Simulation of metal powder packing behaviour in laser-based powder bed fusion
Haapa, Erik (2023-03-12)
Simulation of metal powder packing behaviour in laser-based powder bed fusion
(12.03.2023)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
avoin
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2023031531775
https://urn.fi/URN:NBN:fi-fe2023031531775
Tiivistelmä
Laser-based powder bed fusion (L-PBF) is a method of additive manufacturing, in which metal powder is fused into solid parts, layer by layer. L-PBF shows high promise for manufacture of functional Tungsten parts, but the development of Tungsten powder feedstock for L-PBF processing is demanding and expensive. Therefore, computer simulation is explored as a possible tool for Tungsten powder feedstock development at EOS Finland Oy, with whom this thesis was made.
The aim of this thesis was to develop a simulation model of the recoating process of an EOS M 290 L-PBF system, as well as a validation method for the simulation. The validated simulation model can be used to evaluate the applicability of the used simulation software (FLOW-3D DEM) in powder material development, and possibly use the model as a platform for future application with Tungsten powder. In order to reduce complexity and uncertainties, the irregular Tungsten powder is not yet simulated, and a well-known, spherical EOS IN718 powder feedstock was used instead.
The validation experiment is based on building a low, enclosed wall using the M 290 L-PBF system. Recoated powder is trapped inside as the enclosure is being built, making it possible to remove the sampled powder from a known volume. This enables measuring the powder packing density (PD) of the powder bed. The experiment was repeated five times and some sources of error were also quantified. Average PD was found to be 52 % with a standard deviation of 0.2 %.
The simulation was modelled after the IN718 powder and corresponding process used in the M 290 system. Material-related input values were found by dynamic image analysis, pycnometry, rheometry, and from literature. PD was measured with six different methods, and the method considered as most analogous to the practical validation experiment yielded a PD of 52 %. Various particle behavior phenomena were also observed and analyzed.
Many of the powder bed characterization methods found in literature were not applicable to L-PBF processing or were not representative of the simulated conditions. Many simulation studies were also found to use no validation, or used a validation method which is not based on the investigated phenomena. The validation model developed in this thesis accurately represents the simulated conditions and is found to produce reliable and repeatable results. The simulation model was parametrized with values acquired from practical experiments or literature and closely matched the validation experiment, and could therefore be considered a truthful representation of the powder recoating process of an EOS M 290. The model can be used as a platform for future development of Tungsten powder simulation.
The aim of this thesis was to develop a simulation model of the recoating process of an EOS M 290 L-PBF system, as well as a validation method for the simulation. The validated simulation model can be used to evaluate the applicability of the used simulation software (FLOW-3D DEM) in powder material development, and possibly use the model as a platform for future application with Tungsten powder. In order to reduce complexity and uncertainties, the irregular Tungsten powder is not yet simulated, and a well-known, spherical EOS IN718 powder feedstock was used instead.
The validation experiment is based on building a low, enclosed wall using the M 290 L-PBF system. Recoated powder is trapped inside as the enclosure is being built, making it possible to remove the sampled powder from a known volume. This enables measuring the powder packing density (PD) of the powder bed. The experiment was repeated five times and some sources of error were also quantified. Average PD was found to be 52 % with a standard deviation of 0.2 %.
The simulation was modelled after the IN718 powder and corresponding process used in the M 290 system. Material-related input values were found by dynamic image analysis, pycnometry, rheometry, and from literature. PD was measured with six different methods, and the method considered as most analogous to the practical validation experiment yielded a PD of 52 %. Various particle behavior phenomena were also observed and analyzed.
Many of the powder bed characterization methods found in literature were not applicable to L-PBF processing or were not representative of the simulated conditions. Many simulation studies were also found to use no validation, or used a validation method which is not based on the investigated phenomena. The validation model developed in this thesis accurately represents the simulated conditions and is found to produce reliable and repeatable results. The simulation model was parametrized with values acquired from practical experiments or literature and closely matched the validation experiment, and could therefore be considered a truthful representation of the powder recoating process of an EOS M 290. The model can be used as a platform for future development of Tungsten powder simulation.