aware of this issue.

Maybe I am seeing a slanted view of the issue because I do process
troubleshooting and perhaps see a disproportionately high number of problem
cases or failures.  Then again, maybe not.  We are seeing more and more
product being returned from the field with electrochemical failures because of
this issue.  Ron Daniels, editor of Circuits Assembly magazine has also noted
the same trend.

I have my own viewpoints on the ability of the Bellcore specification (or J-
STD-001 for that matter) to guard a manufacturer against electrochemical field
failures, but that is another issue and a dragon I am not going to joust with.
At least today.

Thus endeth this missive.

Doug Pauls
Tecnical Director
CSL

################################################################
TechNet E-Mail Forum 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's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################


ATOM RSS1 RSS2