Published on: September 12, 2022
One way that woodworking differs from metalworking is that with woodworking, no coolant is used during machining to remove chips and dust, as the absorption of liquid into the wood is undesirable. Another important difference is the high cutting speed involved with woodworking. Both of these factors lead to the problematic distribution of chips in the machines and equipment, plus dangerous dust exposure for machine operators. Chip collection hoods and vacuuming systems can reduce these wood deposits and emissions. However, these methods are very material-, machine-, tool- and process-specific, and therefore require a long development time, which may not be cost-effective for smaller companies.
This particular case also presents many difficulties in the modelling stage. A rotating tool removes material, but also circulates the surrounding air. At the same time, the tool is in direct contact with the workpiece, which modifies the flow and redirects colliding wooden chips. Additionally, the wood chips present a wide range of sizes and shapes, so they are ejected at different positions and directions.
Both fluid flow and particle dynamics play an important role in the design process. Chip size and velocity determine whether the chips follow the flow or their own inertial path defined by the ejection from the cutting tool. By colliding with surfaces inside the hood, the wood chips decelerate and are more easily captured by the vacuuming system. This process is further improved, if the impact surfaces also guide the particles towards the suction. The design of cutting tools, in particular their chip space, influences both the induced flow around the tool and the chip ejection and thus the dispersion of the particles, which is crucial for successful chip collection.
In this study, DEM and CFD simulations were coupled using Ansys Fluent and Rocky DEM. While Fluent handles the fluid flow around the saw and the suction inside the hood, Rocky solves the wooden chip trajectories while considering collisions between chips and the surrounding surfaces. The influence of the local particle concentration on the flow inside the hood is obtained by taking advantage of the two-way coupling available in Rocky.
During the DEM simulation, the particles are initialized in the chip space, and are accelerated by the contact with the saw. This leads to a model of the particle ejection that’s closer to the real process, compared to a modelling approach, where particles are injected outside the chip space with a given initial velocity.
An even better description of the particle ejection is achieved when using the API-based bond-particle module. Particles are initialized in the cutting area and are bonded by virtual beams with tensile strength defined by the material properties. Once the allowable strength is achieved, the bonds break, releasing the particles. This allows for a good representation of the milling process, without the need to completely model the milling process. Some of the ejected particles also retain part of their bonds, resembling larger non-spherical particles seen in real life experiments.
The characterization of the particle trajectories and the fluid flow inside a hood enables a better understanding of chip ejection and collection. This information is useful for the improvement of existing hoods. Walls can be adapted to help guide larger particles in the suction channel direction, improving the overall efficacy and efficiency, without changing the suction pressure and the associated energy costs.
Study conducted by Technische Universität Dresden. TU Dresden is supported by Rocky DEM Channel Partner CADFEM.
PhD Student at the Chair of Fluid Mechanics, TU Dresden
Thomas E. Hafemann works on the numerical modeling of wood chips ejection and collection applied to the grooving process aiming at the improvement of existing chip collection hoods. He holds a Master’s degree in Mechanical Engineering, and has been focused on the modeling and simulation of multiphase flows ever since.
