The influence of filter type and gating system

design on the machinability of vertically parted

grey iron castings

The influence of filter type and gating system design on the machinability of vertically parted grey iron castings
must be located at the ingates otherwise the runner
Introduction system cannot completely fill especially during the early
stages of the pour.
This report describes the effects that well established
different filter types and running system designs have on Sprue cross-over
the machinability of vertically parted moulded castings. A cross-over describes a design in which the downsprue
Foundries with vertically parted moulding lines normally “crosses over” from the swing plate side to the ram plate
want to avoid applying filters because their use can, in side before the filter. The advantage is that the liquid
some cases, lead to extended cycle times. Consequently, metal does not impinge directly on the filter and a slag
the use of filters has been relatively modest. Recently, chamber can be located directly in front it. This is a
however, increasing quality demands has led to a technique that is routinely used on horizontally parted
renewed interest from foundries. moulds. Another advantage of this technique is that the
runner fills more quickly and reduces the likelihood of gas
This paper will deal with grey iron castings such as brake entrainment.
discs, brake drums and clutch plates. The results were
obtained from studies, theoretical expertise and foundry Inclined downsprue
field trials. The paper aims to demonstrate that A technique to reduce the velocity of the metal is to use
problematic issues regarding filter use on vertically inclined downsprues with an inclination of about 10 to
parted moulds can be overcome whilst simultaneously 15°. This technique will significantly reduce the velocity of
improving the properties of the castings. the liquid metal (figures 1 and 2).

Filter location options on vertically

parted moulds

If the pattern plate has two or more levels, there is often

no option other than to locate the filter in the upper part
of the mould. A filter in this position provides a
mechanism for slag retention, but much of the flow
modification benefits will be lost due to uncontrolled flow
of the metal after the filter.

If the filter is located in the bottom of the mould, the

increased ferrostatic pressure can result in higher metal
velocity through the filter. This increases the risk of slag
being forced through the filter and into the casting cavity.

Figure 1 Standard gating system

With these problems in mind, the design of pouring
systems for vertically parted moulds was revised. The
following modifications were found to be helpful in
ensuring that moulds fill in a non-turbulent manner.

Runner system design options for

vertically parted moulds including filters

Basics
For bottom gated mould cavities, the system choke
should be located in the downsprue after the filter. The
cross-sectional areas of the runners and ingates should
then be increased to reduce the velocity of the metal as it
enters the mould cavity. For side-gated moulds, the choke
can again be located in the downsprue. The ingate cross-
sectional area should then be increased to reduce the
Figure 2 Gating system with cross-over and inclined downsprue
velocity of the metal. If the casting is top gated, the choke

07
 The effects on the mould filling can be dramatic,
especially during the critical first few seconds when liquid
metal enters the mould cavity (figures 3 and 4).

Figure 5 Inclined ingates

Figure 3 The metal hits directly against the mould cod

Figure 6 Ingates with 45° angle

Machinability trials
To test some of the above theories, a machinability study
Figure 4 The metal enters the mould cavity at slower speed and was conducted in co-operation with one of the leading
does not hit the mould cod directly
European DISA Foundries and a US university.
Design of the ingates Machinability is related to many factors and the effects
Castings such as brake discs and drums are almost may vary from foundry to foundry, and as such, this study
exclusively bottom-gated and the behaviour of the metal can only provide a rough overview on the various factors
as it enters the mould cavity is critical. Back pressure that may have an influence.
will not be present at this point and the metal can enter
the mould cavity as a “fountain”. This can cause defects An unventilated brake disk was chosen for the trials. This
due to re-oxidation and sand erosion. To minimise this was done in order to avoid effects related to the
effect, the flow direction of the metal can be interrupted positioning of cores.
by the incorporation of an ingate with a 45º angle
(figures 5 and 6). Trial details
Trials were conducted using a pressed device with a hole
diameter of 1.5mm, extruded device in 300 csi, and 30ppi
SEDEX* filters. All filters were 50x50x15mm.

