Surface optimization of trapezoidal thread spindles—tribologicalanalysis and application

Trapezoidal threaded spindles are ideal for heavy loads and are often used in lifting systems. As part of a cooperation project between the Institute of Production Engineering and Machine Tools (IFW) and the company Bornemann Gewindetechnik, a modification of thread whirling was researched to optimize the tribological properties of threaded spindles. The use of whirled microstructures can reduce friction loss by 25.5% and significantly reduce adhesive wear.

1 Introduction

For the production of long threaded spindles, which are commonly used in lifting systems, various manufacturing processes are in competition, including forming and machining [1]. When designing the manufacturing process for standardized machine elements such as trapezoidal thread spindles, the application behavior and service life are not considered [2].

When designing surfaces that are loaded by sliding friction, it is relevant to consider the friction regimes – boundary friction, mixed friction and fluid friction – which are classified using the Stribeck curve [3]. In the hydrodynamic state, the force is generated by the internal friction of the lubricant [3-5]. This is proportional to the contact area, the viscosity of the lubricant and the shear rate of the lubricant film thickness. In the case of mixed friction, the lubricant film is interrupted at certain points by individual roughness peaks, which then contribute to load transmission [4]. Particularly at low sliding speeds, this is a disadvantage in terms of establishing and maintaining the hydrodynamic lubrication state [6].

The direct contact between the two triboelements can lead to increased energy consumption [7]. Microtextures can have an advantageous effect through various mechanisms. For example, microtextures can serve as lubricant reservoirs [8], and localized elevations in the lubricant film can lead to an increase in bearing pressure when lubricant flows over them [9]. A hydrodynamic pressure can build up in the defined microtextures. Both cavitation effects [4] and the build-up of bearing pressure in successive microtextures play a role here [10,11]. These investigations indicate that the surface topography of trapezoidal thread spindles offers great potential for reducing friction through microtextures due to highly loaded sliding contact. Microtextures to reduce friction losses in sliding contacts can be induced by manufacturing processes such as flycutting or laser texturing [10,12].

Thread whirling as a machining process offers great potential for process-integrated microtexturing, as it achieves significantly higher productivity than thread milling, for example [1,13]. Studies on surface texturing by whirling have, so far, only been conducted as an additional texturing process and not as a process-integrated functionalization of the surface texture [14]. Various approaches exist for microtexturing as a separate manufacturing process [12, 15], but these require high integration efforts due to an additional process step. In a study, Denkena et al. demonstrated an improvement in the tribological properties of surfaces subjected to high thermomechanical loads, such as in cylinder liners, through microtexturing [12]. The whirling process, as a machining production method, presents specific potential for functionalization due to the surface textures created on the thread flank (Fig. 1).

A specific challenge when using metallic triboelements in sliding friction is the occurrence of adhesion in cases of insufficient lubrication or overloading. This ca lead to adhesive wear of friction partner with lower hardness and a material transfer to the friction partner with the higher hardness in the sliding contact [7,17]. The surface topography and the real contact area in sliding contact play crucial roles in the previously described risk of adhesion with insufficient lubrication [7,16,17]. The direct contact of the two friction elements is a fundamental prepequisite for adhesion and is defined by the total number of micro contacts. In sliding contacts that are subjected to a high surface pressure, the lubricant film thickness is reduced, which leads to a shift from mixed friction to boundary friction and is accompanied by an increase in the coefficient of friction [3].

Denkena et al. also showed that surface textures with low depth can lead to a reduction in the coefficient of friction and a shift in the mixed friction regime, where both solid and fluid friction coexist [12]. For steelbronze triboelements in sliding frictions, a surface texture depth of 2–5μm has proven particularly favorable for reducing friction [16]. Another phenomenon that may occur with insufficient lubrication or low relative speeds is the stick-slip effect. This results in the triboelements sticking together for a short time before a sliding movement occurs again [18]. Frequent occurrence of the stick-slip effect due to a lack of lubrication leads to increased adhesive wear and, consequently, early component failure [19].

The influence of microtextures on the loaded thread flank of trapezoidal threaded spindles with high loads has not yet been researched, but offers great potential for increasing efficiency by reducing friction losses. There is also as yet no knowledge of the influence of microtextures on wear in the spindle-nut system. This paper therefore aims to address this limitation and investigate the influence of surface textures generated by thread whirling on the tribological behavior of highly loaded trapezoidal threaded spindles in heavy-duty lifting systems under real conditions. The study will examine a specifically developed whirling process for process-integrated microtexturing.

