Taking another look at test bar molds: the mechanical properties of test bars cast in both an ASTM B108 test bar mold and a modified mold were evaluated, and efforts were made to create an entirely new test bar mold.

Permanent mold cast test bars have a number of applications, such as examining metalcasting techniques, metal quality, heat treatment and the properties of an alloy. However, these bars cannot be used to predict the quality of a simple casting with a good gating system. During production, casting properties can be inconsistent because production castings range in structure, have rapid changes in cross-section, and the location of gates is determined by trial-and-error methods, experience and casting geometry. Separate sections of a casting can greatly differ in soundness, thus inhibiting metalcasters from observing the alloy's characteristics. Therefore, it is necessary that a test bar mold provide bars with reproducible structures and properties under normal operating conditions with limited production, tooling and testing costs.

The ASTM B108 mold has become a household item in many aluminum permanent mold casting facilities. For more than 25 years, facilities have adapted the procedures and filling characteristics to fit their needs, but they also have uncovered negative factors (mold temperature, mold coating, metal temperature, tilt and number of bars rejected prior to testing) that could affect the alloy properties.

Investigations were perforated to compare test bar properties from an original ASTM B108 mold with those from a modified ASTM mold. Further, a newly designed mold was created to emulate both good and bad casting practices.

This article also details the results of comparing the ASTM mold with the modified mold and the creation and fallacies of a potential new mold for test bar analyses.

Mold Prep

The first investigation included two different ASTM B108 molds. One was made in 1977 and remained unchanged. The other (made in 2001) was modified with additional gates, a larger sprue and a larger bottom runner to allow for higher flow rates of molten metal (Fig. 1). The use of simulation software indicated that the modified mold design would reduce metal velocities and surface turbulence.

[FIGURE 1 OMITTED]

Thermocouple wells were drilled to within 0.09 in. (2.5 mm) of the cavities in both molds. Because the molds had large plates covering a quarter of both sides as part of the hinge (hinged side, [Fig. 2]), the wells were drilled toward the other casting cavity, referred to as the unhinged side.

[FIGURE 2 OMITTED]

Three different types of coatings were applied to hot molds: alumina, alumina with graphite on the bar gauges and boron nitride with graphite on the gauges. These coated molds were then stored in an oven and soaked at 752F (400C) until they were needed. Mold temperatures fell by 77-158F (25-70C) until pouring began.

The test bars were cast with three different alloys: A356; A356 with 9% silicon; and "dirty" A356 with 9% silicon (using defected metal). Also, three pouring cycles were used: as quickly as possible (150-sec. cycle); a 200-sec. cycle and a 220-sec. cycle. Molds were poured either upright or on angles.

Properties Found

Mechanical properties were measured after the test bars were cast to evaluate the effects of mold temperature and coating, metal temperature, the effect of tilting the mold and the differences between the hinged and unhinged sides.

During the pouring process, once mold temperatures reached 734F (390C) at the sprue for both the ASTM and modified molds, properties remained mostly unaffected by increases in mold temperature (Table 1). As pouring proceeded, mold temperatures near the unhinged test bar gauge reached 824F (440C) in the ASTM mold and 869F (465C) for the modified mold, but for the most part, properties were not statistically different. The ASTM B108 mold will not completely fill at low mold temperatures.

Although the modified mold will fill at lower temperatures, the properties will be erratic. Hence, when used in metalcasting facilities, bars should be poured and rejected until mold temperatures are adequate. The practice of topping up the riser to preheat the mold to higher temperature can still be used, but the bars should be rejected.

Also, during the pouring stages, an alumina insulating wash was applied over the boron nitride-coated molds, but the graphite-coated gauges were left untouched. It was found that ductility will be harmed if a thin coating of alumina is applied over the graphite on the gauge once the casting is poured. Thus, it is ideal when using test bar molds to apply a conductive coating (graphite) on the gauge and a more insulating coating (alumina) on the rest of the mold.

Similar to mold temperature, the effect of the molten bath temperature was found to have little impact during the evaluations. Thirteen castings were poured at several temperatures. Nine were poured with a molten bath temperature near 1,328F (720C) and the last four at 1,382F (750C). Although there was marginal improvement in some properties, the average properties of any set was within one standard deviation of the mean (13 of the 14 bars) for all bars (Table 2). Therefore the experiment did not find an effect.

