Hey Steve ;
hearing about the "custom in-house" dual pads does explains it all :
I'd say your chances of attaining the hero status were very remote indeed ;
the tones of balls likely washed out from the 1206's on pads which could
take 0805's as well would warrant paste mixing division (AKA Matt subplant)
.
Andy ; the dispensing paste 's got 85% of metal ; but true Steve wouldn'
like the price (grading) and touch more flux .
Slump 's no problem ; agree with the "polymer paste"; tested amtech ; worked
well .
see you paul
PS
stop this false whining Steve ; you know very well it's best fun on bottom
budget starting up ; ain't foolin' anybody !
Guys like Bev have to caffeine prop eyelids all day long (kiddin'!)
PPS
Hope yo've got something out of this saga Frank .
IF (and only) you have this ball decease ; than check the profile before you
go to attack printer or stencils .
There are as many solutions as variables ; try the "easy" ones first ;
profile tweaking can do wonders most of the time (takes time too) .
If you bottom down the componentry gently they won't spew ; even without
going to eclectic arrow pad designs (nice if you can redesign it all)
> ----------
> From: Stephen R. Gregory[SMTP:[log in to unmask]]
> Sent: Saturday, 30 January 1999 4:36
> To: [log in to unmask]
> Subject: Re: [TN] Design considerations for No-Clean processes
>
> In a message dated 1/29/99 6:10:33 AM Pacific Standard Time,
> [log in to unmask] writes:
>
> << Paul Klasek raises some interesting points. If you can't change
> stencils,
> then it may be possible to use a softer squeegee (less than 90durometer)
> with a slightly higher print-pressure setting, to 'scoop out' the amount
> of
> paste you need. However, if you then try and slowly introduce
> reduced-aperture stencils, you'll have to mix and match print-processes
> across the production floor.
>
> I think there may be an opportunity for the solder paste suppliers out
> there to come up with exactly what you are talking about: a high solids
> content flux, with a reduced metal loading (i.e. viscosity/rheology stays
> pretty much the same). The problem with this is that hot-slump becomes
> much
> worse, so it may not actually buy you much reduction in the solderballing
> -
> I'm prepared to be convinced otherwise. May be a niche market for some of
> those 'polymer pastes' out there?
>
> Alternatively, inert reflow, giving improved solderability to all metal
> surfaces, may be another way of reducing 'squeezeballs/capillary
> balls/sideballs'. I'd be interested if anyone has any experience with
> this...
>
> Andy >>
>
> Good Morning Ya'll!
>
> Say, this is getting interesting! Let me elaborate a little more
> about
> things during that process change...and no Paul, I won't stick by my guns
> and
> say buying new stencils was the ONLY way I could eliminate squeeze balls,
> solder beads. But it was the quickest way I could be sure that I wouldn't
> be
> having a problem with them. I needed to decide on something pretty
> quickly,
> 'cause during the busy months we needed to build anywhere from 800,000 to
> a
> 1,000,000 modules a month come hell or high water.
>
> You see, back when I joined that company, the date had already been
> set
> when the company was going to go no-clean. In fact, that was something
> that
> was discussed at my interview as being part of my responsibility and why I
> was
> being hired, to take the process from water soluble to no-clean. I told
> them;
> "No problem! I'm looking forward to it!".
>
> After the interview while I was walking back to my car, I heard this
> little voice in my ear, kinda' sounded like Ren (you Ren and Stimpy fans
> will
> appreciate this..hehehe); "You EEDIOT! WHAT WERE YOU SAYING IN THERE?"
>
> I'm not trying to malign anybody, but we're talking about a memory
> company here. You know; "Hurry-up and build the darn things so we can ship
> em'
> before the prices drop again!" was the battle-cry we heard constantly. So
> I
> had quite a task ahead of me. Not making any excuses, but I didn't have a
> heck
> of a lot of time to experiment much.
>
> Although I did try what Andy mentioned, which was using plastic
> squeegees. I experimented with durometers from 80 to 100, and I could get
> rid
> of practically all of the balls, but not without also causing insufficient
> problems with the other parts...stands to reason don't it? I had to use
> the
> high squeegee force to scoop out the passive pads, so I'm gonna be
> scooping
> out the other pads as well, maybe not to the same degree with the narrower
> openings, but it doesn't help when most everything else is 50-mil pitch
> SOJ
> footprints, and you've got caps that sit beneath them...they'll scoop out
> pretty good too.
>
> Even thought about lowering the metals content too, but I was in a
> catch-22 there. Lower metals content means more flux residues, and that
> was
> one of the things that was a critera when selecting the paste we were
> going to
> use. The paste needed to have fairly little residue, it could have some,
> but
> we didn't want a lot. The residue also had to stay nice and clear through
> double reflow as we did a lot of that, and it also had to come out of
> reflow
> with the residue dry and hard, not tacky or sticky because of all the
> routing
> after the panels are built. Another problem with the residues that came up
> after we started building product with no-clean, was with the cache'
> modules
> not passing test until after they were cleaned. They use Syncronous DRAM
> and
> seemed to be real sensitive to the residues. It wasn't a HUGE problem, and
> didn't happen with a lot of boards, it was just big enough to be noticed.
> But
> I imagine if I had more residue, that might've increased the problem? I
> don't
> know, I'm not much of a test guy...
>
> One last little curve that went into the equation deciding whether or
> not
> the stencils needed to be replaced or not, was this in-house "custom"
> footprint that was incorporated into many of the designs for the
> decoupling
> caps by a layout guy a few years before I got there. What he had done was
> to
> lay out pads that you could put either a 0805 or 1206 on. I imagine it was
> a
> real keen idea back when you didn't know what package cap you could buy,
> but
> with the way the pads were laid-out so a 0805 could reach both pads, a
> squeeze-ball was guaranteed when you placed a 1206...I don't even remember
> how
> many assemblies had that footprint, but it was a LOT!
>
> Believe you me, there was a lot of questioning going on when I told
> them
> that we needed new stencils, and if I had any sorta clue what to do to
> keep
> from buying them, I coulda been a hero. But as it is, I'm just Steve, poor
> ol'
> guy workin' at a start-up...(GRIN)
>
> -Steve Gregory-
>
> ################################################################
> 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
> ################################################################
>
################################################################
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
################################################################
|
