Tribological Analysis and Thermal Behavior of Gear Tooth Flanks in Polymer Gears During Dry Operation

Polymer gears applications are growing day by day in some specific mechanical applications for their valuable advantages, such as the cheaper, easier, and faster production, in addition to its lower noise running, in comparison to metal [1,2]. Their capability to run without lubrication or with low or different mediums of lubricant makes them very useful and some important application that requires special conditions of operation, which makes them one of the useful counterparts of metallic gears [3]. Nonmetallic gears can be widely found in many different mechanical applications nowadays, starting from stationary printers to kitchen appliances, ending with cars that are powered mainly with electricity [4]. A technical project aimed at replacing certain metal gears and the housing of the transmission system in the Visio.M electric vehicle demonstrated a significant reduction in both operating costs and weight, along with improvements in the production process [5]. In terms of raw material production, polymers are more environmentally friendly than metals [6]. Moreover, for the drawback of lubrication in metal gears, their polymer counterparts are the good choice in medical and laboratory machineries where lubricants are not preferred [7]. Similar consideration applies with any dry-running machinery requirement [8].

For all the advantages of polymer gearing applications, they are taking over metal gears in many specific mechanical applications. The average number of polymer gears produced in 2005 was around 2 billion, and this figure is steadily growing thereafter, competing their metal counterpart [9,10]. On the other hand, there are still some technical limitations that limit polymer gearing applications to low load capacity applications, some of which is the lack of knowledge of thermal behavior on high-load running, as it is known that polymer gear tooth is more affected by thermal factor than load and stress concentrations [11,12]. Another important factor is wear and wear-rate on gear tooth flank. Some investigations were carried out to understand these parameters and their effect on operations at different conditions [13–17]. Prior studies have highlighted that polymer gears are more sensitive to thermal effects than metallic counterparts, with temperature playing a critical role in gear wear and failure [18,19]. However, the focus has often been on short-duration experiments or static measurements, with limited emphasis on real-time thermal behavior under continuous operation. Although good work has been carried out, this research area is still under development, while more investigations are required for the benefit of the advantages of such mechanical part. Some polymer gear standards were established to help with design and rating, but they were partially driven from metal gear standards, and lack of temperature calculations, which leads to some rating limitations [20]. Thermal investigation was undergoing in 2015 [21], while polymer gear tooth flank thermal behavior under running condition is more required.

While substantial research has been conducted on the wear behavior and mechanical performance of polymer gears [22–24], relatively few studies have focused on the real-time thermal behavior of polymer gear tooth flanks under continuous, long-duration running conditions. Early work [25] explored the mechanical contact behavior of polymer gears, contributing to our understanding of load-bearing capacities under dry-running environments but did not investigate the dynamic interaction between temperature and wear behavior over prolonged operational periods. However, a significant research gap persists concerning the interaction between temperature and wear behavior, particularly when subjected to prolonged dry-running conditions at higher loads and speeds. This study bridges that gap by presenting continuous thermal data and analyzing its impact on wear mechanisms.

The early investigation was carried out to understand the wear-rate behavior of polymer gears made of different materials [25]. This paper will focus more on the experimental continuous readings of polymer gear tooth surface temperature while in continuous running condition in line with tooth flank wear and wear-rate. The main aim of the study is to understand the surface heat build-up phenomena at certain time of contentious running. The data were logged, analyzed, and presented with general discussions.

This study aims to address this gap by presenting real-time, continuous thermal data during the operation of polymer gears, focusing on the relationship between surface temperature and wear-rate. This study combines high-resolution thermal imaging with tribological analysis, building upon the foundational research of Singh et al. [12] and others, while providing fresh insights into the failure mechanisms and thermal limits of polymer gears during extended operation. This research contributes to both the empirical and theoretical understanding of how temperature influences the performance and durability of polymer gears, particularly in dry-running conditions, where high surface temperatures can lead to accelerated wear or even thermal stabilization effects under certain conditions. By analyzing temperature profiles and wear-rates simultaneously, this work demonstrates how surface heat build-up influences wear mechanisms, including self-lubrication and material softening, which can reduce wear-rates at certain critical conditions.

