By James Mullins, Mullins Professional Services
DIS Technical Director
Fortunately for ductile iron producers and consumers, carbides are becoming an increasingly rare problem. This is mostly due to diligence on the part of most foundries, because ductile irons are much more prone to carbide formation than gray iron, even though the composition may appear to be similar. The treatment alloys that we use to convert gray to ductile iron are usually a big part of the problem.
Carbides will cause machinability complaints and may even increase the shrinkage tendency of an iron, since some of the carbon needed to form graphite which will produce part of the expansion needed is not available.
Treatment alloys contain varying amounts of magnesium and rare earths, both of which are quite strong carbide promoters. So it very necessary to control the final levels of these elements according to the cooling rate of the casting sections being produced. Magnesium and cerium are additive in creating nodules, so only looking at magnesium levels will only tell part of the story. Thin sections require much less magnesium and tolerate more cerium than heavy sections. Some foundries use too much of each and many foundries use more of one or another than they need. Doing this can lead to carbide problems, especially when cooling rates are high and inoculation additions may be marginal. See figures 1 and 2.
 Figure 1 Eutectic or chill carbides. Usually found in rapidly cooled areas such as corners, where fins attach to the casting, at chill locations, etc.
 Figure 2 Inverse chill carbides in the center of the casting section. Note the shape of the carbides - acicular or needle like.
The two most common carbides are eutectic or chill carbides and inverse chill carbides. Eutectic carbides (figure 1) are usually found on the casting surface and near fins and corners. They occur because of rapid cooling and insufficient inoculation to over come the undercooling created during treatment. Of course the carbon equivalent level plays a part if carbon and silicon contents are too low.
Of course other elements, such as chromium, molybdenum and titanium can also be responsible, but they often cause a different type of carbides to form and these are usually found in the grain boundary areas and are called segregation carbides. These are normally found in areas where the cooling rate is slow, but with high pouring temperatures even quite thin sections can have these, when carbide promoting element concentrations are too high. See figure 3.
 Figure 3 Segregation carbides. Found in heavier casting sections in grain boundaries. May have to use a higher magnification power (200 X) to see them clearly.
Eliminating the problem
The first step is careful monitoring of charge materials. It is very common these days for some alloyed steel scrap to creep into the scrap stream. More and more steel is made from recycled scrap as well and many of the carbide promoting elements are hard to oxidize out.
Chromium and molybdenum should have a maximum content of 0.05% in ferritic irons. Titanium should not exceed 0.025% as it has other consequences of promoting graphite deterioration as well. Even manganese can segregate when concentrations go above 0.40%. The problem is not with the concentration in the melt; it is the fact that these elements move in the solidifying iron and can cause concentrations in the intercellular regions to be up to 10 times higher, hence the carbide promotion capability.
The important base elements in controlling chilling tendency are carbon, silicon and sulfur. C and Si should be correct and maintained for the section size and grade of iron. The base S content should always be above 0.005% before treatment. Low S contents lead to poor nucleation and low nodule counts.
Magnesium is also a strong carbide-promoting element. Consistent magnesium treatment is important to controlling it to as low a level as required to produce a good nodular structure. Many, many producers use too much and with normal variation in the process, the upper end of the range may easily be too much.
Sufficient inoculation will most often cure many chill problems. Even segregation carbides can be reduced when the nodule count is increased to an appropriate level for the section size. Inoculation additions should be split to insure uniformity. Add some with treatment alloy when treating with Mg Fe Si, and add the balance upon transfer to a pouring ladle. Also inoculation fades, actually more quickly than the magnesium does. Nodularity and nodule counts are easily improved with a late inoculation step. For this reason in-stream and in-mold inoculation processes have gained considerable ground. Very few irons are over inoculated. Thin sections (fast cooling) require more inoculant than heavy sections.
Increasing nodule counts through inoculation and possibly using stronger inoculants, which fade less rapidly, will always be helpful. Some foundries add a very small amount of bismuth (0.01%) along with their normal cerium addition to increase nodule counts.
Finally, higher pouring temperatures are usually better in reducing carbide problems by slowing the cooling rate of a given section, even though the inoculation fade rate is increased. Gating into thin sections is useful as this increases the effective section size as the mold media is heated.
|
