Correlating inclusion sizes with aluminum casting: by following current NDT quality standards for permanent mold and sand castings, investigations contrasted the industry's detection capabilities for molten aluminum inclusions.

Metal caster have difficulty accurately assessing aluminum melt cleanliness prior to pouring. Realistic and practical techniques to predict and assess the damaging effects of inclusion size concentrations on final cast products are limited. Furthermore, little information is available to show a correlation between the inclusion size present in the molten bath prior to the pouring and the quality acceptance criteria for commercial castings.

Better understanding of inclusion defects would help aluminum metalcasting facilities relate practical inclusion control with molten metal handling and treatment process parameters. Technology that would enable better assessment of inclusion content by providing realistic quantitative information on the inclusion sizes would help metalcasters determine if a particular batch of the molten metal could be used without causing castings to fail quality requirements. The long-range benefits eventually would lead to total process control in molten metal.

This article will provide a different perspective on typical inclusion particle sizes and their effect on final casting quality, thus asking: will the casting pass or fail the quality inspection requirements? The article is the result of a recent study on detecting inclusions in the Aluminum melt and includes several metallurgical observation, such as discovering inclusions smaller than 60 microns (0.0024 in. or 0.06 mm) did not necessarily create scrap castings and that particles in the melt can be much larger than thought without causing quality rejections.

Realistic Assessment

While the existing types and sizes of non metallic and intermetallic inclusions present in melting furnaces will vary from each facility, their removal is essential for proper molten metal cleanliness. A number of commercially accepted melt treatment techniques are being used by aluminum metalcasting facilities to remove and separate inclusions from the molten aluminum alloy prior to casting. These include various methods of fluxing, degassing and filtration in the furnaces and in the gating system.

Most metalcasters understand that these techniques have an impact on the melt cleanliness of the molten aluminum alloy prior to pouring. However, the effectiveness to evaluate their removal not only relies on the melt cleanliness measurement technique being used, but also on the casting quality acceptance criteria and the final casting scrap rate.

Over the last 35 years, several techniques have been developed and used for assessing the cleanliness of molten aluminum alloys. These include qualitative, quantitative and analytical laboratory procedures as well as online and off-line techniques. Typical types, sizes and morphologies of inclusions that may be present in molten aluminum alloys have been documented in great detail.

Past investigations have found that:

* inclusion size may vary from less than one micron to a hundred microns and larger;

* inclusions may occur in a variety of particulate and film morphologies, such as spherical or angular particles, films, skins and/ or clusters;

* the inclusion concentration may be in the range of parts per million (ppm) to fractional percentage (by volume).

Although many inclusions originate during the manufacturing of aluminum alloy ingots prior to metalcasting, most inclusions are generated on the shop floor during subsequent melting, handling and pouring operations. The most common sources of inclusion defects are oxide particles and films.

There appear to be two different schools of thought with respect to defining inclusion limits in molten metal for foundry applications. The first believes that the inclusion content in molten aluminum alloys has to be several volume parts per billion and that the average particle size in the population can be no more than 50 microns (0.0020 in. or .05 mm) to produce quality castings. The second idea supports the more practical notion from the foundry perspective. The extent in which inclusions render a molten alloy "unfit for use" is considered a function of the casting application and therefore of suitable molten metal practice.

Inclusion Sizes vs. Radiographic Standards

An ultrasound measurement technique was used to assess molten metal cleanliness in an aircraft and automotive aluminum facility, having reverberatory and crucible furnaces and pouring a variety of aluminum alloys. Metal cleanliness evaluations were made before and after rotary degassing and refinement additions, according to operating procedures on 319, A356, C355, A357 and A206 aluminum alloys, under different melting and handling practices. The metal temperatures that were evaluated ranged from 1,280-1,555F (649-846C)

During the assessment, different particle sizes were measured and monitored. The inclusion sizes and distributions were found to be affected by the melting, handling and/or processing technique. Due to the limitations of ultrasound technology equipment and/ or equipment settings, particle sizes (inclusions) more than 60 microns could not be differentiated either by size or particle size distribution. In consequence, all of them were detected as if they were measuring 60 microns. On the other hand, inclusions smaller than 60 microns were easily differentiated by particles' sizes, distributions and concentrations.

