Surface Durability of 3D-Printed Polymer Gears

Abstract

This paper proposes a series of experimental determinations carried out with the aim of generating new conclusions regarding the ability of 3D-printed gears to be integrated into mechanisms without lubrication. The main factors that influence the appearance of wear in non-lubricated contact are sliding speed, material hardness, surface finish, surface geometry, and material microstructure. The tests considered the type of material from which they were made and the 3D printing technology type. For testing the gear wheels, a mechatronic experimental setup was made consisting of two shafts with adjustable axial distances, a wheel loading system gears, an electric motor, and a command-and-control system. In terms of materials, four types of materials were monitored: PA (polyamide), PLA (polylactic acid), ABS (acrylonitrile butadiene styrene) and PP (photopolymer). The evaluation of the gear wear was carried out by checking the gearing on two flanks (Frenco ZWP 06) and by scanning with the ATOS CORE 135 3D scanner. The PA and PP gears failed to meet the structural integrity standards after the tests. The PLA gears exhibited superior resistance to abrasive wear compared to the ABS gears, whereas the ABS gears generally demonstrated stronger structural integrity.

1. Introduction

3D printing, also known as additive manufacturing (AM) or rapid prototyping (RP), provides several advantages over traditional energy-intensive methods. These include the ability to produce complex geometries as single units without joints, reduced material and labor costs, lower energy requirements, and simplified processing. However, a prominent limitation of this technology manifests in the precision and surface texture of the produced components.

Gears made using 3D printing technology represent a good and efficient alternative for mechanisms and machine elements intended for specific mechatronics applications, like industrial robots, robotic surgery systems and diagnostic equipment, printers, scanners, digital cameras and optical equipment, unmanned aerial and ground vehicles, etc. Studies and research in the field are at the beginning, which is why there are no regulations to establish the precision and wear limits of gears obtained through 3D printing technologies. Current studies generally present experimental attempts with good results for determining the mechanical properties of parts made by such technologies [1,2,3]. The studies evaluate, in addition to the 3D printing technology, the printing parameters and the type of material used [4,5,6]. The thickness of the printing layer influences the mechanical parameters of the printed parts. A greater thickness of the printing layer generates a better overall strength of the structure [7,8,9,10]. The size of the printing layer is inversely proportional to the friction force in the case of FDM–ABS/PLA 3D-printed parts [11] (fused deposition modeling–acrylonitrile butadiene styrene/polylactic acid). If we consider the lower friction surface in the case of a lower printing resolution, this conclusion is normal, although there are studies with contradictory results. The friction force in the samples made of PLA is higher compared to those made of ABS [11]. As a general conclusion regarding the influence of the thickness of the printing layer, we can state that the lower the printing resolution, the better the mechanical and tribological performance. Fewer bonds between layers generate superior mechanical properties [12,13]. However, we must consider that the thickness of the layer directly proportionally influences the printing precision, and therefore, the precision of the parts. Moreover, in the case of gears with toothed wheels, the influence of the printing resolution on the degree of wear must be determined. For FDM technology, a minimum coefficient of friction correlated with a degree of wear has been identified. This shows a linear character for a filling percentage with material of 50% and a layer thickness of 0.1 mm in the case of ABS parts, and a percentage of filling of 50% correlated with the thickness of the printing layer of 0.15 mm for those made of PLA [14]. In the general formulation, through a multi-objective optimization process, the optimal structure specific to the application can be identified from the printing parameters. Another important aspect is represented by the influence of the melting temperature of the material on the tribological properties of the samples. In the case of FDM-PLA printing technology, the most widespread method–material combination currently used, according to the coefficient of friction, the printing temperature has a non-linear influence, obtaining an optimal value for temperatures between 210 and 220 °C depending on the type of material [15]. This value can be correlated with an optimal level of finishing for the surface [16,17]. Another particularly important and not sufficiently well-documented aspect up to this point is represented by wear resistance. As mentioned before, the studies show better results in the case of samples with a greater thickness of the printing layer subjected to wear tests. Based on the material, in general terms, parts made of ABS have less wear compared to PLA. These results find their foundations in the glass transition temperature values that characterize these materials [18]. However, wear resistance requires additional research because the problem has a nonlinear character relative to the specific application. A high print resolution is equivalent to higher frictional forces but not necessarily to high wear under any conditions.

