Strengthening ABS , Nylon , and Polyester 3 D Printed Parts by Stress Tensor Aligned Deposition Paths and Five-Axis Printing

In most fused filament fabrication systems, all filament laydown paths are at constant Z height. This creates a weak direction in the resulting parts, as the interlayer adhesion between melted and solidified material is much weaker than the tensile strength of the bulk material. For example, a hemispherical dome pressure vessel endcap will fail easily along these Z=constant cleavage planes. We resolve this problem by proposing a 3D printing system that does not limit the nozzle positioning to a single Z layer at a time, or to constant pitch and yaw angle, but instead lay down extrusions more closely aligned with the stress tensor within the part (but requiring 5 simultaneous axes of motion). To verify this, we have constructed a working 5-axis fused-filament fabrication 3D printer and produced a number of test parts in ABS, nylon 645, and T-glase polyester. Using a commercial hydrostatic pressure system, we have tested these parts to destruction and find a typical strength improvement of 3x to 5x over conventional 3-axis parts printed to the same specification, in the same machine, from the same spool of polymer; the only thing changed was the extrusion pattern. An approximate calculation to translate this into the material’s ultimate tensile strength shows that the 5-axis FFF parts are within a factor of two of the ultimate tensile strength of typical professionally injection-molded ABS material. International Solid Freeform Fabrication Symposium This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. Copyright c © Mitsubishi Electric Research Laboratories, Inc., 2016 201 Broadway, Cambridge, Massachusetts 02139 Strengthening ABS, Nylon, and Polyester 3D Printed Parts by Stress Tensor Aligned Deposition Paths and Five-Axis Printing William S. Yerazunis (*), John C. Barnwell III, Daniel N. Nikovski Mitsubishi Electric Research Laboratories {yerazunis, nikovski}@merl.com * correspondence author Abstract: In most fused filament fabrication systems, all filament laydown paths are at constant Z height. This creates a weak direction in the resulting parts, as the interlayer adhesion between melted and solidified material is much weaker than the tensile strength of the bulk material. For example, a hemispherical dome pressure vessel endcap will fail easily along these Z=constant cleavage planes. We resolve this problem by proposing a 3D printing system that does not limit the nozzle positioning to a single Z layer at a time, or to constant pitch and yaw angle, but instead lay down extrusions more closely aligned with the stress tensor within the part (but requiring 5 simultaneous axes of motion). To verify this, we have constructed a working 5-axis fused-filament fabrication 3D printer and produced a number of test parts in ABS, nylon 645, and T-glase polyester. Using a commercial hydrostatic pressure system, we have tested these parts to destruction and find a typical strength improvement of 3x to 5x over conventional 3-axis parts printed to the same specification, in the same machine, from the same spool of polymer; the only thing changed was the extrusion pattern. An approximate calculation to translate this into the material's ultimate tensile strength shows that the 5-axis FFF parts are within a factor of two of the ultimate tensile strength of typical professionally injection-molded ABS material. In most fused filament fabrication systems, all filament laydown paths are at constant Z height. This creates a weak direction in the resulting parts, as the interlayer adhesion between melted and solidified material is much weaker than the tensile strength of the bulk material. For example, a hemispherical dome pressure vessel endcap will fail easily along these Z=constant cleavage planes. We resolve this problem by proposing a 3D printing system that does not limit the nozzle positioning to a single Z layer at a time, or to constant pitch and yaw angle, but instead lay down extrusions more closely aligned with the stress tensor within the part (but requiring 5 simultaneous axes of motion). To verify this, we have constructed a working 5-axis fused-filament fabrication 3D printer and produced a number of test parts in ABS, nylon 645, and T-glase polyester. Using a commercial hydrostatic pressure system, we have tested these parts to destruction and find a typical strength improvement of 3x to 5x over conventional 3-axis parts printed to the same specification, in the same machine, from the same spool of polymer; the only thing changed was the extrusion pattern. An approximate calculation to translate this into the material's ultimate tensile