Browse Prior Art Database

Precision No-Clean Fluxing Of Flip Chip Devices

IP.com Disclosure Number: IPCOM000116754D
Original Publication Date: 1995-Nov-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 166K

Publishing Venue

IBM

Related People

Slesinger, KA: AUTHOR

Abstract

Many circuit board designers have begun using Flip-Chip Attach (FCA) technology to optimize performance versus available space. Demand for such products is growing rapidly. A manufacturing line has been installed for no-clean assembly of FCA devices onto organic carriers (MCM-L, PCMCIA, and various sizes of FR-4 circuit boards).

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 33% of the total text.

Precision No-Clean Fluxing Of Flip Chip Devices

      Many circuit board designers have begun using Flip-Chip Attach
(FCA) technology to optimize performance versus available space.
Demand for such products is growing rapidly.  A manufacturing line
has been installed for no-clean assembly of FCA devices onto organic
carriers (MCM-L, PCMCIA, and various sizes of FR-4 circuit boards).

      Though they provide a key advantage over other fluxes, no-clean
flux has several negative properties with respect to FCA.  The tiny
size of most FCA chips requires precise control of the volume of flux
applied to each device.  This article discusses some of the
challenges
of implementing a repeatable and economical no-clean FCA fluxing
process.  It also describes the implementation of one such method.

      Due the small size of the chips and the behavior of the flux
itself, several significant challenges were associated with fluxing:

      Low Adhesive Strength - Flux is applied to chips and/or boards
prior to chip placement.  Its tackiness holds placed chips in
position on the carrier until solder reflow is performed.  This
disclosure provides excellent solderability over a wide range of
reflow process parameters.  Unfortunately, it provides little
adhesive strength.  Chips must be placed while flux is still wet.
Due to the small volumes of flux involved, dry out occurs within a
few seconds of application effectively requiring flux dispensing and
chip placement to be integrated within the same workcell.

      Machine Vision - A high-accuracy, twin-camera placement robot
is required to meet accuracy requirements.  An up-facing camera
locates solder bumps on chips and a down-facing one locates
corresponding pads, fiducials or other repeatable features on the
board.  Chips are then placed at the "best fit" location computed
from the vision results.

      Complicating this scenario is the fact that the presence of
flux prevents reliable imaging by the vision system.  Wet flux
distorts feature shapes and dimensions and dried flux reduces the
light/dark contrast required between features and background.

      Hence, imaging must be performed prior to fluxing.  But the
flux application method cannot physically contact the chip after
vision.  To do so could shift its known location and produce
inaccurate placement.

      Volume Control - IBM* engineering specifications allow rather
large flux volume tolerances:  50% nominal amount per unit of
chip footprint area.  However, the tiny size of most chips results in
flux volumes in units of nanoliters.  Achieving volume repeatability
in such volume ranges prohibited the use of most commercially
available dispensing systems.

The solution to this follows.

      Equipment - An Ultra One cartesian robot was selected for chip
placement due to the high-accuracy positioning and associated machine
vision capabilities required for FCA.  System integration was
perform...