This preview shows page 1. Sign up to view the full content.
Unformatted text preview: The fluid surface conveys much
more information to the viewer than do the particles, in
particular depth perception and variations in the surfaces
are much easier to identify. In this case, it can be seen that
the fluid separates from the front wall of the cylinder as it
enters the runner producing a thinning sheet of fluid that
flows into the gate. At the same time it spreads sideways
and flows up the side walls of the runner to the top
surface, giving a U-shaped profile in a cross section
through the runner parallel to the gate. Figure 1: Cross section of the die.
Concentrating on the flow after 28 ms when the horizontal
section of the die cavity is beginning to be filled, the two
views show the highly fragmented nature of the fluid and
the three dimensionality of the fluid flow pattern. These
features continue as the die cavity is increasingly filled,
with many voids in the fluid. These voids move in an
irregular and complex pattern and would lead to
significant porosity in the final cast component.
Die Cast Object
The geometry of the die, the gate, the runner and the
cylindrical plunger section is shown in Figure 3. In real
die casting, the cylinder orientation would be horizontal
and the part vertical but for the modelling we use a
reference orientation as shown here. The fluid initially
fills the cylindrical column and is pushed downward by a
plunger at the top of the fluid moving at 15 m/s. In the
simulation, a resolution of 1 particle per millimetre was
used giving a total of 243,576 particles. Figure 5: Liquid metal surface (coloured by fluid
velocity) as the fluid passes through the gate visualised
using a surface mesh calculated from the SPH particles. Figure 3: A 3D mould with cylindrical plunger on the left
leading to a divergent runner, through a curved gate into
the die of a 3D machine component. Using another visualisation option, the particles are
mapped to a rectangular grid which is then used to display
the fluid surface. Figure 6 shows four views of the flow
as die cavity fills. Note the introduction of the staircase
artefacts in the surface. These are produced by the
visualisation grid and are not present in the SPH data.
This indicates that mesh based descriptions of the surface
better represent the fluid than interpolation to an
underlying grid. Two perspective views of the filling pattern at 3 different
times are shown in Figure 4. The fluid particles are shown
coloured by their speed. The first frame at 1 ms shows the
system just as the fluid enters the runner. The second
frame at 2.8 ms shows the fluid having mostly filled the
runner and the leading fluid splashing through the gate.
This leading material consists of fast moving fragments
and droplets generated by splashing when the leading fluid
flowed around the right angle turn as it entered the runner.
In the final frame (4.52 ms) the fluid has entered the die
proper and has split into separate jets around each of the 439 CONCLUSION
In this paper, we have described the SPH method and the
application of SPH to simulate the 3D die filling in high
pressure die casting. The met...
View Full Document
- Fall '13