The results articulated in this investigation enhance the comprehension of the thermal and tribological properties exhibited by polymer gears. The findings are expected to inform future design and material selection for polymer gears, particularly for high-load, dry-running applications where temperature management is critical. This work not only builds on prior research but also fills key gaps by offering continuous, quantitative data that enhance both the empirical and theoretical understanding of polymer gear performance under real-world conditions.

To satisfy the main aim of the research, a specified research method was designed, including a selection of testing and recording equipment and tools. The primary test rig, schematically shown in Fig. 1, was designed to provide continuous measurement of gear tooth thickness during operation (features and specifications were provided in deep details in previous work [25]). Rig's function applies the gear back-to-back layout, where two metal gears engage with polymer gears through two primary shafts, with one of the shafts featuring a conical clutch for initial positioning adjustment. Tested gears are loaded using the counterweight arm set that guarantees a constant moment load through the testing time, regardless of the change in gear tooth shape. In case of polymer gear teeth full damage, loaded arm will descend to the final end of the base platform allowing the assembly block to touch the cutoff switch for safety complete stop of the rig. While running, the movement of the assembly block around the pivot joint represents the change in tested gear tooth shape and thickness. This change is measured and recorded using the linear variable differential transformer (LVDT) that is attached above the block after being calibrated using slip gauges calibration to the accuracy of 0.1 µm. The LVDT measures the vertical displacement of the block (neglecting the horizontal component), which directly represents the amount of tooth surface wear or deformation to gear tooth bitch line direction (wear axis) (Fig. 2). The displacement in µm is logged continuously to the computer and plotted against time in microsecond, which by differentiation gives the wear-rate of the tooth flank surface at any certain period of time.

As can be seen in Fig. 2, the measured displacement may include some tooth deformation due to the applied load with the surface wear reading. This deformation is not removable from wear measurements with the current test setup. After number of tests, it was realized that this deformation occurs mostly at the early stage of running and reduced later after teeth are settled down. To reduce this effect of wear misreading from the results, the initial period of running was neglected from the result. In addition, wear-rate (per number of cycles) was defined from the nearly linear period of wear readings for more accuracy and the elimination of deformation overlap.

For boarder understanding of polymer gear wear-rate and surface heat build-up for different materials, two of the most commonly used polymer materials in gearing applications were chosen, which are nylon (PA) and acetal (POM). Gears were designed and in-house manufactured using a machining (MC) method. Two sets of samples were designed and prepared as per Table 1. These materials are commonly chosen for different previous studies, including pin on disk and nonconformal rolling sliding disk methods, which make the result validation and comparison more reasonable with previous studies. Gear specifications were chosen according to standards to satisfy the test rig requirements (Table 2). The gear specifications were determined based on the dimensions and requirements of the test rig to ensure compatibility and accurate data collection. These specifications were aligned with the mechanical properties and operational conditions set by the rig, allowing for optimal testing of the polymer gear materials.

The surface roughness of the machined gears was not delineated with precision, as it was deemed non-essential for the objectives of this investigation. During testing, the gear surfaces were polished once in contact and during running, which provided sufficient surface quality for the analysis. The focus of the testing was on the wear and thermal behavior of the polymer gears rather than surface roughness, as the polishing effect during operation ensured that the surfaces were sufficiently smoothed.

The gears were exposed to three unique loads—6.5 N m, 8.5 N m, and 9.0 N m—with the aim of assessing wear and thermal efficacy throughout various operational conditions. All tests were conducted at a constant rotational speed of 1500 rpm to replicate high-speed applications typical of polymer gears. The selected materials, PA and POM, were manufactured using the machining method. This resulted in different sample sets, as summarized in Table 1. Each experimental arrangement was conducted in triplicate to ensure the statistical strength and reproducibility of the results.

The gear tooth surface flank temperature was continuously measured while in continuous running using a high-definition thermal imaging camera (FLIR T425). Because of the gear's high-speed running, the continuous recording was measured at a rate of 1000 frame per second. The highest temperature readings were recorded at several time intervals and averages were defined. The camera is capable of measuring temperatures in polymer materials from −20 °C to 1200 °C, with a thermal sensitivity of below 0.05 °C, at ambient conditions with an accuracy of ±2%. It was installed directly above both gears (Fig. 3) and connected to a computer for continuous data recording using flir research ir software. The initial parameters are provided in Table 3. Both driving and driven gear highest surface temperature were measured simultaneously and results were plotted against gear cycle number.

