Lorenzo Pedrolli

Many industrial processes require a consistent material flow in the form of powders, typically achieved through pneumatic conveying. This work presents a dataset of high-speed video recordings capturing the horizontal pneumatic conveying of metallic powders in thin capillary pipes. Additionally, pressure and photodiode recordings are included. The experimental setup is described in detail, with suggestions for potential improvements. In the video recordings, each particle is individually distinguishable and can be tracked across frames.

Directed Energy Deposition (DED) processes necessitate a consistent material flow to the melt pool, typically achieved through pneumatic conveying of metal powder via thin pipes. This study aims to record and analyze the multiphase fluid–solid flow. An experimental setup utilizing a high-speed camera and specialized optics was constructed, and the flow through thin transparent pipes was recorded. The resulting information was analyzed and compared with coupled Computational Fluid Dynamics-Discrete Element Modeling (CFD-DEM) simulations, with special attention to the solids flow fluctuations. The proposed methodology shows a significant improvement in accuracy and reliability over existing approaches, particularly in capturing flow rate fluctuations and particle velocity distributions in small-scale systems. Moreover, it allows for accurately analyzing Particle Size Distribution (PSD) in the same setup. This paper details the experimental design, video analysis using particle tracking, and a novel method for deriving volumetric concentrations and flow rate from flat images. The findings confirm the accuracy of the CFD-DEM simulations and provide insights into the dynamics of pneumatic conveying and individual particle movement, with the potential to improve DED efficiency by reducing variability in material deposition rates.

Directed Energy Deposition (DED) processes necessitate a consistent material flow to the melt pool, typically achieved through pneumatic conveying of metal powder via thin pipes. This study aims to record and analyze the multiphase fluid-solid flow. An experimental setup utilizing a high-speed camera and specialized optics was constructed, the flow through thin transparent pipes was recorded. The resulting information was analyzed and compared with coupled Computational Fluid Dynamics - Discrete Element Modeling (CFD-DEM) simulations, with special attention to the solids flow fluctuations. The paper details the experimental design, video analysis using Particle Tracking, and a novel method for deriving volumetric concentrations and flowrate from flat images. The findings confirm the accuracy of the CFD-DEM simulations and provide insights into the dynamics of pneumatic conveying, which could be used to enhance the efficiency of DED processes.

Pneumatic conveying of fine powders is essential for many industrial processes, including Laser Metal Deposition (LMD), a Direct Metal Additive Manufacturing (DMAM) technology that builds solid objects layer-by-layer using a laser to melt metal powder. To optimize the process, it is necessary to have a correct understanding of the powder’s behaviour under the process condition. The coupled Computational Fluid Dynamics - Discrete Element Modelling (CFD-DEM) and MultiPhase - Particle In Cell (MP-PIC) are two popular Eulerian-Lagrangian models to simulate particle laden flows. This study compares them to analyse powder behaviour in a small channel of LMD machines. Results from the two methods differ significantly, with CFD-DEM offering a more accurate representation of the physical reality, while MP-PIC is more computationally efficient. The study finds that the CFD-DEM method produces higher fluctuations in the solids flow rate due to the formation of particle clusters, while MP-PIC displays a smooth and essentially uniform flow. The results suggest that CFD-DEM should be used for more accurate and detailed studies of solids flow rate in pneumatic conveying systems, while MP-PIC can be used for preliminary studies and design optimization.