There were no dramatic differences in properties when tilting the ASTM mold either. Tilting the mold at a 30-degree angle from the vertical position had no significant effect on the yield and ultimate tensile strengths of the bars, though elongation was better in the vertical position. Increasing the angle further to 60 degrees also had no effect on properties of bars produced in the ASTM mold, though filling time increased and turbulence decreased. However, there were benefits of pouring the modified mold at an angle, such as improved strengths and reduced scatter, while elongation increased slightly (Table 3, Fig. 3). For the modified molds, the tilt-pouring methods were more efficient in obtaining similar properties of the ASTM mold.

[FIGURE 3 OMITTED]

Despite the similarities with several other properties, the three alloys used in the investigation (particularly the 8.5% silicon) revealed some differences in yield strengths, ultimate tensile strengths and elongation (Table 4). However, the hinged- and unhinged-side properties were similar (Fig. 4).

[FIGURE 4 OMITTED]

Creating a New Mold

After the original and modified molds were studied, an attempt was made to produce a new mold with multiple cavities that would not only duplicate the conditions of the ASTM mold, but also indicate the range of properties that can be expected from castings made with a typical gating system and a poor system. This meant producing a mold with both good bars (with a quiescent fill) and bad bars (very turbulent filling).

First, an ASTM mold was machined from (vertically) laminated cast iron blocks, and test-bar castings were produced and tested against the ASTM mold. Once it was confirmed that the laminations or potential thermal discontinuities had no effect on properties, a four-bar mold (with two good bars and two bad bars) was produced. The mold was tested and reconfigured four times. The gating system had to be enlarged to achieve quick filling with no hesitation when a bad bar filled. Changes to the gating caused changes in filling of different test bars, which affected the properties. When different alloys were used, the hierarchy of mechanical properties was different from the ASTM B108 and modified molds.

A problem with the four-bar design was that the mold temperature at the two central bars was the same as the sprue and 86F (30C) hotter than the outer bars. These central bars filled first and froze last, which meant filling, feeding and cooling were different from the outer bars regardless of good or bad design. Incorporating gaps (similar to those that were used in another mold for ceramic filters placement) between the sprue and the inner bars might reduce the impact of the extra heat from the spree.

Further work is required to develop a mold that can produce bars with good and bad properties. While the four-bar mold design attempted to produce instant good and bad test bars, future designs should concentrate on two bars to avoid inconsistent filling.

Although the ASTM B108 mold does not fill properly below 698F (370C), the properties of bars poured above that temperature are not affected. Pouring temperature and mold orientation have little or no effect. However, while an alumina coating is best, the coating on the gauge section must be much thinner or coated with graphite. A mold modified with larger sprees and gating systems should be poured with the mold temperature at the sprue base at least 734F (390C) with the mold inclined.

Even with efforts to augment an original mold and create a new one, these overall investigations prove that the ASTM B108 test bar mold is effective and foolproof.

This article was adapted from a paper (04-032) presented at the 2004 Metalcasting Congress.

Table 1. Mechanical Properties for Bars Cast in Molds Tilted at
30-Degree Angles Under Rapid Pouring Conditions

Bar No.    Gauge     Fill Time                Unhinged
         Temp. (F)    (sec.)
                                 YS--MPa  UTS--MPa  Elongation %

              ASTM Mold--Tilted 30 Degrees From Vertical

  1-3       653     [+ or -] -8   92.7     141.7        1.7
  4-6       741          8        85.9     161.2        2.6
  7-9       779          8        86.8     158.4        2.7
 10-11      808         8.5       87.2     166.0        3.5

              Modified Mold--Tilted 30 Degrees From Vertical

  1-3       622          8        90.5     160.5        2.5
  4-6       737          7        85.6     146.2        2.1
  7-9       788          7        82.8     145.2        2.2
 10-11      831          8        81.4     163.6        3.9

Bar No.              Hinged

         YS--MPa  UTS--MPa  Elongation %

            ASTM Mold--Tilted 30
            Degrees From Vertical

  1-3     95.0     170.3        3.5
  4-6     82.4     178.0        5.2
  7-9     83.5     182.5        6.1
 10-11    84.3     177.9        6.0

