Testing 1-2-3: Eliminating Veining Defects
Simulation software and lab testing methods were used to pinpoint why the defect occurs and how to avoid it.
A MODERN CASTING Staff Report
(Click here to see the story as it appears in July's Modern Casting.)
Casting veining or finning defects appear as projections in the form of veins generally occurring perpendicular to the casting surface, either singly or in networks. They are not situated along primary parting lines. Veining can occur in any alloy but is primarily seen in either ferrous-based or copper-based metals. Although various studies have examined the defect, the basic causes have been elusive.
Researchers at the University of Northern Iowa Metal Casting Center recently conducted studies to determine the mechanisms that cause veining defects in iron and steel castings. Advanced testing methods along with computer casting simulations were used to evaluate various sand mixtures for the propensity to form casting veins. The results were shared in the paper, “Causes and Solutions to Veining Defects in Iron and Steel Castings,” by Jerry Thiel and Sairam Ravi of the University of Northern Iowa, Cedar Falls, Iowa.
What sand factors contribute to veining defects and how can the defect be avoided?
According to previous research, the veining defect in cores and molds stems from a tensile stress exerted by a combination of contracting sand at the mold metal interface and subsurface expanding sand. This situation is made possible by the loss of sand volume after reaching 1,063F (573C). The stress is created by an imbalance in the temperature at various points in the sand at differing distances from the liquid metal heat source that creates differences in the thermal expansion and related strain.
When the forces exerted on the mold or core’s surface exceed the high temperature strength at the surface, a tensile mode failure cracks the sand and allows liquid metal to enter. This differentiates the defect from the scabbing and buckling defects that exhibit overlap of sand surfaces from increased surface sand volume.
Much of the veining defect research has emphasized the determination and development of sand additives to reduce or eliminate the defect. Engineered sand additives work by causing two effects. The first takes advantage of a high temperature phase change that occurs in sand at approximately 1,598F (870C). Four main phases of quartz impact veining (Table 1). The first phase is alpha quartz, which is stable from room temperature up to around 1,063F (573C). The second phase is beta quartz. This phase of silica sand is less stable than alpha quartz, and there is a decreased viscosity from a solid indicating some surface softening. This change occurs regardless of binder type. Losses in volume during this stage can range from 50 to 100% of the original length of the sample.
If there is sufficient softening on the surface of the sand grains to create a liquid, tridymite will be formed. Materials such as sodium, lithium or aluminum can force the phase change that is associated with linear change three times as great as the sands original alpha/beta expansion. One Engineered Sand Additive (ESA) that contains lithium has been shown to force tridymite transformation on sands resulting in high temperature increases in volume. This resulting increased volume of up to 12% has been shown to reverse surface strain and thereby effectively eliminate veining defects in iron castings.
Steel castings are poured at higher temperatures and therefore the sand temperatures at the mold metal interface are higher. As the sands treated with ESA are heated above 1,922F (1,050C), they soften and lose volume (Fig. 1). This loss of volume at higher temperatures mimics the strain induced in plain silica sand setting up high surface stresses that cause cracking and veining in steel castings. Although very effective in iron castings, the tridymite phase change and resulting increase in volume occur at too low of a temperature to prevent veining in steel castings. Looking at the expansion of silica sands which have additions of iron oxide, it can be seen that although they experience softening, they do not change phase from beta quartz to beta tridymite. The loss of volume continues with increased temperatures until a change to the fourth phase of silica.
This phase consists of the change from beta quartz to beta cristobalite and is associated with a volume change up to 14.7% occurring around 2,678F (1,470C). This increase mimics the effect of ESA at iron casting temperatures and reduces veining by reducing the tensile stress induced at the surface of the core or mold through expansion. Along with the cristobalite transformation and associated expansion, the sintering point of the sand decreases. Unfortunately, the reduction of veining with the use of iron oxide may come at the high cost of room temperature tensile strength.
