Having reliable simulations of 3D printed parts requires understanding the effect of individual print parameters on the performance of the part. One parameter that is often used when tuning a 3D printer is the material extrusion rate, also known as “flow”. Flow in Fused Filament Fabrication (FFF) is the amount of material that is pushed out of a nozzle over a given interval of time. There are many different settings in a slicer program that influence the flow such as layer height, layer width, and print speed. When a user changes one of these settings, the slicer has to change the rate of extrusion so the part is not dramatically over-extruded or under-extruded.
The images shown in Figures 1 and 2 demonstrate the effect of material flow on the opacity of two printed parts. This helps us visualize the effects of flow. The part printed with 70% flow has much larger voids than the part printed with 100% flow. These voids result in a dramatic change in how these two parts look, but also how they will hold up when put under load. In addition to having effects on strength properties, flow also affects the dimensional accuracy of these parts making them harder to place in assemblies.
Figure 1. PC test print at 70% flow.
The part shown in Figure 1 is under-extruded and has large voids between printed rows of extruded material. This results in an opaque appearance when printed with a clear polycarbonate material.
Figure 2. PC test print at 100% flow.
The part shown in Figure 2 is printed with adequate extrusion resulting in tightly packed rows of extruded material. This results in the part having minimal voids and a relatively more transparent appearance. The minimal voids result in a much larger bonding surface between printed rows and improved strength properties.
To capture the effect of material flow on mechanical properties of printed parts, the following experiment was conducted. Six sets of ASTM standard dog bones were printed using 3DXTech ABS material on an Ultimaker S5 printer. The only difference between the six sets of samples was the material flow: all other print parameters were held constant to isolate the effect of flow. The XY90 print orientation shown in Figure 3 was selected for the test samples. The XY90 print configuration allows us to measure the bond strength of the printed rows oriented perpendicular to the applied load.
Figure 3. Test specimen with XY90 print geometry.
Figure 4. Test specimen yield strength vs. material flow.
Figure 5. Test specimen stiffness vs. material flow.
The samples were then tested on a universal testing machine, and the yield strengths and stiffnesses were computed from the experimental data. Figure 4 shows the yield strength vs. flow % and Figure 5 shows the part stiffness vs. flow % for flow values between 60% and 110%. The most significant take away from both plots is that the stiffness and yield strength plateau when the material flow is at least 85%. This is an excellent trend because it indicates that once you reach a reasonable extrusion rate the stiffness and strength remain relatively constant. When printing parts, it is important that this level of flow is reached so that the parts will have minimal voids and the expected material properties based on the material the part is printed with.
Figure 6. A visual inspection of the gage region of 70% & 100% flow test specimens.
The visual inspection in Figure 6 demonstrates the difference between an under-extruded part and a part printed with adequate extrusion. The 70% flow part has gaps that are clearly visible between the individual rows that make up the structure of the part. This has a dramatic effect on the strength as shown in Figure 4. These voids explain why the stiffness and strength decrease as the flow decreases. The 100% flow sample has strong bonding between individual rows, with no visible gaps or voids.
Figure 7. Visual inspection of 70% & 100% flow fracture surface.
Figure 7 above shows the fracture surface of the test specimens after testing. The 70% flow has a clean fracture surface that follows bonds between the printed rows of material. That clean fracture between rows is a delamination failure. The bond between the printed rows delaminated before the bulk strength of the material was reached. The 100% flow samples failed in a completely different manner, creating a rough surface with failure occurring across many different rows of printed material. This rough surface indicates that the failure relates to the bulk material properties of the material. These test specimens did not fail completely due to bond strength, but due to a combination of bond strength and bulk material failure.
Perhaps the most interesting result from this study is that parts printed with adequate flow settings have similar strengths and stiffness, regardless of whether the flow is increased or decreased by a few percent. Most technicians only adjust the material flow up or down by a few percent to achieve higher reliability when printing parts, and usually never need to lower the material flow below 90%. While this data is not conclusive, it does give us insight into developing a more complete understanding of how these individual printer settings influence part performance.