2 Experimental setup

2.1 Tribological testing

The test rig shown in Fig. 2 is used to investigate the influence of the surface topography on the thread flanks concerning the tribological behavior. Test spindles with a Tr80× 10mm thread were oscillated at a frequency of f= 0.81Hz through an angle of ν= 15°. The rotation of one load change corresponds to a distance of 0.42mm translational stroke of the thread. With this setup, the threaded spindle was loaded with a weight force of FG= 91.3kN during lifting and lowering. This corresponds to a surface pressure of p= 5.0N/mm2, which is within the maximum load range for trapezoidal screw drives in heavy-duty lifting systems. For this application, the threaded nut material of G-CuSn 7 ZnPb was selected. The grease DGM HTF 940, which is particularly suitable for sliding bearing applications, was used as lubricant. The lubrication in the test sequence was based on a maintenance interval of one month. This interval corresponds to the lubrication of the spindlenut assembly every 167 cycles. The test rig was set up by Sincotec, which is also the manufacturer of the force sensor Interface 125kN and the torque sensor SincoTec 1200Nm.

The wear condition of the flank surfaces was analyzed to characterize the friction properties of different surface topographies and their corresponding surface textures. For this purpose, the spindles were subsequently cut open using a cut-off grinder. With 20,000 load cycles, a service life of 10 years was experimentally mapped for the examined threaded spindles.

To investigate the tribological behavior, four threaded spindles were prepared, whereby different process parameters in the whirling process were compared with a rolled spindle (Table 1). For the whirled threaded spindles, the process parameters were selected to achieve defined gradations between the heights of the surface texture on the loaded thread flanks of the threads. Test series 1 includes rolled threaded spindles produced through the continuous rolling process. The specific process parameters are based on Bornemann’s experience. However, test series 2 was whirled with process parameters corresponding to the state of the art and is considered as an additional reference for the newly developed surface texture.

In test series 3 and 4, the surface textures developed in this study were produced in two stages, to increase both the surface texture height yf and the surface texture length sf. An increase in these two texture parameters leads to an increase in the lubricant retention volume in the surface. Test series 3 shows a theoretical texture height of yf= 2.31μm with a texture distance of sf= 3.67μm. In test series
4, the texture parameters are even more pronounced with yf=3.61μm and sf= 4.59μm. This clearly shows the limited adjustability of the texture’s due to the whirling process: an increase in the texture height yf inevitably leads to an increase in the texture’s length sf.

2.2 Surface topography measurement

The Duo Vario optical measurement system from Confovis GmbH is used to analyze experimentally-generated 3D surface topographies. The surface topographies are detected with confocal white light microscopy. The confocal measurement was carried out with a Nikon 20x/0.45 NA lens. To analyze the flank surfaces a lateral measurement resolution of 0.20μm was used. The measured area with a width
of 2.26mm and length of 8.71mm was detected with a resolution of 0.28μm. The experimentally produced thread spindles were orientated in an orthogonal measuring position to the thread flank.

3 Application and characterization of surface topography

The following results show how the surface topography on the thread flank can be specifically adjusted by the whirling process and how it differs from the limited topography of the thread rolling process. Figure 3 shows the surface topographies of the thread flank of a rolled trapezoidal threaded spindle in comparison to a whirled thread flank.

The surface topography of test series 1 shows that no surface texture is produced during thread rolling. The whirled thread spindle from test series 2 show no significant differences to the surface topography in comparison to test series 1 and no surface texture is recognizable either. By specifically adapting the whirling process, it was possible to produce significantly more pronounced surface textures in test series 3 and 4. In these test series, the texture height yf and the texture length sf were successively increased to produce surface textures with a low proportion of peaks, minimizing the direct solid contact between the surfaces of the triboelements. The greater distance between the surface texture peaks makes it possible to retain lubricant in the valleys of the roughness profile.

Due to the occurrence of stochastic roughness effects during machining, which affects the technical surface as fourth level shape deviations, the less pronounced surface texture of test series 2 is overlaid. Because of this superimposition, the texture parameters are difficult to determine and have a low texture hight yf= 0.82μm and texture length sf= 1.63mm. Effectively, with this low surface texture, test series 2 does not differ from the rolled sample. The stochastic roughness effects during whirling are mainly caused by the roughness on the cutting edge of the whirling tool, also known as chipping. This is formed as a negative on the newly created surface during machining [20]. The topography of test series 3 and 4 shows characteristic, recurring grooves generated by chipping. If the texture height yf is lower than the roughness Rz on the cutting edge of the whirling tool, the texture is superimposed and the characteristic texture of the whirling process does not occur. This is evident in the roughness profiles on the thread flanks (Fig. 4).

