Diecasters look to improve casting porosity and surface finish.

Demonstrating the growth of new ideas and equipment within their industry, the North American Die Casting Assn. (NADCA) held its 19th International Diecasting Congress & Exposition in Minneapolis on November 3-6. Attended by more than 500 diecasters and suppliers, the conference provided a product exposition and 61 technical presentations with topics ranging from various die treatments, process engineering and environmental control to magnesium, aluminum and zinc.

Following is a discussion of four presentations from the conference that detail improvements in casting porosity and surface finish.

SOLDERING OF ZINC DIE CASTINGS

The critical nature of surface finish on zinc die castings allows for little soldering (sticking) of the casting to the die components. In the past, diecasters approached the soldering problem with zinc in a similar manner as with aluminum. But as shown with research performed by Donald Argo and R.J. Barnhurst, Noranda Technology Centre, Pointe-Claire, Quebec, and William Walkington, Walkington Engineering, Cottage Grove, Wisconsin, the soldering occurring in zinc has different causes than in aluminum, and therefore different solutions.

The two companies, using extreme soldering conditions for their trials, conducted experimental runs of 100-300 shots. The objective was to examine the effect of various diecasting parameters, such as fill time, gate velocity, metal [TABULAR DATA FOR TABLE 1 OMITTED] pressure and die temperature on soldering, and to determine their relative importance for further investigation in the next stage of high-production tests.

For these runs, a special die - intended to generate soldering conditions - was designed and built. Both pressure and temperature indicator pins were used in the die to measure cavity conditions. The conditions for the four trials are shown in Table 1.

The Trials

The first trial was performed with a die design that had the metal from the gate impacting on an inclined ramp 45 [degrees] from vertical. The result was no soldering on the ramp, but instead a very rapid increase in the rate of die erosion where the metal from the gate impinged on the ramp.

In the second trial, the slope of the ramp was increased (60 [degrees] from the horizontal) for an even greater degree of metal impingement. Although the ramp was not heated as much, the result was no significant soldering in the area of metal impingement. Zinc deposits, however, were noted on the back sides of the die in rough, low draft areas.

For the third trial, to induce a galvanizing reaction, the gate was made narrower and thicker to increase the amount of metal impinging in one location on the ramp. The results were the same as in the second trial - significant soldering where expected - although there also was soldering in the areas shadowed from the metal flow on the back and sides of the die.

For the fourth trial, the front of the die directly across from the gate was made with zero draft. This time there was some soldering on the pin, in addition to the usual deposits on the sides and back of the die.

After the results of the four trials were analyzed, die temperature and draft angle were found to be the most significant variables, while gate velocity and aluminum content of the alloys did not show any significant effect. Figure 1 summarizes these results. The lack of influence of gate velocity was substantiated by the soldering that occurred in areas away from the main metal flow. For the most part, the soldering was confined to the eddies or low-pressure areas in the metal flow pattern.

Another significant observation from the trials was the dependence of soldering on die temperature. If a low draft angle is to be run, the die temperature should be low to lessen soldering, although the surface finish may suffer and a shorter fill time may be required.

DESIGN CRITERIA FOR VACUUM DIECASTING

Diecast aluminum components, due to the weight reduction they provide, have been targeted for extensive future use by the U.S. auto industry. But the weight advantage diecast aluminum castings hold is clouded by their difficulty with heat treatment. Although the commonly used diecasting alloys have enough copper or magnesium for heat treatment, the large amount of gas entrapped in die castings causes severe blistering during solution treatment.

To solve this problem, vacuum has been applied to remove air from the die cavity prior to metal injection, which produces parts with a low level of gas and porosity that can usually be heat-treated without blistering. The vacuum applications commonly used to evacuate the die cavity gases are called the chill block method and the valve method. This study, as performed by Naoyuki Tsumagari and Allen Nitz, Briggs & Stratton Corp., Milwaukee, developed basic design criteria for vacuum-assisted diecasting, focusing on the fastest evacuation and the lowest vacuum possible. The level of gas that can be included in a die casting for positive heat treatment had been set at below 1cc per 100g of casting at 273K and 1 atmosphere.