strength shows that the 5-axis FFF parts are within a factor of two of the ultimate tensile strength of typical professionally injection-molded ABS material. Introduction: Fused Filament Fabrication (FFF) systems generally operate by melting and extruding a polymer, placing the extrusion in a series of laydown paths in an XY plane of constant Z. When every section of the desired part at that constant Z has been extruded, the Z motors are activated to raise the extruder a small distance (usually on the order of 0.2 to 0.5 mm) and the next set of laydown paths is extruded. The advantage of this mode of operation is that it is relatively simple to write the software that converts a 3D CAD model into a common format (typically the “STL”, a triangular surface tesselation format), then converting the STL to layers of constant Z, and then converting the 2-D layers of constant Z into one-dimensional motion-along-a-line extruder path commands (typically in the form of G-code). At each stage of this tool chain, semi-formal or formal standards exist which make it possible to support a wide choice of different software packages and methodologies along the tool chain. This simplicity of design has yielded a wide variety of software systems (Slic3r, Skeinforge, Cura, MatterControl, Synplify, Marlin, Teacup, etc.) and hardware for 3D FFF printing (Reprap, LulzBot, Ultimaker, Prusa, etc), many of which are completely “open source” everything about them is freely available to anyone with Internet access. Unfortunately, this simplicity of design and partitioning of the problem into 3D→2D→1D causes the side effect that all volume elements at a particular level Z are filled before any volume elements at Z+ε are filled. This means that at each Z=constant level within the final part is a possible cleavage plane, held together only by the adhesion of hot plastic extruded against solidified plastic, rather than by continuous melt. Therein lies the problem with FFF printingthe literature reports strength variations as much as 15:1 from the injection molded or continuous melt strength along the axial fiber length versus the interlayer adhesion strength [1][2][3][4]. This high level of anisotropy has a major impact on the usability of 3D printed parts; some part designs with a tall, thin geometry are simply unprintable because the printed parts will break during removal from the print surface. Those with skill in the use (some would say “black art”) of FFF 3D printers know the importance of orienting parts during the build setup process so that they have the majority of the in-use stress is in the XY plane. Our research agenda is to (1) reproduce and quantify the above anisotropy for a commonly useful yet weak part, (2) create strong parts with the same physical geometry as the weak parts, by aligning the strong-axis paths of laydown extrusion with the in-use stress tensor’s maximum tension direction and (3) is to prove (or disprove) the hypothesis that it is possible to make strong parts in FFF by changing the laydown paths so that the laydown paths align with the direction of stress in the part. If this hypothesis is true, the long-term goal is to optimize FFF part strength by automatically aligning the high-strength laydowns in the part with the direction of the stress tensor within the part, when the part is in actual use. Relationship with Established Testing Methods: ASTM and other organizations publish standard testing methodologies for determining the strengths of materials. In particular, ASTM D3039 and D638 would on the face of them seem applicable to help us verify our hypothesis. Unfortunately, both D3039 and D638 postulate that the material will be stressed linearly, and with the axial fiber length controlled to be constant with the direction of the imposed stress. This situation applies only in the case of pure tension members (which are “uninteresting”, in the sense of 3D printing, because bar and rod stock of most extrudable materials is already available at a lower price than the cost of 3D printer filament). Ahn et al [3] approaches this question by testing bars with mixed 0/90 extrusion with reasonable results, but again, this orientation would apply only in linear tension members, not in arbitrarily shaped parts. Therefore, we decided to approach the testing problem with a specimen design that is inherently stressed in three dimensions simultaneouslya hemispherical pressure cap. The hemispherical pressure cap has a number of useful properties: 1. Hemispherical pressure caps have a simple, analytical stress tensor – the stress at any point in a thin hemispherical pressure cap is radially symmetric tension in the local plane of the cap, plus an impulse of compression from t