The material attributes specified in Table 1, encompassing tensile strength (TS), flexural modulus (FM), and particular characteristic factors (static coefficient of friction (SCF) and dynamic coefficient of friction (DCF)), demonstrate no correlation with temperature. The melting temperature (MT) values are provided in the table for reference, offering context for material behavior under thermal conditions. The gear specifications, detailed in Table 2, remained constant throughout all tests, with no change in axis distance observed due to wear. The gears were manufactured in two quality grades (46 and 66), ensuring consistency across materials and manufacturing methods.

To strengthen the discussion, additional analysis was included to correlate temperature profiles with wear-rates over multiple load levels. Figures presenting both temperature trends and corresponding wear-rates have been updated with annotations to highlight key findings. The discussion now explicitly addresses how thermal effects influence wear progression and explains the conditions under which thermal stabilization occurs, making the novelty of this work more evident compared to previous studies.

The load capacity of the tested gears was defined using a set of step loading tests commenced previously [25]. From which, the maximum gear load capacity was defined, leading to a good indication of gear test parameter design. A general conclusion was drawn that more long running and endurance test were required to gain more understanding of gear tooth wear behavior and surface tribology. In addition, more thermal investigation was required to further understand the failure modes on tooth flanks. Fatigue life of the fabricated gears was not explicitly measured in this study, as the scope was limited to investigating wear behavior and surface thermal effects under long-duration, high-load conditions. While fatigue life measurements are outside the study's objectives, the results presented here provide essential insights into the tribological and thermal behavior of polymer gears, which are critical for understanding their operational limits and durability.

This work satisfies part of these requirements by testing certain materials of polymer gears that were chosen depending on the previous work results on certain test parameters as follows:

This type of polymer gears was tested at a load straight before the peak point to understand surface failure behavior, by continuously measuring wear and surface temperature of gear tooth flank while running. Figure 4 illustrates the surface wear thickness of the tooth of MC acetal gear while operating continuously at 1000 rpm and maintaining a constant load of 7 N m. Pictorial views of wear-out gears can be seen in this figure. These figures illustrate the progression of wear and the corresponding surface tribological changes during operation, providing valuable insights into the wear mechanisms observed in the tested gears.

Normally, tooth wear curve grows through three different stages, starting at the running-in stage, then to the linear stage, and finally to the high increased wear and failure. The most important stage reveals the important results about the surface wear behavior and thermal phenomena in the linear stage. For that, test was interrupted during this stage to keep the surface shape and look for further surface tribology investigation.

By differentiating the linear part of the curve, it was determined that the wear-rate for machined acetal gears operating at 1000 rpm with a torque load of 7 N m is 8.55 × 10⁻⁷ mm per cycle. This result reasonably compromises with the step loading test which was done previously [25], where the result at the same load was 9.66 × 10⁻⁷ mm/cycle. During the running time, gear debris were falling in a very small amount at both stages. Gear tooth thickness for both gears was measured and compared with the original using the shadowgraph device (enlarged up to 100 times) (Fig. 5). The figure also shows the sliding direction in both teeth. Thickness reduction measurement was compared with the wear measurement and showed small difference, which thought to be due to the tooth deflection measured with wear readings.

It can be seen from Fig. 5 that sliding occurs by small amount at pitch point of tooth flank, which slightly deviates with the theoretical rolling and sliding assumption that sliding at such point is around zero. This is thought to be due to the deflection of polymer gear teeth while under loading during mesh period in addition to the high Hertzian contact, especially with the polymer material.

The gears were weighed before and after testing to determine mass loss with an accuracy of 0.001 mg. This measurement was used to validate and compare the direct wear measurements, helping to assess the impact of tooth deflection on wear measurement readings. In addition, gear tooth thickness reduction measured and showed in Fig. 5 proves more validation, with the difference of 0.20 mm from wear reading, which proves the amount of tooth deflection.

