I’m about to finish my second internship, this time at ASRC Federal in Huntsville, AL. I’ve spent the summer in the Rocket City, contributing a small part to the development of additive manufacturing technologies for rocket engines. Additive manufacturing is a type of 3D printing. The 3D printing familiar to most people is the desktop-sized machine that melts a thin filament of plastic and builds a part layer by layer by depositing a melted plastic bead. The type of 3D printing we’re doing is different. It’s called selective laser melting (SLM), where a laser fires at a bed of metal powder, melting the metal and fusing it together into a solid product. The process is complex and has its challenges, but with additive manufacturing, we can make products with complicated geometry that is impossible to create with conventional manufacturing methods like casting, forging, and subtractive manufacturing. This video from Siemens gives a good general overview of how SLM works (starting around 0:25).
My project this summer has focused on the pre-processing of the files we send to the machines. First, someone uses CAD software to design the part to be printed. Then, the file is converted to STL format. This is because the original CAD file formats use smooth geometries (circles, curves, etc.), and the SLM machines can’t handle that. The curves need to be broken down into many flat surfaces, created by triangles. This is what the STL format does – it uses thousands of small triangles to create a mesh that represents the smooth geometry “close enough” to the original. This is usually fine, but it isn’t perfect. The computer-automated process of converting file formats makes mistakes sometimes, especially when the geometry is especially complex. The parts we’re making are complex with a lot of small, important details that can be difficult for the STL file to accurately represent. Rocket engine parts are complicated and require extremely accurate manufacturing, so this is a problem for us. Fortunately, we have software that helps clean up the STL files before sending them to the machine to be printed. This software has a toolbox of options for us to fix the errors that are introduced during the file conversion. The trouble is that each file is unique and requires a specialized combination of tools to get the fixes right (some “fixes” can actually make the problems worse in some files but work great in others).
I’ve spent the summer experimenting with different fixing procedures on different files, communicating with engineers at the software company, and identifying weaknesses in our current fixing procedures. We’re having some quality issues that are suspected to be a result of problems with the STL files. My work has confirmed that STL errors are contributing to quality issues, and I’ve also developed procedures to address the many types of problems encountered in the file pre-processing required for additive manufacturing.
In the process, I’ve developed a number of new skills. I learned how to identify and assess problems in a pre-existing system, develop solutions to those problems, and document the process. I also learned a little bit about comparative analysis with CAD models, how to adjust the course of a project according to new information, and how to maintain motivation throughout an individual project. Though I had the support of multiple mentors, I was the only intern on this project, and this was the first time I’d had a large project all to myself. It was a little intimidating, but I’ve learned a lot from it.
When I wasn’t working on that project, I was doing tensile testing on GRCop-84 (an alloy invented at NASA Glenn Research Center), helping out with hardness testing for a graduate student intern’s project to develop a heat treatment for Monel K500 (a nickel-copper alloy), and helping with density measurements using image analysis. I was able to use procedures I learned in my MSE 2331 Structures and Characterization lab course to improve ASRC Federal’s image analysis methods.
This has been a productive summer. I’ve learned a lot, and I’ve enjoyed my time in Huntsville. I got the chance to volunteer at the U.S. Space and Rocket Center on a couple of weekends, and I also went on my second tandem skydive this summer! It inspired me to get working on an A license for skydiving so that I can jump by myself someday. I don’t quite feel ready to leave Huntsville yet, but I’m excited to be heading to Houston in a few weeks to start a semester-long internship at the NASA Johnson Space Center. I’m looking forward to applying the skills I’ve learned here to my projects at NASA. For now, though, I still have one week left to say I’m doing rocket science (maybe a bit of a stretch) at ASRC Federal – or at least I’m surrounded by rocket scientists.