TECHNET Archives

March 1998

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
SteveZeva <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Fri, 20 Mar 1998 18:24:10 EST
Content-Type:
text/plain
Parts/Attachments:
text/plain (72 lines)
Jim,

     In the past I have had some problems with the solder joints on QFP's when
the temps start getting that high. It was really baffling at the time that we
first noticed something was wrong. What was occuring was that these boards
were failing when they went into ICT. Up until that point everything looked
fine, and we looked at them too...we had a pretty decent SPC program going at
the time. We looked closely at the joints that ICT caught (they always were on
QFP's), and there indeed were open joints.

     But what was weird with these opens, was that the fillet was completely
and cleanly separated from the pad. It was still attached to the foot of the
lead, and the shape of the fillet indicated that there was good wetting taking
place during initial reflow on both the lead and the pad...you could tell that
by the angles of the fillet that had formed on the sides, heel, and toe.

     So then we started looking VERY, VERY closely after SMT reflow and no
problems were seen, and we did the same after wave solder, again no problems.
But test was still getting boards that were failing for the exactly the same
thing. I went back and double-checked my reflow profile, and dragged a mole
behind the board while it was going through wave. I had the thermocouples at
one of the QFP's that was failing. I saw just what you've been seeing, temps
up there pretty high...not liquidous yet, but not too far away from it...maybe
between 165-175.

     When the boards came out of the wave, I looked at the joints again, they
looked great...about that time my ulcers started getting a little bothersome.
In the meantime, test was still failing boards for the same thing...they even
failed one that I had hooked the mole to and looked at after it came out of
wave. I knew that was one of the boards I looked at because I had marked the
edge of the board with a "sharpie"...my ulcers were bleeding big time now.

     I was down in test drinking some Malox after looking at one of the latest
failures, and I noticed something that the test tech was doing. After he would
place a board in the fixture, he would take a sheet of conductive rubber and
drape it over the assembly, and then start the test. (The sheet of rubber is a
common thing I've seen used with a single-sided fixture to help vacuum the
board down to the pins...) But sometimes he wouldn't get contact on all the
pins, and then I saw him pushing all over on top of the board to help make pin
contact...THAT WAS IT! He was flexing the board and that's what was causing
the fillets to separate from the pads! But I still had a problem, we
discovered by tracking each failure that it was only happening on a few
locations...there were other QFP's on the board that didn't have one single
failure...so what's up with that? I was wondering...

     Turns out that the QFP's that had been failing were the ones that had
via's right at the ends of each pad...there may have been a little bit of
trace between the via and the pad, but not much. Heat transfer was causing the
joint to start to get in the plastic range, and somehow weaken the bond at the
lead and pad interface...not enough to be able to see anything, but enough to
cause it to easily separate if there was any stress put on it at all. What we
wound-up doing to fix things was to use temporary mask to cover the via's
until they could redo the gerbers to tent them on the bottom...which by the
way, is the best side to tent vias on, don't tent the topside!!

                                      Sorry this was so long...I can go on
sometimes, can't I?

                                                          -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 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
################################################################


ATOM RSS1 RSS2