Most of the consumer electronics I have torn apart are manufactured with out
flux removal. I am not aware of any definitive studies. In the lab, where I
once worked, we saw many examples of failure from flux contamination, almost
always poorly executed aqueous processing. I don't recall seeing a case
involving true L0 class low residue fluxes.
In my current position we did see returned boards with corrosion adjacent to
L0 low residue fluxes. Turns out our customer was using the assembly in a
condensing environment. As described by Brian below, the result is not
acceptable.
Traditionally these assemblies were made with WS fluxes and cleaned. Now, we
are seeing MEMs relays and MLD packaged charge pumps. We simply cannot clean
under them. The maker of the charge pump recommends no-clean flux because
they have easily swamped high impedance inputs. . . . We are looking at
coatings and solvent cleaning.
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Hernefjord Ingemar
Sent: Wednesday, May 23, 2007 6:26 AM
To: [log in to unmask]
Subject: Re: [TN] Flux residue, burnt flux
Thanks, Brian,
those were indeed worst case stories. However, limited to one customer and
one place. Some boards are produced in numbers of millions, for instance
electronics for cars. Or PWBs for microwave ovens, or parabol antenna cards
etc. I have never heard about recalling thousands of boards because of flux
residues. I would like to know how many boards are made yearly with flux
removal vs. non removal. Maybe flux removal is still dominating?
Inge
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Brian Ellis
Sent: den 23 maj 2007 11:07
To: [log in to unmask]
Subject: Re: [TN] Flux residue, burnt flux
I have seen many cases of massive failures due to flux residues. The most
spectacular was in the late 1970s. I had a client making NC machine tools
and he used boards made by a very large German company. These were soldered
with a DIN 8511 (?) F-SW32 flux, which many Germans considered as "no-clean"
as they were halogen-free (and they were hell to clean, anyway). They were
rosin fluxes, typically 15-25%, activated with alkane carboxylic acids. The
theory was that the rosin held the activators harmless in its matrix. The
practice was far different. I had a look at some circuits that had been in
service for a year or two. The copper had all but disappeared on several of
them. I was asked to find out why, as this was the first time this problem
was recorded and it was not just one circuit but many. I visited my client's
client and found the machine in question was installed in a large workshop
where there also a number of honing machines using a water soluble cutting
oil. The visibility in the shop was about 20 m, as there was a mist of this
cutting oil. I needed no further knowledge; the circuits were operating in
an aqueous aerosol and this was reacting with the flux residues to produce a
nice acid solution. The rest is history. I recommended my client that he
clean-air-purged the NC electronics cabinet and replaced all the existing
circuits and this cured the problem.
Another spectacular case, about 10 or 12 years ago. A client, making very
high-power electronics, asked me for advice. He had a number of failures
from a recent batch of units that he had been making for a number of years.
Tracks connected to three-phase 400 V power had sinply and suddenly
volatilised, even though they were about 7 mm wide in 105 µm copper, between
the input connector and the 100 A fuse holders, a distance of about 5 or 6
cm, all three of them. To cut a long story short, no-clean wave-soldering
flux residues had their activators sublimate as the conductors warmed up and
they condensed on the top edge of the fuse-holders. These were,
unfortunately, spaced by only about 3 or 4 mm and, after some time, a
short-circuit occurred sufficient to start an arc between two phases and one
of the tracks overheated to explosively splutter molten copper around like
there was no tomorrrow, causing the other two tracks to volatilise in turn.
It transpired there had recently been a change in the formulation of the
flux that was used, which was why it had never happened before. I
recommended that they used a W/S flux with cleaning and they redesigned the
PCB for more spacing between the fuse holders. No further problems, AFAIK.
Many other cases of leeser massiveness.
Brian
Hernefjord Ingemar wrote:
> What I say now may cause some to think that I'm member of
> Ku-Flux-Klan, which is not the case. So, we follow the general habit of
removing 'all'
> flux residues. However, I can't deny, that I think this flux removal
> hysteria is little exaggerated. I've been in the business for so long
> a time, and I have not seen many reported failures that have been
> caused by flux residues. In theory, flux residues have many
> ingredients that can cause corrosion, leakage current, decreased
> insulation etc, but it seems as that does not happen in reality. 100%
> cleanliness is satifying and beautiful, but costs a lot to obtain.
> Just a thought. Would be very interesting if anyone could describe a
> case with flux residues causing massive failures.
> Inge
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Brian Ellis
> Sent: den 23 maj 2007 09:38
> To: [log in to unmask]
> Subject: Re: [TN] Flux residue, burnt flux
>
> I'm sure that an eminent co-contributor to this forum will tell you,
> "it depends". If the flux is truly burnt, i.e., black, this indicates
> pyrolysis, which is a fancy word meaning decomposed by heat. Pyrolysis
> indicates that the residues have split apart into numerous compounds,
> leaving carbon-rich stuff. Elemental carbon can be an electrical
> conductor; do you want conducting particles in your assembly? They may
> appear fixed in place now, but will they remain so during the life of
> the equipment?
>
> More important, WHY are they there? It may be because the operators
> don't keep the bits of their irons clean. Do they wipe them on a wet
> sponge before each joint is made? It may be that the time/temperature
> conditions of the joint being made are far from optimal. It may be
> lack of adequate training of the operators. I can't tell. Whatever,
> prevention is better than cure; a lttle research into the causes may
> give you the answer.
>
> As to flux flow, maybe your solder wire simply has too much flux. Some
> manufacturers allow you to choose the percentage. Yes, it is easier to
> solder with an excess. It's a compromise.
>
> What you have not told us is the essential information: what kind of
> assemblies are you making. You can obviously be more tolerant of
> imperfections if you are making toys than if you are making inertial
> guidance or satellite systems. Probably you are somewhere between thes
> extremes. "It depends"!
>
> Brian
>
> Sue Powers-hartman wrote:
>> We fight a constant battle with operators leaving burnt flux in
>> joints. Maybe only a small speck, but drives the inspectors nuts.
>> The
>
>> way I read JStd-001D, if they can not see it at referee inspection
> power, they have to accept it.
>> How dangerous is this burnt flux to the PWB? If it's not seen at
>> inspection power and left on the board, what happens. Also, what
> about no clean flux?
>> Our solder training video says that if no clean flux runs out to far
>> and is not heat activated, it can cause problems. Operators watch
>> this
>
>> video, but somehow do not get this. They say that it's no clean, they
> can leave it all on.
>> I keep saying that this can be a problem, and then they ask me, how
>> far out can the flux be away from the joint before it's unacceptable.
>>
>> Wow, I'm glad I found this forum, I have so many questions to ask you
> guys.
>> Anyway, thanks for the help on this subject.
>>
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
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/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------
|
