Skip to main content

Thermal images collected during large-scale 3D printing with Ingersoll MasterPrint

  • Villez, Kris | Oak Ridge National Laboratory
  • Tomlinson, Scott | University of Maine Advanced Structures and Composites Center
  • McGlone, Aidan | University of Maine Advanced Structures and Composites Center
  • Bailey, Ben | University of Maine Advanced Structures and Composites Center
  • Chase, Tyler | University of Maine Advanced Structures and Composites Center
  • Chowdhury, Dhruba | Oak Ridge National Laboratory
  • Jacobs, Nicholas | University of Maine Advanced Structures and Composites Center
  • McClain, Monique | Purdue University
  • Slattery, Lucy | University of Maine Advanced Structures and Composites Center
  • Webster, Morgan | University of Maine Advanced Structures and Composites Center
Download dataset
Overview

Description

This data set contains images produced to test the performance of anomaly and fault detection methods in the context of additive manufacturing. Two print jobs were executed using the Ingersoll MasterPrint with the Model 30 Strangpresse extruder. The material used was Techmer compounded polylactic acid (PLA) with wood flour as a filler (80/20 PLA/WF by weight). The first print job consists of the 3D printing of a hexagonal cylinder with a two-bead wall. This print was sliced at a gantry velocity of 3000 millimeter per minute, and an extruder screw speed of 68.14 rotations per minute. This are considered the normal operating conditions. A second hexagonal cylinder was printed using a reduced extruder screw speed, 15% lower than under normal operating conditions. The images were collected with a Teledyne FLIR Lepton 3.5 infra-red camera, a small form factor radiometric long-wave infrared camera with a spectral range of 8 µm to 14 µm. Sensor resolution was 160x120 pixels, with a pixel size of 12 µm, a temperature range of -10 - 450°C, and an accuracy of +/- 10°C in its low gain configuration. Images in this data set were collected with cameras oriented at the printer nozzle. The nozzle camera setup consisted of two cameras located 12.5cm from the nozzle center. These were mounted directly to the print head, so that the camera positions relative to the print direction would remain constant as it rotated around its C-axis to follow the print path. One camera was placed ahead of the nozzle to capture the previous layer immediately before being covered by the new layer of material, while the second camera was placed behind the nozzle and captured the freshly extruded bead. Images are collecting during each phase of the print: (a) idle (i.e., no material deposited), (b) extrusion (i.e., to prime the extruder), and (c) printing (deposition of material to manufacture the hexagonal cylinder).

Funding resources

DOE contract number

DE-AC05-00OR22725

Originating research organization

Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Other contributing organizations

University of Maine, Purdue University

Sponsoring organization

Office of Science (SC)

Details

DOI

10.13139/OLCF/2441473

Release date

September 25, 2024

Dataset

Dataset type

IP Still Images or Photos

Other contract number(s)

4000174848

Acknowledgements

Papers using this dataset are requested to include the following text in their acknowledgements:

*Support for 10.13139/OLCF/2441473 is provided by the U.S. Department of Energy, project N/A under Contract DE-AC05-00OR22725. This research used resources of the Oak Ridge Leadership Computing Facility, which is a DOE Office of Science User Facility.

Category

  • 36 MATERIALS SCIENCE

Keywords

  • Additive manufacturing,
  • Composite materials,
  • Machine vision