The fundamental studies are complemented by specific research that presents structures and systems made of 3D-printed components. Theoretical and experimental analyses have been undertaken to investigate the mechanical characteristics of gears fabricated using 3D printing technology across varied operational parameters [19,20,21].

By monitoring the temperature of a gear made by 3D printing, according to the vibrations induced by it, it was found that gears made of PLA showed better performance compared to those made of ABS [22]. In another investigation [23], an examination of PLA, ABS, and nylon gears revealed that the PLA gears exhibited fracture during the tests, while the nylon gears demonstrated superior performance.

Considering the fundamental research that shows better characteristics for parts made of ABS compared to PLA, we can see the complexity of the problem generated by the flexibility of the technology.

The final objective of this research concerns the introduction of standards in a field that presents a spectacular dynamic. The International Organization for Standardization (ISO) established the technical committee ISO/TC 261 for additive manufacturing (AM), whose role is to develop common standards that can be applied globally [24]. These regulations are to establish a general technical basis, and the basis is the research in the field carried out worldwide, which, at this moment, is in an early phase [25].

2. Materials and Methods

The study that is the object of this paper aimed to generate conclusions regarding the influence of the material and the manufacturing technology on the wear performance of gears obtained through additive technologies. The tests carried out were based on the premises that the gears will be integrated into mechatronic/opto-mechatronic systems where precision is a necessity. The maximum loads used were generated by the conditions of the domain.

2.1. Sample Preparation

There were four 3D-printed gears considered, manufactured from PA (polyamide), PLA (polylactic acid), ABS (acrylonitrile butadiene styrene) and PP (photopolymer). All four gears ($z_w$) had the same modulus, m = 1 mm, number of teeth, z = 56, and gear thickness, b = 12 mm. FDM (fused deposition modeling) and SLA (stereolithography) were used as the printing technologies. In the case of FDM technology (Figure 1), the Anycubic Mega S printer produced by HONGKONG ANYCUBIC TECHNOLOGY CO., LTD., Hong Kong, China was used. The Anycubic Mega-S printer is capable of achieving layer heights as low as 0.05 mm (50 microns), allowing for relatively high precision prints. Through FDM, three gears were made from different materials: PA (polyamide), PLA, and ABS. Regarding the structure, the printing parameters were identical in the case of the three samples, namely a degree of filling of 100% and size of the printing layer of 0.2 mm. A filling degree of 100% was adopted to avoid errors introduced by possible changes in the basic structure of the evaluated models.

The structural resistance of the tested models was not the object of this research; the aspects related to the wear of the teeth were strictly of interest.

The printing and bed temperatures (

Table 1. 3D printing parameters.

Material	PA	ABS	PLA	PP
3D printing technology	FDM	FDM	FDM	SLA
Printing temperature	250 °C	240 °C	210 °C	NA
Bed temperature	110 °C	105 °C	50 °C	NA
Radiation (nm)	NA	NA	NA	405

) were set according to the specific recommendations for each material [26,27,28] and by printing samples with different printing temperatures.

In the investigation of PA gears, the research underscores the significance of the moisture levels in relation to the tensile strength, surpassing the influence of the printing temperature [29]. Taking into account the extruder’s maximum operating temperature of 260° Celsius, the printing temperature for the PA toothed wheel was set to 250° Celsius.

Most studies recommend 215 °C as the best printing temperature for PLA. This is because the best tensile performance of PLA is achieved at this temperature [30]. After conducting several 3D printing experiments on the Anycubic Mega-S, a temperature of 210 °C was determined as optimal for printing the PLA gear to mitigate issues of excessive stringing/oozing.

For ABS filaments, the recommended printing temperature typically ranges between 220° and 250 °C [30,31], with variations based on specific printer models, filament quality, and desired print characteristics. A commonly recommended starting point is around 230 °C to 240 °C. The experimental tests conducted determined the optimal printing temperature to be 240 °C.

In the case of the toothed wheel executed by SLA, an Anycubic Photon printer with ANYCUBIC resin photosensitive to radiation with a wavelength of 405 nm was used. Both the printer and the resin are products of HONGKONG ANYCUBIC TECHNOLOGY Co., Ltd., Hong Kong, China.