            Modified Mold--Tilted
           30 Degrees From Vertical

  1-3     90.5     162.9        2.6
  4-6     82.4     149.1        2.3
  7-9     79.6     146.5        2.3
 10-11    82.5     159.0        3.4

Table 2. Mechanical Properties of Bars Cast at Various Mold
Temperatures in the ASTM B108 Mold

Bar No.     Sprue      Bath                Unhinged
          Temp. (F)  Temp. (F)
                                YS--MPa  UTS--MPa  Elongation %

  1-4        737       1,328     87.0     158.6        3.0
  5-9        737       1,331     85.9     156.7        3.0
 11-14       748       1,382     85.1     163.9        3.6
13 of 14     739       1,348     85.7     158.9        3.2

Bar No.               Hinged

          YS--MPa  UTS--MPa  Elongation %

  1-4      82.1     159.0        2.7
  5-9      84.0     155.8        3.0
 11-14     86.4     166.2        3.8
13 of 14   84.7     159.0        3.2

Table 3. Mechanical Properties of Bars Cast in Both Molds with
Graphite Coating on the Gauge and Alumina Elsewhere

Timing    Gauge               Unhinged
(min.)  Temp. (F)
                   YS--MPa  UTS--MPa  Elongation %

                   ASTM Mold-Vertical

 3:00      694      97.5     202.7        7.0
 3:05      797      92.2     195.0        6.7
 3:05      836      92.2     195.2        7.0
 3:35      831      89.5     195.7        7.5

        ASTM Mold-Tilted 30 Degrees From Vertical

 3:05      800      93.3     191.1        5.7
 3:05      831      91.7     192.2        6.1
 3:15      836      94.1     195.1        7.1
  *         *       95.2     191.8        5.7

                 Modified Mold-Vertical

 3:00      718      93.9     185.5        4.6
 3:05      836      88.8     181.1        5.3
 3:05      871      88.3     183.4        6.2
 3:35      869      88.1     183.2        6.0

             Modified Mold-Tilted 30 Degrees
                    From Vertical

 3:05      840      92.4     189.4        7.0
 3:05      867      89.2     187.2        6.8
 3:15      871      94.5     186.4        6.8
  *         *       90.3     184.4        5.8

Timing            Hinged
(min.)
        YS--MPa  UTS--MPa  Elongation %

              ASTM Mold-Vertical

 3:00    96.0     195.3        6.1
 3:05    93.3     189.2        5.9
 3:05    93.8     190.4        6.5
 3:35    92.2     190.2        6.3

           ASTM Mold-Tilted 30 Degrees
               From Vertical

 3:05    92.6     187.7        5.7
 3:05    93.1     189.8        6.0
 3:15    93.0     192.0        6.1
  *      97.2     192.4        6.1

          Modified Mold-Vertical

 3:00    91.1     186.7        5.4
 3:05    90.4     181.5        5.3
 3:05    90.4     182.0        5.6
 3:35    91.9     179.5        5.0

         Modified Mold-Tilted 30
          Degrees From Vertical

 3:05    91.1     188.2        6.1
 3:05    94.0     185.0        5.7
 3:15    95.9     186.0        6.2
  *      91.1     184.9        5.9

Table 4. Mechanical Properties for Bars Cast in ASTM Molds with
Three Different Alloys

Alloy                    Unhinged

                YS   UTS-MPa  Elongation %

A356           89.8   166.9       3.3
8.5% Si        95.4   204.6       6.9
8.8% Si Dirty  83.7   174.1       3.9

Alloy                     Hinged

                YS   UTS-MPa  Elongation %

A356           92.0   176.3       4.1
8.5% Si        94.8   207.3       8.2
8.8% Si Dirty  82.4   175.0       4.3

For More Information

"Influence of Casting Conditions on Mechanical Properties of A356 Aluminum Alloy Permanent Mold Cast ASTM B108 Test Bars," F. Chiesa, T. Boisvert, T. Houde, D. Lavoie, 2004 AFS Transactions, Paper No. 04-077.

Daryoush Emadi and Laurence Whiting are research scientists, and Mahi Sahoo is the manager of casting technology at the Materials Technology Laboratories (MTL)-CANMET/, NRCAN, Ottawa, Ontario, Canada. Daniel Larouche is a professor in the Dept. of Mining, Metallurgy and Materials Engineering at Laval Univ., Quebec, Canada.