Maintaining competitiveness through tech development: three research projects provide practical information for metalcasters to remain on the cusp of technological advancement.

Firms that constantly explore new metalcasting methods, practices and applications will eventually find a way to remain competitive in the marketplace. Investing in the research of new technology and process development takes time and resources, but once the results have been analyzed, new techniques or applications previously thought unattainable can be implemented.

This article details three research projects conducted by the AFS Thin Wall Iron Group (TWIG), Magnesium Div., and Melting Methods and Materials Div. that provide metalcasters with practical information to maintain competitiveness by improving current methods and processes and developing new applications. The research is funded by industry, AFS, the Defense Logistics Agency (DLA) through the American Metalcasting Consortium (AMC), and the U.S. Dept. of Energy Office of Industrial Technology (DOE-OIT) through the Cast Metals Coalition (CMC).

Dimensional Tolerance Challenge

Background--Although aluminum has a density 34% that of cast iron and magnesium has a density 67% that of aluminum, design factors such as fatigue life, stiffness and elevated temperature strength often dictate that component dimensions be significantly increased when using aluminum or magnesium, thereby diminishing the benefits of lower densities.

Table 1 shows engineering alloy families competing for transportation applications where light weight is a primary design objective.

Goal--Many efforts to consistently produce thin-wall iron castings represent significant departures from traditional green sand molding techniques. But an empirical statistical study of the sensitivity of green sand process parameters on both dimensional variation and nominal dimension had not been conducted. TWIG studied the feasibility of reducing weight in iron castings by improving dimensional control and curtailing practices of over-design.

Approach--Four principal green sand casting parameters considered to have the largest effect on casting dimensional variations were selected following preliminary trials and extensive review of anecdotal information. The four factors--clay content, pouring temperature, sand compactability and shake out time--also were chosen based on the ability to control and monitor the process in a pilot production facility. Two levels (low and high) were chosen as bracketing levels representative of the industry (Table 2).

High volume horizontally-parted mold production equipment was used with a cluster of six test castings. The thickness of each of the 12 fins on the star test casting (Fig. 1) was measured midway between the vertices of the fins along the outermost edge. Both dimensional variability and nominal dimension of the star test casting fin thickness were analyzed as a function of clay content, pouring temperature, sand compactability and shakeout time. Process effects were analyzed separately in four groupings:

* the horizontal fins located across the parting line;

* the vertical fins formed in the cope;

* the vertical fins formed in the drag;

* all of the fins together.

[FIGURE 1 OMITTED]

Results--A trend was found between the average variation and the average nominal dimension of the test casting fin thickness (Table 3). The average dimensional variation increased nonlinearly as the nominal dimension increased. If this nonlinear trend was extrapolated to thicker and thinner nominal dimensions, the coefficient of variation was approximately the same--1.5% below 0.26 in. (6.5 mm) and approximately half as large and shrinking above 0.26 in. (6.5 mm).

The results indicated that control of clay content and pouring temperature are of utmost importance for reducing feature variations. Surprisingly, an increase in sand compactability resulted in only a slight increase in the dimensional variation for most cases and had a very small and varying effect with location (drag side, parting line and cope side) on nominal dimensions. Dimensional variation increased as the distance of the test casting from the down gate increased and toward the center of the casting where multiple fin faces intersected.

The results also showed that proximity to adjoining fins has a significant impact on dimensional quality (Fig. 2). As the distance from the location of merging features increased, the variation decreased. However, average deviations of 0.004-0.02-in. (0.1-0.5-mm) from many features demonstrated a capability not normally associated with the green sand process for gray iron, 0.03-0.05 in. (0.8-1.3 mm). The development of additional data to technically support the publication of a cast iron dimensional tolerance standard and recommended practice with automated equipment will provide designers and buyers confidence to adopt the attractive stiffness, strength and fracture properties economically found in cast iron.

[FIGURE 2 OMITTED]

Magnesium in Automotive Applications

Background--The AFS Magnesium Div. 6 developed its technology roadmap to complement the U.S. Automotive Materials Partnership (USAMP) North American strategic vision for automotive weight reduction through the application of magnesium wrought and cast components.

Cast magnesium structures have the potential to reduce 220 lbs. (100 kg) of vehicle mass, which could reduce emissions by 5% and increase fuel economy by 1 mpg (0.425 km/l). Light metal alloys have greater recycling value with reduced energy consumption versus plastics (including melting, machining, handling and transportation energy requirements). Health and environmental issues for workers are reduced during light metalcasting operations when compared to polymer molding operations.

Goal--The Structural Cast Magnesium Development Program, sponsored through USAMP-AMD and the DOE through a cooperative research and development agreement, aimed to address consistent quality in automotive volumes for economical cast magnesium components.

Approach--A new high temperature creep resistant alloy, AE4, was evaluated by bolt load retention testing. These results demonstrated that engine cradle attachments would meet vehicle performance requirements. A galvanic corrosion mitigation strategy developed in bench testing proved successful during platform vehicle validation testing. Furthermore, the magnesium engine cradle passed all validation requirements and is now in volume production on the 2006 Z06 Corvette (Fig. 3). Finally, a database initially launched in an earlier USAMP-DOE light metals process development program incorporated computer aided engineering properties from cradle structural cast magnesium development testing. The old database was improved with a new navigation system and additional property inputs.

