Science and Research 64 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 Introduction In innovative industrial applications, especially in the automotive and drive technology sectors, the demand for sealing systems is continuously increasing. High circumferential speeds are necessary to meet new performance requirements, which in turn leads to significant thermal stress within the tribological system - the radial shaft sealing system (Figure 1). During dynamic operation, friction between the system components - the radial shaft seal (RSS), the shaft surface, and the lubricant - converts the mechanical energy generated by the contact partners into heat, with the highest temperature occurring directly within the contact zone (sealing gap). Increased contact temperatures can pose a significant thermal challenge for both the lubricant and the sealing material. In sealing systems that are not optimally designed, high contact temperatures can lead to lubricant degradation - such as the formation of deposits at the sealing edge - and damage of the sealing material, including hardening of the elastomer (Figure 4). The presence of deposits, along with a hardened sealing edge, can significantly influence the mechanical properties of the elastomer. This, in turn, can disturb the reverse pumping effect of the RSS (Figure 2), which is essential for a well performing sealing system. An impaired reverse pumping Optimization of Radial-Shaft-Sealing systems for the use at high circumferential speeds Adrian Heinl, Erich Prem, Christian Wilbs, Fabian Kaiser, Daniel Frölich* Presented at GfT Conference 2025 Radial Shaft Seals (RSS) are critical components to seal technical applications and are increasingly challenged with high circumferential speeds. In many innovative applications, e.g. within the automotive industry and drive technology, high speeds are necessary to meet performance requirements. High circumferential speeds increase tribological and thermal challenges, potentially leading to reduced sealing performance and shortened system lifetime. Therefore, this paper evaluates possible modifications to optimize the RSS-system, such as adjusting the radial load and selecting suitable elastomer materials. Using model-based approaches, contact temperatures are estimated and experimentally validated to prove the effectiveness of the modifications Keywords Simmerring, radial shaft seal, lubricant, elastomer, testing, contact temperature, temperature estimation, high-speed, optimization, ExACT, Theta, Gümbel curve Abstract * M.Sc. Adrian Heinl, Erich Prem, M.Sc. Christian Wilbs, Dr.-Ing. Fabian Kaiser, Dr.-Ing. Daniel Frölich Freudenberg FST GmbH Höhnerweg 2-4, 69469 Weinheim Figure 1: Radial-Shaft-Sealing System [1] Figure 2: Reverse pumping effect [1] effect can ultimately result in premature failure of the sealing system and leakage. To effectively reduce contact temperature during operation, minimizing friction within the sealing system is essential. This can be achieved through targeted optimization strategies, such as adjusting the shaft surface topography and selecting high-performance lubricants. In the context of radial shaft seals, further potential for improvement lies in modifying the radial load, selecting suitable elastomer materials and optimizing the seal design. Modeling Approaches for Contact Temperature While dynamic testing is the most direct method for evaluating the effectiveness of optimization measures, it is often resource-intensive and time-consuming. The contact temperature itself is not easily measurable. It requires, e.g. a thermocouple to be installed in the shaft, some way of transferring the data to the stationary measurement system and careful adjustment as the sealing contact is typically less than 0.2 mm wide. Furthermore, the shaft surface needs to be re-ground after every test due to the wear on the shaft, which can only be done a limited number of times per shaft. All of this makes it impractical to measure the contact temperature for larger studies. To enable a more efficient preliminary assessment of potential optimization solutions, model-based approaches for estimating contact and oil sump temperatures can be used. These methods allow for an early-stage evaluation of the fundamental suitability of various seal modifications without the need for dynamic testing. To estimate the contact temperature in sealing systems, two methods are used: ExACT Equation The ExACT equation (Extended Approximation of the Contact Temperature) was developed at the Institute for Machine Elements (IMA), as a generalized method for estimating the contact temperature of the RSS systems [2]. This equation describes a linear relationship between the temperature increase in the sealing gap ΔT - defined as the difference between the contact temperature T contact and the oil sump temperature T oil sump - and the frictional power P R referenced to the shaft diameter d. The proportionality constant R 0 represents the thermal resistance of the system and characterizes its ability to dissipate heat to surrounding components such as the shaft and oil sump. Since the oil sump heats up during dynamic testing due to the frictional heat generated - and this affects the contact temperature - the oil sump temperature is also estimated. The estimation is performed iteratively to account for the thermal influence over time, continuing until a stable thermal condition is reached. To improve the accuracy of this estimation, Freudenberg developed a simplified heat transfer model specifically tailored to the oil chamber of the RSS test rig. Based on natural convection, the model describes heat transfer from the oil sump through the chamber housing to the surrounding environment. A more precise determination of the oil sump temperature enables a more reliable estimation of the contact temperature. Thermal Network The thermal network is based on the simulation tool “Theta”, developed as part of FVA project 574 IV [3] at the Institute for Machine Elements and Gear Technology (MEGT). This tool enables the digital replication of the RSS test rig oil chambers. By defining thermal nodes and specifying the