Abstract
Support structures are required for overhanging surfaces in additive manufacturing to prevent part deformation. However, these support structures are undesirable because they are removed during postprocessing after the part is built. It is critical to remove metal support structures because they can occasionally leave residue on the part. Among the various support-removal techniques, manual removal and wire electrical discharge machining (EDM) are the most frequently used. While most manufacturers recommend the manual method due to its economic benefits, it may be challenging for thin walls, small-scale components, and brittle materials. The purpose of this study is to evaluate and compare these two methods, taking into account both their advantages and disadvantages. The results indicate that while manual support-removal is quick, it results in a degraded surface topography and increased surface roughness due to support tooth remnants and residue. The wire EDM method of removing support resulted in a more uniform surface topography and a superior surface finish. However, as a result of thermal damage, a heat-affected zone developed on the machined surface.
1. Introduction and Literature Review
AM is a collection of technologies and procedures for creating parts by layering materials [1, 2]. Many processes and technologies are classified as AM, but they all use the same basic technique to create the physical model. The methods for constructing layers vary depending on the process.
There are various classification systems for AM processes and technologies. Liquid-based material, solid-based material, and powder-based material are three typical techniques [3, 4]. Another prominent method is to categorize AM technologies based on the processes they employed. According to ISO 17296–1 [5, 6], there are seven major processes in AM technologies as shown in Figure 1.
The major and most prominent technologies that represent powder bed fusion (PBF) procedures are selective laser melting (SLM), selective laser sintering (SLS), and electron beam melting (EBM) [7–9].
SLM and EBM are two PBF techniques for employing AM technology to fabricate metal parts. The sole difference is the type of energy source and machine setup in both processes. Unlike EBM, SLM uses a laser to melt the powder. The entire process takes place in a vacuum in EBM, whereas in SLM, the laser beam is less impacted by atmospheric conditions and an inert gas, such as argon, nitrogen, or occasionally helium is employed. EBM-produced items have exceptional mechanical qualities and a unique microstructure because of the unique nature of additive manufacturing [10].
As previously said, an AM technique creates objects by layering materials together. This can lead to mistakes, especially if a new layer does not have a footprint that is distinct from the prior layer (Figure 2). Errors can also occur when a new layer is larger than the previous layer or is printed in the air, as in fused deposition molding and stereolithography technologies. During the manufacturing process, an overhanging structure is a portion of a component that is not supported by solidified material or a substrate at the bottom.
(a)
(b)
The different AM processes and technologies have different reasons for requiring support structures. Some technologies need support to resist deformation or even collapse caused by gravity. Some require, such as the structure, to resist deformation caused by any generated thermal gradients during the manufacturing process, especially in metal processes. Also, the support structure may be used to balance a printed model so that it can securely attach to the build platform during the building process. Table 1 parents some AM technologies and their need for support.
Support structures are nonfunctional parts that are removed after the building process. In powder bed fusion additive manufacturing (AM), support structures are mechanically removed by applying force or cutting (often manually) [12]. In some cases, an inability to remove support structures causes an inability to manufacture preferred structures. Depending on the involved material and geometry, metal support can be removed in a different ways. Wire electrical discharge, saws [12, 13], Dremel handheld power tools [14, 15], and pliers [15] are all frequently used methods for removing support structures from PBF parts [16]. Sandpaper effectively removes the marks left on the bottom by supports. For metal supports, wire EDM, milling equipment, and handsaws may be required to separate the support part from the baseplate, and then separate supports from the part [4].
