An Improved Two-Level Support Structure for Extrusion-Based Additive Manufacturing

https://doi.org/10.1016/j.rcim.2020.101972Get rights and content

Highlights

  • Two-level support is an effective support for Extrusion Based Additive Manufacturing

  • An improved strategy for generating a two-level support is proposed

  • Experiments validate that the improved strategy saves more support volume

ABSTRACT

Extrusion-Based Additive Manufacturing is one of the most commonly used Additive Manufacturing techniques. However, the technique requires that the overhang features in a model should be provided with supports from below. The added supports contribute to a major source of material waste during the manufacturing process. In order to save support materials, considering 3D printing directions as a constraint of adding supports, we propose a two-level support structure where Level 1 consists of a truncated beam-structure and Level 2 consists of a slim tree-shaped structure. The main contribution of this paper is the modification of the shape of Level 1 support and optimization of the topologies of the tree-supports in Level 2. The approach developed in this paper leads to material savings up to 31.01% and printing time savings up to 13.10% in comparison to the existing literature. Several comparison experiments were conducted on benchmark models to validate the developed approach and have been reported in this paper.

Introduction

Additive Manufacturing (AM) is a process in which an object is fabricated by joining materials in a layer-by-layer manner. AM technologies have been widely applied in the medical, aerospace, and automotive industries. Among these technologies, Extrusion-Based Additive Manufacturing techniques such as fused deposition modeling (FDM), fused filament fabrication (FFF) and melt extrusion manufacturing (MEM) are widely exploited by common users due to their low cost [1,2].

During the deposition process, an overhang may collapse if there is no support structure beneath it [3]. Adding support structure for a 3-axis printing system with a unique building direction [4] as well as printing with varied building directions using a multi-axis printing system [5], [6], [7], [8], [9], [10], [11] are two major approaches for preventing material collapse. Adding support structure is widely used since it requires fewer hardware settings. However, it costs more material. In order to reduce the support materials, several approaches have been proposed, including optimizing build orientation [12], [13], [14], [15], segmenting a model into several support-free sub-parts [16], [17], [18], [19], and optimizing its support structure. In the following, the review work is focused on the literature related to support structure optimization with fixed build direction, which is the concern of our proposed approach in this paper.

Optimizing a support structure refers to the optimization of its topology connection and geometric parameters. In recent years, more ingenious support structures, such as cellular structure, tree-shaped structure, and strut structure, have been exploited in Extrusion-Based Additive Manufacturing.

In terms of cellular structures, Lu et al. [20] proposed a honeycomb-cell structure (Fig. 1 (a)) based on a Voronoi-diagram algorithm. The structure was further subdivided to maximize the hollowing space while maintaining the structural strength of the model. Lee et al. [21] proposed an inner support-free structure generation algorithm based on a 3D block partitioning method to reduce the manufacturing time and material. Vaidya and Anand [22] discretized the free space of the model into a set of unit cells and employed Dijkstra's shortest path algorithm [23] to generate cellular support structures that can minimize the support volume. Strano et al. [24] proposed a graded cellular support structure where dense cells were placed under the heavy overhangs, and sparse cells were placed under the light overhangs.

In terms of tree structures, Zhang et al. [25] proposed an internal supporting frame based on a medial axis tree. The tree-shaped branches rooted on the medial axis. Dense leaves supported the regions where a large structural strength was required, and coarse leaves supported the other parts of the model. Vanek et al. [26] proposed a greedy algorithm based on the geometric feature to grow the tree-support from top to down. Schmidt et al. [27] proposed a top-down generation of the branching support structure, starting from the support point obtained by using a combination of Watershed and Poisson surface sampling strategies. This technique has been implemented in the free software tool, Autodesk MeshmixerTM. Zhang et al. [28] proposed a local barycenter based tree-support algorithm that has nice property of higher efficiency and less material consumption. We proposed a hybrid of PSO and greedy algorithm to minimize the support volume of the tree-shaped support structures (Fig. 1 (b)) for 3D printed models [29,30].

In terms of strut structures, Wang et al. [31,32] proposed a self-supporting skin-frame structure to reduce the material cost in fabricating a 3D printed model and designed a sparsity optimization framework to ensure the structural strength. Dumas et al. [33] proposed a scaffolding support structure (as shown in Fig. 1(c)) composed of horizontal bridges and vertical pillars. Jiang et al. [34,35] proposed a two-level support structure (Fig. 1(d)) generation strategy that takes into account the influence of the print path on the cost of supporting materials.

In this paper, considering the effect of the print path, we propose an improved algorithm for the two-level support structures of [34,35]. Level 1 support consists of a set of beams beneath the overhanging surface and it is orthogonal to the print direction, and Level 2 support consists of a set of trees for supporting the bottom of Level 1.

The rest of this paper is organized as follows: Section 2 presents the existing two-level support structures and the details of our improved approach. Section 3 presents four comparison experiments to validate our approach. Finally, this paper ends with some conclusions in Section 4.