The measurement is additionally distorted by subtracting the mathematically approximated thread form from the surface topography, resulting in the textures at the edge of the measuring range having a slightly lower texture height (Fig. 3 and 4).

4 Tribological assessment

The effects of the whirled surface textures on the tribological system of screw and nut in trapezoidal screw drives for heavy-duty lifting systems were analyzed using the test setup described in Sect. 2.2. In the design of trapezoidal screws, the coefficient of friction is often assumed to be constant for simplicity. However, the coefficient of friction cannot be regarded as a material property, as it is influenced by all components of the tribological system. To be able to characterize the influence of the surface topography on this system, all influencing variables are kept constant. Only the surface texture induced by the manufacturing process is varied in the test series.

4.1 Investigation of the coefficient of friction

In heavy-duty lifting systems, the movement threads are usually arranged vertically, whereby a loaded thread flank is loaded during both the lifting and lowering movements. For the calculation of the coefficient of friction in a trapezoidal thread, the friction on the inclined plane can be used [21]. The normal force FN can be calculated from the weight force FG and the frictional force FR from the frictional torque required for the rotational movement via the geometric relationships. This leads to a significantly higher required frictional torque when lifting. As the frictional torque is measured in the test setup, and the weight force FG is controlled to a constant value, the coefficient of friction μ can be calculated from these variables. The separate determination of the coefficient of friction μ for lifting and lowering is made possible by considering the different orientations of the force components. The exemplary course of the coefficient of friction for these two forms of motion for the rolled spindle in test series 1 is shown in Fig. 5. Considering the different force orientations, both forms of motion show a similar course for the coefficient of friction. Despite a higher drive torque, which is required for lifting, a characteristic friction coefficient behavior can be seen, which occurs with a trapezoidal threaded spindle over 20,000 load cycles. The average coefficient of friction for both types of motion increases significantly at the beginning of the test, reaches a maximum at about 2000 cycles, and then drops to a constant level at about 8000 cycles. This behavior is referred to as the running-in phase and describes the general change in the coefficient of friction throughout the test, depending on the tribological system [22]. The oscillating movement over a small angular distance emulates a short distance of lifting and lowering in a load change. The resulting speed profile within a load change consists of an acceleration phase, a phase of constant speed and a deceleration phase, each for lifting and lowering. The stick-slip effect, which occurs at low speeds, is particularly favored at the point of direction reversal. This effect manifests itself in a brief sticking of the surfaces, followed by a sudden sliding of the triboelements. This movement can lead to vibrations, which are reflected in a greater deviation in the coefficient of friction [19]. In Fig. 5, this effect can be observed in the form of a greater deviation within the first 5000 load cycles.

A constant coefficient of friction for the investigated trapezoidal thread Tr80× 10mm is achieved after approximately 8000 load cycles. During the running-in phase, there is an initial increase in the coefficient of friction when testing the rolled trapezoidal threads, which is also accompanied by a high deviation of the coefficient of friction. The highest coefficient of friction is associated with the greatest deviation, t indicating an increased occurrence of the stickslip effect, which can lead to higher adhesive wear. Due to the similar friction coefficient curves of both forms of motion, only the more force-intensive lifting movement is considered in the following.

There is a clear difference in the friction coefficient curves for the surfaces shown in Fig. 3. While test series 1 and 2 differ in manufacturing processes – thread rolling and thread whirling, respectively – both samples exhibit a similar surface topography with almost no surface textures (Fig. 3). This similarity is also reflected in the friction coefficient curves (Fig. 6). While the coefficient of friction of test series 1 has a significant maximum, in test series 2 no maximum can be observed over the test duration. However, a constant level is reached in both tests after a similar number of load cycles (L8000).

The mean coefficient of friction μm over the entire test duration also assumes a similar value for both tests. A running-in phase occurs for both the whirled and rolled threaded spindle, which subsequently leads to a very constant coefficient of friction with low deviation at a level of μ= 0.1. In test series 3, with a texture height of yf= 2.31μm, a significantly shorter running-in phase is observed, which is completed after approx. 4500 load cycles. The average coefficient of friction μm can be reduced to μm= 0.098 in test series 3, compared to the rolled (μm= 0.115) and the untextured, whirled thread flank (μm= 0.112), which corresponds to a reduction in the coefficient of friction of 14.6%. Aneven more significant reduction is shown in test series 4 (yf= 3.61μm), in which the introduced surface texture both reduces the running-in behavior by approx. 44% and permanently lowers the coefficient of friction to a lower level of μm= 0.085, which corresponds to a reduction of 25.5% compared to the rolled threaded spindle. A complete reduction of the running-in phase was not possible with this thread in the described tribological system.

