Conveyor Dynamics: Turn slag waste into money. Learn more about our new slag quenching conveyor system.

Design Considerations for 3-D-Printed Cores and Molds

Developments in additive manufacturing mean engineers have a new set of considerations when designing molds and cores.

Kip Woods and Sairam Ravi, Univ. of Northern Iowa, Cedar Falls, Iowa

(Click here to see the story as it appears in the May issue of Modern Casting.)

The growth of 3-D printing capabilities has become a hot topic in manufacturing. For metalcasters in particular, what was a way of producing prototypes and niche tooling has grown into a process capable of producing large numbers of molds and cores with excellent dimensional accuracy and mechanical and physical properties.

Trials with a 3-D sand printer were conducted at the Univ. of Northern Iowa to explore different design aspects that have to be considered before printing sand molds or cores. Although the research was conducted on a particular 3-D printer, the design considerations discussed in this paper can be used effectively for sand additive manufacturing technology on the whole.  

The production of molds and cores with 3-D printers offers advantages compared to conventional molding methods. A design engineer has more freedom in designing parts and does not have to consider limitations such as drafts or undercuts. Also, multiple-piece parts in nobake or coldbox systems can be printed in a single piece with a 3-D printer to reduce stacking tolerances. The finished parts also have excellent dimensional accuracy and mechanical properties, due to the bonding of sand at every thin layer.

Design Considerations

Once solidification modeling is complete, the initial positioning or orientation of the part can be determined. In the trials, solidification design was run using a commercially available simulation software package for the metalcasting industry. A solidification simulation was first run on the casting to determine possible porosity. Risers were placed at appropriate spots and the solidification process was repeated until a sound casting was obtained. The rest of the casting system was designed and a filling simulation was run to determine metal velocity and air entrapment.

After the initial casting system design was completed, different design considerations were taken into account for 3-D-printed cores and molds. Several projects from the Univ. of Northern Iowa illustrate the different design considerations.

3-D printing can print complex cores in single pieces, rather than a multi-part assembly. Figure 1 shows a core for a train’s airbrake that was produce in eight different sections in a coldbox setting. Previously, this part was produced by gluing eight sand cores together. Now, with the capabilities of the 3-D sand printing technology, this part can be printed as a single core (Fig. 2). This process also reduces labor and stacking tolerances, while improving the part’s dimensional accuracy. An important point to remember in the conversion process is to provide or design a method to remove the uncured sand from within the cores themselves, often using compressed air or vacuum.

Critical Surface Finish in the Drag

3-D printing also allows designers to place the 3-D model in any desired orientation. This allows for specific placement of critical surface finish sections in the drag. By placing a critical surface face down in the drag, the inclusions and air bubble difects will rise to non-critical finish areas higher in the casting.

Placement of Risers

While traditionally produced molds and cores need draft and do not allow for undercuts, with 3-D printing, risers can be placed in the ideal location for a sound casting. Figure 3 shows the porosity results from a solidification model of the sculptural table legs with no risers or gating system. In Figure 4, a riser was placed centrally to aid in feeding the legs of the table. Risers also were needed at the top of the legs to remove porosity present in the initial solidification (Fig. 4).

Vent Placement

Venting in 3-D designed molds is a relatively simple process because the need to integrate vents into a hard tooling pattern is eliminated. Vents are created in the 3-D mold model, which allows for improved precision and efficient placement of the vents based upon the air entrapment and air pressure results obtained from the filling simulation.
Unbonded Sand Removal &

Parting Lines

While the main application for 3-D sand printers is industrial part production, this technology can be used to create new works of art. Dan Perry, a sculptor from Northern Iowa’s art department, collaborated with the Metal Casting Center to produce a piece, shown in Figure 5, which highlights the complexity of 3-D printed parts while showing the limitations of sand removal from complex designs.

The traditional method of using horizontal cutting planes proved to be insufficient for complete removal of uncured sand, which can lead to casting defects. In this case, the residual sand had prevented the normal passage of the molten metal, leading to a misrun defect in that particular area (Fig. 6). A large amount of unremoved sand in the casting also can act as a local chill in certain areas, especially those with thinner sections. This misrun was due to ato insufficient cleaning techniques, which prevented the removal of loose sand after printing.

Due to the complex nature of the feather mold, traditional parting lines weren’t sufficient for proper sand removal. Surface contour offset parting lines were designed to allow complete access to the inside of the mold for sand removal. The surface offset parting line also eased refractory coating application to the final surface. Vertical and horizontal parting lines were designed to split the runner in half to allow complete removal of all uncured sand from the runner and gates, reducing the possibility for uncured sand to be in the mold.

Vertical parting lines introduce the possibility of metal runout during pouring. This risk can be reduced by two ways: having sufficient sand volume below the runner to quickly reduce any runout to solidus temperature and/or having sufficient mold overlap with drafting for a near-perfect mating surface. The interlocking surface traps any runout in place and the large volume of sand around the stationary runout quickly freezes the metal. The yellow path in Figure 7 is the short distance of travel at 6.8 in.

Bonded sand is the most expensive consumable required for 3-D printing. 3-D printing facilities often charge based upon the mold volume of the bonded sand. The flexibility of being able to design and print the 3-D model allows design engineers to efficiently use sand. Figure 8 shows the original mold before removing sand from non-critical areas. It weighed 124.68 lbs. (56.67 kg). Figure 9 shows the reduced mold after having sand removal techniques applied with a weight of 113.16 lbs. (51.43 kg), a weight reduction of 10.18%. Mold weight reduction allows for easier assembly and reduces the shipping cost of molds and cores.

Orientation of Parts

Part orientation to the printing head ensures the part’s final dimensional accuracy, especially in smaller cores with tight tolerances. When the Z layer is too thick, it produces a stepped contour that reduces the roundness of circular features (Fig. 10). This jagged appearance cannot be entirely eliminated by decreasing the layer thickness, but can be corrected with proper orientation to the printing head.

As the layer thickness decreases, the roundness of the part will increase. The resolution along the X, Y and Z axes and the orientation of the part based on the required tolerances are especially critical in smaller parts with tighter tolerances.

Model degradation is unavoidable when printing complex parts in 3-D printers. Figure 11 shows a sample triangle with a Z layer thickness of 0.011 in. (0.28mm) and an angle incident of 15 degrees. The printed layers are green, light blue and blue. These layers are the geometry that remain after printing. The red sections are not printed and the cause of model degradation.

The amount of model degradation is based upon the angle of incident. Figure 12 shows an example of the same simple triangle in Figure 11, but the angle of incident is increased to 30 degrees. The amount of unprinted material is reduced to 0.018 in. (0.48mm) wide instead of the 0.04 in. (1.04mm) in the first example. This is an area of continuing research as there currently isn’t a method to predict the amount of model degradation based on part geometry and Z layer resolution.

When any new technology comes into production, limitations must be found before the full capabilities of that technology can be exploited for the greatest gain. With 3-D printing technology, the freedom of being able to think outside the box is certainly an advantage, but a few limitations must be and currently are being addressed for the technology to be more widely adopted. 3-D printing will never entirely replace traditional mold production methods. However, it can reduce core stacking tolerances and increase dimensional accuracy. These printed cores can be integrated into existing production tooling, allowing metalcasting facilities to produce better castings more affordably.   

EZX Wheel