George
Before going to the meat of your question, is the 20 years of continuous
service requirements realistic? Cast your mind back 20 years and think
what electronics were available. VLSI was in its infancy then, with
chips like the 8086 in PCs, operating at 4.77 MHz, with 640 kb max of
memory. Today a state-of-the-art machine would be over 500 times more
powerful on both counts. In 20 years time, it is probable that the
printed circuit will not even exist as we know it today, let alone
today's semiconductors. In the "Energy and Automation" field, are you
still using boards you designed 20 years ago? What is the real life
expectancy of your products?
That having been said, I have always considered that using "no-clean"
chemistry for long life-time applications is hairy. Let us consider what
may be left on the boards after soldering:
- unaltered or thermally altered residues from the bare PCB and
components themselves, as they are delivered. These may be hygroscopic
(and therefore not necessarily measurable in an ionic contamination
tester) and/or ionic
- unaltered or thermally degraded residues from packing, handling,
assembly equipment, undefined dust, air pollution, maybe even salt if
you are close to the sea (in Tennessee?) or close to a road that is
salted in winter
- unaltered flux/paste residues
- thermally degraded flux/paste residues
- metal salts (by definition, ionic) formed by the reaction between the
flux activators and metal oxides and carbonates on the solder, the PCB
lands and the component leads or metallisation
The above together is a totally undefined porridge over which you have
little control. With "no-clean" products, the process that makes them
reasonably reliable does not end after soldering but may continue for
months afterwards as the ionic flux activator residues slowly sublime.
If you coat before this sublimation takes place, then you are capturing
these products, which may reduce the reliability, into the polymer
matrix of your coating. Not only is this a source of potential danger to
reliability, it may also upset the polymer matrix itself by changing the
cross-linking characteristics.
Now, the next question is should you clean "no-clean" products? My
opinion is an unequivocal no for most applications. Fluxes that are
designed not to be cleaned are designed to be as innocuous as possible
if left in place and their chemistry is often such that cleaning is
either not possible, at all, or, at least, very difficult. Why use a
stick to beat yourself over your own back? Use a flux/paste that is
designed for easy cleaning and use the cleaning method recommended by
the manufacturer.
Yes, I know that there are many who clean "no-clean" fluxes/pastes and
there are many who coat over "no-clean" residues. In other words, these
guys will contradict what I say. But do these same guys guarantee that
their products will work in the summer temperature and humidity of
Florida for 20 years? Do you even have facts and figures of the
reliability of their products? Are their products working under similar
electrical conditions, such as impedance, frequency, characteristic
impedance, voltage gradients etc.?
You have a choice; you may take a big step into the risky unknown,
remembering that no one can really advise you as your products and
conditions form a unique combination, or you can use the braces
(suspenders, for you) and belt technique and work according to the best
known and proven technology, which you would have to qualify under your
conditions of working, to obtain maximum reliability, knowing that you
can sleep in all tranquility on both ears. Any departure from the latter
can only be at your own risk, but you may be able to get away with it --
on the other hand, you may not!
What I've written above is not an exhaustive analysis of the situation
but simply generalities. If you wish to go into the particularities of
your case, please contact me off-list.
Brian
Carroll, George wrote:
> I've read through the archives, but I'll ask again anyhow.
>
> Washing no-clean boards before conformal coating: Is it necessary for
> extended reliability? The polyurethane conformal coated, mixed technology
> board in question will be in a housing but not sealed from the outside
> environment of daily and seasonal temp and humidity variations (SE US).
> Customer is looking for 20 years of nearly continuous service.
>
> First, adhesion is not an issue. Our process has coated tens of thousands
> of no-clean assemblies without coating adhesion issues. We use a ORL0 flux
> and REL1 type paste. I've Omegameter tested incoming bare boards and they
> are <2 micro g. / sq.in. Our assembly operators are instructed to handle
> boards with gloved hands. The question is - is washing prior to conformal
> coat required for long term or, for that matter, does it even improve (or
> degrade) the long term prospects.
>
> At this point it would appear that if board washing were mandated, it would
> probably be with saponifier followed by D.I. rinses and a bake to dry the
> assemblies. Our conformal coating always measures (flat coupon) > 1 mil and
> averages 1.8.
>
> My own opinion from lack failures of coating over no-clean in the field is
> that coated no-clean boards will survive and function as needed. However
> when a customer asks, "20 years?", it's hard not to waiver. Do I really
> need a belt AND suspenders?
>
> I will be coating and having B-25A's tested to 830B in our lab in a month
> (and setting up accelerated aging tests of these boards later on) to look at
> this and other questions but, for the moment, I'm looking for the opinion of
> the experts.
>
> George Carroll
> Process Engineer, Siemens Energy & Automation
> P.O. Box 1255
> 3000 Bill Garland Road
> Johnson City, TN 37605
> (423) 461-2948
> [log in to unmask]
>
> ---------------------------------------------------
> Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
> To unsubscribe, send a message to [log in to unmask] with following text in
> the BODY (NOT the subject field): SIGNOFF Technet
> To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
> To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
> Search the archives of previous posts at: http://listserv.ipc.org/archives
> Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
> information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
> -----------------------------------------------------
>
>
>
---------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-----------------------------------------------------
|
