Message-ID: <[log in to unmask]>
Date: Tue, 27 Jan 1998 21:36:42 +0000
From: Gabriela Bogdan <[log in to unmask]>
X-Mailer: Mozilla 4.01 [en] (Win95; I)
MIME-Version: 1.0
To: [log in to unmask], TechNet <"<mail list>"@ipc.org>
Subject: BGA VOIDS
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello, Bob!
Just a while ago we purchased your tape about BGA's.
Afterwards I read the requirements for voids in BGA's in Bellcore's last

revision.
The statement was:An average of 1%in volume,with a maximum of 5%in
volume for an individual ball.
At first,I thought that a 0 is missing.But this would be too much for
the 5%.
Then I x-rayed our first experimental boards from an angle and cross
sectioned them vertically.
The results of both tests showed that the voids are concentrated at the
interface between ball and component. And they are 10% average in the
cross section.I could try to use the computer's imaging system for
calculations,
but the last time I tried , I saw that it is very much influenced by my
settings.
The question is, how to be able to prove that the percentage is
accurate,
and how to improve the actual situation.Maybe,the requirement is for a
single void,not for the accumulate volume?
Also, if all the voids are at the interface,isn't it worse than if
distributed
in the whole ball?
How do in line X-ray machines without tomography cope with this?
Thank you,
Gaby