Keywords
material; extrusion; parameters; surface; texture; roughness;
This article is included in the HEAL1000 gateway.
The material extrusion process has been widely used to manufacture custom products. However, the surface texture varies due to the additive mechanism of the process, which depends on the layer height and surface orientation, resulting in varying average surface roughness values for inclined, flat and vertical surfaces. Different strand welding conditions result in non-uniform internal stresses, surface distortions, layer traces, weak bonding, non-uniform pores and material overflow. This paper comprehensively examines material extrusion process achievements in surface texture quality and studies and summarises the most influential processing parameters. Parameter effects are critically discussed for each topic; flat, inclined, and vertical surfaces. The results of this research help reduce post-processing.
material; extrusion; parameters; surface; texture; roughness;
The selective deposition of thermoplastic material by hot extrusion in sequentially patterned slices (layers) using computerised numerical control positioning is broadly known as the filament material extrusion (ME) process.1,2 ME is a fully controlled thermomechanical process.3,4 It consists of the following elements: (i) a machine control unit (MCU: micro-computer) which reads the g-code technological program,5,6 (ii) three step-motors with the belts for the motion requirements in X, Y and Z axes,7,8 (iii) the extruder mechanism (head carriage: filament, feed wheels, nozzle, fans, etc.),9,10 and (iv) the bed in which the 3D-printed parts are created (see Figure 1).11,12 The slicing (CAM) software handles the orientation of the STL model,13 the slicing parameters,14 and the selection of the infill structure and finally generates the g-codes.15 ME technology is also known as fused filament fabrication (FFF),16 fused deposition modeling (FDM) or material extrusion (ME).17,18
Due to the cost-effectiveness of the ME process,19 the timely manner and the customized orientation of the product,20 ME 3D printed parts have been tested in many applications in the automotive,21,22 aerospace,23,24 electrical motors,25 microwave absorption applications,,26 electronics,27,28 maritime,29 medical,30,31 neurosurgery,32 orthotics,33,34 dental,35 antifouling,36 heritage,37,38 jewellery,39 textile and fashion,40,41 forensic,42 and other sustainability and recyclability issues.43 Surface texture (ST) parameters play an important role in surface engineering and are very important in determining the mechanical properties of ME 3D printed parts.44,45 For example, higher surface roughness values of flat dogbone surfaces reduce the mechanical properties of the elements.46 The inferior ST is a characteristic of all additive manufacturing techniques, compared with machining processes,47 and post-processing methodologies have been proposed in the literature.48,49 Therefore due to the progressive material deposition, additive manufacturing parts need post-processing with machining, abrasives, chemicals or cold spray coatings.50,51
Since the ME 3D printing process has gained tremendous popularity due to its simplicity,52,53 extensive materials,54 organic infills,55,56 fiber infills,57,58 environmental friendliness,59,60 and low budget,61,62 several researchers have reviewed its achievements in material and strength issues,63,64 benchmarking issues,65,66 shape accuracy,67,68 laser post-processing,69,70 machining post-processing,71 small-sized reliefs,72,73 dimensional accuracy,74 surface hardness,75 friction and wear issues,76 stair-stepping effect,77,78 and flat inclined, horizontal and vertical surface roughness issues. Table 1 summarises and concludes the literature’s influential parameter investigations regarding the ME surface texture cases (sloped, flat and vertical)79–97 between the end of 2000 and the start of 2023.
