Beata Bączkowska
Additive Manufacturing Engineer — LPBF · Materials & Process Engineering
Warsaw, Poland
Additive Manufacturing Engineer specializing in LPBF process development and qualification for aerospace structural components. Responsible for developing and qualifying printing parameters for Ti-6Al-4V and AlSi10Mg on EOS platforms, designing support strategies for complex geometries, and leading AS13100-compliant qualification programs. Increasing focus on in-situ process monitoring and defect detection.
Expertise
- Laser Powder Bed Fusion (LPBF) process development
- Ti-6Al-4V and AlSi10Mg parameter optimization
- build orientation and support strategy
- AM part qualification per AS13100
- in-situ process monitoring
Technologies
Work History
2025-01
In-situ melt pool monitoring implementation — connected EOS M290 in-situ monitoring outputs (melt pool intensity, area, temperature) to a Python analysis pipeline for real-time layer-by-layer defect detection.
Challenge: Melt pool monitoring signals have high variability even in defect-free builds — distinguishing genuine defect signatures from process variation required training a classifier on labeled reference builds with known defect locations (confirmed by CT after printing).
Learned: In-situ AM monitoring classifier training requires CT-confirmed ground truth for both defect and defect-free examples. Training on monitoring signals alone without CT validation produces classifiers that learn process variation patterns rather than defect signatures.
2024-04
AS13100 qualification program for additively manufactured bracket — coordinated coupon manufacturing, mechanical testing, CT inspection, dimensional verification, and surface roughness measurement across 3 build campaigns.
Challenge: CT scan defect acceptance criteria for LPBF parts were not defined in AS13100 — the standard references conventional casting defect limits that do not translate directly to LPBF porosity morphology and distribution. Had to develop internal CT acceptance criteria in collaboration with the quality team.
Learned: AM qualification requires developing new inspection acceptance criteria, not just applying existing casting or machining standards. The defect types, sizes, and distributions from LPBF are fundamentally different from conventional manufacturing — the qualification program must establish part-specific acceptance thresholds.
2023-09
Support structure design optimization for a complex impeller geometry with overhanging vane surfaces — iterated support topology using Materialise Magics to minimize support material while maintaining dimensional accuracy and preventing warping.
Challenge: Thin vane trailing edges (<0.5mm) were distorting during building due to thermal stress despite support structures. ANSYS Additive Print predicted distortions of 0.3mm. Required a compensation strategy: pre-distort the nominal geometry in the opposite direction so post-build distortion lands on the nominal shape.
Learned: Distortion compensation for thin AM walls requires a validated thermomechanical simulation. Pre-distortion without simulation is guesswork — the distortion magnitude and direction are non-intuitive for complex 3D geometries.
2023-03
LPBF process parameter development for Ti-6Al-4V structural bracket — designed a 25-run DoE (layer thickness, scan speed, laser power, hatch spacing) on EOS M290. Optimized for density >99.9% and tensile strength >1000 MPa.
Challenge: Keyhole porosity appeared at high energy density combinations that had given good density in literature sources — our EOS M290's laser focus mode was different from the reference study. The optimal parameter window was 15% narrower than expected, requiring a finer DoE resolution.
Learned: LPBF process parameters are machine-specific, even for the same model. Differences in laser focus mode, scan strategy, and gas flow uniformity make literature parameters a starting point only. Always verify on the specific machine with the specific powder batch before narrowing the DoE.