The lack of data on the elevated temperature performance of unbonded sand is most likely due to the lack of the availability of appropriate equipment and methods to allow the thermal expansion of unbonded sand to be measured. To address the problem of measuring the thermal expansion of foundry sands, the University of Northern Iowa’s Metal Casting Center designed and built an updated dilatometer, capable of accurately measuring the thermal expansion characteristics of heterogeneous foundry sands. The dilatometer was designed robustly to provide the temperature range and degree of accuracy required. Thermal expansion tests were run on bonded sand samples utilizing the university’s high temperature aggregate dilatometer (Fig. 2).
An experimental method to measure surface softening of small solid particles heated to high temperatures uses a dilatometer to measure the rate of deformation of a granular material and calculate the viscosity. The calculated viscosity is useful in describing the sintering characteristics of a granular material. An extension to the this method was developed to better understand the surface softening of sand particles. The surface viscosity measurement is based on the compaction of the sand particles while under a compressive load and constant heating rate. Sand particles, being porous, initially expand with temperature but subsequently contract due to softening and sintering at inter-granular contact points. This behavior results from surface softening and deformation where the compressive load is concentrated.
Since veining defects are common in silica sand cores and/or molds, a casting simulation software model was created based on the expansion and contraction of silica sand as it was heated up to a high temperature to simulate surface strain.
Step cone iron and steel castings produced in molds consisting of various additives were tested and compared with the predictions of the casting simulation software.
3. Results and Conclusions
Results from the tests showed that two forces act on the surface of the sand that can either contribute to or reduce the veining defect. The linear expansion of bonded silica sand causes the volume of sand to increase sharply until 1,063F (573C), where it changes phase from alpha quartz to beta quartz. Upon further heating, the sand loses its volume due to loss of binder volume and softening and rearrangement at the surface of the sand grains. This loss of volume at temperatures above 1,063F is the main cause of veining defects. As the temperature of the mold or core surface increases, the length and volume of the sand decrease. The cooler sand directly beneath the surface increases in volume as it passes through the alpha to beta quartz transformation. The combination of contracting sand on the surface with expanding sand directly beneath the surface creates tensile failures that fill with liquid metal forming the defect classified as veining. Sand additives that reduce veining defects provide liquid on the surface of the sand grain and favor formation of tridymite or cristobalite and greater expansion of the sand. This secondary expansion reduces the negative strain at the surface of the core and prevents tensile failure and the associated cracks. Through the fluxing action of sand additives, the surfaces of the particles adhere to each other, increasing the tensile strength of the sand on the surface of the core or mold. This increase in strength reduces the tensile failure on the sand’s surface and reduces veining defects.
The accepted methods of reducing veining defects in castings are:
Using low expansion aggregates including chromite, zircon, olivine and ceramics to eliminate the difference of expansion rates of surface and subsurface sections of cores and molds. These materials generally exhibit linear thermal expansion and little if any phase transformations. Their refractory value is higher than silica sand, and therefore they show minimal softening and volume loss. The strain values at the mold metal interface closely match the subsurface strain values and therefore eliminate any mechanical forces that would cause tensile failures and veins. Blends of low expansion aggregates with silica sand have been successfully used to reduce or eliminate veining defects.
Using sand additives containing fluxes, such as iron oxide and lithium-based products. Fluxes decrease the temperature at which the silica starts to soften and provides liquid on the surface of the grains, increasing the reactivity and lowering the transition temperatures for tridymite and cristobalite. These transitions force increases in volume of the subsurface sand and reduce the vein strain on the surface of the core or mold. These fluxes can also sinter the sand together at high temperatures, effectively increasing their resistance to tensile failures as shown by increasing viscosities of bonded sands.
Using sand additives containing organic materials such assaccharides or dextrin to provide a slight cushioning effect as the sand goes through the alpha/beta transformation, but mainly to act as a carbon source for a high temperature bonding of the sand. Any available oxygen in the mold cavity is depleted shortly after it fills with liquid metal. In the absence of oxygen, the organic materials break down to primarily carbon, which bonds to the surface of the sand grains, increasing the viscosity and tensile strength of the surface sand. This increase in tensile strength resists the vein strain and reduces veining defects. Often these materials are blended with fluxes and oxides to increase their effectiveness.
This article is based on a paper (14-030) presented at the 2014 Metalcasting Congress.