Phil
Welcome to Board Talk. This is Phil Zarrow and Jim Hall, the
Assembly Brothers. Today we're joined by Joe Belmonte. Jim,
what's today's topic?
Jim
Phil, we get a lot of questions about soldering and solder paste for surface
mount reflow. There's a tremendous amount of development going on in paste technology. New materials to deal
with lead-free, head-and-pillow and so
forth.
Bottom line is it's important
that most everyone look at what's becoming available and be able to evaluate
and qualify a new paste. Joe is such an expert in this field, so Joe expand on the
importance and the general procedure for qualifying a new paste.
Joe
Here's my approach and my philosophy on qualifying a new solder paste.
First, I know that changing
solder paste is a very significant activity. Once you lock into a paste that works for you, and you feel comfortable with it, it's performing as you hope, it's giving you the yields you want.
However, solder paste, as Jim mentioned, is advancing all the time. There's new solder paste coming to the
market, so how do we periodically evaluate those new solder pastes?
When I first started in the industry, what we would do is
replicate all
the laboratory tests that were done by the world-class suppliers and do all those tests over again
in the
lab.
However, my philosophy now is buy into those test results. Believe in them
because they're done by
world-class, reputable organizations.
What we focus on is the performance tests. I'll call those those the
process tests. How will the new paste perform in
your process? What about printability,
shelf life, tack etc..
So my philosophy is to focus
on those tests and not replicate all the tests that are done by the paste manufacturer lab.
And the way to execute those tests is a
formal design of experiments.
Set up a
brainstorming session, put together the right people, including
the supplier, maybe a customer; whatever the right cast of folks is. Get in a room and we decide what
characteristics in your process are critical for that paste to work
for
you. Look at things like
tack time, shelf life, storage, solderability, so
forth.
And
then do a design of experiments (DOE). There is one DOE that's been done many times for paste. It was actually developed by Cookson
Electronics many years ago called a 27-board challenge.
It allows you to test several parameters with
a very small sample size. It's an
economical and practical test. It's been
in the industry for many years; used by many customers as a way to
identify on the shop floor, in your process floor, what paste is going to give you
the performance characteristics that you need.
Phil
I think the thing that Joe emphasizes is
that these tests are done on the shop floor, in the environment that the solder paste will be subjected to and that's something
that cannot be done generically.
Jim
We could go into infinite detail, but I think that's a really good,
basic, very heartfelt endorsement and recommendation from Joe.
Phil
Well you have just wasted five minutes with Jim, Phil and Joe, the Assembly Brothers, who by day,
go as ITM Consulting.
Jim
Whatever you
do, don't solder like these guys.
Phil
Yeah,
don't solder like these brothers.
Great topic, been there a few too many times. I agree with not re-performing every test and trusting your vendor testing. I would only add that IPC-TM-650 TEST METHODS MANUAL remains a good reference, and I continue to perform these test but as you mention, perform these in your production environment. Also I would add to pay close attention to board cleanliness between the paste/flux samples, when newer BTC type components are used. Cheers!
Jeff Hempton, BAE Systems, USA