Initial evaluation of the cleanliness data would leave the impression that the molten metal was full of damaging inclusions. Nevertheless, all of the several hundred different automotive and aircraft castings (with a variety of gating systems) passed their respective ASTM, AMS, MIL-A-21180 standards, and/or customer material specifications. All of the castings that were subjected to the stringent liquid penetrant and radiographic inspections per MIL-STD-6866 Type 1 method A and MIL-STD-2175 class 1 grade A or B also passed the requirements. In addition, the castings that were required to satisfy mechanical properties from separately cast test bars and from designated areas met the requirements. Moreover, some of these castings required pressure test qualifications over 2,500 PSIA.

The casting quality requirements were met regardless of the particle size ranges (smaller or larger than 60 microns) detected in the molten metal just before the pouring operation. These results differed from common assumption because, in theory, such inclusions would have indicated low quality molten metal to start with, and therefore, scrap castings would have been anticipated. With respect to the inclusion sizes in the range of 20 microns (0.0008 in. or 0.02 mm) to 60 microns there was no doubt in accepting the evidence that such inclusion sizes did not affect the final quality of the castings. With respect to the inclusions more than 60 microns, the only practical conclusion was to acknowledge that the castings passed the inspection criteria because of a single or combination of factors including a proper gating system (in-mold filtration and/or non-turbulent filling) and the fact that that the inclusions more than 60 microns regardless of their actual size were smaller than the foreign material discontinuity given by the ASTM E155 reference radio-graphic plate #1.

Additionally, all the castings met the X-ray requirements for shrinkage, porosity, gas holes and foreign material. Some of the automotive castings, which did not require radiographic inspection, were X rayed for the purpose of evaluating their internal soundness against the inclusion sizes detected in the molten metal prior to pouring. The X-rays of these castings were found to be free of inclusions, as well.

Standard industrial X-ray qualification inspection doesn't require knowing the actual size of the inclusion defects. However, in this particular case, the foreign material discontinuity sizes in the ASTM E155 reference radiographic plate levels 1 to 3 were measured by evaluating the greatest length and width of each inclusion. In all cases, the more dense exhibit was used. Thus, the inclusion sizes for the first three plates were considered to be:

* Plate 1 between 762 microns (0.030 in. or .762 mm) to 1,524 microns (0.060 in. to 1.524 mm);

* Plate 2 between 2,032 microns (0.080 in. or 2 mm) to 3,429 microns (0.135 in. to 3.4mm);

* Plate 3 between 1,778 microns (0.070 in. or 1.7 mm) to 3,734 microns (0.147 in. or 3.7 mm).

More to Know

Permanent mold and sand aluminum metalcasters may have inclusion scrap defects after radiographic or fluorescent penetrant inspection. However, the majority of the inclusion scrap defects that a facility experiences typically occur after the machining operation. Such castings will be rejected because of poor machinability, such as hard spots and/or because of failure to meet stringent cosmetic requirements on machined surfaces like cylinder heads, pistons and structural castings. Cosmetic requirements may cause a casting to be scrapped if inclusions were larger than 400 microns (0.0157 in. or 0.4 him). Such size is considered to be about the smallest size defect that could be seen by the naked eye on a machined surface. Thus, most inclusions are only discovered when the customer complains. Unfortunately, troubleshooting after the customer complains is still one of the main ways of establishing melt quality control, therefore costing the facility more money to produce a quality component.

Regarding potential harmful discontinuity sizes, past studies have revealed that porosity defects of 100 microns (0.004 in. or 0.1 mm) start affecting mechanical strength and fatigue life. At more than 1,000 microns (0.040 in. or 1 mm), fatigue life can be severely affected. Although an inclusion and porosity of the same size might cause different effects, it is interesting to point out the magnitude of the porosity discontinuity sizes. By this concept, it seems unlikely that controlling or monitoring inclusions lower than 100 microns would provide economical and practical benefits to the metalcasting industry.

