Steel foundrymen aim for better castings through melt control.

Efforts to eliminate oxide macroinclusions have helped develop new methods of manipulating furnace and ladle conditions.

It is estimated that 20% of the direct cost of casting a steel component is incurred in the cleaning room. Much of that is spent in repairing oxide-related defects. That is why steel foundries have spent quite a bit of time and energy over the past several years pursuing "clean steel."

Clean steel is defined as the absence of oxide macroinclusions - oxide inclusions that can be seen with the naked eye [ILLUSTRATION FOR FIGURE 1 OMITTED]. There are other impurities, such as sulfide and nitrogen control, but oxide inclusions have been cited as the most problematic issue for steel foundries.

Besides cleaning room expense, macroinclusions cause machining problems. Clearly, macroinclusions pose one of the bigger quality challenges steel foundries face.

The Steel Founders' Society of America (SFSA) recognized the challenge in 1985 and, with cooperation from the U.S. Depts. of Commerce and Energy, initiated major research efforts to combat oxide macroinclusions. According to Malcolm Blair of SFSA, the aim of the Clean Steel Research Program is "characterizing macroinclusions, how they are formed and what steps can be taken to mitigate their occurrence."

With the program in its third phase, the steel casting industry has made a three-pronged attack on macroinclusions, looking at the areas of pouring/metal delivery, filtration and melt control. Of these, melt control is the most widely practiced form of clean steel technology, and has shown the most substantial results.

Clean Heat/Dirty Heat

Steel cleanliness can be measured in the number of inches of oxides that need to be removed in the cleaning room or in the frequency of weld repair as indicated by the use of weld rod per ton of castings. In studying this data, SFSA has noted that macroinclusion occurrence varies far more on a heat-to-heat basis than on a casting-to-casting basis. That is, from the first poured to the last poured, castings of the same heat tend to have a similar incidence of macroinclusions, whereas, that number varies greatly from one heat to the next. "For whatever reason, 20-25% of our heats qualify as dirty," said Raymond Monroe of SFSA.

There are two important impurity sources that influence cleanliness. One is the entrapment of slags, refractories and molding sand, which can be minimized through careful adjustment of foundry practices. Studies have shown, however, that 83% of inclusions are the result of reoxidation - the reaction of deoxidizers in the steel with oxygen. It is standard practice for foundries to deoxidize the melt with the addition of aluminum. Deoxidation particles generally measure about 5 microns. Particles of 10+ microns are termed macroinclusions.

Reoxidation is a local condition within the melt in which reintroduced oxygen combines with and swells the aluminum particles to form large inclusions. It can be caused by exposure of the steel to air, poor refractories, slags high in FeO, or mold binders with a high oxygen content.

"What we have to remember," said Monroe, "is that we have to control all sources of oxygen that come in contact with the melt, including refractories. If another source, like air entrainment during pouring, was the dominant cause, there would be greater in-heat variation in inclusion occurrence."

Calcium Wire Injection

Since the mid-1980s, many steel foundries have employed some form of calcium (Ca) treatment to help minimize reoxidation - a concept borrowed from steel mills. Ca is the strongest deoxidizer used in steel making. When added to steel, it chemically combines with oxygen in solution and in nonmetallic inclusions within liquid steel. According to Nayyar Iqbal, Texas Steel Co., Fort Worth, Texas, Ca allows control of the "composition, melting point and distribution of nonmetallic inclusions - oxides and sulfides - that remain in solidified steel."

Ca treatment also helps to reduce any blockage in the pouring nozzle of the ladle. Blockage encourages reoxidation by exposing more of the stream surface to the atmosphere. Oxygen lancing, the common method of relieving blockage, compounds the problem by contributing oxygen to the molten steel. As a liquid at steel temperatures, Ca aluminates improve the metal flow and lessen the blockage, allowing a tighter pouring stream with less surface to reoxidize.

However, with a density of one-fifth of steel and a boiling point of 2718F (1492C), careful attention needs to be given to the method of Ca addition. "If added in bulk as metal or alloys to the tap stream," said Iqbal, "the Ca tends to float on the surface, burning profusely and reducing the efficiency of addition and the predictability of recoveries." If simply added to the ladle, Ca will stay atop the higher density slag cover, curtailing effectiveness.

