Underfill Misalignment Predictor (Download
.zip
or .tar.gz
This code is designed for cases where one wants to determine the upper and
lower pad displacements across a flip-chip geometry solder joint (standoff
height and misalignment) given known normal and shear forces on the joint.
This is particularly relevant to study of misalignment for a flip-chip that
supported only by the solder joint(s), such as during reflow of a flip-chip
on a board. Those interested a code where the standoff height and
misalignment are considered known and the equilibrium normal and shear
forces are calculated should follow this
link.
This page is a description of a down-loadable Surface Evolver code including
.fe files and .cmd files. The code was developed to compare experimental
and simulated flip-chip displacements. The novelty of the code is the ability
to predict both normal and lateral displacements. Below are some images
of experimental cross sections and simulated flip-chip solder joints. Experimental
work was carried out with varying solder volume, pad sizes and weights
to obtain a range of vector misalignments. Fitting of simulation to experimental
results gave optimized values for the contact angle and surface tension
of the solder which were consistent with measured and literature values.
Disclaimer
Users of this code are requested to separately cite the paper describing
implementation of this code (D. Josell, W.E. Wallace, J.A. Warren, D. Wheeler
and A.C. Powell IV, "Misaligned flip-chip solder joints: prediction and
experimental determination of force-displacement curves", to appear in Journal of
Electronic Packaging) and this website address in publications containing
results from code based substantially on this code. Users are also asked
to note the website address (only) in presentations.
Experimental cross-sectioned joint
Simulated joint
The code includes a group of evolver files which predict the lateral
and normal
displacements given a vector force. In most cases this code will need
to be
modified to model specific problems. Most of the modifications with
respect to
geometry and initial conditions will occur at the beginning of the
file force.fe
Included in underfill.tar.gz are the following files:
|
readme.txt
bracketx.cmd fixonconstraint.cmd
forcez.cmd
iterate.cmd report.cmd
bracketz.cmd force.fe
goon1.cmd
minimumz.cmd pev
evolve1-md.cmd forceb.fe
goon2.cmd
onpad.cmd
evolve1.cmd forcex.cmd
iter.cmd
param.cmd
To run an example test case type:
evolver force.fe
and then at the evolver command prompt type:
evolve
This will find the displacements for a particular vector force.
To see the results use the following commands:
printf "%g %g %g %g %g %g %g %g %g %g\n",facet_count,hgt,total_energy,xoff,ac_xoff,tilt,prop_off_pad_2,prop_off_pad_1,gamma,theta_out1
| "cat >> tilt.dat";
printf "{%g,%g},",tilt,xoff | "cat >> m.dat";
Look in tilt.dat and you will see the x-offset (xoff) and stand-off
height (hgt) for a given tilt angle (tilt).
The vector force is given from the equations on lines 29&30 in
force.fe.
Also included is a shell script (pev) for running sequential evolver
processes. This can be used to
plot force displacement curves.
If you have any problems using this code please email me E-mail:Daniel.Wheeler@nist.gov
Download the code as a .zip
or as a .tar.gz
|