Dimensional Tolerances of 3D-Printed Sand Cored Iron Castings
The use of 3D-printed sand casting is growing in the production environment, and the initial feedback is that the process results are comparable to the precise sand casting processes. Currently, 3D sand printing is mainly used in production via a hybrid approach, where the mold is made with the conventional green sand process and the complex core assembly is redesigned with a consolidated 3D-printed core. However, little has been studied and is known in the public domain about the dimensional tolerances achieved with this toolingless precision sand casting process, especially the potential of achieving truer position and better internal feature tolerances.
A recent AFS research project aimed to identify and provide guidelines for improved dimensional tolerances with 3D-printed sand iron castings to design engineers.
Generally speaking, the metalcasting industry intelligentsia does know some facts about 3D printing and production. We know 3D printing can allow us to change shapes quickly while prototyping. We know 3D printing all but eliminates the need for tooling storage. And we know 3D printing allows us to make shapes we wouldn’t be able to make with a single core with conventional tooling.
What we don’t know is the overall dimensional capabilities of 3D-printed cores and whether the industry can turn added control into added value. What if designers can reduce machine stock, in turn making a casting that is cheaper to machine? What if foundries can ensure minimum wall thickness with a thinner target value that would lead to making a lighter finished part?
Using side-by-side trials and dimensional studies, the current research study set out to test and quantify the capabilities of 3D sand-printed cores to eventually enable design engineers to achieve lightweight, quality parts with confidence.
Ultimately, the study’s intent is to give OEM design engineers the standard process capability of 3D-printed sand molds and cores. Table 1 compares the capabilities of certain casting processes according to ISO 8062. As shown in Table 1, conventional green sand casting spans grades 8-14, while investment casting ranges from grades 4-9. It was expected that the results for 3D-printed sand cores would live somewhere between, from grades 7-9. This gives engineers the ability to make the whole design lighter with better properties.
A complex iron housing already in production at Waupaca Foundry was chosen for the study. It is produced using a six-piece core assembly, which is made via the coldbox chemically bonded process with complex hard tooling. The assembly is inserted into a green sand mold on a vertically-parted high pressure molding line.
This part has been in production for several years, so the researchers had a wealth of CMM measurement data to compare the 3D-printed process to the conventional process.
For the study, the six-piece core assembly used in the conventional sand casting process with conventional tooling was redesigned as a three-piece printed core. Production runs of 25 sets of cores were conducted at five participating 3D printing core suppliers:
- Hoosier Pattern
- Humtown Products
- University of Northern Iowa—Additive Manufacturing Center.
The cores were primarily made of silica sand, furan bonded, plus one set each of ceramic sand, furan bonded, and silica sand with a phenolic binder.
The cores were washed in select areas, similar to the existing practice. And the castings were poured and processed in the same production environment as the conventionally-cast part with the six-piece core assembly.
As part of the study, the following build and machine parameters were captured from each 3D-printed core assembly provider:
- Layer thickness, build rate/time.
- Binder type and % addition.
- Substrate—silica or ceramic with GFN, shape, purity, etc.
- Strength, permeability and LOI of 3D-printed core material.
- Build box layout.
- Type of machine used.
The idea was to see if there is a correlation with any machine parameters impacting overall dimensional tolerance.
Each printed core was serialized for traceability to maintain the one-to-one relationship to the castings poured. And each core was inspected by the core suppliers for overall dimensions before shipping.
Once all 150 3D-printed cores were produced and delivered to Waupaca Foundry, a total of 175 castings were poured—25 with the current conventional six-piece assembly. Process parameters were captured for each mold, and every casting was inspected for critical dimensions by CMM.
Statistical analysis of the measured dimensions was performed, and overall dimensional assessment with regard to repeatability, reproducibility, and variability/tolerance were conducted and then compared to the conventional 6-piece practice.
Residual stress and distortion were estimated using production conditions with 3D-printed furan cores, and the researchers compared the key dimensions with the actual.
Overall, the study indicated that 3D-printed sand cores can be predicted to be one grade tighter in dimensional tolerance over conventional sand castings.
The ceramic substrate was more dimensionally stable in the project, but more data is required to quantify the properties.
The 3D-printed core assembly performed better than the conventional assembly on some critical features, such as the oil cooler cored hold opening. However, the spread of dimensions was larger than expected—partly due to other factors, such as desanding/core handling/coating, which are manual post operations after 3D printing.
Further research is needed to quantify the root cause of the dimensional variability in 3D sand printing, along with the exploration of automating the desanding process to reduce the piece-to-piece variability.
More results and data from individual foundries are needed to make a case to the ISO 8062 community to add separate linear dimensional casting tolerance grades 5-7 for 3D-printed precision sand casting.
This research is sponsored by the AFS 3D Sand Printing Technical Committee under the Additive Manufacturing Division of AFS and funded partially by AFS with in-kind cost share support from Waupaca Foundry, PDA LLC and 3D-printed core suppliers ExOne, Voxeljet, Hoosier Pattern, Humtown Products, UNI-AM Center, and Emerson’s AM Center.