The most efficient method of introducing Ca is feeding steel clad calcium wire through the slag cover [ILLUSTRATION FOR FIGURE 2 OMITTED] and deep into the molten steel beneath, where it will melt in fewer than three seconds. To further ensure a good Ca recovery, Iqbal has found that keeping a tight fit cover on the ladle during and after injection keeps the element from dissipating [ILLUSTRATION FOR FIGURE 3 OMITTED].

Iqbal cites several things the Ca does once in the liquid steel:

* modifies alumina clusters into small globular calcium alumina silicates;

* spheroidizes alumina galaxy-type oxides;

* transforms high-melting alumina silicates into low-melting alumina silicates for better fluidity;

* modifies hard alumina particles into soft calcium alumina silicates for easier machining.

Ca wire injection has proved to be one of the most effective ways to reduce oxide macroinclusions in steel castings. Many foundries report an average of 40% reduction in defects, and in one case selective Ca wire application produced a 67% inclusion reduction.

Argon Use

If one of the major causes of reoxidation is exposure of the molten steel to oxygen in the ambient air surrounding it, it is logical to try and protect the metal from that oxygen. To that end, many foundries have employed argon gas to insulate the steel. Argon, which is heavier than air, creates a layer of controlled atmosphere between the steel and the ambient air, limiting reoxidation from that source.

Several methods of argon protection have been tried. One is to shroud the pouring stream in argon. Unfortunately, no truly efficient method of shrouding has been discovered.

But one beneficial use has been to blanket the surface of the melt in the furnace or ladle with argon. This can be accomplished by adding liquid argon directly to the steel, where it will evaporate into the atmosphere directly above the bath. The other way is to pump, argon already in gaseous form directly to the area.

Waukesha Foundry, Inc., Waukesha, Wisconsin, has been using that method since extensively testing it in 1991. According to Waukesha's Vijay Talwar, "The constant goal of these trials was to reduce atmospheric oxygen adjacent to the melt from approximately 20.9% to below 1.0%." To accomplish that, the foundry used an average of 5 cu ft of argon to melt one lb of steel - an expensive process.

To try to cut treatment costs, the foundry created its own, homemade delivery system [ILLUSTRATION FOR FIGURE 4 OMITTED], which allows a controlled atmosphere over the surface of the induction furnace bath. However, this system reduced the effectiveness of the argon and oxygen levels jumped to as high as 10%. Fortunately, Waukesha determined that the result - scrap and cleaning room times - were not significantly affected, while the amount of argon used per lb of metal melted dropped to 0.625 cu ft. In this way, the foundry realized significant benefits at a fraction of the cost previously incurred. Those benefits include:

* a 9% reduction in slag-related scrap - about $175,000;

* cleaner furnace and increased refractory life;

* and a reduction in late alloy additions.

While the cost of argon use may be prohibitive to some foundries ($1.75 per lb of alloy melted in this case), Waukesha has certainly found it to be an excellent tool for obtaining cleaner steel, and saving money from scrap and rework.

Plasma Ladle Furnace

Another method of melt control that uses argon is employing a plasma ladle furnace (PLF). The concept is to perform a significant portion of the melting and refining process in a closely controlled vessel separate from the melting furnace - enabling chemistry and atmospheric control and production versatility.

The first foundry application of the refiner was introduced at Maynard Steel Casting Co., Milwaukee, Wisconsin, in 1993. Maynard's unit consists of a cylindrical graphite electrode mounted on a 4-1/2 ton bottom-pour ladle fitted with a special three-plate sidegate nozzle system [ILLUSTRATION FOR FIGURE 5 OMITTED]. In operation, molten steel is transferred from the electric arc furnace to the refiner, where the electrode strikes an arc with the molten bath.