A special experimental device was developed to measure evacuation performance, as shown in Fig. 2. A vacuum pump continuously drew air from a 60-gal tank, with an air-actuated valve installed between the tank and the test chamber. This valve prevented the vacuum tank from losing its vacuum.

The test chamber consisted of two segments. The volume of segment V1, which contained the ports for tube connection, was 243.5 cu. in., and the volume of the other, V2, was 157.4 cu. in. The tubes used to evaluate evacuation resistance were from 1/83/4 in. in outer diameter (OD), and 0.065-0.376 in. in internal diameters (ID). Three different lengths were provided for each tube - 4, 8 and 12 in. A metal plate that had a slit (0.017x2 in.) at the middle was inserted between V1 and V2 to simulate the effect of a gate during the evacuation.

Evacuation Results

A typical evacuation profile is shown in Fig. 3. This profile was obtained when a tube, 5/8 in. OD and 4-in. long, was connected to the test chamber without a metal plate between V1 and V2. With the volume of air to be evacuated at 401 sq. in., two evacuation stages were found during the measurement.

During the first stage, which is shown by the steep pressure drop, air in the test chamber was quickly transferred to the vacuum tank through the tube. The pressure in the test chamber equilibrated with that in the vacuum tank at the end of the first stage. After equilibrium was reached, the evacuation solely depended on the pumping speed.

The second stage of the profile resulted from evacuation by the pump. When a 1/8-in. OD and 4-in. long tube was used, the evacuation of the test chamber became difficult. The rate of evacuation through the thin tube was slower than the pumping speed, resulting in no second stage for the profile. Figure 4 illustrates the various profiles observed from different OD tube sizes (all tubes were 4 in. long).

In another scenario, a 0.375-in. thick plate, which had a 0.017 in. x 2 in. slit, was inserted between V1 and V2, with a 0.5- in. OD and 4-in. long tube connected. The size of the slit was designed so that it could simulate the thinnest gate size. The evacuation profiles from both segments showed the evacuation of V2 was slightly slower than that of V1. Although the delay in V2 was small, the effect of a gate during evacuation was considered important.

The required level of vacuum in the die casting cavity was determined to range from 8-18 torr depending on the metal fill percent in the shot sleeve. But to attain the level of vacuum required for heat treatment, the system must be designed so that the smallest opening is still big enough for fast evacuation. To design vacuum runners and valves, the formula that describes the time required to evacuate a cavity to 100 tort is: [t.sub.100]=A[x.sup.B], where A and B vary depending on the size of the cavity to be evacuated and the target vacuum, and x is the minimum area of opening in the vacuum conduit.

LOW-IRON ALUMINUM ALLOYS

As the demand for castings of various sizes and properties continues to rise, the diecasting industry is working to develop new alloys to meet those demands. For Alois Franke and Hubert Koch, Aluminum Rheinfelden GmbH in Germany, this demand was fulfilled with the development of its Silafont-36 and Magsimal-59 low-iron alloys.

The company's objective in the development was to provide alloys that react sensitively with cooling rate, offer good castability and provide excellent corrosion cracking resistance. The company determined that these requirements could only be met by low-iron alloys that avoid unfavorable metallic [TABULAR DATA FOR TABLE 2 OMITTED] phases and have more than about 25% by volume eutectic content in the microstructure.

Table 3. Chemical Composition Limits of Magsimal-59 (in % by weight)

         Si      Fe      Cu     Mn     Mg      Zn      Ti     Other

Min.     1.8                    0.5    5.0            0.1
Max.     2.5    0.13    0.05    0.8    5.5    0.08    0.2     0.06

Silafont-36

Silafont-36 is a low-iron aluminum-silicon-magnesium (AISiMg) alloy with Si content between 9.5-11.5%, Mg between 1.5-5%, iron content kept below 0.12% and manganese (Mn) between 0.5-0.8% (see Table 2). The Si creates fluidity for casting large parts, while the low-iron minimizing aluminum-ferrosilicate (AlFeSi) phases in the alloy structure. The Mn provides rigidity and reduced die soldering.