Researchers | Description | ME parameters studied | Conclusions |
---|---|---|---|
Mahapatra & Sood, 201279 | ABS P400 Prismatic part. Top, Bottom and side measurements *Ra: 0.3 to 11.8μm | **H: 0.12, 0.17, 0.25 θ: 0, 15, 30 RA: 0, 30, 60 W: 0.4, 0.45, 0.5 AG: 0, 0.004, 0.008 | Most influential parameters: Top: W, H, O Bottom: θ, RA, AG Side: H |
Boschetto et al., 201380 | ABS/ABSPlus/PC/ULTEM Inclined Ra > 15μm Rt > 75μm | H: varieties θ: 0-30 | A neural network to predict Ra and Rt values was trained and suggested. |
Chaidas et al., 201681 | PLA Vertical surfaces of a thin-walled cuboid part. Ra, Rz, Rt, Rsm | T: 210, 220, 230 H=0.2 | The increase in T decreases all roughness metrics. Ra between 14-17μm. Rz beteen 63-84μm. Rt between 68-124μm |
Kim et al., 201882 | PLA Top flat surfaces 0.4: 0.8 mm inside/outer nozzle tip diameter 1.38 μm Ra after optimisation | Variables H, S and Q | By adjusting the H, Q and S can achieve Ra values of about 1.38 μm. |
Vyavahare et al., 202095 | ABS Parts with pyramidal and conical features Ra: 9-38μm | H: 0.1-0.3 S: 30-90 θ: 0, 90, 180 WT: 0.8-1.4 T: 230-250 temperature. | H was the most critical parameter, followed by θ. |
Aslani et al., 202083 | PLA Vertical surfaces of a thin-walled cuboid part. Ra: 12-22μm Rz, Rt: 60-150μm Rsm: 196-321μm | T: 210, 220, 230 Wall Thickness: 1, 2, 3 | The increase in T decreases all roughness metrics. |
Biglete et al., 202084 | ABS Vertical surfaces Ra: 5-30μm Rt: 80-150μm | H: 0.2, 0.3, 0.4 T: 220, 230, 240 S: 40,50,60 | H dominates T moderate S not significant |
Buj-Corral et al., 202185 | PLA PLA Curved surfaces (hemispherical cups) Ra: 7-23 μm | T:195, 200, 205 H: 0.1, 0.2, 0.3 S: 30, 40, 50 EM: 0.4, 0.5, 0.6 ND: 93, 95, 97 | H and ND were the most influential parameters. |
Chaidas and Kechagias, 202286 | PLA wood Vertical surfaces Ra: 13-24μm Rt: 81-132 μm | T: 180, 190, 200, 210, 220 H: 0.1, 0.2, 0.3 | Lower H values decreased the Ra values significantly. On the other hand, wood flour increases the Ra values considerably compared with pure PLA. |
Caputo et al., 202287 | PLA Top surface ironing effect on surface texture and mechanical dynamic response | S: 50,75, 100 T: 170, 195, 220 H: 0.12, 0.2, 0.28 ID: 5, 50, 100 | Surface height variates between 0.326 mm and minus -0.32mm. Ra varieties between 0.8206 μm and 6.5526 μm. T was the most influential parameter for Ra. |
Fountas et al., 202293 | PLA Vertical surfaces Ra: 4.6-6.9μm | T: 210, 220, 230 Shells: 2, 3 ID: 10, 15, 20 Pattern: Diamond, Hexagonal, linear 0.2mm H (constant) | T and Pattern type were the most influential parameters. |
Selvam et al., 202288 | ABS Thin-walled parts Top and Bottom Ra: 2.5-5μm Inside vertical Ra: 1.4-1.85μm | S: 60, 105, 150 H: 0.1, 0.2, 0.3 T: 220, 232.5, 245 | Vertical surfaces: S and H affected the Ra values significantly, while T was unimportant. Top/Bottom surfaces: All parameters significantly affect the Ra values. |
Rakshit et al., 202289 | PLA Inclined surfaces | RA and θ as variables | They achieved the lowest 5.3μm Ra value at 9o θ and 0o RA. |
Rajesh et al. 202297 | ABS Flat surfaces of dogbone specimens Ra: 26-38μm | H: 0.15-0.25 T: 235-245 PlaT: 95-110 | T was the most influential parameter, followed by H. |