Argon flowing through a small hole down the center of the electrode is used to form and stabilize the plasma arc. Simultaneously, argon gas is introduced through a porous plug in the bottom of the ladle to stir and deoxidize the molten steel. The argon flow is varied - a higher stir rate for uniform heating and desulfurization, a lower rate to float oxide inclusions to the surface of the melt, where they are picked up and held by the slag cover.

The plasma arc can be changed from a negative polarity, which is used to initiate heating and melt fluxes, to a positive polarity plasma to enhance sulfur and oxygen removal and aid alloy recovery. The tightly sealed cover on the ladle, along with the combination of argon use and plasma polarity manipulation, helps the foundry produce very clean steel. According to Maynard's Alastair Davidson, the ladle successfully achieves desulfurization, a low total oxygen level, and tight chemistry and temperature control.

"Using the PLF," said Davidson, "Maynard has successfully produced steel with very low oxygen, nitrogen, sulfur and phosphorous content. There is a marked improvement in inclusion type and severity in steels treated with the PLF compared to those without secondary refining." The foundry couples the PLF with other treatments to eliminate oxide inclusions. Calcium is injected into the heat late in the process for sulfide and inclusion shape control.

Beyond better control of oxide inclusion and overall melt chemistry, there are productivity benefits to using a PLF. It offers steel foundries the flexibility to produce basic steel, even though initial melting is done in an acid electric arc furnace. It also enables the production of two separate grades of steel from the same initial heat.

RELATED ARTICLE: Air Entrainment Research

While melt control is a means of reducing heat-to-heat variations in steel cleanliness, in-heat variations are also a cause for concern.

Some of the most interesting current research in that area focuses on air entrained in the metal stream during the pouring process. In standard pouring practices, 1/2 cu ft of air is entrained for each cu ft of metal poured. Once within the mold, 1 cu ft of air will cause 1/4 cu in. of oxide inclusions.

The amount of air entrainment is dependent on the pouring system and sprue/runner/gating system. To study the former, researchers at the University of Alabama-Birmingham have conducted pouring simulations using water instead of molten steel. They focused on the differences using bottom-pour, teapot and quadrant-pour ladles.

According to Lesley Brown, John Griffin and Charles Bates of UAB, when studying the bottom-pour ladle, nozzle throttling to control the pour rate actually increased the amount of air entrained. With the teapot and lip-pour ladles, they found that the geometric configuration of the spout produced the only notable variations.

In his analysis of all available water modeling data, Christoph Beckerman of the University of Iowa-Iowa City, concluded that "the impingement angle between the jet [pouring stream] from the ladle and the spree is the most significant flow variable affecting air entrainment."

He recommends "a ladle design that produces a large diameter circular jet with the least velocity and turbulence, and a trajectory that causes the jet to impinge on the spree wall with a minimum angle."

RELATED ARTICLE: Steel Foundries: A Vital Corner of the Industry

Talk to a steel founder right now and he'll probably tell you business is humming. Benefiting from the general surge in demand for railroad and construction equipment, steel foundries in the U.S. have just closed out a year that saw casting shipments rise up to 13% over 1993 - 11% higher than had been projected.

The nation's 245 sand casting steel foundries employ about 31,300 people. The railroad industry, particularly rail car wheels, makes up 40-50% of their production, while the mining industry accounts for another 20%. Other markets include construction equipment, trucks, pumps and valves. The last two appear to be good sources of opportunity for high-alloy casting producers, as the need for corrosion-resistant components rises with increasing environmental concerns.

Steel foundries have to contend with competition from two main sources - ductile iron castings and steel fabrications. While ductile casting production is roughly twice that of steel, it is near capacity, and there may be opportunities for steel castings pick up any shortfall in market production. Current difficulties in the availability and pricing of ductile iron may lead many customers to consider steel as an alternative.

Any opportunity, however, is contingent upon steel foundries' ability to increase high-production capacity, which had been cut back severely as many foundries closed in the 1980s. Thought must also be given to improving steel casting design to take advantage of steel's unique properties and make it more attractive to a wider variety of customers.

However, SFSA predicts a bright outlook for the next year. Though largely dependent on railroad car production, overall shipments are expected to rise another 2% in 1995, making a total of over 35% growth since 1992.