High Resolution Imaging for DMLS Inspection

Transcription

1 High Resolution Imaging for DMLS Inspection Institute of Imaging and Computer Vision RWTH Aachen University Germany 1

2 Agenda The BIGS Research Project Image Acquisition Setup Application Example: Material Qualification Image Analysis Outlook Dipl.-Ing. Stefan Kleszczynski Dipl.-Ing. 2

3 BIGS Research Project: Objectives Proof of Part Quality and Errors Documentation Understanding Optimization 3

4 BIGS Research Project: Approach Feature Extraction Classifier CCD Camera EOSINT M 270 Part CAD Data Reference Database Reference Image Data Material 4

5 Image Acquisition Setup EOS EOSINT M 270 5

6 29 megapixels Camera Pixel: 5.5 µm x 5.5 µm (sensor: 36 mm x 24 mm) 6

7 Weld Seams: 90 µm 1 pixel: µm 7

8 Application Example: Material Qualification Identification of optimum beam expander position 1 2,5 4 0,5 2 3,5 0 1,5 3 8

9 Application Example: Material Qualification Poor compound of melt traces Smallest position 4 2 Chosen setting 0 9

10 Laser Scanning Velocity High Resolution Imaging for DMLS Inspection Application Example: Material Qualification Laser Power A B C D E 10

11 Application Example: Material Qualification Hatch 11

12 Image Analysis: Sample Build +20 % +40 % -20 % -40 % Scanning velocity Laser power Hatch : Increased energy input : Decreased energy input Reference Skip 1 layer 2x exposure Red. overlap 12

13 Impact on Tensile Strength Volume energy density E v : E v = P l h v s d P l : laser power h: hatch distance v s : scanning velocity d: layer thickness Kleszczynski et al. - Mechanical properties of Laser Beam Melting components depending on various process errors 13

14 Microscopic Analysis vs. Imaging System R m = 868 MPa R m = 833 MPa R m = 813 MPa R m = 892 MPa 14

15 Microscopic Analysis vs. Imaging System R m = 868 MPa R m = 833 MPa R m = 813 MPa R m = 892 MPa 15

16 Sample Build: Elevation of Contour Regions Power +40 % Power -40 % 16

17 Sample Build: Elevation of Contour Regions Power +40 % Power -40 % 17

18 Image Analysis: Layer Comparison Manual analysis of multiple layers Synchronized display of selected regions z/mm 18

19 Automatic Detection of Differences High Resolution Imaging for DMLS Inspection Texture Analysis

20 Outlook Detection of areas with low energy input Construction of knowledge database, link images to: process parameters part properties 20

21 Summary Inspection of powder and melt result at microscopic scale (weld seams) Existing machines can be equipped with imaging system Images enable analysis of part quality Development of automatic image analysis 21

22 References Kleszczynski, S.; zur Jacobsmühlen, J.; Sehrt, J. T. & Witt, G. Mechanical properties of Laser Beam Melting components depending on various process errors. NEW PROLAMAT (to appear) zur Jacobsmühlen, J.; Kleszczynski, S.; Schneider, D. & Witt, G. High Resolution Imaging for Inspection of Laser Beam Melting Systems. I2MTC (to appear) Kleszczynski, S.; zur Jacobsmühlen, J.; Sehrt, J. T. & Witt, G. Error Detection in Laser Beam Melting Systems by High Resolution Imaging. Proc. 23rd Solid Freeform Fabrication Symposium,

23 High Resolution Imaging for DMLS Inspection Contact Details Dipl.-Ing. Stefan Kleszczynski Tel Institute for Product Engineering University of Duisburg-Essen Duisburg, Germany Dipl.-Ing. Tel Institute of Imaging and Computer Vision RWTH Aachen University Aachen, Germany 23