94

REDUCING OXIDATION PROBLEMS IN MEDIUM

Early coreless furnaces (main or line frequency The thinnest steel scrap may go from room
furnaces - run as heel melters) usually had just two temperature to glowing cherry red color within just one
major sources of oxidation problems: rusty steel scrap or two minutes in these high power furnaces
and overheating of scrap in a preheater. Buying a increasing the amount of slag formed. All of this
cleaner scrap and/or keeping a dust removal system oxidation produces a highly reactive FeO slag.
in good working order can minimize the rusty scrap
problem. Overheating in the preheater can be reduced It is no wonder that so many foundries have reported
or eliminated by reducing the time and/or temperature, increased oxidation problems since switching to
using a «jog» system to move the scrap, adjusting the medium frequency melters. The answer is not to go
burner profile, or using heavier/thicker scrap. In either back to line (main) frequency furnaces, since the
of these conditions, the main source of oxides came newer medium frequency furnaces do indeed have a
from somewhere OUTSIDE the furnace. number of installation and operational advantages.
The answer is to find ways around the oxidation
In today’s, high power density, medium frequency problem to make these new furnaces work even
furnaces the problem is just the opposite. When the better.
furnace is run as a batch melter, the primary source of
oxidation is INSIDE the furnace. 1. Explore the option of running as a heel melting
furnace rather than batch melt. Just because
When steel scrap is charged into a heel melt furnace, these furnaces can be run as batch melters,
it is drawn into the bath rather quickly, especially if doesn’t mean they have to run this way. This
there are pig and returns on top to push them down. accomplishes several things: Surface oxidation is
The steel then heats up in the presence of the high reduced, since the steel scrap now «dissolves»
carbon iron bath. As it heats up, there is some into a molten bath and power input levels are
diffusion of carbon into the steel, thus lowering the higher, since the «ramp up» normally associated
melting point and allowing it to melt at a lower with batch melting is eliminated. (In most batch
temperature in the bath. melters it usually takes 10-15 minutes before the
furnace is pulling maximum power (Figure 2)).
In batch melt, medium frequency furnaces, most of the
steel is suspended with air inside the furnace. As the In addition, alloy recovery, from carbon and
induction field raises the temperature of the scrap, it silicon carbide, should be higher and more
now must go all the way to the melting point of steel, consistent. Introduction of these materials in
because there is no carbon present to lower the batch melt operations is a problem. If added too
melting point. Hence, this requires more energy and early, the alloys tend to «fuse» to the bottom. If
time for the initial melting. In addition, once the steel added too late, there is more chance for them to
reaches a temperature of about 1300°F (700°C) the be lost to the slag or dust collector and they may
increase in oxidation becomes dramatic and during not have enough time and stirring to get into
solution. Since the larger batch melters are in the
the heat up from 1300° to about 2800°F (1540°C), 7-12 ton range, they are running large charges
the surface of the steel scrap continues to oxidize at a with very large alloy additions. It is not unusual to
higher and higher rate. See Figure 1. Once molten, see several hundred pounds of SiC and carbon
the droplets of steel will continue to oxidize as they fall being added in one big «lump» in these furnaces.
down the charge until they reach the bottom of the Running as a heel melter will automatically
furnace and join the hopefully higher carbon iron bath. reduce the amount of alloys being added at
The carbon in the bath stops the oxidation of the iron. anyone time.
 If these furnaces are run as heel melters, some In all cases, batch or heel melt, there are two
sort of scrap dryer/preheater will be needed in the steps that should be taken first to start with the
system. This represents additional cost of least amount of external sources of FeO. First,
installation and operation, but gas almost always always buy the cleanest, most dense steel scrap
represents the lowest cost energy up to about possible to eliminate rust and minimize the effects
1000°F (540°C) and the increased melt rate of preheating. The amount of surface area and
rust that thin scrap can present is very high.
should help to offset this additional cost.

2. If the furnace is run as a batch melter, there are We recommend the use of high purity iron and
still several steps that can be taken to reduce SiC in the charge. The use of some SiC in the
oxidation: charge part of the high carbon charge will also provide for reduction in the FeO
metallics (pig iron or remelt scrap) first. These content as well as reduce slag attack on the
will melt faster and at a low temperature and form refractory, reduce base iron chill and increase Mg
a high carbon «pool» (heel) in the bottom for the recovery in Ductile Iron. Adding high purity iron to
steel above to dissolve into. Then try to optimize the bottom of a furnace reduces oxide input by
the charge density. Thicker steel scrap will result creating a high carbon low melting point bath,
in less surface area, minimizing oxidation or reduces chill, produces more consistent and
increasing the amount of pig will certainly help. better chemistry, and will also eliminate the
This will also reduce total melt time, since the «elephant footprint» defect in the furnace bottom,
ramp up at the beginning of the melt should be which is caused by oxidation of the steel scrap
reduced. Finally, if necessary, it is possible to and the reaction of this slag with the refractory.
employ an inert gas shield on the top of the By following these guidelines, you will get longer
furnace to almost totally eliminate oxidation in the lining life, less total slag volume, fewer slag
furnace. This is fairly costly, but may still be worth defects in your castings, reduced kWh/ton making
while if casting scrap due to slag is high and/or for a higher melt rate, and (in Ductile Iron) should
refractory life is low. see more consistent chemistry and higher
magnesium recovery.
Thanks to Mr. Ken Copi from Exolon ESK, for allowing
us to use information from his paper, which was given
to the AFS Induction Melting Conference, January
1999.

REV – March 2006