thermal conductivities of individual components, it is possible - given the input of ∆ T = T contact − T oil sump = R 0 ∙ P R π d Science and Research 65 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 Figure 3: FKM seal after dynamic testing, in very good condition Figure 4: FKM seal after dynamic testing, with deposits, hardening and discoloration Method 1: Gümbel Curve Based Estimation The first method calculates frictional power using the coefficient of friction, radial load of the RSS, shaft speed, and shaft diameter. The coefficient of friction is derived from the Gümbel curve, which describes the relationship between the friction coefficient and the Gümbel number G. The latter characterizes the lubrication condition of an RSS system and is determined by the dynamic viscosity η of the lubricant, the angular velocity of the shaft ω and the contact pressure pm by the seal on the shaft [2]. To establish this relationship, friction torque measurements were conducted on RSS test rigs at Freudenberg, as shown in Figure 5. Science and Research 66 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 the heat source - to estimate the contact temperature as well as the oil sump temperature through heat conduction. Experimental Determination of Friction Power and the Gümbel Curve The estimation of the contact temperature is based on the modeling approaches described in the previous section. Both approaches require the input in the form of a heat source generated by the frictional loss of the sealing system. Two established methods were used to determine this frictional power. Figure 5: Cross section of the RSS friction torque test rig [1] Figure 6: Test procedure for the friction torque measurement Figure 7: Gümbel curve - Relationship between the Gümbel number (G) and the friction coefficient (μ) These measurements served as the basis for subsequent contact temperature calculations. The test series included two FKM materials - 75 FKM 585 and 75 FKM 170055 - and two different RSS geometries: the standard BAUM seal design and the Premium Sine Seal (PSS). All tests were performed using plunge-ground shafts and mineral oil (VG 220). The rotational speed range covered both low speed (0.35 m/ s) and high-speed conditions (6.3 to 18.8 m/ s), as shown in Figure 6. Figure 7 shows the relationship between the Gümbel number G and the friction coefficient μ, based on dynamic friction torque measurements under varying operating conditions. A quadratic approximation function, as proposed by Feldmeth [2], was used to formulate a regression equation that serves as the basis for determining friction torque. Method 2: Friction Torque Based Estimation The second method bypasses the use of the Gümbel curve entirely and calculates frictional power directly from the measured friction torque. This approach provides a more direct and potentially more accurate estimation of the heat source, but it is only applicable to the tested sealing system. In contrast, the Gümbel curve can be used to estimate frictional power of similar sealing systems. Estimation und validation of the contact temperature for RSS optimizations As previously discussed, several optimization strategies have been implemented to enhance the performance of the radial shaft sealing system, each targeting specific components of the system. The modifications summarized in Table 1 are primarily aimed at reducing the contact temperature through systematic adjustments to the RSS. The RSS modification number reflects the sequential application of these measures, beginning with a baseline configuration consisting of a standard seal design (BAUM), a standard garter spring (STD), and the elastomer 75 FKM 585. Subsequent modifications include: • Modified spring to reduce radial load (MOD) • Selection of a high-performance elastomer (75 FKM 170055) • Friction-optimized sinusoidal sealing profile (PSS) These design changes are intended to mitigate thermomechanical stress at the sealing gap and to significantly improve system reliability and efficiency, particularly under high circumferential speed conditions. The estimation of the expected contact temperature, as well as the subsequent validation experiments, was carried out using similar test parameters as those used for measuring the friction torque and determining the Gümbel curve. The applied test parameters are listed in Table 2. Various circumferential speeds were investigated. Prior to each test, the oil sump was preheated to 70 °C and subsequently not actively cooled, to allow self-heating due to the friction. Since direct temperature measurement at the sealing contact is not possible, the measured oil sump temperature can be compared with the theoretically estimated oil sump temperature derived from the approaches of the ExACT Equation and Thermal Net- Science and Research 67 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 Condition Value Unit Speed 1000, 1500, 2000, 2500, 3000 rpm Circumferential Speed 6.3, 9.4, 12.6, 15.7, 18.8 m/ s Medium Mineral gear oil ISO VG 220 Temperature 70 - self heating °C Cycle 20/ 4 - 20h speed / 4h pause Duration 504 h n 2 RSS Table 2: Overview of the test conditions for temperature estimation and validation RSS-Modification-No. RSS Elastomer Spring Dimension 1 BAUM 1 75 FKM 585 STD 3 120-150-12 2 BAUM 1 75 FKM 585 MOD 4 120-150-12 3 BAUM 1 75 FKM 170055 MOD 4 120-150-12 4 PSS 2 75 FKM 585 - 120-150-12 1 BAUM = Standard Seal Design 2 PSS = Premium Sine Seal 3 STD = Standard garter spring 4 MOD = Modified garter spring for reduced radial load Table 1: Summary of Radial Shaft Seal (RSS) optimization strategies Complementary to the oil sump temperature analysis, Figure 9 shows the estimated contact temperature for the same range of RSS modifications and speeds. The results indicate a consistent increase in contact temperature with rising speed as well. The estimations indicate that the baseline configuration at a speed of 18.8 m/ s leads to a contact temperature of approximately 200 °C. This value significantly exceeds the permissible temperature limits for both the lubricant and the elastomer, representing a critical thermal load for both