A visual process simulation for novice users of desktop 3D printers to reduce part defects
Date
2023-12
Authors
Renner, Alex Raymond
Major Professor
Advisor
Winer, Eliot
Dorneich, Michael
Frank, Matthew
Li, Beiwen
Stone, Richard
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Abstract
Over the past decade, the emergence of the desktop 3D printing sector has revolutionized the Additive Manufacturing (AM) industry with record sales of goods and services. While the AM industry has enjoyed consistently high growth rates every year since first recorded in 1988, the first sustained years of exponential growth coincided with the first sales of desktop 3D printers (those that cost less than $5,000). Manufacturers of low cost pre-assembled “kits” targeted the consumer market, helping 3D printing to go mainstream. Almost overnight, the general public knew what 3D printers were and how to purchase them. While desktop 3D printers were marketed as easy-to-use, they did not produce parts as reliably as industrial AM systems, and many of the claims to consumers fell short of expectations. Desktop 3D printers produced parts with noticeable visual defects defined by consumers who had little or no 3D printing experience. Customer support for novice users did very little to reduce or eliminate defects in printed parts or resolve complaints about how long it took to print parts. While desktop 3D printer manufacturers worked to increase the usefulness of their machines, making it easy for novice users to print is still a significant obstacle. Many of the manufacturers provide software offerings to help users choose print settings to reduce errors, as well as the number of times a part must be printed before it is usable. However, these offerings often present a myriad of settings to adjust without a clear indication of the outcomes. These settings must be simulated and presented so a user can understand and avoid common print failures.
A review of defect reduction approaches was performed to determine if methods used in commercial or open-source software could be implemented in a simulation. Even if the cost of commercial software could be reduced, process simulations often performed Finite Element Analysis (FEA) requiring solid CAD models that many novice users would not have (i.e., most users download 3D mesh models from the internet). Open-source software that analyzed 3D mesh models for problems that would cause defects in printed parts (e.g., holes in the 3D model), required users with previous 3D printing experience to understand the analyses. Furthermore, the predominant cause of part defects is from printing processes which requires a 3D printing expert to fix by modifying print settings. Academic research has shown that multiple print settings relate to part quality, by evaluating the mechanical properties of parts produced without defects. Meanwhile, consumers continue to struggle to print parts without defects, since software offerings do not provide features that relate print settings to part quality. Reducing part defects would allow greater accessibility of 3D printing to consumers who want to utilize this manufacturing technology to produce their own parts. A new method to help novice users understand print settings was needed utilizing easy-to-understand visualization techniques in a software simulation of the printing process.
The first iteration of the simulation was developed to apply color to extruded shapes according to the type of print settings utilizing G-code exported from 3D printing software. In this iteration, the base application framework was developed for a virtual reality (VR) system, as well as a cross-platform desktop application. The simulation contained all the components of an example 3D printer, moving as they would during a physical print job, with the extruded shape be displayed for each print movement. A user can rotate, pan, and zoom into the 3D models of the printer and printed part, and start or stop the print process at any time. The fully immersive VR aspect of the simulation provided users with a unique perspective of the 3D printing process, allowing examination of extruded shapes on a large scale in collaborative environment.
To help users relate print settings to print quality the second iteration of the simulation was developed with a real-time thermal model to address the most prevalent issues encountered during printing. An extrusion process was simulated by displaying a printed road shape during and after a G-code print command that accurately represents the position, size, and shape of a 3D printed part’s roads, based on changes in nozzle position, orifice diameter, and layer thickness. The Lumped Capacitance heat transfer method was used to determine each road’s temperature with a corresponding color legend. A thermal imaging camera attached to the print head of an actual 3D printer monitored temperature for simulation validation. This simulation, in conjunction with a usable 3D software interface, provides an easy-to-use tool for novice 3D print users to explore setting tradeoffs and view resultant outcomes with any combination of 3D printer and 3D printing software. The software was evaluated based on its ability to help a user fix defects in parts with geometric features that are challenging to create using AM. After printing three different parts with defects, the software was used to identify print settings that would keep roads above the glass transition temperature in and around potential defects. These settings were used to successfully print the parts without defects.
The thermal model used in the simulation has been validated for three test parts with geometric features that are challenging to produce without defects, for any combination of 3D printing hardware, material, and software. As fewer machine manufacturers produce their own software, it was even more important to validate the accuracy of road temperatures in the simulation with respect to temperatures from a thermal imaging camera attached to the print head of a low-cost 3D printer. Methods to simulate any model and type of 3D printer were outlined, in conjunction with improvements in G-code parsing methods that expand the simulation’s use to include the consumer’s 3D print preparation software preference. The road temperatures calculated in the simulation were compared to thermal imaging camera image data for all three test parts using default print settings. Print settings explored in the simulation were then used to print parts with reduced defects. The road temperatures recorded using the thermal imaging camera validated the simulation temperatures, before and after print settings were explored. Experimental results confirm the use of the simulation application as a decision tool for users of desktop 3D printers, to reduce or eliminate defects before and after 3D printing parts.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Mechanical Engineering
Type
article