2.2. Experimental Setup for Testing Gears

In order to identify the level of wear that appears on the gears made by additive technologies, an experimental setup was designed and realized. The stand consisted of: an ARDUINO type development board, a bipolar stepper motor with a gear reducer that supported a maximum torque of 3 N·m; a TB 660 driver with a maximum current of 4A; and a mechanical system for gearing and requesting the gears, consisting of two shafts (Figure 2). A pulley-type wheel with a diameter of 150 mm was fixed on the driven shaft, on which a wire was wound. At the free end of the wire, the weights were fixed. The setup allowed for loading two gears with a maximum torque of 1 N·m. Changing the weights allowed for strict control over the load. Considering the gears, a metal reference gear and gear made by additive technologies were used. The reference gear was made in the Q = 7 quality class, with a modulus, m = 1 mm, number of teeth, z = 56, and the same gear thickness as the workpieces, b = 12 mm. The reference steel workpiece gear was manufactured on a special gear-cutting drill machine.

This test method was adopted to monitor wear relative to a reference element. The program was designed so that changing the direction of rotation stressed each flank of each tooth of the gear 100 times.

Regarding dynamics, each flank of each tooth was loaded 5600 times (

Table 2. Test parameters.

Material	PA	ABS	PLA	PP
3D printing technology	FDM	FDM	FDM	SLA
Operating cycles	5600	5600	5600	143
Torque (Nm)	0.75	0.75	0.75	0.75

). Considering the gear evaluation methods, the algorithm of the program was designed so that there was uniform stress on the gear.

3. Test Results and Discussion

A comparison method was used to monitor the gears’ abrasive wear. The gears were evaluated before testing (BT) and after testing (AT). The toothed wheel made by SLA did not withstand the imposed loads (Figure 3). This was due to the properties of the material from which it was made—material that gives it increased rigidity under the conditions of low molecular bonds.

For gears made of photopolymer by SLA, research can be oriented toward obtaining models that can be integrated into micro/nanostructures where the loads are low.

The evaluation of the gears that resisted the imposed loads was carried out on two distinct systems: double flank gear roll inspection machine ZWP 06 FRENCO and 3D scanner ATOS CORE 135.

3.1. Gear Evaluation on ZWP 06 FRENCO Machine

The ZWP 06 is a double-flank gear roll inspection machine specially designed for high-precision gears. FGIpro 2018 Software by FRENCO was used to record the data.

The software determined the following parameters (Figure 4):

• The total radial composite deviation Fi” is the difference between the largest and the smallest center distance Cd” during one rotation of the inspected gear (DIN3963 [32]);
• The tooth-to-tooth radial composite deviation fi” is the largest difference between the center distance Cd”, which occurs at an angle of rotation corresponding to the duration of engagement of one tooth (DIN3963 [32]);
• The runout deviation by the composite test Fr” is the value of radial runout of the gear between the maximum and the minimum radial distance from the gear axis as observed by removing the short-term or undulation pitch deviations and analyzing the long-term sinusoidal waveform;
• The short-wave component fk”.

During double-flank gear rolling inspection, two gears are rolled together free from backlash, which is suitable for plastic gears (Figure 5). This determines gear roll variations, tooth-to-tooth composite deviations, composite runout deviations, and the short-wave part and displays them clearly. Immediately after the measurement, it can be seen whether the gear is acceptable or not. From the colors, the visual comparison of the actual values and the reference value is made easier. For each deviation, different colors are used, and the operator can easily observe the wrong deviations.

The gears were evaluated on the ZWP 06 FRENCO before the load tests (BT) and after the load tests (AT). The master gear ($z_m$) had, of course, a modulus of m_m = 1 mm, number of teeth z_m = 48, gear thickness of b_m = 15 mm, and gearing angle of α = 20° (

Table 3. Geometric parameters of master gear and workpiece gears.

Element	zm	zw
m, modulus (mm)	1	1
z, teeth number (-)	48	56
d, pitch diameter (mm)	48	56
x, displacement coefficient (-)	0	0
dw, rolling diameter (mm)	48	56
ha, height of addendum (mm)	1	1
hf, height of foot (mm)	1.25	1.25
da, diameter of addendum (mm)	50	58
df, foot diameter (mm)	45.5	53.5
db, base diameter (mm)	45.105	52.623
αp, gearing angle (°)	25.56386	24.86658
h0a, height of addendum coefficient (-)	1	1
h0f, height of foot coefficient (-)	1.25	1.25

).