Gear teeth surface and debris samples were further investigated under the scanning electron microscope (SEM). Figure 6 shows different SEM images for the driving and the driven gear teeth surfaces, at different locations. SEM around pitch line reveals friction as a result of sliding, which deviates from theoretical assumption that sliding around pitch line is zero, while rolling is the highest. The occurrence of such phenomena was thought to be a result of tooth deflection, which is more likely to increase at nonmetallic gears. Both gears showed nearly similar tribological phenomena with slight differences due to the sliding direction. At the driving gears, debris was observed to accumulate and migrate across the tooth surface. Specifically, debris originating near the pitch line appeared to spread both toward the root and tip regions of the tooth due to the contact pressures and sliding motion. Conversely, debris generated at the root and tip regions of the tooth was observed to move toward the pitch line under similar conditions. This movement is attributed to the combined effects of gear meshing forces and the direction of sliding contact during operation, which facilitate the redistribution of wear debris along the tooth surface. Tooth surface SEM indicates that due to the surface contact between the driver and driven, surface micro-chips are formed gradually and their edges are separated and move along the direction of motion. Such formation was thought to be due to the repetitive Hertzian contact at the surface contact area. While moving along the two contacted surfaces, those edges are rounded and twisted forming debris with micro-fiber shape with the size of around 10–50 μm, as can be seen in Fig. 6. Micro-fibers were primarily found concentrated around the pitch line area of the driven gear, whereas they were located at the root and tip edges of the teeth on the driving gear. Interestingly, debris at the tip edge of the driving gear was found highly pressed forming more flattened pieces than at the root side, which confirms the theory that pressure at the tip point of the tooth is higher than the pressure at the pitch point. Another explanation of the re-flattened pieces is the high surface temperature as a result of the contentious repeated high-pressure friction. Overall, it can be seen from all SEM that scratches wear is the most occurred type of wear along the surface contact area while running an acetal–acetal gear test, which is thought to be the result of rubbing the separated micro-fiber debris among running gears. Additionally, pitting wear was observed at some places on the contact area and attributed to the continuously repeated Hertzian loading throughout the long running time.

Although the opposite sliding direction on the driven gear differentiates the surface tribology phenomena in tooth flank in term of debris movement toward the pitch line of the tooth, gear tooth surface wear showed similar types. Scratch wear was discovered throughout the surface due to the presence of debris movement at the interface of the two rubbing materials. Adhesive wear was also noted around the tip area of the tooth, as a result of the high thermal effect in this area, which is thought to be because of the high surface pressure contact at the beginning of gear meshing cycle. This high-pressure contact caused more severe wear behavior at the addendum side than wear at the dedendum side of the tooth.

It was discovered previously [25] that PA gear showed some wear-rate variation with respect to load change, which was concluded that this unsteady wear-rate in such material is highly sensitive to surface thermal behavior. Therefore, to be able to understand the functionality and durability of such material in case of mechanical gearing applications, deep surface thermal investigation is required. In this work, the required investigation is satisfied by contentiously measuring the tooth flank surface temperature while in continuous running.

The step loading test done previously showed an interesting critical point where wear-rate slightly decreased with respect to load increase. This critical point is hypothesized to result from a thermal stabilization phenomenon, where the softened surface layer functions as an internal lubricant, reducing wear despite increased load. For further thermal investigation and more understanding of this phenomena, long run tests were commenced with load variation around the critical point. Figure 7 illustrates the tooth wear on a PA gear pair operating at 1000 rpm under loads of 6.5, 8.5, and 9 N m. Tests were left to run for around 3.6–4 × 10⁵ before they were halted at the nearly linear wear stage for further investigation of the surfaces using SEM. Once stopped, gear teeth seemed to regain part of its original shape after deflected under the applied load.

Figure 7 shows that the wear-rate of PA gear rose with increasing load between 6.5 and 8.5 N m. This increase was followed by a slight decrease in wear-rate with the increase of load by 0.5–9 N m. Higher sound and vibration were discovered at 8.5 N m load test than at 9 N m. These observations were based on aural inspection and subjective assessment conducted during the experiments. While quantitative measurements were not performed in this study, the aural observations provide preliminary insights into the behavioral differences between the two loading conditions. This aspect will be explored further with dedicated sound and vibration measurement tools in future studies to provide more quantitative evidence. Although wear-rate was higher at 8.5 N m, this trend was slightly decreased after running more than 3 × 10⁵ cycles, which thought to be because of two factors driven by the reduction in gear tooth thickness after worn out: first is the increase on gear contact ratio and second is the change in pressure angle.