08
 The influence of filter type and gating system design on the machinability of vertically parted grey iron castings
The following pouring systems were tested. These were
designed to test the concepts discussed earlier.

1. Running system without a filter. Ingate controlled with

slag trap. The pattern layout for these trials was
provided by the foundry

2. The same system as 1, but without slag trap

3. Filter positioned in the upper part of the mould,

downsprue controlled (figure 7)
a) Pressed device with 1.5mm hole diameter
b) Extruded device 300 csi
Figure 9 Gating system with filter at the bottom of the mould
c) SEDEX 30 ppi filter and cross-over

4. Filter positioned at the bottom of the mould, All trials were conducted during a single morning to
downsprue controlled (figure 8) ensure conditions were as consistent as possible. With
a) Pressed device with 1.5mm hole diameter each layout approximately 100 castings were made and,
b) Extruded device 300 csi after finishing, the surface defects were counted. The
c) SEDEX 30 ppi filter average amount of surface defects on the castings was
then calculated.
5. Filter positioned at the bottom of the mould with
additional cross-over, downsprue controlled (figure 9) Table 1 shows the results:
a) Pressed device with 1.5mm hole diameter
System Pouring Pouring Number Surface Surface Defects Average
b) Extruded device 300 csi Temp Time of Defects Defects TOTAL Defects
c) SEDEX 30 ppi filter [°C] [s] Castings Side 1 Side 2 per
Casting
1 1410 10 80 278 178 456 5.7
2 1405 7.5 90 583 636 1219 13.5
3a 1410 6.0 86 215 303 518 6.0
3b 1417 5.4 72 177 245 422 5.9
3c 1408 5.4 92 285 341 626 6.8
4a 1403 5.0 79 56 49 105 1.3
4b 1402 5.2 85 33 24 57 0.7
4c 1411 5.4 82 22 26 48 0.6
5a 1406 5.6 83 46 79 125 1.5
5b 1404 5.6 85 19 11 30 0.4
5c 1404 5.2 85 13 10 23 0.3

Table 1

Figure 7 Gating system with filter in the

upper part of the mould Trial observations

❑ The redesigned gating (incorporating a filter)

significantly reduced the pouring time

❑ The design and the connection to the slag chamber in

trial 1, does not allow faster pouring times without
sustainable quality losses

❑ The position of the filter greatly influences the

average amount of surface defects per casting. This
could be a reason why many foundries do not notice
any scrap reduction when filters are placed in an
ineffective position. Additionally, the possibility of
Figure 8 Gating system with filter at the
bottom of the mould reducing pouring times is often not considered

❑ The best result was achieved with cross-over and the

SEDEX foam filter.

09
 After the evaluation, the machinability of The machinability index is defined as tool wear in inches (in.) for a
the castings was tested. Because of the certain amount of cut metal in cubic inches (in³).
considerable high costs involved only 4
systems out of 11 were machined. They The results were recorded and graphed as described in the graph below
were then compared to the castings made (figure 11).
by pouring systems without a filter and
slag trap. Figure 12 - 15 present the results of various gating and filter types.

For the machinability study metal tools

were used. Surface speed was constant at
1500 SFM (7.62m/s).

Machining procedure:
❑ The remains of the ingates were
removed and these areas were
machined

❑ The brake disc was then mounted on

the CNC machine and machined from
the inside to the outside as described
below:

- 2 slices of 0.5mm each were

machined off. These slices will
hereafter be referred to as "SKIN".
After each cut the wear of the
tools was measured and recorded Figure 11

- 4 slices of 0.5mm each were Comments on the graph:

machined off. These slices will • The x-axis shows the amount of cut metal in cubic inches
hereafter be referred to as "BULK.
Again, the wear of the tools was • The y-axis shows the tool wear in inches
again measured and recorded.
• As reference point a tool wear of 0.020 inches was set and in
This was repeated until a maximum tool comparison the amount of cut metal was identified
wear of 0.025 inch was attained for "SKIN"
and for "BULK". Three tools were applied • The dotted line shows the variation (standard deviation)
per batch (figure 10).
• The steeper the straight line the worse the machinability.