Section snippets

Two-level Support Structures

To illustrate our improved two-level supports, we shall introduce the concepts of the longest printable bridge length (LPBL), the printable threshold overhang angle (PTOA), and the longest printable length of a bar, and the most recent two-level support structure proposed in [35] in section 2.1. The rest of this section is organized as follows: section 2.2 presents the most recent two-level support structure proposed in [35]; section 2.3 presents our improved two-level support structures.

Comparison Experiments

To validate our proposed approach, we compared the support material reduction of our approach with the most recent result presented in [35]. Note that the 3D printing materials are homogenous, which means that the weight is proportional to the volume. Therefore, we compared the performance based on volume consumption. In addition to fabricating the same set of models (“U”, “O”, and “A”), we further validated our approach by using one more example of “T”. The dimensions of these parts are shown

Conclusion

The support structures play a critical role in a 3D printing task by preventing the overhanging features from collapsing. However, support structures are removed after the 3D printing task and are therefore a major source of material waste. Addressing the problem of saving printing materials, a vast body of literature exists and has been devoted to the problem of lightweight design of support structures. In the recent research [35], considering building direction as a constraint of adding

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China (No. 51605290) and Shanghai International Science and Technology Cooperation Fund No. 18510745700.

References (46)

  • B.N. Turner et al.

    A review of melt extrusion additive manufacturing processes: I. Process design and modeling

    Rapid Prototyp. J.

    (2014)
  • B.N. Turner et al.

    A review of melt extrusion additive manufacturing processes: II. Materials, dimensional accuracy, and surface roughness

    Rapid Prototyp. J.

    (2015)
  • J. Jiang et al.

    support structures for additive manufacturing: A Review

    J. Manuf. Mater. Process.

    (2018)
  • K. Lee et al.

    Slicing algorithms for multi-axis 3-D metal printing of overhangs

    J. Mech. Sci. Technol.

    (2015)
  • H. ming Zhao et al.

    Inclined layer printing for fused deposition modeling without assisted supporting structure

    Robot. Comput. Integr. Manuf

    (2018)
  • C. Wu et al.

    RoboFDM: A robotic system for support-free fabrication using FDM, in: 2017 IEEE Int

    Conf. Robot. Autom

    (2017)
  • C. Dai et al.

    Support-free volume printing by multi-axis motion

    ACM Tran

    (2018)
  • C. Wu et al.

    General Support-Effective Decomposition for Multi-Directional 3-D Printing, IEEE Trans

    Autom. Sci. Eng

    (2019)
  • X. Zhang et al.

    Perceptual models of preference in 3D printing direction

    ACM Trans. Graph

    (2015)
  • J. Jiang et al.

    Optimisation of multi-part production in additive manufacturing for reducing support waste

    Virtual Phys. Prototyp

    (2019)
  • R. Hu et al.

    Approximate pyramidal shape decomposition

    ACM Trans. Graph.

    (2014)
  • X. Chen et al.

    Dapper: Decompose-and-Pack for 3D Printing

    ACM Trans. Graph. Artic

    (2015)
  • X. Wei et al.

    Toward support-free 3d printing: A skeletal approach for partitioning models, IEEE Trans

    Vis. Comput. Graph

    (2018)
  • Cited by (20)

    • Non-contact/contact hybrid support based on the slice for ceramic stereolithography

      2023, Journal of Manufacturing Processes
      Citation Excerpt :

      Another situation is that the printing layer has a strong stiffness to stop the blade from moving and the recoating cannot continue, so the printing is aborted, so the whole printing process is unstable with the warpage problem. For other materials and printing technologies, some researchers have carried out segmentation of the part [19], support structure [20,21], or building orientation optimization [22]. However, there is some incompatibility when these methods are directly applied to ceramic printing.

    • Grid-tree composite support structures for lattice parts in selective laser melting

      2023, Materials and Design
      Citation Excerpt :

      Then, the genetic algorithm is used to optimize the support structure and trim many beams to have lower material consumption. Although scholars have proposed various support structure generation methods [16–24], few scholars consider support structure generation for lattice parts. For the lattice parts that include the lattice structure and the frame, in addition to the support structure for the overhang area of the frame, it is also necessary to add the support structure for the overhang area of the lattice structure.

    • Continuously varied infill pattern (ConVIP): improvement of mechanical properties and printing speed of fused filament fabrication (FFF) 3D printing

      2022, Journal of Materials Research and Technology
      Citation Excerpt :

      Printable threshold overhang angle (PTOA) and the longest printable bridge length (LPBL) were defined to optimize support structure in slicing before printing. Feng R., et al. [31] proposed the idea that improves [30]. The two-level support structure was developed and achieved a 13% of reducing in printing time.

    • A kinematics-aware part clustering approach for part integration using additive manufacturing

      2021, Robotics and Computer-Integrated Manufacturing
      Citation Excerpt :

      Moreover, these support structures tend to be sealed since they are hard to be touched. Although there are many sound works dedicating to support structure research [6,52,54-58], the automatic and efficient method, detecting whether a support structure (upcoming) anchored on any place of a model can be cleaned before fabricating, is still rarely discussed [19,52,59,60]. In view of that the support structures often can be effectively cleaned when they can be seen directly, an oriented bounding box (OBB) [61]-based detection method is designed to detect whether any support structure will be sealed after fabricating.

    View all citing articles on Scopus
    View full text