Following from the Electronics Manufacturing Productivity Center in 
Indianapolis, prepared by Bill Goers:

"The basic statement or premise of "If there is a difference between the 
board's coefficient of expansion and the part, the solder joint will crack" 
is incorrect.  The volume of solder in the joint determines the cracking 
resistance of the joint.  High temp solder joints act as "compliant lead".

Plastic BGAs have a very similar CTE to FR4 type boards.

The EMPF is involved with JPL in a BGA reliability study (plastic/ceramic 
and metal).

We will be happy to work with Hitachi to develop their BGA manufacturing 
processes. Please contact the HelpLine at 317.226.5616 or 
[log in to unmask]
 ----------
From: TechNet-request
To: technet
Cc: anderson_p
Subject: Protection of BGA Solder Joints
Date:  15 Feb 96 5:51PM

Return-Path: 
<[log in to unmask]>
Resent-Date: Thu, 15 Feb 1996 19:55:02 -0800
Old-Return-Path: <[log in to unmask]>
Mime-Version: 1.0
Date: Thu, 15 Feb 1996 17:51:42 -0600
Message-Id: <[log in to unmask]>
From: [log in to unmask] (ROGER HELD)
Subject: Protection of BGA Solder Joints
To: [log in to unmask]
Cc: [log in to unmask] (PHIL ANDERSON)
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part
Resent-Message-ID: <"faKtH2.0.JW7.5az8n"@ipc>
Resent-From: [log in to unmask]
X-Mailing-List: <[log in to unmask]> archive/latest/2692
X-Loop: [log in to unmask]
Precedence: list
Resent-Sender: [log in to unmask]
 ----------------------------------------------------------------------------  
 --
     At Hitachi, we are considering moving to BGA technology during our
     next model design phase.  We are not very familiar with the
     manufacturing process for this technology.  We have come up with a
     question we have not yet been able to resolve (as follows).

     Solder joints on QFP's resist cracking due to thermal expansion
     because the leads have some give due to a "spring" effect.  With BGA's
     there is no give between the board and the part except for the solder
     joints.  If there is a difference between the board's coefficient of
     expansion and the part, the solder joint will crack.

     1. Has anyone experienced this problem?  How do you resolve it?

     2. We are considering some sort of sealant material to protect the
        solder joints and provide some thermal relief.  Does anyone have
        any good ideas for materials?

     Anyone who can do a data dump on this for me will be greatly
     appreciated! Thanks in advance,

     Roger Held
     Hitachi Computer Products
     Norman, OK
     [log in to unmask]