|
|
THE EFFECT OF METALLIC CHARGE |
The Effect of Metallic
Charge/Melt History on Nucleation Potential in Ductile
Cast Irons
A paper
presented at the Ductile Iron Society meeting,
June 1998.
By James D. Mullins & Eugene C. Muratore;
Rio Tinto Iron & Titanium, Inc.
The melt history, including the type of charge material,
chemistry and molten metal processing of a cast iron melt
has a pronounced and measurable effect on the final
structure and properties of the castings poured. The
assessment of the nucleation potential of a graphitic
cast iron before treatment and/or inoculation has been
practiced at some level or degree for a long time.
CHILL WEDGES
See Figure1
Chill wedge testing of base
cupola iron gave the operating foundryman a qualitative
measurement of the graphitizing (nucleating) potential of
that iron. See figure 1. An iron that has a large chill
value (tendency to form carbides upon solidification)
means that it possesses a low nucleating potential. So by
measuring the width of the chilled (carbidic) portion of
the wedge, changes to the charge amount or type of
material could be made. Pouring another wedge after the
inoculation step could assess the effectiveness of the
inoculant. Before the advent of affordable and timely
chemical analysis, and certainly before the development
of computerized cooling curve analysis, the dependency
upon wedge testing for the assessment of suitability of
an iron for pouring was mandatory. Melters soon realized
that changes to the charge and/or thermodynamic changes
within the cupola manifested dramatic changes in the
chill wedge values of cast iron.
For example, when pig iron was introduced into the
charge, the chill value typically decreased. As the steel
portion of the charge was increased, the chill value
increased. As the melt conditions moved to more oxidizing
conditions, the chill value increased. Cupola well depths
and iron dam heights were carefully measured and
controlled in order to maximize the nucleating effect of
the coke. All of these changes to the nucleation
potential were seen even though the chemistry most often
did not change.
With the increasing popularity of induction melting
furnaces as primary melters, the utilization of the wedge
test has fallen out of favor. Since the chemical analysis
could be much more closely controlled, it was incorrectly
assumed that the nucleation potential was also being more
closely controlled.
DUCTILE IRON
FINAL WEDGE
See
Figure 2
Ductile Iron foundrymen
oftentimes saw nothing other than white iron (100%
carbidic) fractures in their base iron wedges, typically
because of the lower content of silicon and sulfur, and
also in the final wedges. So they too abandoned the use
of chill wedge testing for evaluation of the nucleation
potential.
There are a number of factors that affect the nucleation
potential and metallurgical quality of cast irons. They
are: the metallic charge, the type of melting equipment
employed, melting and holding temperatures, dwell time
(holding time), chemical composition, and inoculation.
Each of these factors will now be explored further.
As I mentioned earlier with cupola melted irons, the
metallic components of the charge exert a large effect on
the nucleation potential of the melt. The reason for this
effect is the steel component of the charge contributes
very little in the way of nuclei for the growth of
graphite. Likewise, the Ductile Iron returns portion of
the charge, being quite deoxidized during treatment and
inoculation, also contribute little nucleation.
EFFECT OF REMELTING
See
Figure 3
As an example, when returns
are repeatedly remelted, even just two times, the
solidifying iron can become all carbidic. To reduce this
effect and renucleate, additions of some pig irons,
graphite, silicon carbide, and other ferrosilicon alloys
are made to liquid melts.
In order to produce a cast iron melt that responds well
to inoculation and exhibits the lowest potential for
carbide formation during solidification, the returns
should be limited to no more than 50%, the steel
component should be limited to 40% maximum, and
consideration should be given to utilizing some pig iron
in the charge. Figure 4. Effect of rusty scrap on chill
in induction melted 4.1% C.E. gray cast iron with no
inoculation
EFFECT OF RUST ON CHILL VALUE
See
Figure 4
The cleanliness of the
charge material also plays a role in determining the
chill value. If the charge material is heavily oxidized,
the resulting iron will exhibit a much higher chill
value. We have seen the opposite to also be a problem.
Several foundries have shot blasted all of their charge
materials to remove rust and sand. They found very high
chilling tendency in this iron and as a result more
shrinkage defects. So having a small amount of oxygen in
the base melt is necessary.