In terms of mechanical properties of castings, the alloy provides the following:

* 20% elongation or 340 Mpa ultimate tensile strength can be attained upon heat treatment;

* a limit of endurance greater than 12.5 ksi (86 Mpa);

* stress cycle curves that outperform A356.

The company's breakthrough experience casting the alloy was for the Audi A8 space frame body. There are 31 nodes of this alloy mounted on the frame with weights of 0.88-4.4 lb and thicknesses of 0.08-0.20 in. A second part diecast was the cross member for the Porsche Boxster. Its mechanical properties after heat treat are 180 Mpa yield strength, 290 Mpa tensile strength and 14% elongation.

Magsimal-59

The objective in the development of Magsimal-59 was to create an alloy system that reacts sensitively with the cooling rate, eliminating heat treatment. The alloy is in the Al-Mg-Si-Mn family with iron content below 0.13% and a eutectic fraction of 25% (see Table 3).

For the alloy, its mechanical properties increase with thinner wail thickness and higher cooling rates. As an example, a wall thickness of 3 mm provides 18% tensile elongation, 27 ksi (185 Mpa) yield strength and 45 ksi (310 Mpa) ultimate tensile strength. The alloy has a limit of endurance of 14.5 ksi (100 Mpa) at a probability to fracture of 5%.

One Magsimal casting example is the cross member of a sedan. The mechanical properties of the finished casting (which wasn't heat treated) were 26.1 ksi (180 Mpa) yield strength, 40.6 ksi (280 Mpa) tensile strength and more than 8% elongation. The part was cast on a 450-metric ton diecasting machine with a forced venting system.

Both the Silafont and Magsimal low-iron alloys have provided a first step toward high mechanical properties in finished castings with minimal or no surface treatment.

Table 4. Density vs. Porosity

Casting #                    Density (g/cc)     Porosity (%)

1                                2.7404             1.92
2                                2.7418             1.87
3                                2.7400             1.94
Average of Low Pressure          2.7407             1.91
4                                2.759              1.27
5                                2.7537             1.45
6                                2.7513             1.53
Average of High Pressure         2.7547             1.41

EFFECT OF SOLIDIFICATION PRESSURE ON POROSITY

The application of pressure to a solidifying diecasting alloy has two purposes - to feed additional liquid alloy into the casting cavity (reducing porosity associated with solidification shrinkage) and to reduce the volume of any insoluble gases present in the solidifying alloy [ILLUSTRATION FOR FIGURE 5 OMITTED]. Andrea Kay, General Motors Powertrain, John Wronowicz, General Motors Casting Advanced Developing Center, and Angela Wollenburg, Jerry Brevick, and Carroll Mobley, Ohio State Univ., teamed to investigate the effects of applied pressure on porosity distribution in a given casting.

In their experiment, a diecasting system with the following characteristics was set up:

* a single gate located at one end of the casting;

* a relatively long fill distance or casting cavity length;

* overflows and vents at the end of the casting opposite the gate;

* the ability to change the intensification pressure.

The selected commercial test casting had a ring gate at one end and vented overflows at the other end, with a length of 17 in. Six castings were made sequentially, holding all the production variables constant, except pressure during solidification. Three of the castings were made with 6730 psi and three with 11,210 psi. Once cast, the parts were weighed and sectioned, and density and porosity were measured.

Results

The density and corresponding porosity data given in Table 4 and shown in Fig. 6 confirmed the initial thought that the application of pressure during solidification decreased porosity. The average gas content of all castings for groups was 33cc per 100 grams of alloy, which is equivalent to 0.28% and 0.17% porosity for the high- and low-pressure castings.

In addition, the following conclusions were made:

* the weight of the trimmed castings increased by 2%, while the weight of the gate decreased by 3.9% with an increase in the solidification pressure from 6730 to 11,210 psi;

* the average porosity dropped from 1.9% to 1.5% due to the increase in applied pressure;

* the volume of the trimmed casting increased by 69 cu. cm due to the increase in pressure;

* the porosity due to the contained gas dropped from 0.28% to 0.17% with the increase in pressure;

* solidification shrinkage decreased from 1.6% to 1.3% with the increase in pressure;

* the porosity levels were reduced uniformly across the length of the casting with the increase in pressure, making the feeding distance for the casting at or above 17 in.