Vinoth Babu et al., 202292 | CF/PLA composites Top surfaces of dogbones samples Ra>28μm | ID: 20, 40, 60, 80 H: 0.08, 0.25, 0.64 infill pattern: (a) rectilinear, (b) triangular and (c) hexagonal | The samples were excessively rough due to the carbon fibers. |
Spahiu et al., 202390 | PLA Top surfaces Ra: 2.4-5.5μm | T: 210, 220, 230 H: 0.15, 0.2, 0.25 S: 40, 50, 60 | S was the most influential parameter. T was not significant. Lowering T values, the Ra improve slightly. |
Bruijn et al., 202391 | PEI Ultem™ 9085 Vertical bars | H: 0.254 ±45 lines style ID: 100 T: 185 | They measured Ra values before annealing at about 17.3 and after annealing at 1.2μm and Rz values at about 70.4 and 6.3μm, accordingly. |
Francis et al., 202396 | ABS Flat surfaces of dogbones | H: 0.25 T: 230 PlaT: 110 ID: 100 | They have improved surface texture and hardness with chemical and thermal treatments. They reduced the Ra to 0.819mμ (from 9μm) with 50s immersion time on acetone solvent NC1. |
Kechagias & Zaoutsos, 202394 | PLA Dental implants | H: 0.2-0.16 ID: 0, 35, 70 PlatT: 50, 55, 60 T: 210, 220, 230 | Optimising Ra and hardness of a dental implant (achieved Ra lower than 10μm). |
* Ra (μm) is the arithmetic mean of the absolute values; Rz (μm) is the mean value of the sum of the highest and the deepest profile deviations; Rt (μm) is the difference between the highest peak and the deepest valley; Rsm (μm) the mean value of the width of the profile elements.
** Layer Height (H, mm); Orientation (θ, deg); Printing Speed (S, mm/s); Printing Temperature (T, oC); Flow speed (Q, mm/s); Raster angle (RA, deg); Raster width (W, mm); Air gap (AG, mm); Infill density (ID, %); Nozzle Diameter (ND, mm); Extrusion Multiplier (EM, %); Platform Temperature (PlatT); Wall thickness (WT, mm).
The reader can be seen the variability of the measured mean and max height of the ME part’s surface texture (Ra and Rt). Ra values between 1-50μm and Rt between 50-150μm have been reported showing how important parameter tuning was in the ME process.
Analytically, for upper surfaces of dogbone parts, Ra values were between 0.8206 μm and 6.5526 μm for pure PLA,87 whereas for CF/PLA composites measured Ra>28μm.92 More, vertical Ra values were between 0.3 to 11.8μm for ABS P400 parts,79 where for incleaned ABS/ABSPlus/PC/ULTEM parts measured Ra values higher than 15μm and Rt higher than 75μm.80
It is noted that none of these studies summarises the surface texture in terms of inclined, vertical and flat surfaces as presented here. The surface texture of ME parts depends entirely on the arrangement of the processing parameters (perimeter, infill and top/bottom),98,99 which in turn are affected by ‘filament properties’,100 desired ‘strand welding’,75 the ‘stair-stepping’ effect,101 and part details.102
Figure 2 uses a ring-type hexagonal part to illustrate all cases of surface texture types engaged in the ME 3D printing process (the 3D model was designed in ‘3D builder’ and slice in ‘Creality v4.8.2’). In summary, we encounter flat, vertical or inclined surfaces. In addition, the inclined or flat top and bottom surfaces can meet without or be covered by a support structure, depending on the end user’s choice.103,104
Source: Author’s work.
