Phil,

I agree with Jerry's info below. Are you using top side preheat? If so, what is 
your top side temp is just before your boards hit the wave? Is this too high?

Another thing I would look at is the height of the wave.

The best,

Charlie Barker
Input/Output, Inc.
Stafford, Texas 
http://www.i-o.com
[log in to unmask]

To: Hintpwb1 @ aol.com @ Internet
cc: TechNet @ ipc.org @ Internet (bcc: Charles Barker/IO-US)
From: jcupples @ iphase.com (Jerry Cupples) @ Internet
Date: 05/15/96 06:11:05 PM
Subject: Re: Assy: opens after 2nd solder

Phil Hinton asked:

>Soldering of mixed technology boards:
>   I have a problem in which the SMDs show random opens at the solder joint
>after the wave soldering of the components that go into the holes.

The first question in my mind is: were they ever soldered at reflow?
Your wording seems to imply that they were, and that somehow wave solder
opened the solder connection. My guess would be that the joint was never
formed at reflow, or formed very poorly.

>   The SMD solder connections that are most likely to have opens are those
>which have a short heat path thru the via holes to a land that contacts the
>solder wave.

And this I would attribute to insufficient heating of the joint during
reflow due to the "sinking" effect of the metal connection to the via and
VCC/ground internal planes, rather than to heating during wavesolder.


>   Suggestions have been:
>      (1)  Tent or plug all of the holes that contact the wave except
>             those that have lead in the hole.

No objection to this, but it may cost you a buck or so per board. Assuming
you have some control over design of the PWB, I'd say make sure that there
is a good thermal relief pad design used for all plane connections to the
vias.

>       (2)  Use a nitrogen blanket during wave soldering

IMO, useless unless you are using no-clean flux, and I can't see how it
would help the problem you describe. You don't want lower surface tension
of the topside connections, (assuming that they are getting enough heat to
melt), you would probably see less solder pullback with higher liquid
solder surface tension.

>       (3) Flood the top side of the board with flux during wave
>                        soldering so that any connection that is reflowed
>does              not oxidize and will form a new solder joint.

See above.

>       (4) After wave soldering, flood the top side of the board with
>                    flux and reflow it.

A longshot. If solder has wet the lead and the pad, surface tension will be
working to keep the metal apart. Flux might "bridge", and help to form a
connection between the pad and lead.

>Is there any other solutions,  the nitrogen blanket is a long term fix for
>the assembly shop but not a present fix.

Two places I've worked have installed nitrogen for their reflow furnaces
only to shut it off after seeing minimal improvements.

If I were you, I'd be pretty certain whether the joint formation was good
at reflow, then failing at the subsequent wavesolder, or whether you aren't
getting a joint during the first reflow.

If you are reasonably sure the SMT joints are "opening" during wavesolder,
then I'd say speed up the conveyor on the wavesolder, crank down the
topside preheat, of narrow the wave contact; any approach so as to transfer
less heat to the board and avoid melting the topside joints during
wavesolder. And if you don't have a good thermal relief pad on the via to
internal plane connections, this could again cause more heat transmission
to your pads than you'd like.

cheers,

Jerry Cupples
Interphase Corporation
Dallas, TX USA (how 'bout them Rangers?)
http://www.iphase.com