In analyzing the results, changes were generated by wear in the case of all spur gear samples (

Table 4. ZWP 06 FRENCO evaluation results.

Gear	Fi” (mm)		Fr” (mm)		fi” (mm)		fk” (mm)
	BT	AT	BT	AT	BT	AT	BT	AT
PA	0.146	0.225	0.125	0.209	0.103	0.104	0.055	0.065
PLA	0.262	0.261	0.183	0.146	0.192	0.17	0.102	0.07
ABS	0.284	0.223	0.205	0.142	0.096	0.104	0.087	0.076
PP	0.298	NA	0.109	NA	0.291	NA	0.124	NA

).

Considering that the objective of the evaluation was not the inclusion of gears in a precision class but to establish the degree of wear, the results were processed in order to monitor the evolution of the reference parameters of the wheel after testing.

Table 5. Fi”/fi” interval values before tests (BT) and after tests (AT).

Material	Fi”/fi” (mm)
	BT (Before Tests)		AT (After Tests)
	MIN	MAX	MIN	MAX
PA	−0.015	0.03	−0.102	0.124
PLA	−0.066	0.196	−0.077	0.183
ABS	−0.117	0.166	−0.099	0.124

shows the extracted and averaged results obtained after the evaluation of the gears on the ZWP 06 FRENCO system.

The gear made of PA and subjected to the wear test showed an evolution of the total radial composite deviation (Fi”) with a non-linear character. Both the positive deviation and the negative deviation compared to the reference increased in absolute value. Based on the values compared to the reference wheel, there was an increase in the total radial deviation by approximately 55%. The increase in the maximum positive deviation was not an expected result. The most important aspect is represented by the character of the changes that took place. The general results indicate significant modification of the tooth profile (Figure 6).

Visually, the toothed wheel showed, in certain areas, small detachments of material at the tip of the tooth on the laterals of gear. Those small pieces of material can explain the obtained results. The accentuation of the negative deviation suggests the appearance of wear in this area.

The PLA gear showed linear changes in the measured parameters. Wear introduced a small difference in the total radial composite deviation Fi (0.262 mm before vs. 0.261 mm after). Theoretically, the gear after the load tests showed a slightly improved performance. Furthermore, the polar view generated through the FRENCO system (Figure 7) showed a constant repositioning of the MIN–MAX interval for Fi”/fi” relative to the reference. The interval offset had an average value of −0.01 mm and uniform behavior on the entire circumference of the gear. This decrease in deviation suggests uniform wear of the toothed wheel. Moreover, the Fr” factor had an improved value for the gear subjected to the load test (0.183 vs. 0.146). This improvement is justified by the elimination through wear of the execution errors generated by the technological process of making the gear. During the 3D printing process, inhomogeneity of the filament and variation in the environmental conditions generates extra material on the printer nozzle. The deposition of that material at the level of the teeth is equivalent to the appearance of the maximum deviation point during the evaluation of the gear. In this case, due to wear, the peaks of the total radial composite deviation that correspond to these errors decreased in value after the stress tests. In the real case of gearing between two gears obtained through the same 3D printing technology, the problem is more complex. However, these aspects are strictly related to the capabilities of the technologies.

The gear made of ABS shows a similar evolution to the PLA toothed wheel (Figure 8). At the level of values, the shift in the variation range of deviation was approximately −0.02 mm, which suggests more pronounced wear. Regarding wear, the gear made of ABS showed weaker characteristics compared to the gear made of PLA, but without elements generated by errors of the printing process.

The degree of wear of the ABS gear shows perfect uniformity, but higher compared to the gear made of PLA. This aspect does not perfectly align with initial expectations but is in accordance with the results of fundamental research on the material.

3.2. 3D Scanning Evaluation

To objectively quantify the effects of wear, the wheels were checked using an optical scanning system. For the evaluation of the samples, the 3D ATOS CORE 135 scanner was used. GOM INSPECT 2022 software was used for data recording and processing.