Debris falling in all tests were monitored and recorded, which showed some variation. Under 6.5 N m load, debris were fallen in comparably smaller amount. Interestingly, gear debris were falling with higher amount at 8.5 N m than 9 N m. At 8.5 N m, debris started to fall in small amount at the running-in stage and followed by higher amount of debris with a small and thin profile. On the other hand, during the running-in stage of the 9 N m test, a smaller amount of debris was observed initially, which was succeeded by a larger amount of debris with a more substantial profile. Debris for each test load can be seen in SEM images, along with a comparative analysis of their profiles and quantities. These visual aids illustrate the observed trends and substantiate the discussion on debris behavior under different loading conditions. The color of debris observed at 9 N m test showed different color than the gear material color at first place, with slightly yellowish thin shield. This change in color was believed to result from surface melting at the contact areas due to high thermal effects. Thermal camera measurements support this hypothesis, as the recorded surface temperature approached the glass transition temperature of the PA material. Specifically, during the test, peak surface temperatures were observed to exceed 75 °C, aligning with conditions conducive to thermal softening and potential melting. These findings indicate that localized thermal effects contribute to material transformation at the contact areas, resulting in the observed color change. The tested gears were weighed, and their weights were compared to the pretest values to identify and minimize wear measurement errors caused by tooth deflection.

Similar results were reported by Hooke et al. [26], who conducted tests with two PA 66 discs in a non-matching contact configuration. They found that the wear-rate decreased with increasing slip ratio or load, although a general increase in wear-rate was observed with higher loads, except at the defined “critical point” where the rate of wear significantly decreased. They assessed the friction between the discs and used theoretical calculations to determine the peak surface temperature, including both average and flash temperatures. Their findings indicated that the sudden change in wear-rate was caused by thermal effects. To examine the reduction in wear-rate of PA gears with higher loads, surface temperatures were recorded for the 8.5 N m and 9 N m tests using a thermal imaging camera. Further analysis will be provided in the next section.

The increase in wear with increasing load for PA gears has been attributed to thermal effects and material softening during operation. At higher loads, surface temperatures approach the material's glass transition point, leading to thermal softening. This softening alters the material properties, causing increased wear due to reduced hardness and higher frictional forces. At the same time, thermal stabilization effects at specific load levels (e.g., 9 N m) reduce wear-rates by forming a softened surface layer that functions as an internal lubricant. These observations align with prior studies, such as those by Hooke et al. [26], which demonstrate the dual role of temperature in influencing polymer gear wear.

Figure 8 illustrates the highest surface temperature of the gear teeth for MC PA gears running at 1000 rpm with an 8.5 N m load. The figure shows that the maximum surface temperature increased significantly during the wear running-in stage (the first 70,000 cycles), with both driving and driven gears reaching nearly identical temperatures, peaking at around 80 °C. Following this, there was a general rise in maximum surface temperature from 70,000 cycles to 200,000 cycles, reaching up to 120 °C, with notable temperature fluctuations during this period. Afterward, the maximum surface temperatures for both gears stabilized near 120 °C with a fluctuation of about 6 °C. Additionally, at certain points during testing, the driving gear's surface temperature was roughly 5 °C elevated relative to the driven gear. Once the test was stopped, the peak surface temperature of the gears decreased quickly within 30 min, returning to the initial running temperature of approximately 40 °C.

PA gears subjected to a 9 N m load exhibited distinct thermal behaviors. Figure 9 depicts the peak surface temperature of a MC PA gear pair operating at 1000 rpm with a 9 N m load. During the initial 20,000 cycles, the maximum surface temperature rose rapidly, reaching around 75 °C, which was more pronounced compared to the 8.5 N m test. Following this, the temperature stabilized between 70 °C and 75 °C from 50,000 cycles to 220,000 cycles. Between 220,000 cycles and 350,000 cycles, there was a gradual increase in maximum surface temperature with more significant fluctuations, eventually reaching 110 °C. After halting the test at 350,000 cycles, the gears were set aside to cool for a period of 30 min, resulting in a quick drop in maximum surface temperature to 40 °C. Throughout the test, the driven gear generally exhibited a maximum surface temperature approximately 5 °C exceeding that of the driving gear.

