Have you considered using tented vias (soldermask goes right over vias on
both sides of board)?  This will save you solder.  The seal may or may not
be adequate depending on the soldermask and the via size.  I have had good
luck with 0.015" vias, SMOBC with LPI mask (though I didn't specify a
material part number).

Seth Goodman
Goodman & Associates
7687 Leta Way
Verona, WI 53593
Tel (608)833-7687
FAX (608)827-2631

-----Original Message-----
From: TechNet [mailto:[log in to unmask]]On Behalf Of Greg Parke
Sent: Friday, November 20, 1998 10:54 AM
To: [log in to unmask]
Subject: [TN] Screenprint problems


Hi All,

I recently switched screenprinting processes from an semi-auto machine to a
full
auto machine.  upon changing over we began to see an increase in
shorting/bridging for fine pitch components such as QFP's, SSOP's, etc.
Upon
further investigation it was found that the problem was confined to two
sided
boards.  Our test process dictates that vias are filled in order to pull the
required vacuum during test .  We normally paste the vias on the circuit
side
mainly due to fact that many of our boards have BGA's and we can't paste
them on
the component side due to the obvious concern with shorting.  For one sided
BGA
boards we have the extra process step of pasting via on the circuit side.
Two
things are occuring which are contributing to the problem:

Increased Volume - It seems that the switch to the automated printer has
resulted in a higher volume of solder being deposited on the board.  The
increase is approximately 15-20%.

 Poor Gasketing - When the circuit side gets reflowed the result is a "dome"
on
the component side of the board which results in a poor gasket during the
print
stroke which in itself can lead to shorting.  When mixed with the increased
volume being put down to begin with you end up with a mess.

 The big question in my mind is the increased volume necessarily wrong?
After
all, the old process could have been depositing marginal paste volumes. I am
working to put together a multivariate experiment to minimize the variation
between the actual volume of paste deposited and the expected amount of
paste
based on the aperture size, stencil thickness, and some release efficiency.

Now that you're all asking yourselves "just what the heck I'm talking
about?", I
have a few questions. Has anyone out there had a similar issue with these so
called "domes"?  How did you solve the problem?  Has anyone gone through any
experimentation similar to what is outlined above or do you feel that it is
overkill?  If you would like to discuss this off-line feel free to call me.

Thanks for the help.

Greg Parke
Teradyne, Inc.
(617) 422-3192

################################################################
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
################################################################