EXPERIMENTAL GENERATION OF SUPPORTS BASED ON LIGHTNING INFILL PRINCIPLES FOR FDM 3D PRINTING

This paper presents the development and experimental validation of a new type of supports for additive manufacturing using the FDM/FFF method — Lightning Support. The approach originated as a logical evolution of the Lightning Infill method, where the idea of branching into “critical” zones was transferred from internal infill to external support structures. The relevance of this work is justified by the drawbacks of traditional support generators such as Grid, Lines, and Tree, which often lead to material overuse, increased printing time, and deterioration of surface quality at contact areas. The proposed method was implemented on the open Cura/CuraEngine architecture, enabling the creation of a dedicated support generator module with configurable geometry parameters. The algorithm includes: detection of overhangs, formation of anchors on the build plate and side anchors on model walls, branch generation with angle and collision control, adaptive adjustment of horizontal and vertical gaps, generation of interface layers, as well as heuristics for optimizing the number of branches and fallback to Grid/Tree in complex cases. Comparative experiments on benchmark models demonstrated reduced printing time compared to Grid and improved detachment predictability compared to Tree supports. At the same time, material consumption remained moderate relative to models without supports. Limitations of the method were observed in the case of long bridges and steep overhangs, where fallback or parameter adjustments are required. The discussion provides tuning guidelines, proposes hybrid strategies combining Lightning and Tree, and outlines perspectives for optimization, including the use of path-planning algorithms, adaptation to cooling airflow, and machine learning methods for support selection. The practical value of the approach lies in saving time and energy, preserving surface quality, and simplifying removal, making Lightning Support highly suitable for rapid prototyping and educational tasks. It is concluded that the developed method provides a balanced compromise between speed, quality, and cost-effectiveness, is technically integrated into Cura/CuraEngine, and is suitable for hybrid printing profiles. Future work should focus on large-scale testing with various materials, high-speed printing modes, and standardization of profiles for open benchmarking.