End

Start

Metal tool

Figure 10 CNC device with tool

10
 The influence of filter type and gating system design on the machinability of vertically parted grey iron castings
Skin Bulk
0.025 0.025

0.020 0.020

0.015 0.015

0.010 0.010

0.005 0.005

0.000 0.000
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Volume of Metal Removed (in3) Volume of Metal Removed (in3)

Figure 12 Gating systems without filter and slag trap

Skin Bulk
0.025 0.025

10,0
0.020 0.020

0.015 0.015

0.010 0.010

0.005 0.005

0.000 0.000
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Volume of Metal Removed (in3) Volume of Metal Removed (in3)

Figure 13 Gating system with pressed device (1.5mm hole diameter) and cross-over

Skin Bulk
0.025 0.025

0.020 0.020

0.015 0.015

0.010 0.010

0.005 0.005

0.000 0.000
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Volume of Metal Removed (in3) Volume of Metal Removed (in3)

Figure 14 Gating systems with extruded device (300csi) and cross-over

Skin Bulk
0.025 0.025

0.020 0.020

0.015 0.015

0.010 0.010

0.005 0.005

0.000 0.000
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Volume of Metal Removed (in3) Volume of Metal Removed (in3)

Figure 15 Gating system with SEDEX 50x50x15/30ppi and cross-over

11
 Discussion

❑ The castings without a filter and slag

trap have more surface defects. The
“SKIN” therefore exhibits a much
worse machinability than the “BULK”

❑ When extruded and pressed filters are

used, some improvements are
observed. “SKIN” machinability is
improved but the results are variable

❑ When SEDEX filters are used the

machinability of the castings are
significantly improved. The values for
the “SKIN” and the “BULK” are almost
Figure 16 Comparison of the amount of removed metal to a tool wear of 0.020in.
identical. This suggests that very few
surface inclusions are present.

Summary
In explaining the differences between
pressed, extruded and foam filters, it is
The graph clearly shows that when using SEDEX filters machinability is
thought that pressed and extruded filters
improved by 30% compared to castings produced without filters.
can quickly create a back pressure, that
will prevent dirt from the pouring system
The design of the gating systems and the positioning of the filter has a
from entering the mould cavity. However,
great influence on the casting result.
these filter types are not as effective as
foam filters when it comes to removing
Additional positive side effects when using filters can be: reduced
inclusions throughout the duration of the
pouring time, improved productivity and improved yield, since no slag
pour. This means that over the whole
trap is necessary. These points should not be neglected when calculating
pouring time foam filters will retain more
the overall costs, attributes such as these are commonly overlooked
inclusions than their pressed and extruded
when considering the use of filters.
counterparts.

The results for the machinability of SKIN References

and BULK were compared to the system
without a filter and slag trap. The results Casting Plant and Technology International, 04/2006
are presented in the following graph New possibilities for pouring systems with vertical mould partition
(figure 16). Andreas Baier and Günter Strauch FOSECO GmbH, Borken

12
All statements, information and data contained herein are published as a guide and although believed to be accurate and reliable
(having regard to the manufacturer's practical experience) neither the manufacturer, licensor, seller nor publisher represents or
warrants, expressly or impliedly:
(1.) their accuracy/reliability;
(2.) that the use of the products(s) will not infringe third party rights;
(3.) that no further safety measures are required to meet local legislation.
The seller is not authorised to make representations nor contract on behalf of the manufacturer / licensor. All sales by the
manufacturer / seller are based on their respective conditions of sale available on request.
*FOSECO, the logo, and SEDEX are Trade Marks of the Foseco Group of companies and used under licence.
Editorial policy is to highlight the latest Foseco products and technical developments. However, because of their newness, some
developments may not be immediately available in your area. Your local Foseco company or agent will be pleased to advise.