In the 8.5 N m test (Fig. 8), the surface temperature of the material neared its glass transition point of 98 °C (as shown in Fig. 10) after around 115,000 cycles. This temperature increase is believed to have contributed to the observed rise in wear shown in Fig. 7 during and following this period, preventing the wear from stabilizing into a nearly linear phase. Conversely, in the 9 N m test (Fig. 9), the temperature remained stable during the nearly linear wear stage up to around 220,000 cycles. It was only after this point that the surface temperature began to increase slightly, not reaching the glass transition temperature until approximately 350,000 cycles. This delay may account for the minor increase in wear readings observed in Fig. 7 after this period. However, since the test was stopped thereafter, we could not investigate the continued wear behavior.

Figure 11 displays the peak surface temperature of an MC PA gear pair running at 1000 rpm under a 6.5 N m load. Throughout the running-in phase (0–150,000 cycles), the maximum surface temperature rose from 26 °C to an average range of 60–70 °C, with the driven gear's peak temperature being about 3 °C higher than that of the driving gear. During the nearly linear wear-rate stage depicted in Fig. 7 (from 150,000 to 400,000 cycles), the maximum surface temperature gradually increased to between 68 °C and 78 °C and then remained stable. It is clear from this test that the gear surface temperature did not reach the glass transition point for the PA material (as shown in Fig. 10), which explains why the wear-rate stayed low and did not escalate to a level that would cause failure.

To better understand the relationship between wear-rate, temperature changes, and varying loads, additional surface analyses were carried out using SEM to examine the wear characteristics on the teeth of the gear pairs subjected to testing. Figure 12 displays SEM images showing the tooth surface of the machine cut PA gear pair subjected to testing at 1000 rpm with an 8.5 N m load. The images reveal various tribological features and wear patterns in different areas of the tooth surface.

At the driving gear, tooth surface debris were forming around pitch line area, then moved toward tooth root and tip sides due to the sliding direction. Debris rubbing between the two surfaces forms surface scratches and fretting wear, whose severity increases toward tip area of the tooth. In addition, plastic deformation was observed on the addendum portion, which was thought to be due to reaching of glass transition temperature at the surface for the high-speed friction, leading to adhesive wear. Surface cracks were found around the pitch line area. On the other side of the tooth surface (dedendum side), more debris was found, but with less effect, as lower pressure occur. Adhesive wear and plastic flow were common here. With the adhesive wear, small chips were detached from the surface and fused to the tooth surface of the driven gear, leading to the increase of wear-rate and tooth thickness reduction.

The opposite sliding direction of the driven gear and different load amount and distribution lead to different surface tribology phenomena. Debris move from tooth root and tip toward pitch line, where they gathered in two different forms of long-rounded and chips. No microcracks were detected in this region. The addendum area of the tooth experiences high-pressure contact during the access stage of the engagement, which forms high surface damage in the form of adhesive and plastic flow, as can be seen in Fig. 12(b). Tooth dedendum side is subject to adhesive wear and form more long-rounded debris from the small chips separated from the surface.

Increasing the torque by 0.5 N m from 8.5 N m to 9 N m leads to some wear-rate decrease. Therefore, SEM investigation may show some useful data for this test. Figure 13 presents SEM images of the tooth surfaces from the MC PA gear set, assessed at 1000 rpm with a 9 N m load. The images illustrate diverse tribological features and wear patterns observed in different regions of the tooth surface.

On the surface of the driving gear tooth (Fig. 13(a)), adhesive wear is most prominent as surface temperature approaches the glass transition point for the machined PA material. The area around the pitch line showed two different surface characteristics. The top part is more of an adhesive wear, where the lower part forms debris with the function of adhesive wear. No microcracks were observed in this area. At the dedendum part of the tooth, more softened layer of material can be seen on the surface. This observation was supported by SEM imaging, which revealed areas with smoother, less-defined surface features compared to other regions, indicative of material softening under repeated high Hertzian contact pressures. Furthermore, the thermal camera data demonstrated elevated surface temperatures near the glass transition temperature of PA, correlating with the presence of the softened material. This soft layer is common at most places, which thought to be functioning as an internal lubricant. This internal lubricant could be a contributing factor to the reduction in surface wear and the rate of temperature under this particular loading value as a reason of some specific conditions, such as, the increase of surface contact pressure leading to the spread of the separated debris on the surface. Debris was minimal in these SEM observations, further supporting the hypothesis of debris integration into the softened layer.