Figure 3 illustrates all factors affecting ME surface texture using a cause-effect diagram. The texture of the perimeter, infill and top/bottom surfaces should be optimised according to filament properties, welding and staircase conditions.105,106 It is noted here that some 3D printing processing parameters affect the perimeter, infill, and texture of the top/bottom surface differently.107 For example, the width of the strand is not so important for vertical surfaces, but necessary to adjust the texture of the flat surface (top and bottom). Therefore, the quality and quantity of flat surfaces can affect the strand width choice, affecting the part’s time, energy consumption and mechanical strength.108
Unfortunately, a comprehensive examination and detailed study on the surface texture of the ME 3D printing process in the context defined in Figure 3 has not yet existed in the literature. This study simultaneously considers the parameters’ effects on the material extrusion process’s surface texture regarding flat, vertical, and inclined surfaces. Though material type, component orientation, thermomechanical parameters, and total weld area affect surface types differently, this work shows the importance of optimizing the material extrusion process for different materials and component orientations, which is not emphasized in the existing literature. In addition, it summarises the most influential analytical and experimental work that has been done to study welding, stair-stepping, and filament effects, focusing on inclined vertical and planar (top and bottom) surface texture. Last, it communicates surface texturing issues and prospects for further study by the 3D research community. The results can help achieve a uniform surface texture and reduce post-processing and anisotropy of material extrusion parts.
Each part manufactured by the ME process contains successive flat layers. Separately layer includes one or more perimeter lines and a defined infill structure according to the pattern selected by the end user (see Figure 4).109 Infill follows the same configuration (pattern) along the Z-axis except for the flat top and bottom layers.110,111 The user can select zero, one or more top and bottom layers.112 In addition, support lines are created to support inclined sections or “island” type geometry.113
Source: Author’s work.
Infill structures are deposited within the perimeter of each layer following several different deposition routes, e.g. lines,114 zigzag,115 grid,116 and tri-hexagon,117 to name a few (Figure 4a). The infill density determines the final part weight (porosity),118,119 while the pattern deposition angle determines the durability of the infill structure (static and dynamic mechanical response in the X and Y fabrication direction).120,121 Different infill styles have been invented to reduce component weight and achieve controlled mechanical strength,122 see gyroid infill structure.123,124 Last but not least, layer thickness, filament properties (base and filler material),125 nozzle inner and outer tip diameter (see Figure 4b),126 and infill density determine the mechanical strength in the Z direction, as it has been shown in the literature that a lower layer thickness is beneficial due to a reduction in the porosity of the component (see Figure 4c-d).127
The top and bottom surfaces are characterized by 100% infill because they should be solid and allow no gaps between the perimeters and infill lines.96 The flow speed (Q, mm/s), also known as flow rate (Fl, %), should be adjusted in cases of gaps or overflow material on top and bottom surfaces. Therefore an upper surface is affected by welding conditions the most, i.e., the material flow, strand width, deposition temperature, and printing speed; see Figure 5a-c. On the other hand, the bottom surface reflects the texture of the deposited surface and is also affected by the flow conditions of the material and the platform Z-leveling calibration; see Figure 5d-f.128,129 Figure 5 shows the top and bottom surfaces of PLA parts 3D printed with different welding conditions; printing temperature (T), deposition speed (S) and strand height (H). Again, variable surface texture quality is evident, even in the same sample case (Figure 5c).130 Last but not least, in Figure 5a, the outer diameter of the nozzle is formed on the top surface due to the concentric filling (from the outside to the inside).
Vertical and inclined surfaces primarily depend on the perimeters’ welding quality and on the infill characteristics. Bonding of the perimeters, infill lines, and successive layers of slices is achieved by applying thermal cycles between lines and layers,131 which are directly affected by material properties,132,133 processing parameters,134,135 part size,136 and orientation within the build bed.137,138 Therefore, the nozzle, platform and vat temperature must be adjusted appropriately and according to the part volume and orientation.139,140 Figure 6 illustrates the overflow problems during the ME process on different surfaces of the same part and with the same processing parameters. It is evident that some surface textures exhibit excellent quality and others are unacceptable due to material overflow.141
This section reviews the above surface texture case studies according to the factors influencing deposition and orientation, e.g. the stair-stepping, overflow, material properties, and layer trace cases.