components. Furthermore, it can be observed that the applied optimizations measures lead to a significant reduction in contact temperature up to 20 °C consistently across all estimation methods. Science and Research 68 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 work. This enables improved validation of the contact temperature estimation. Figure 8 shows the estimated oil sump temperature as a function of the RSS modification across the investigated speeds. The figure includes three estimation methods, complemented by measured temperatures obtained from the friction torque tests. The results indicate that the experimentally determined temperatures lie between the calculated values, suggesting a realistic approximation by the models. Furthermore, a clear increase in temperature with higher speed is observed, along with a temperature reduction of up to 20 °C due to the applied modification measures. Figure 8: Estimated oil sump temperature results for various modifications under different circumferential speeds - friction torque test Figure 9: Estimated contact temperature results for various modifications under different circumferential speeds - friction torque test This correlation is also reflected in the temperature estimations derived from the long-term validation tests, see Figure 10. In this case, the estimation was only based on the Gümbel curve. For the baseline configuration, a contact temperature of approximately 200 °C is estimated, confirming the previously identified critical thermal load. However, through the implemented RSS modification measures, the temperature can be significantly reduced - depending on the method, by approximately 30 to 70 °C. A similar trend is observed in the estimation of the oil sump temperature. Notable, the predicted values align well with the measurements, supporting the validity of the contact temperature estimations. The contact temperatures estimated during the test series are clearly confirmed by the analysis of the sealing edges. Figure 11 shows the sealing edges of the different optimized RSS configurations after testing at 18.8 m/ s. In the baseline configuration, significant hard deposits are visible at the sealing edge, indicating thermal degradation of the lubricant. Additionally, the elastomer shows discoloration toward both the air-side and oilside contact area, further suggesting thermal overload. Leakage was observed as well. The radial load optimized RSS variant, with a modified spring, shows significant reduction in deposits and a narrower seal wear band width. No leakage occurred in this test. This optimization supports the previously estimated contact temperature, as the reduced spring load results in a temperature decrease of approximately 30 °C to 125 °C. Another measure, the use of the high-performance elastomer 75 FKM 170055, lead to another substantial reduction in deposits and seal wear. The switch to the PSS profile demonstrated a significant improvement in terms Science and Research 69 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 Figure 10: Comparison between estimated contact and oil sump temperatures and the measured oil sump temperature over the 504-hour long-term validation test Figure 11: Seal wear band resulting from the applied modifications after 504 hours validation tests at 18.8 m/ s Post-test analysis of the sealing edges supports these findings: while the baseline configuration shows significant deposits, discoloration and leakage, the optimized variants clearly show reduced deposit formation, narrow wear band width, and no leakage. In particular, the highperformance elastomer 75 FKM 170055 and the PSS profile demonstrated the lowest estimated contact temperatures for the long-term validation test and the best overall condition after test. A promising outlook arises from the potential to estimate the model-based contact temperature estimation based on some basic measurements. It enables early assessment of sealing system functionality and predictive analysis of thermal and tribological stresses - before critical conditions such as leakage or material failure occur. As a result, the diagnostic value of test data is significantly enhanced, supporting the development of robust and high-performance sealing systems for high-speed applications. Reference [1] Freudenberg FST GmbH [2] Feldmeth, S.: Simulative Bestimmung der Temperatur im Dichtkontakt von Radial-Wellendichtungen. Dissertation, Universität Stuttgart, 2025. [3] Bohnert, C.; Heilemann, J.; Thielen, S.: Th ermische Simulation Dichtkontakt. FVA-Projekt 574 IV, Satus: Vorläufiger Abschlussbericht, Forschungsvereinigung Antriebstechnik e.V. (FVA), 2025. Science and Research 70 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0022 of deposits formation and seal wear. This profile also exhibited the lowest estimated contact temperature for the validation test. Summary and Outlook Radial shaft sealing systems used to seal rotating components face significant thermal and tribological challenges at high circumferential speeds. These stresses can lead to degradation of the lubricant and damage of the elastomer, resulting in premature failure of the sealing system. The aim of this paper was to reduce the thermomechanical load and improve the performance of the system through targeted modifications of the RSS, including spring modification, selection of high-performance elastomer, and optimization of the sealing-lip profile. To evaluate the effectiveness of these optimization measures, model-based approaches were used to estimate contact and oil sump temperatures - specifically the ExACT [2] equation and a Thermal Network [3]. Both methods enable reliable predictions of temperature development in the sealing contact and were validated using experimental data. The results show that the baseline configuration can reach a critical temperature of up to 200 °C at high speed (18.8 m/ s), exceeding the thermal limits of both lubricant and elastomer. Through the applied optimization measures, the contact temperature was reduced by up to 70 °C depending on the modification - an effect confirmed both by contact temperature model and experimental validation.