Jim,

I believe that Steve is on the correct path. We have experienced problems
that sound similar where there is a very clean fracture that occurs at the
pad surface. What I think happens is that due to the different thermal heat
transfer that is present when some pads have vias nearby and others don't,
the pads on the topside go through liquidous at different times. The hottest
pads melt and resolidify fine, but the ones that approach liquidous later
are now under increased stress since they are supporting all the mismatch
between board and part. As they get close to liquidous they loose strength
and the joint fractures right at the pad interface. It now loses any
additional heat transfer and never melts. These type fractures are very
difficult to pick up because the fracture is so clean that you can't easily
see it visually and a lot of times it will make contact during ICT and pass.
I am somewhat confused by your statement that some of the defects you are
seeing are the leads pulling out of the solder joint. This sounds different
than the fractures we experienced at the pad surface. Maybe there is a
coplanarity issue with the part and the local stress is enough to cause a
similar fracture at the lead / joint interface. Regardless the solution in a
perfect world is to redesign so that there is insufficient thermal transfer
to the topside to melt the solder joints (longer trace lengths to the vias).
In practice, masking off the bottomside vias with a liquid mask or kapton
tape will work. One word of caution is that some of the peelable soldermasks
can leave undesirable residues that could impact reliability in use. We have
tried to minimize the heat transfer through the vias by lowering our solder
pot temperature and changing our design rules to provide longer trace
lengths.

Thanks,
Robert Furrow
New Product Engineering
Lucent Technologies
978-960-3224    [log in to unmask]

> -----Original Message-----
> From: Guenter Grossmann [SMTP:[log in to unmask]]
> Sent: Friday, December 15, 2000 3:50 AM
> To:   [log in to unmask]
> Subject:      [TN] Antw: [TN] Solder Failure During SMT/Wave Solder
>
> Jim
>
> I can't tell you why this happens since I didn't see any of the parts but
> I believe Steve with his never ending story of experience in production
> could have given you the right input ( Well, to be honest I don't see the
> connection between a via consuming heat and the degradation of a solder
> joint that used to be OK. I mean the worst that can happen is that the
> Joint gets partially melted and solidifies again ) . Just two remarks:
>
> - Steve, what do you mean by via too close. Do you have a number how far a
> via should be away of the pad?
>
> - Jim, In my opinion it doesn't matter if the solder is heated up. Above
> 20 deg.C solder is in its plastic stage ( metallurgically speaking) and I
> don't know that it's loosing ist strength when recristallizing ( The metal
> is allready cristallized whenever it is solid ). The recristallization
> process takes place anyway whithin the first weeks of operation of the
> assembly and I think it would be very sad if all the leads could be lifted
> off easily by than.
>
> Have a great day
>
> Guenter
>
>
>
> Guenter Grossmann
> Swiss Federal Institute for Materials Testing and Research EMPA
> Centre for Reliability
> 8600 Duebendorf
> Switzerland
>
> Phone: xx41 1 823 4279
> Fax :      xx41 1823 4054
> mail:     [log in to unmask]
>
> --------------------------------------------------------------------------
> -------
> Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
> To unsubscribe, send a message to [log in to unmask] with following text in
> the BODY (NOT the subject field): SIGNOFF Technet
> Search previous postings at: www.ipc.org > On-Line Resources & Databases >
> E-mail Archives
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for
> additional
> information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
> ext.5315
> --------------------------------------------------------------------------
> -------

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------