Ceramic breaks the mold: when rising prices forced a steel caster to shop around for an alternative molding aggregate, it opted to give ceramic a try.

AmeriCast's Amite Foundry and Machine Inc. used zircon sand at its nobake casting facility in Amite, La, for specialty applications, but its rising costs due to increasing global demand had Junior Chavers, plant operations manager, looking for an alternative.

The type and quality of molding aggregate can mean the difference between a defect-ridden casting and a showroom-worthy component, so metalcasting facilities take great care in choosing their molding media and catering their operations to the chosen aggregate.

As a basic ingredient for most metalcasting operations, sand's role in casting is anything but simple. Molding sand must fit the metalcasters' operations in a number of ways, and the following factors weigh in on the sand choice:

* thermal and dimensional stability at the temperature of casting;

* particle size and shape;

* distribution of particle size;

* chemical reaction to molten metal;

* chemistry, pH and level of cleanliness;

* availability and cost;

* presence of volatiles that could produce gas defects in castings.

Finding the best sand for a metalcasting operation can help reduce defects, save money and improve quality and productivity. Most firms use one of four major types of naturally-occurring sand: silica, olivine, chromite and zircon.

Silica sand is the most common sand. Because it's the second-most abundant mineral in the Earth's crust, it is inexpensive and readily available. Silica sand does not offer the best hardness or abrasion resistance, nor is it the most chemically inert material, but for many casting firms, its properties are adequate for casting production.

Olivine is a magnesium silicate found in Washington and North Carolina. Its specific heat is the same as that of silica sand, but it has lower thermal expansion. It's used mostly in nonferrous metalcasting facilities.

Chromite sand has high heat transfer and low expansion, so it often is used for heavy steel applications where controlling the solidification of the metal is important.

Zircon sand has high refractory value and low thermal expansion, so it often is used in applications where better dimensional reproducibility is desired.

According to Daryl Hoyt, president of Foundry Sand Technology, Marseilles, Ill., when choosing a molding sand, metalcasters generally should consider four main factors:

1. The cost of the sand, which is mostly determined by location since a large chunk of money goes toward transportation costs.

2. The longevity of the supplier. A supplier that has firm roots will have good reserves of sand in case of emergencies and better control of quality of the product.

3. Technical service support of your supplier. Changing sand will affect the whole operation, so technical help will be a lifesaver.

4. Angularity of the grain and properties of the sand. This includes the grain fineness number, surface area, thermal expansion, fusion temperature, pH difference and particle size distribution.

Amite Foundry's Quest

Amite Foundry casts steel parts typically in the 3,500-50,000-lb. range for the mining, construction, marine, rail, turbine valves and industrial markets, pushing 6 million lbs. of new and reclaimed sand (mostly silica) a month. For some of Amite's specialty castings, zircon sand was used to give the metal a good surface finish and because its low thermal expansion aided shakeout. The decision to switch sands would not be made lightly.

Amite tried a cheaper grade of zircon, but that resulted in burn-in on the castings. Plus, the sand would lock up on the casting for tougher shakeout. Chavers also looked at using more chromite sand in the plant, but chromite picks up a higher level of metal elements during casting and puts it back into the sand during reclamation, which Amite wanted to avoid.

Eventually, Amite turned to a molding media that wasn't sand at all; the metalcaster opted to step into ceramic.

Worth a Shot

Ceramic is a manufactured molding media that looks like sand but has been produced under controlled process limits engineered to meet precise design specifications. The particles have a rounded, spherical shape due to the forming, drying and sintering processes in its manufacture, although the finer-sized products do show a slight decrease in shape properties. In the manufacturing process, alumino-silicate raw materials are reduced to a fine particle size and formed into rounded spherical pellets using water and binder. The pellets then are dried, pre-sized and sintered in a rotary kiln. The sintered material is re-screened to assure the desired sizing is achieved.

The rounded, spherical shape enhances the flow characteristics of the media, while the sintering process leads to the development of corundum and mullite crystalline phase assemblages that provide hard, durable and thermally stable attributes. Its physical properties lend it to low expansion and high reclaim characteristics, which Amite needed in order to successfully cast its large, steel parts.

The price per pound of the ceramic molding media was comparable to zircon sand, but with a 34% lower density, the ceramic presented a significant cost per cubic foot advantage. Plus, the ceramic media supplier, Carbo Ceramics, had one of its six manufacturing facilities nearby in New Iberia, La., helping to curb transportation costs and foster an easy supplier-customer relationship.

Intrigued, Chavers began experimenting with the ceramic media in the coreroom in Feb. 2006. Unfamiliar with the quirks of the ceramic's properties, Chavers started small.

"We started with smaller jobs of 15,000 and 18,000 lb.-pours, and the cores just peeled out," Chavers recalled. "Then we started experimenting with other jobs and eventually got up to a 91,000-lb. pour. Even with that pour, the core just fell out."

According to Chavers, the only adjustment needed was the amount of binder required. After a few sample runs, it was determined that a minimal extra percentage of binder was needed over the zircon sand the facility had been using.

Results from the use of ceramic in replacement of zircon are still preliminary for Amite Foundry, but initial reaction has been positive. Chavers said he sees the biggest advantage in the cleaning room, where cores made with the ceramic media break out cleanly and quickly. Adding the use of ceramic molding media in the molding room for certain applications potentially could cut cleaning times further.

Moving On

The early success of the ceramic media in core production has made Chavers eager to see it used in mold-making, but he's not without his reservations. Although the transition to ceramic media in the coreroom has been positive so far, he is concerned about the new media's affect on the plant's sand system. Any sand switch might bring about unexpected glitches in the system to which it will take time to adjust. Right now, the new media has shown no damage to the system, but Amite is waiting to see how it will react when the ceramic is used in the molding room.

The coreroom's split-hopper mixer, which only needs an adjustment of the amount of binder in order to switch molding aggregates, was able to handle the addition of ceramic; however, the metalcasters' molding operation is without the dual hopper, leaving the use of ceramic in molding to be relegated to addressing casting hot spots. Amite Foundry has plans to install a new mixer which will allow the ceramic to be used in full molds.

For More Information

"Synthetic Ceramic Media Provides a Unique Alternative to Silica Sand for the Metalcasting Industry," C.A. Krause, M.E. Hoover, 2003 AFS Transactions (03-060).

Shannon Kruse, Senior Editor

Table 1. Physical and Chemical Properties of Various Molding Media
at 1,1000

                                       Intermediate
                                         density      Low density
                                         ceramic        ceramic

Average specific gravity                   3.23          2.68
Grain fineness number                       40            110
Loose bulk density (lbs./cu.ft.)           113            94
Packed bulk density (lbs./cu.ft.)          125            109
Thermal expansion (% linear change)       0.708          0.667
Thermal conductivity (W/m-C)               0.70          0.74
Heat capacity (cal/g-C)                   0.291          0.292
Shape                                    Rounded        Rounded

                                        Zircon    Chromite     Silica

Average specific gravity                 4.65       4.51        2.65
Grain fineness number                     50         50          60
Loose bulk density (lbs./cu.ft.)         168         163        100
Packed bulk density (lbs./cu.ft.)        189         183        110
Thermal expansion (% linear change)     0.510       0.970      1.808
Thermal conductivity (W/m-C)             0.63       0.94        1.14
Heat capacity (cal/g-C)                 0.197       0.235      0.284
Shape                                  Rounded     Angular     Varied