Strand height (H) and surface orientation angle (θ) are the two directly influencing parameters (see Figure 7). In case the deposited material is assumed to have rectangular edges, the Ra and Rsm values are calculated with the following formulas (see explanation in Figure 7a):
Source: Author’s work.
According to Eq. 1, higher H values and orientation angles between 5-30 degrees result in higher Ra values (Figure 7b). Figure 7b shows the theoretical (Eq.1) trend lines for 100, 200 and 300 μm H concerning the orientation angle (θ) and the corresponding experimental values (exp 100, 200 and 300μm). Figures 7c-e show the specially designed ABS-FFF test part and the oriented surfaces on which the average roughness (Ra) has been measured experimentally. The observed values follow the trend line only for 100μm (0.1 mm) H. More; lower Ra values were observed on the bottom surfaces, while the top surface texture appears to have significant Ra values. Finally, for 200 and 300μm H, the highest experimental Ra values were observed for the orientation angle of 45 degrees. Similar results have been marked by Boschetto et al.,80 who applied artificial neural networks to predict surface roughness parameters (Ra, Rt, etc.).
The overflow effect is observed in all types of welding flat surfaces as well as on vertical surfaces. Top and bottom overflow is presented in Figure 5, where it is affected by the deposition conditions, i.e., the strand width, line distance, flow rate, nozzle inner and outer diameter, strand height, printing temperature, and printing speed. It is also possible to select zero top and bottom layers. In this case, the top and bottom layers follow the infill pattern grid style (see Figure 8). Therefore, the infill parameters must be adjusted appropriately so that voids are not observed (solid parts, 100% infill rate).
Source: Author’s work.
On vertical or inclined surfaces, overflow occurs at the edges or start and end points of a line when the previous or next line changes direction due to the acceleration or deceleration of the nozzle at those locations (see Figure 6).142 When the nozzle takes the printing velocity (S) the vertical and inclined surfaces have the profile of Figure 9a. This profile has gaps with max distance related of the layer height (H), material properties, and the flow rate (see Figure 9b). The gap height is about the same as the surface parameter Rt (difference between the highest peak and the deepest valley) and mean weave width about the Rsm surface parameter (see Eq. 2; the mean value of the width of the profile elements) which represents the layer height H for vertical surfaces (see Figure 9b).83
Source: Author’s work.
Finally in Figure 9c the case of the layer trace effect due to the deposition instability of strands is highlighted.143
The ME process material can be a pure thermoplastic or a composite with a pure base material and a filler.144 The filler could be carbon fibre,145 graphene oxide,146,147 glass fibers,148 carbon nanotubes,149 carbon black,150 or organic powder,151,152 to name a few.153 Thermoplastic materials are divided into amorphous (ABS,154,155 ABS/HDPE,156 HIPS,157 PET,158 TPU,159,160 etc.) and semi-crystalline (PLA,161,162 PA12,163 PEEK,164 etc.).
The material’s properties determine the welding conditions more as the nozzle, bed and vat temperatures are adjusted accordingly. For example, filament strength and indentation shear through feed wheels play a critical role in material flow conditions and stability issues during the material deposition. In addition, the filler material directly affects the texture of ME 3D printing parts. Figure 10 shows the effects of filament material on vertical surfaces. For example, in Figure 10a (pure PLA), the substrates are smooth with a uniform texture, while in Figure 10b, the filler affects the surface texture by causing defects such as pillows and voids (5 wt.% CNT filler). Further, in Figures 10c and 10d, the effects of wood powder filler on pure PLA are shown with similar defects on the vertical surfaces.
Source: Author’s work.