The type of melting equipment can play a role in iron
nucleation. Iron melted in a cupola is conditioned by the
nucleating effect of the intimate contact between molten
iron and the coke in the cupola well and a relatively
short time at high melt temperatures. Cupola melted irons
usually exhibit a lower chill value and generally require
less inoculation in order to produce carbide free
microstructures. Further the presence of adequate oxides
and sulfides as nucleation sites renders cupola melted
iron as one with a high metallurgical quality.
As more experience was gained throughout the 1950's and
1960's with melting gray irons and ductile base irons in
induction and arc furnaces, note was made that these
irons exhibited higher chill and more shrinkage tendency
even while having identical chemical compositions as
cupola melted irons. The reasons for this are several: In
electric furnaces there is no coke contact as well as
more stagnant bath conditions, higher melting
temperatures are used to dissolve carbon, and longer
holding times and often times there are lower oxide
contents. This leads to higher base iron chill (low
nucleation values) . For these reasons, electric furnace
melted irons generally require different charge ratios
and additional amounts and often times more potent
inoculants.
I have already mentioned something about temperature, but
there is more. In the case of cast iron melting in
electric arc furnaces, the temperatures attained near the
arc tip may exceed 5000oF. Irons thusly treated are
called "fried" irons, because all the nuclei
have been cooked out, leaving an iron that will not have
a low chill value.
EFFECT OF SUPERHEATING
See
Figure 5
As the temperature of any
melt is increased above the normal melt temperature (high
superheat), the nucleation is reduced. This loss of
nucleation or reduction in metallurgical quality is
manifested with virtually no change in chemical analysis.
The measured chill depth may change from an acceptable
level to all white wedge over a 200oF temperature range
or less. This reduction in metallurgical quality requires
the use of greater amounts of inoculant(s) in order to
produce acceptable final microstructures. It may not be
possible to correct this iron. It is therefore advisable
to melt and hold iron at as low temperatures as
practical.
The effect of long dwell or holding times on the
nucleation potential of cast irons is similar to the
effect of high melting (superheating) temperature. The
longer the hold times, at any temperature, the greater
the loss of nucleation. The higher the temperature during
this holding period, the worse is the loss. The most
prevalent instance of this phenomenon is known in the
trade as "Monday morning iron". It has long
been recognized that irons held over a weekend exhibit
very different solidification behavior than normal. These
irons exhibit higher shrinkage tendency and have more
carbide due to this loss of nucleation. Irons that have
not been renucleated by the addition of "fresh"
iron or nucleating agents exhibit a much higher chill
level. This issue is so important, that the AFS Molten
Metal Processing Committee has begun a research project
to show foundries this holding effect on iron
properties/defects and what can be done to reduce or
eliminate this problem.
The chemical composition can alter the nucleating
(graphitizing) tendency of cast irons to a certain
extent. As the carbon equivalent is lowered the tendency
to solidify with a more carbidic microstructure
increases. As the level of carbide stabilizing elements
is increased, the same effect is seen. Even at the same
carbon equivalent and residual element levels, changes in
carbon/silicon ratio can alter the metallurgical quality
and physical properties. Generally speaking increasing
carbon content reduces shrinkage tendency in cast irons
and increasing silicon content reduces carbide formation,
but these effects are lost due to the loss of nucleation.
Ductile Irons treated with magnesium ferrosilicon alloys
often begin as base irons of very low silicon (often
times less than 1.2%) content. These low silicon base
irons may exhibit an all white chill unless an adequately
large wedge test core is used. Of course, a low
nucleation level may also contribute to an all white
chill value. So using the correct size wedge (see ASTM
A367) is important, as is a good sampling procedure in
order to achieve the correct result. Done properly, the
chill test can be very helpful to assess ductile iron
base metal to see that is has been well processed and has
a low chill value.
MAGNESIUM vs. MODULUS
In magnesium treated irons,
high magnesium content acts to promote carbidic
microstructures and increase shrinkage. The magnesium
level must be controlled carefully to the cooling rate of
the casting to avoid increased chilling tendency. This
cooling rate is described as the modulus, which is a
ratio of casting volume to cooling surface area. Thus
modulus is a more accurate way to describe the cooling of
a casting section than just measuring the section(s)
size. See
figure 6.