[FIGURE 3 OMITTED]

Results--Although high integrity castings, corrosion and joining were addressed in this case study, process and production simulation, galvanic corrosion, joining and enhanced casting process development remain as cost issues limiting the application of cast magnesium automotive structural applications. A future goal is to cast a magnesium cradle with other processes, such as low-pressure permanent mold. This would demonstrate that aluminum casting technology can be transferred to magnesium casting at a low facility cost. The metallurgical and geometric limitations of diecasting addressed by low-pressure permanent mold offers other automotive opportunities.

Melting Methods and Materials

Background--Melting Methods and Materials Div. 8 sees five approaches to providing information to stimulate innovation to lower cost or improve quality of hot metal:

* Process control tools or methods;

* Direct and indirect charge materials or charge methods;

* Reduction of energy consumption;

* Reduction of all process emissions, wastes and scrap;

* Beneficial reuse of either waste produced by metalcasting operations or society waste streams.

Metallic charge materials represent a large cost in the production of ferrous melting operations. The cost of the charge materials can vary, which considerably affect the quality and competitiveness of the melting process.

Goal--An AFS-funded project that grew into a multi-year project funded by the DOE evaluated the metallic and alloy recovery of different charge materials for a coreless induction furnace.

Approach--Replicate covered and uncovered heats were melted with both rusty and clean charges to statistically evaluate their effects on alloy recoveries. Pig iron and cast iron gear blanks with thin and thick steel were melted in a 200-1b. (90.7-kg) coreless induction furnace. Each heat was charged to maintain a 4.00 carbon equivalent (CE) so the cast iron had sufficient fluidity for completely draining the induction furnace.

Common knowledge suggests that covered heats and clean scrap should give a higher metallic recovery. Charge materials react with the oxygen in the air and oxidize. However, during initial meltdown, air is present in both covered and uncovered furnaces. Since the oxidizing reactions occurred in covered and uncovered furnaces, it was not surprising that both melting conditions resulted in similar metallic recoveries of approximately 98.5%.

Covered melting with an inert or reducing atmosphere would reduce burning carbon monoxide and ambient oxygen, which reacts with molten metal, in the furnace. Analysis of the furnace atmosphere during melting reveals what happens during melting, so meaningful economic estimates can be made of the costs and the benefits of this approach.

Results--Because high-temperature low-density gases in the furnace quickly rise when the furnace lid is removed for charging, microwaves offer the advantage of all the energy being put into the solids, with none being used to heat gas. As far as metalcasting applications go, the microwave process could be used to make direct reduced iron that could then be charged to a melting furnace.

Carbon recovery is not statistically affected by the melting conditions, but it is affected by the carbon content of the charge materials. For instance, higher carbon-level charge materials have better carbon recoveries. The metallic recovery for the cast iron scrap and pig iron is higher than with the steel charge materials. Similarly, silicon recovery is higher for higher silicon charge materials and conversely lower for lower silicon-level charge materials.

Covered melting and clean charge material significantly increase the silicon and manganese recoveries (Fig. 4-5). The reduced oxygen content of the iron will react with silicon and manganese to form silicates and manganese oxide slag components.

For More Information

"Magnesium Casting Process Development," S. T. Robison, D. Weiss, G. Woycik, M. Marlatt and B. Cox, 2005 AFS Transactions (05-217).

"Metallic Recovery of Ferrous Charge Materials," W.M. Nicola and V.L. Richards, 2001 AFS Transactions (01-126).

The AFS Research Board evaluates, approves and coordinates research plans proposed by AFS Technical and Management Committees. Steve Robison is senior technical director and Joe Santner is director of research at AFS, Schaumburg, Ill.

AFS Research Board Steve Robison and Joe Santner, AFS, Schaumburg, Ill.

Table 1. Relative Physical Properties Among
Alloy Families

Alloy             Density           Elastic Modulus
                  (Kg/cu m)              (GPa)

Cast iron (1)          7,870              138
Aluminum               2,700               67
Magnesium              1,810               44

Alloy             Modulus to Density
                  [[(m/sec).sup.2 x [10.sup.6]]

Cast iron (1)          17.5
Aluminum               24.8
Magnesium              24.3

(1) Class 20 gray iron at zero strain or
in compression

Table 2. Low and High Ranges for Four Green Sand
Process Parameters Hypothesized to be Dimensionally
Significant

Process Parameters          Low Level    High Level

Clay Content (%)                 6            9
Pour Temperature (F)         2,550        2,750
Sand Compactability (%)         32           42
Shakeout Time (min.)            30           60

Table 3. Average Dimensional Variation and Nominal
Dimension Relationship in Iron

Fin Location               Dimensional           Nominal
                       Variation ([micro]m)   Dimension (mm)

Extrapolate to 20 mm           104                 20
Horizontal fins                96                 12.5
Drag fins                      89                 6.5
Cope fins                      80                 5.5
Extrapolate to 3 mm            54                 3.0

Fin Location           Coefficient of
                       Variation (%)

Extrapolate to 20 mm           0.5
Horizontal fins                0.8
Drag fins                      1.4
Cope fins                      1.5
Extrapolate to 3 mm            1.8

Fig. 4. This graph depicts silicon recovery by melting condition.

Uncovered 11.44           91.3
Covered 3.40              96.3
Clean 3.83                94.7
Rusty 11.59               93.2

Note: Table made from bar graph.

Fig. 5. This graph depicts manganese recovery by melting condition.

Uncovered 7.71         89.4
Covered 3.49           96.3
Clean 3.79             94.9
Rusty 8.49             91.1

Note: Table made from bar graph.