The cost of postprocessing must also be considered when designing a support structure. For example, reducing the number of supports results in fewer surface marks and reduces part cleanup and postprocessing labor [4]. It is critical to optimize the support structure in order to fabricate parts successfully. Accurately selecting a support structure design is critical, as it has an impact on the production costs, time requirements, and quality of the build. Applying a large contact area support to the overhanging area, for example, would complicate support-removal and may cause damage to the part surface [7, 17]. Calignano et al. examined the effect of block support structures made of aluminum and titanium alloys via selective laser melting [18]. The results indicate that support structures are easier to remove from titanium samples than they are from aluminum samples. Additionally, the support design has an effect on the detachability of support structures [18]. By selecting a support structure that maximizes part removal and maintains the component's surface quality, Järvinen et al. [19] compared the properties of two types of support structures (web and tube) with manual support-removal that utilizes pliers to evaluate their removability. The results indicate that web supports offer superior removability as compared to tube supports [19]. Hildreth et al. demonstrated dissolvable metal supports for three-dimensional direct metal printing [20].
In this study, the metal supports were fabricated by exploiting small differences in the chemical and electrochemical stability between metallic alloys. A stainless steel bridged structure with a 90° overhang was fabricated by using a carbon steel sacrificial support that was later removed through electrochemical etching [20, 21]. Chou et al. proposed a method for designing and fabricating contact-free support structures for overhanging EBM surfaces that eliminated one or more unmelted powder layers between the overhang's lower and upper horizontal surfaces [22]. Through the unmelted powder, heat was conducted from overhanging surfaces to the support structure. The support structure improved heat transfer and minimized thermal gradients caused by the unmelted metallic powders’ low thermal conductivity [22]. Jiang et al. proposed a support-interface method for easy part removal from the substrate, without additional postprocessing [23]. The strut-array pattern acted as a sacrificial layer, and design considerations and process parameters were discussed. The proposed method could potentially enhance the process’s efficiency and automation [23].
2. Experimental Procedures
To evaluate techniques for removing support structures, a ledge overhang with a 10 mm width and block support structures was designed as shown in Figure 3.
The EBM process (ARCAM machine) was used to fabricate the designed specimens from the Ti6Al4V alloy. Additional information about the procedure is available elsewhere. [24, 25]. The specimens and support structures were manufactured using standard process parameters. Following that, these specimens were subjected to a powder recovery system to remove any remaining unmelted powder within support structures. These fabricated specimens are shown in Figure 4 after the unmelted powder has been removed.
The fabricated specimens were then subjected to the support-structure-removal process. Two methods (manual and wire EDM) to remove support structures were incorporated and evaluated. A simple plier was used for the manual method, whereas a brass wire electrode with a 50-V discharge voltage and 1.8-A discharge current were used for the wire EDM machine. To evaluate the removability of support structures, the removal time was recorded and analyzed. The pattern of support-structure-removal was maintained for all specimens to ensure consistency, and the operator was given a rest period following each specimen to avoid fatigue. After removing support structures, the surfaces’ topography and roughness were evaluated with a scanning electron microscope (SEM) and a Dektak XT surface profiler (Bruker, Germany). The effects of the support-structure-removal method on the mechanical properties of the supported surface, microstructures, and chemical composition was evaluated.
It is difficult to conduct a tensile test on small samples (overhang structures) to evaluate mechanical properties, so micro-Vickers hardness values were used to estimate the yield strength of fabricated overhang structures, using the relation (σy ≈ HV/3) [26, 27]. The ZHVµ micro-Vickers hardness (Zwick/Roell, Ulm, Germany) with a 500-gf force and a 10-s dwell time was used for the hardness test. Multiple firmness measurements were taken along the edge, and averages were then computed. Specimens were ground, polished, and etched, to evaluate their microstructure and chemical composition. Hardness tests were conducted close to the bottom edge of overhanging structures (by the supported surface).
3. Results Analysis
3.1. Removal Time
To evaluate the effect of the support-removal method on the elimination of support structures, the time consumed by removing them for each specimen was recorded. To maintain uniformity, the support-structure-removal pattern was maintained.
The results show that the required support-removal time increases drastically with the use of wire EDM as opposed to the manual method. Manual support-removal with pliers resulted in a 3.08-min removal time, whereas a wire EDM use resulted in a 9.88-min removal time, as shown in Figure 5.