Of course, all of the carbide stabilizing elements should
be kept to relatively low levels to minimize their effect
on chill (carbide) promotion. Doing this will then allow
more of the available carbon to transform to graphite.
Many foundries have reinstituted melt assessment through
chill wedge testing and /or thermal (cooling curve)
analysis programs because they are simple and
inexpensive. The wedge test can be used to verify the
results of the cooling curve.
Magnesium concentration effect on shrinkage
Inoculation is the final and the most important step in
molten metal processing. Although not all of the problems
addressed above can be compensated for with inoculants,
several facts stand out. Foundries that pour thin-section
castings, tapped at elevated temperatures, may be able to
produce acceptable castings with very good inoculation.
Without it, this would not be possible. The correct use
of inoculants and preconditioning agents can also allow
for the utilization of irons held over weekends and
holiday periods, if the iron has not deteriorated badly.
Despite the rigid control of residual elements in many
foundries, some percentage of deleterious elements is
usually always present. The employment of adequate
amounts of and effective inoculants enables the seasoned
foundrymen to produce acceptable castings from these
irons.
ELEMENT SEGREGATION TENDENCY
The production of heavy
section castings also requires adequate nucleation and
inoculation in order to shorten the intercellular spacing
so that strong segregation of carbide stabilizing
elements is avoided. Even at low concentration levels,
these elements are known to segregate to the last to
freeze areas and contribute to grain boundary carbides
and deteriorate the mechanical properties, as well as
machinability.
When we look at the tendency to segregate; elements with
numbers greater than 1 tend to segregate into the
intercellular regions and those elements with numbers
less than 1 tend to increase their concentration around
the graphite nodules. As an example, from the slide, the
molybdenum concentration can be up to 25 times more in
the intercellular region than it is in the rest of the
iron. Conversely the concentration of copper around the
nodule will be higher than the concentration of silicon
and neither one will have much of a presence in the
intercellular regions See figure 7.
Element Segregation
Factor
Mo..........................25.3
Ti..........................25.0
V..........................13.2
Cr..........................11.6
Mn..........................1.7
- 3.5
P..........................2.0
Si..........................0.7
Co..........................0.4
Ni..........................0.3
Cu..........................0.1
Supporting Work
The metallurgical quality or nucleation state of the iron
has been studied and published by many authors. Vern
Patterson who wrote Foote Foundry Facts - devoted several
issues to the importance of measuring chill wedge values
and the effects of processing variables on the nucleation
level of cast irons. The benefits of an established and
practiced wedge control program are a recurrent theme
throughout the issues.

See
Figure 8. Preconditioning Effect on BHN Hardness

See
Figure 9. Preconditioning Effect of Elongation

SeeFigure
10.
B.C. Godsell, in his AFS Transactions paper,
"Preconditioning of Ductile Iron" describes one
foundry's method to adjust the base nucleation state of
ductile base iron before treatment. Before utilizing a
preconditioning program, the foundry was unable to
produce castings to an acceptable hardness or elongation
range. After the institution of a preconditioning
program, which normalized the nucleation state of the
iron before treatment, ductile iron castings could be
produced as cast with properties consistent to those of
heat-treated castings.
J.M. Frost and D.M.
Stefanescu in their paper "Melt Quality Assessment
of SG Iron Through Computer Aided Cooling Curve
Analysis" ran a designed experiment where it was
shown that several processing variables had pronounced
effects on nodule count and chill depth.
As the percentage of pig iron is increased and the
superheat and pouring temperatures are decreased the
nodule count is increased.
As the superheat time or temperature is decreased, the
nodule count is increased and the chill depth decreased. See
figure 11.
Further, decreasing
superheat temperature and increasing pig iron content had
the effect of reducing the chill depth, while reducing
the pouring temperature had little effect. See
figure 12.
In conclusion, the metallic charge and melting history of
cast iron melts have a significant effect on the final
metallurgical structures obtained. These structures
affect mechanical properties, shrinkage behavior and
machinability in these castings. A base iron that has a
low chill value or is preconditioned to have a high
nucleation state will tend to have less magnesium and
inoculation fading. This usually means that shrinkage
problems will be reduced. Assessment of this nucleation
condition is important to producing consistently high
quality castings.?
References
Ductile Home Officers & Directors Back Issues Contact Us Legal