In a message dated 02/21/2000 8:46:30 AM Central Standard Time,
[log in to unmask] writes:

<< Now back to the work/no clean thing:

 I am asking a BIGGG favor!

 I need to take a serious look at switching our process from a water soluble
flux
 soldering operation to a no clean one. Normally, I would put several cored
wire
 solders, liquid fluxes and solder pastes through a rigorous evaluation
process.
 That could be considered re-inventing the wheel, since many of you have
probably
 already done that.

 If any of you have done this and could share the results (OFF-LINE, of
course),
 I would be greatly obliged.

 I would like to know which brands, which model numbers, which alloys and
under
 what conditions they were tested, as well, of course, as the results.

 THANKS A BUNCH!!!

 Charlie B.
 [log in to unmask] >>

Hi Charlie!

At a past company I was heavily involved in the switch from water soluble to
no-clean. We focused primarily on the solder paste to make a selection, it
was important that the vendor also have the core solder, and other fluxes
that would be compatible with their pastes...we didn't want to have to go to
this vendor for our paste, that vendor for our core solder, etc..and by the
way, we didn't do any wave soldering there, it was all SMT...what little PTH
we had we hand soldered (memory company, we did do some PCMCIA modems and
stuff like that). It's been 3-4 years ago, and the formulation may have
changed since them, but I'll email you who we settled on .

I first narrowed the brands down to three. One of the vendors was our current
water soluble vendor (we gave their no-clean a try), and I picked two others.

Basically we used one of the more challenging assemblies that we built as our
test vehicle. It was a double-sided PCMCIA assembly that a 15.7-mil QFP on
it. They were also finished with OSP. We wanted a paste that had enough
activity in the flux to eat through the OSP. Which brings-up a little
"gotcha" that we didn't think about. Thought enough about making sure the
flux would be active enough to eat through the OSP, but didn't think a whole
lot about what do you do with your mis-printed OSP coated boards? We learned
that you clean, re-print, and run them as quick as you can! (learned that the
hard way, I might add...)

Then it was just a matter of building the boards. We looked at printing
qualities, what the joints looked like after reflow, and the characteristics
of the post reflow residue. We wanted a non tacky and clear residue. After
that I was able to eliminate two brands, one for poor print characteristics,
and the one other for a yellowish color to the residue.

The last thing was SIR testing, we took the final panel and ran it through
standard SIR testing to make sure the residue was truely benign.

Actually, selecting the soldering materials was the easy part, the hard part
was the change of culture that everybody had to go through. From the way
material was purchased and the checks that needed to be done at
incoming/receiving, to the way everything had to be handled, to the change in
the technique of hand soldering and touch-up was performed...it WAS a real
change! There was a lot of grumbling going on in the beginning.

All of the stuff we built was class-II, so we didn't have the headache of
conformal coating anything, which I understand can be a bear...

There was a few things that kinda' caught management unaware. First thing was
was the thought that once we converted over to no-clean soldering, that we
could throw our cleaners away. Can't do that...still gotta' clean your
misprints, your stencils, and there may at times be some boards that need a
good bit of rework, part replacement from test failures etc., those boards
should be cleaned. Also, what if you get some bare PCB's that for some reason
are found to be dirty (high ionic levels, etc.), do you send them back to the
vendor even if it means missing shipment so they can be cleaned?

Another thing that took a lot of people by suprise was "squeeze-balls" or
solder beads. We had close to 400 different part numbers of SIMM's, DIMM's,
PCMCIA Flash cards, Modems, etc., that were in production. There were some
parts that shared a common fab just loaded differently to change the part
number. But I know we had at least 300 stencils...and all of them were made
when in a cleaning environment. Squeeze balls aren't a problem then, because
they wash off...fact is, (sad to say) that's how a lot of people know if
something's going wrong with the cleaner, the squeeze-balls aren't washing
off...hehehe. But when I told management that we were going to have to spend
well over $100,000 getting our stencils re-made,
they about SH#T!!

It's a doable thing, but I'm sure you know it's not just a matter of
selecting the materials. A lot changes on the floor when using no-cleans. Not
quite sure why they wanted to switch, I just got the word that management
wanted it to happen by such and such a date. Mine is not to question why,
mine is just to "yada-yada-yada",end of story.

But I think the driver behind switching processes was that there was some
thought that there was going to be some money saved somewhere by not having
to invest in another cleaning machine (we only had one at the time and it was
overloaded) and all the other associated costs; power, resin beds (we were
closed-loop at the time). As it turns out, I think we wound-up spending more
money than if we just let it be...(as Paul McCartney would say...)

Just thought I'd share my experiences...

-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>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################