3.2. Topography and Surface Roughness
To find the effect of a support-removal method on the topography and surface roughness, the supported area was viewed under SEM and surface roughness analysis was performed. Figure 6 shows the surface topography comparison of the supported surface after removing support via the manual and wire EDM methods. In general, downward-facing overhang surfaces were found to be rough and porous due to the adherence of partially melted powder particles and the unsupported area, as shown in Figure 6(a). Additionally, edges were found to have support leftovers after the manual support-removal method, resulting in nonuniform surfaces as shown in Figure 6(c). The wire EDM method removed partially melted powder particles and support leftovers, along with support structures, which resulted in a uniform surface (Figures 6(b) and 6(d)).
(a)
(b)
(c)
(d)
The surface roughness analysis of downward-facing overhang surfaces with manual support-removal resulted in an average Ra of 27.25 µm. Figure 7(a) shows the analysis results of the overhang surface with the manual support-removal method. Figure 7(b) reveals that the wire EDM-based support-removal resulted in an average Ra of 4.01 µm.
(a)
(b)
3.3. Mechanical Properties and Microstructures
To study the influence of the support-removal mechanism on the overhang surface, microstructure analysis, and Vickers microhardness tests were conducted on downward-facing overhang surfaces. Microstructural analysis revealed the presence of a heat-affected zone (HAZ) on overhangs with the wire EDM support-removal method. The average HAZ thickness was found to be 8 μm along the supported surface, as shown in Figure 8(a), due to the thermal damage of the wire EDM. However, for manual support-removal, there was no variation in the microstructure, as shown in Figure 8(b), which is as expected due to the absence of heat and cutting forces.
(a)
(b)
For microhardness, Figure 9 indicates that the surface hardness with the wire EDM support-removal technique was found to be slightly lower as compared to the manual method, since the wire EDM induces the thermal gradients that result in reduced hardness.
3.4. Chemical Composition
EDS analysis was performed to identify the effect of the support-removal method on the chemical composition of the overhanging surface. The default chemical composition was intact with the manual support-removal process, as shown in Figure 10(a). EDS analysis of the wire EDM surface identified the percentage of elements that were transformed from wire electrode to the machined alloy surface by the support-removal process. The results indicate the presence of 9.49 wt% copper (Cu) on the wire EDM machined surface, as shown in Figure 10(b). This result is predominantly due to electrode wear during the wire EDM-based support-removal process.
(a)
(b)
4. Conclusion
Support structures are critical in additive manufacturing, particularly when fabricating overhanging structures. They are later removed following the completion of the build. The current study evaluated and compared two of the most frequently used support-removal methods, manual and wire EDM. The following conclusions were drawn from the analysis of the data.(1)The manual support-removal process is a fast and easy way to remove supports. The average support-removal time for the manual method was found be around 3.08 min, whereas in the case of the wire EDM method, it was found to be 9.88 min.(2)Manually removed supports for overhanging surfaces were found to have support structure tooth leftovers and partially melted power particles, which resulted in a poor surface topography and a higher degree of surface roughness.(3)The wire EDM method resulted in a uniform surface topography, due to the removal of partially melted powder particles with support structures.(4)The average surface roughness Ra was found to be 27.25 µm for the manual support-removal method, whereas it was found to be 4.01 µm for the wire EDM technique.(5)Micro-Vickers hardness tests revealed decreased rigidity of the wire EDM surface due to themal effects.(6)The wire EDM method resulted in an average HAZ of 8 µm and an addition of 9.49 wt% Cu on the outermost machined surface, due to thermal damage and electrode wear, respectively.
Data Availability
No data were used to support this study.
Conflicts of Interest
The authors declared that they have no conflicts of interest.
Acknowledgments
The authors are grateful to the Raytheon Chair for Systems Engineering for funding.