The evaluation was carried out in a comparative mode; the gears were evaluated before and after the load test. The scanned models were compared, and the differences were evaluated. The monitored points in the evaluation process were chosen starting from the distribution of the maximum relative speeds at the contact point. For a perfect gearing, the most pronounced wear occurs near the base of the tooth and at its tip (Figure 9).

For the gear made of PA, pronounced wear was observed. On each tooth, at the point considered most susceptible to wear, the registered dimensional differences between the gear before and after the tests have values in the range of [$-0.01 \mathrm{m}\mathrm{m}; -0.14\mathrm{m}\mathrm{m}]$ (Figure 10).

The relevant conclusion after this evaluation is a negative evolution of the dimensional differences. However, during the tests carried out on the ZWP 06 FRENCO system, an increase in the Fi” was found.

Analyzing the 3D-scanned model in detail (Figure 11), the differences between the two conclusions, which, in a first analysis, are contradictory, can be justified. The increase in the “Fi” was registered due to the fact that the load applied during testing generated a high degree of damage to the gear. The thicknesses of the gears from the test stand (the metal gear and the polymer gear under test) are identical and the high loads generated micro displacements of the material at the level of the tooth.

Presenting a high elasticity coefficient, pieces of material from the tooth moved outward toward the edges, generating an area with extra material. The reference gear used on the ZWP 06 system had a much larger thickness than the polymer gear, and the gearing during the evaluation included the areas with extra material as well, thus generating a positive total radial composite deviation.

The results obtained after the evaluation by scanning the PLA and ABS gears support the conclusions obtained after the evaluation of the samples on the ZWP 06 FRENCO machine and provide additional information on the evolution of wear at the level of the teeth. For both gears, PLA and ABS, the highest degree of wear was recorded at the level of the tip of the teeth (Figure 10 and Figure 11). For the PLA gear, the values recorded for wear on the 3D-scanned model show an average of −0.03 mm (Figure 12).

However, we need to consider that a difference generated by wear at a certain point does not translate into wear at the level of the gear due to the fact that the roughness of the surfaces that characterize the gear tooths is very high, and a possible difference at a certain point can cancel a noise type element. Moreover, at the base of the teeth, an increase in the gear hub width can be observed. This phenomenon was also present in the case of the gear made of PA, but in this case, was not so pronounced.

For the gear made of ABS, the conclusions are obvious: there was pronounced wear at the tip of the tooth, and the thickness of the gear at the base circle was not affected by the phenomenon of the outwardly displaced material (Figure 13).

4. Conclusions

The results obtained are important for future research whose objective is to determine the capabilities of gears made by additive technologies.

Manufacturing photopolymer gears by SLA is not a solution for small- or medium-load mechanisms. Future studies need to determine the capabilities of micro/nano gears made using this technology.

Considering the diversity of materials that can be integrated with FMD technology, this remains an option with great potential for manufacturing gears.

The wear test is decisive in any industry, but it has the disadvantage that it requires a long time.

The toothed wheel made of PA did not withstand the imposed load from a structural point of view. The value of average dimensional differences recorded at the level of the teeth between gear before and after tests was 0.05 mm. The behavior of the gear suggests good resistance to wear of the material. Additional tests with lower loads are needed to concretely determine the capabilities of using PA for making gears.

The PLA gear showed the best wear behavior. The abrasive wear generated a modification of the total radial composite deviation Fi” of the gear by only 0.013 mm. Based on the changes at the toothing level of the gear, in the area where the sliding speed had the maximum value, an average dimensional decrease of approximately 0.04 mm was registered. These modifications had a uniform character. Furthermore, at the base of the teeth, material deposits were observed. These deposits generated changes in the radius of the gear foot circle by +0.06 mm. From a structural point of view, the toothed wheel presents certain problems generated by the elasticity of the material. Through a good correlation of the load with the structure, gears made of PLA can be perfectly integrated into a functional mechanism. This study remains open with the aim of correlating the parameters of the gear with the loading of the system, based on the limits imposed by the wear resistance of the material.

The toothed wheel made of ABS had good structural performance but low wear resistance. The modification of the total radial composite deviation Fi” was three times higher than that of the PLA gear. These results indicate that small loads can be tolerated by the material.

Among the four gears tested (photosensitive resin, PA, PLA, and ABS), the PLA gear manufactured through FMD showed the best wear behavior.