Test bars are used in aluminum foundry practice for both sand and permanent mold casting applications. They may be incorporated into the casting gating system, or more commonly, metal is poured into a separate mold. The purpose of the test bar is to give an indication of the mechanical properties achievable with a particular melt composition. However, test bar mechanical properties only correlate with the actual properties in a casting when the solidification conditions and defect population are similar in the test bar mold and a given section of the casting.

The permanent mold test bar design widely used in the U.S. was developed nearly 50 years ago by Ken Whaler at Stahl Specialty Company. He cut molds of various designs from plexiglass blanks. By pouring colored water into the transparent molds, he found a design that produced a smooth, quiescent fill of liquid metal in the mold cavity. Stahl Specialty also produced and sold foundry equipment, including their test bar mold, so it became known as the “Stahl” mold. The mold design was subsequently incorporated into ASTM standard B108.

Whaler studied the effect of aging time and iron content on tensile properties in A356-T6 alloy castings, using the B108 mold. His results are shown in Figure 1. This figure clearly shows how a standard mold can be used to produce extremely useful results.

The original Stahl mold had to run fairly hot, or shrinkage porosity would occur in the gauge section of the bars. (The mold casting temperatures recommended were 800-860F or 427-460C). A hot mold was also required to fully fill the cavity. This was not so convenient in practice—15 to 20 shots were needed to preheat the mold—so foundries sometimes made modifications. The sprue was usually made thicker to produce more rapid (easier) filling; and sometimes a filter was incorporated into the sprue or pour cup. These changes often had the effect of lowering casting quality, as shown in a study by CANMET researchers in 2004: “Revisiting the ASTM B108 Test Bar Mold for Quality Control of Permanent Mold Cast Aluminum Alloys.”

Stahl Specialty overcame these inconveniences by mounting the mold on a mobile cart and placing cartridge heaters into the stationary half of the mold. This cart assembly was placed next to a furnace and plugged into a source of electric power (Fig. 3). The heaters would heat the mold to the desired set point automatically. 

Stahl also incorporated coating procedures for the test bar mold. The mold is sandblasted once a week and then heated to 550F. A thin coating of conductive (graphite) die coat is sprayed on the central gauge section of the test bar, followed by placing a mask over the coated gauge section. The rest of the mold is coated with white (insulating) die coat. The insulating coat is applied in a number of thin layers, which are allowed to dry. The coating should have a “pebbly” surface appearance, which indicates it is insulating. (Coating the mold at 550F mold temperature should produce this effect.) It is also possible to smooth and thin the graphite coating in the gauge section by rubbing gently with fine steel wool. The preferred spray up is shown in Figure 3.

Unfortunately, even with these careful, optimized casting and mold coating procedures, a small amount of microporosity occurs. This is not obvious in X-rays of test bars, but the small pores can be observed in small “slices” cut from test bars (Fig. 4).

As a result, an AFS research project was undertaken to ascertain how clean metal could achieve highest mechanical properties in a separately cast permanent mold test bar. Through casting simulation and validation, a knife-edge ingate into the gauge section was incorporated into the test bar (Fig. 5). This acts as a cooling fin and improves the temperature profile in the gage section. The new design, referred to as the Case mold, allows for better feeding in the gage section by overcoming the two-wave front solidification that occurs in the standard (Stahl) mold. The reduced tendency for microporosity formation with this design allows achievable properties in A356-T6 alloy test bars of 45 ksi tensile, 35 ksi yield, and 12–15% elongation. 

During its development, the Case mold prototype was fabricated by modifying an existing Stahl mold. To incorporate filters, a pocket was cut and a new subinsert with two filter prints was added. Several possible modifications of the Case mold were studied by simulation to improve the permanent test bar mold design. The most important features studied were the sprue thickness, the filter and the knife ingate thickness at gage section. 

Initial mold designs incorporated filters into the gating system to assure melt cleanliness. However, the research demonstrated that with best in-furnace clean metal practice and virgin ingot, applying filters in the test bar mold have minimal effect, and the final mold design submitted and approved by ASTM did not include filters. 

Best practices must be followed to achieve optimum performance of the mold, such as maintaining a consistent mold temperature, proper mold coating techniques and careful pouring practices. These (and other) best practices have been identified in the AFS publication Aluminum Casting Technology.
All results of this AFS research project on the new permanent mold test bar design were reported in the International Journal of Metal Casting, as well as presentations made at AFS Casting Congresses.
The positive results with the newly designed knife-edge ingate were subsequently submitted to ASTM, and the AFS-Case test bar mold was approved as an alternative design for permanent mold separately cast test bars in the B108 standard. (The AFS-Case test bar mold is depicted in Appendix X4, page 19-21 of the new ASTM B108 standard.)

The mold is now being produced commercially and is available to interested foundries. 

Solidification simulation of the Case mold predicted significantly reduced microporosity, largely due to the incorporation of the knife-ingate design. The Case mold demonstrated superior as-cast tensile properties to the standard Stahl mold over a wide temperature range from 400F to 700F (204C to 371C). As would be expected, higher as-cast tensile properties were obtained with A356 virgin melt versus recycled melt. In these experiments, using in-mold filters up to 30 ppi in the test-bar mold had little effect on mechanical properties of the test bar. The use of an in-furnace filter had a very positive effect on reducing inclusions and therefore improving tensile properties in the test bar. 

Employing sufficient clean metal practices enabled recycled metal to exhibit very nearly equal as-cast test bar tensile properties to those from a virgin ingot melt. Initial standard heat treat T6 results show that a very high Quality Index result can be achieved in the new Case test bar permanent mold design, even without further microstructural enhancements—such as grain refining, modification, and SDAS considerations—leading to speculation that even higher levels can be ultimately reached.
While test bar results do not replicate actual properties within a complex casting, they represent a good foundry tool that can predict what should be achievable.    

Click here to view the article in the March 2021 digital edition.