Taking into account that the smallest inclusion defined by the most stringent ASTM E155 reference radiograhic plate (#1) is not far off from 762 microns (0.030 in. or 0.762 mm) and that the maximum inclusion size permissible by cosmetic requirements is no more than 400 microns (0.0157 in. or 0.4 mm), then it is logical to think that an online inclusion assessment should be in the vicinity of such a range between 400 and 762 microns. Metalcasting facilities may not benefit economically from trying to control the smaller inclusion particles that are less than 60 microns. Such small sizes, which remain suspended in the molten alloys at the time (if pouring, should be regarded not as defects in the macroscopic sense but as intrinsic characteristics of proper molten metal preparation. Therefore, efforts must be directed toward a consistent application of proven techniques to remove and separate larger inclusions.

Facilities using less than optimal melting and handling processes will make molten aluminum alloys with inclusion particle sizes more than 60 microns suspended in the molten bath. Even in this case, the melt might still be acceptable if the inclusions are below 762 microns.

Further Studies Needed

The negative effect of inclusions in mechanical property evaluations is commonly noticed during the tensile testing of separately cast test bars. Flaws in such bars due to inclusions in the fracture do not cause rejection of the castings and can be replaced with another bar and retested per ASTM B 557. Moreover, the negative effects of inclusion occurrence in the test bars are almost never related to castings being scrapped due to inclusion defects. From the metalcasters' perspective, the most important factor is that as long as the finished casting is within specification, it will be fit for use, and most importantly, will satisfy the end user.

Mechanical property requirements for separately cast test bars (according to ASTM B108), test bars designated from specific casting locations (according to AMS standards) and customer specifications are being met with molten metal containing inclusions lower than 60 microns. Even if the casting quality specifications were to be improved by 50% (from the present day cosmetic and radiographic inspection requirements), this would mean a critical inclusion size range between 200 (0.0078 in. or 0.2 mm) and 381 microns (0.015 in. or 0.381 mm). Even at this dramatic improvement, inclusion sizes to be monitored would be more than 200 microns.

Yet, the question remains as to what mechanical properties could be obtained by cleaning or controlling the molten metal alloys with inclusion sizes of 762, 381,400, 200 or 100 microns? Without a real time inclusion-measuring device capable of measuring such sizes, those questions may never be resolved.

For More Information

"Ultrasound for On-Line Inclusion Detection in Molten Aluminum Alloys: Technology Assessment," Gallo, Rafael, et. al., AFS international Conference on Structural Aluminum Casting, Orlando Fla., 2003.

Visit www.moderncasting.com to view the entire report, "Contrasting Molten Aluminum Inclusion Size Detection Capabilities with Current NDT Quality Standards for Permanent Mold and Sand Castings."

Finding 40 Micron Inclusions

Illustrated here are equations that reveal the number of inclusions that could exist even at an inclusion concentration of 1 ppm (a typical factor used by quality control personnel). Thus, if the average size of a spherical inclusion is 40 microns (0.0016 in. or 0.04 mm) then, 1 lb. of molten aluminum will contain 5,000 inclusions.

The number of inclusions that could exist at a concentration of 1 ppm in 1 lb. of A356 aluminum alloy is given by the following expression:

1 ppm of inclusions in 1 lb. of A356 aluminum / Volume of the considered inclusion size

If the density of A356 aluminum alloy is considered to be 0.097 lb./[in.sup.3], then 1 lb. of such alloy is equal to 10.309 [in.sup.3].

An inclusion concentration of 1 ppm in 1 lb. of A356 aluminum alloy will be equal to:

(10.309 [in.sup.3]) = 0.000010309 in[in.sup.3] = [10.sup.6]

The volume of a sphere is given by:

4/3[pi][r.sup.3]

Therefore, an inclusion size of 40 microns in diameter (0.0016 in. or 0.04 mm) will have a volume equal to 4/3[pi][(0.0008 in.).sup.3] = 0.000000214 [in.sup.3].

Thus, the number of inclusions would equal: 0.000010309 [in.sup.3] / 0.000000002 [in.sup.3] = 5,155 inclusions.

Calculating 762 Micron Inclusions

If an inclusion of 762 microns (0.030 in.) were considered instead of the previous 40 microns size, then the 762-micron inclusion would have a volume of:

4/3[pi][(0.015 in.).sup.3] = 0.0000141 [in.sup.3]

Thus, the number of inclusions would equal: 0.000010309 [in.sup.3] / 0.000014137 [in.sup.3] = 0.73 inclusions.

Rafael Gallo is a product application manager for Foseco Metallurgical Inc., Cleveland.