The surface texture of the ME part is most affected by the material properties, the orientation of the part in the vat and the welding conditions during strands deposition. This is evident due to the process’s thermomechanical state and the material’s selective deposition following the g-codes positioning approach. Therefore, ME process performance is affected by some common defects, i.e., underflow or overflow, underheating or overheating, layer shifting, gaps between layers-perimeters-strands, curling and irregular corners resulting from welding processing parameters’ unsuitable setting.14 As the ME process becomes more popular and broadens the choice of materials, surface roughness issues could become even more critical in future surface engineering applications, including metals and composites.165 Note that the surface texture is crucial for the component’s durability and functionality in parts’ assemblies and systems such as external and internal thread,166 helical gears,167 remanufacturing of damaged or broken parts cases,168 custom-made anatomical orthotics,169 etc. Thus, many materials that can be extruded in filament form should be tested and investigated for surface texture and geometric quality characteristics,170 roughness,95 conductivity,171 hardness,94 friction properties,172,173 tribological and wear properties,174 and aesthetics,175 of 3D printed parts in all directions, not just on flat surfaces of mechanical test specimens.176
Additionally, intent shear effects of the filament during nozzle feeding, nozzle outer vs inner diameter associated with each material composition, the appropriate flow rate for each nozzle type vs material selection combination, and optimization of welding parameters are just some of the future challenges of the ME process on the surface texture quality issues of 3D printed parts. In addition, stability issues that will improve nozzle effects on layer traces, material overflow effect in edges, and texture weaving approaches could further enhance the surface quality of ME 3D printed parts.177
Considering all the above issues, the surface texture of the ME part can be significantly improved uniformly in all the mentioned cases below 10-15μm, reducing the time and cost of further post-processing by blasting,136 chemical treatment,96,170 or machining.97 For example, Francis et al.,96 improved surface texture to 0.819μm from 9μm with 50s immersion time in acetone solvent.
Last but not least, the effect of the support parameters on the surface roughness of the top and bottom surfaces is another challenge for the ME process (see Figure 2). Therefore, it should be studied concerning the characteristics and integrity of the ME parts.
In the ME process, many parameters affect the surface texture and properties of 3D-printed parts. In this study, the types of surface texture encountered in the ME process were first analysed. e.g., top, bottom, vertical and sloped and after studying the effects of the stair-stepping, overflow and filament material. This study presents for the first time in detail the complex process of surface texture control in the ME 3D-printing process as, for each case, some processing parameters have competing effects and others synergistic effects.
• Surface orientation (θ) and layer height defined by the strand height (H) are significant for the stair-stepping effect.
• Material filament fillers negatively affect all surfaces, increasing the Ra and Rt values. Today, new materials are being investigated, such as metal powders on polymer bases. As a result, researchers are constantly expanding the materials used in ME.
• Strand width (w), printing speed (S) and flow rate affect the overflow effect and are significant for vertical, inclined and flat surfaces.
• Surface texture properties can be significantly improved by 3D printing parameters optimization.
• Post-processing time and cost can consequently be reduced considerably.
Improving the ME surface texture properties will speed up the ME technology in more surface technology applications, such as mechanical applications in systems and assemblies or remanufacturing cases.
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Is the topic of the review discussed comprehensively in the context of the current literature?
Yes
Are all factual statements correct and adequately supported by citations?
Yes
Is the review written in accessible language?
Yes
Are the conclusions drawn appropriate in the context of the current research literature?
Yes
References
1. Baila D, Sanfilippo F, Savu T, Górski F, et al.: 3D printing of personalised stents using new advanced photopolymerizable resins and Ti-6Al-4V alloy. Rapid Prototyping Journal. 2024; 30 (4): 696-710 Publisher Full TextCompeting Interests: No competing interests were disclosed.
Reviewer Expertise: Additive Manufacturing
Is the topic of the review discussed comprehensively in the context of the current literature?
Yes
Are all factual statements correct and adequately supported by citations?
Yes
Is the review written in accessible language?
Yes
Are the conclusions drawn appropriate in the context of the current research literature?
Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: FDM 3D Printing
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | ||
---|---|---|
1 | 2 | |
Version 1 10 Sep 24 |
read | read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)