Matthew Park wrote:
>
> Enza,
>
> I do not agree with your recommendation of overprinting as you
> described (either by wave or stencil apt/thickness increase) to
> relieve insufficient solder on pads with via holes. I had
> experience of trying your techniques years ago. All I got was
> insufficient solder and insufficient solder. I was wandering at
> that time where all excessive solder went. All solderpaste/
> solder filled prior to the reflow went thru black holes and never
> been found.
>
> Later I found some of evidences that these excessive solder
> landed on the bottom heat panel of reflow oven and a lot went
> thru heat holes in the heat panel.
>
> I agree some of responses using off-centered and smallest via
> holes in smt pads. This restricts some degree of molten solder
> going down thru black holes. It is better to have conductive epoxy
> filled via holes or tented via holes from the bottom side.
>
> re
> Matthew Park
> NII Norsat International Inc.
>
> Hello Geoff,
> My experience with "via in pad" has been this...during reflow,
> solder
> tends to scavenge into an unfilled via leaving an "insufficient" or
> starved solder joint. We had overcome this on older board
> designs by
> filling the PCB vias with solder over a wave solder pot to prepare
> the
> PCB for the SMT process. Now, I'm not very familiar with blind
> vias and
> such, but you might be able to compensate for the solder which
> will be
> lost down the hole by over-printing solder paste. Overprinting
> can be
> accomplished by selectively modifying the stencil apertures for
> the pads
> which contain the via so that you do not put excess solder all
> over the
> board (i.e., by using a thicker stencil.)
>
> That's my 2 cents worth.
>
> Enza
>
> ################################################################
> 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 web site (http://jefry.ipc.org/forum.htm) for
> additional information.
> For the technical support contact Dmitriy Sklyar at
> [log in to unmask] or 847-509-9700 ext.311
> ################################################################
>
> ################################################################
> 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 web site (http://jefry.ipc.org/forum.htm) for additional information.
> For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
> ################################################################
Dear Matthew,
I think whether the technique is wrong depends on the application. The
boards that we processed were mixed technology, multilayer with
cu/moly/cu cores for military electronics. We had many, many via in
pads...a rework nightmare after the SMT reflow process because all the
solder went down the hole (for LCC lands). Our solution at that time
was to fill the .012" vias with solder using a wave solder process. You
mentioned that your experience with pre-filled vias was that the solder
would go into a "black hole", never to be found. (Interesting that you
found solder in your reflow oven. I never had that happen.) I had done
an extensive study and had found that the quality of the via fill varied
and that in many cases, vias that appeared filled were merely capped
(i.e. voided inside). During SMT reflow the vias re-reflowed and any
solder from the pad was scavenged. Although the via fill process had
many issues, it did help us increase our yields out of reflow.
Eventually we were able to convince management that PCB redesigns were
the most cost effective solution and we were able to implement tented
vias. I was reading one response to this via in pad issue which
proposes tenting only one side...we couldn't do that because the
potential of flux entrapment in the vias could not be tolerated.
Studies had also been done on epoxy filled vias, but there were
reliability issues because of the extreme environment our products would
have to endure.
When I mentioned modifying the stencil aperture to deposit more solder
to compensate for what will be scavenged by the via in pad, my thinking
was that blind vias don't go all the way through the board (correct me
if I'm wrong), and the boards in question were not as thick as the ones
I had to work with...it could be a possible solution for some
applications.
I also had experience with vias in chip lands. A trick we used was to
modify the stencil aperture so that solder paste was not printed over or
too close to the via. (This appeared to allow the solder to first wet
itself to the component and through surface tension, prevent it from
going down the hole. Whatever the dynamics involved, it worked.
As I mentioned, a solution has to fit the application. I just proposed
ideas from my experience which might help trigger ideas for the
appropriate solution.
Regards,
Enza
################################################################
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 web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################
|