Driven gear tooth (Fig. 13(b)) shows nearly similar surface tribology phenomena, taking into account the sliding opposite direction. Again no surface microcracks were found here. Comparing to the previous test, no debris were found around pitch line area, as another one proves that most debris were pressed between the two rubbing surfaces and formed soft layer that acts as an internal lubricant. Adhesive wear is most common at the driven gear tooth surface. In general, the SEM investigation identified the cause of the abrupt reduction in wear-rate and surface temperature with respect to the increase of loading at a certain value by finding the changes in surface tribological behavior between the two test and the formation of the surface layer as an internal lubricant.

Applying SEM investigation on the long run test revealed some extra findings at extra running time condition. Figure 14 displays SEM images of the tooth surfaces from the MC PA gear set, evaluated at 1000 rpm with a 6.5 N m load over a relatively extended duration (115 × 10⁵ cycles). Gear teeth at this test did not experience failure or fracture, even though long continuous running with relatively high surface temperature (Fig. 11). At the driving gear, surface scratches can be seen, especially at the addendum side. Fretting wear and surface pitting can also be seen. Good amount of microcracks can be observed around pitch line area, which is believed to be the cause of the contentious Hertzian pressure with the long run time.

At the driven gear, the continuous high surface contact pressure at the beginning of the gear engagement mesh leads to highly damaged surface around the addendum side of the tooth. Abrasive wear is more common at this side. On the dedendum area of the tooth, fretting wear can be observed in addition to the abrasive wear. Microcracks are also formed around the pitch line area.

This study investigated the tribological behavior and thermal performance of polymer gears under dry-running conditions using a custom-designed test rig. The following key conclusions were derived from the experimental findings:

• Abrasive wear was predominant in MC POM gears, particularly on the driving gear tooth surfaces. Adhesive wear was more pronounced in MC PA gears, with the additional presence of chip debris, highlighting material differences in wear mechanisms.

• The correlation between applied load and the rate of wear was scrutinized, demonstrating a consistent escalation in wear-rate concomitant with elevated load conditions. However, at specific critical loads, thermal softening resulted in the formation of a lubricating surface layer, reducing wear-rates despite the higher load. For instance, at 9 N m, the wear-rate decreased due to the lubricating effect of the softened material.

• Continuous monitoring revealed that surface temperature significantly influenced wear behavior. At lower loads (6.5 N m), surface temperatures remained below the glass transition temperature, resulting in minimal material softening and wear. Conversely, at higher loads (8.5 N m and 9 N m), surface temperatures approached or exceeded the glass transition temperature of PA, leading to thermal stabilization and reduced wear-rates at 9 N m.

• Extended testing of MC PA gears demonstrated increased wear-rates after prolonged cycles due to persistent Hertzian pressures, elevated surface temperatures, and the formation of microcracks. These findings emphasize the importance of temperature management to enhance gear durability.

• High-resolution SEM imaging revealed critical wear mechanisms, including adhesive wear, pitting, and the generation of micro-fiber debris, particularly near the pitch line and addendum regions. The presence of softened material layers and micro-fiber debris was more pronounced at higher loads, further confirming the impact of thermal effects on wear.

These findings contribute to a deeper understanding of the relationship between thermal effects and wear mechanisms in polymer gears. The observed phenomenon of thermal stabilization challenges traditional assumptions about wear progression and highlights the potential for optimizing polymer gear designs for high-load, thermally demanding applications.

Practical applications for the proposed gears include their use in electric vehicle transmissions, where weight reduction and dry-running capability are critical, and in medical equipment or laboratory devices, where lubrication-free operation is required. These findings can inform future gear design standards, ensuring greater reliability and efficiency in high-performance environments.

There are no conflicts of interest.

The authors attest that all data for this study are included in the paper.

ρ =

material density

DCF =

dynamic coefficient of friction

FM =

flexural modulus

MT =

melting temperature

SCF =

static coefficient of friction

TS =

tensile strength