4.2 Analysis of the adhesive wear

The wear that occurs when spindle and nut come into contact is crucial to the life of the trapezoidal screw drive. To identify the wear mechanisms occurring on the lead screw, the surface of the loaded flank was examined after 20,000 load cycles, with a weigth force of FG= 91.3kN and a regular supply of lubricant (Fig. 7). After the test, the test series show a clearly different wear pattern depending on the surface texture of the loaded thread flank. Test series 1 and 2, which were tested without surface textures on the loaded thread flank, show pronounced adhesive deposits on the loaded thread flank of the threaded spindle. Thisis caused by significant removal or adhesive wear of the nut material. As the height of the texture increases from test series 3 to test series 4, the degree of adhesive wear noticeably decreases (Fig. 7).

To characterize the proportion of the loaded thread flank covered with adhesive deposits, a colour analysis of microscope images was carried out to quantify the predominantly red deposits caused by the copper content in the bearing material G-CuSn 7 ZnPb. The evaluation of the area-related adhesion percentage on the loaded thread flank shows, that the rolled trapezoidal screw has the highest adhesion percentage at 36.3% (Fig. 8). The test series 2, with the whirled thread flanks exhibits the largest deviation of the covered area, with 27.4% and a standard deviation of 13.1%. After the experimentally emulated service life, the surface of this test series is covered with up to 45.4% adhesion residue in some areas. The wear pattern of test series 4 illustrates that a texture height of yf= 3.61μm reduced the proportion of adhesive wear and limited it to the areas of the roughness peaks (Fig. 8a).

Due to the low sliding speed and the high surface pressure in tribological contact, the friction present in the tribological system is classified between solid-state friction and mixed friction.

Adhesive wear on the thread flank decreased from 36.3 to 13.6% in test series 3 and further to 10.1% in test series 4 with the whirled flank surfaces compared to the rolled threaded spindle. Adhesion does not occur between the pronounced roughness peaks with a clearly pronounced surface texture. Due to the characteristics of the adhesive coating, it can be assumed that there is a shift from the boundary friction regime to the mixed friction regime. According to Wang et al., the lower number of roughness peaks, or in this case only the peaks of the surface textures, can lead to a reduction in the coefficient of friction [4]. In these areas, the original surface created by the whirling process is retained.

The successive arrangement of the microtextures can also favor the formation of a lubricant film with low thickness, as described in [10]. Due to the reduced solid-state friction over the entire contact surface, a lower frictional torque is required for the movement of the threaded spindle. Lubricant can be retained in the areas of the pronounced texture, increasing the lubricant film thickness and causing fluid friction, thereby preventing direct contact between the surfaces. Direct contact between the roughness peaks of the triboelements is a necessary condition for adhesive wear [19].

As a result of the more pronounced surface textures (Fig. 9b), the proportion of solid-state friction decreases in contrast to the untextured surface topography (Fig. 9a). A more pronounced surface texture on the thread flank can absorb a larger quantity of lubricant.

When triboelements made of different metals come into contact, material is transferred from the cohesively weaker bonded friction body, in this case, the nut, to the cohesively stronger bonded base body, in this case the threaded spindle [3]. The additional lubricant in the friction contact results in a lower number of contacting roughness peaks. The formation of adhesive deposits on the loaded thread flank is shown schematically in Fig. 9c. The progressive loading of the contact surface of the threaded nut results in the detachment of particles that are deposited on the contacting roughness peaks of the loaded thread flank. These increasingly form a layer on the contacting areas of the loaded thread flank (Fig. 10) and prevent direct contact between the screw and nut material. This process continues until a steady state is reached, and no further material is transferred from the surface boundary layers of the nut onto the loaded thread flank. This process influences the running-in behavior, transitioning to a steady state once the contacting roughness peaks are covered with adhesive deposits. By which the coefficient of friction is influenced. Figure 10 shows that the deposits are approximately 2μm in height and that there is no abrasive wear of the initial surface of the loaded thread flank.

5 Conclusion and outlook

As part of this study, three different surface topographies were adjusted using the whirling process and examined for their tribological properties in trapezoidal threaded spindles. In addition to measuring the coefficient of friction, the wear pattern on the loaded thread flanks of the threaded spindle was examined, and the adhesive deposits were quantified.

In comparison with rolled threaded spindles, it has been demonstrated that surface textures on the loaded thread flank of a threaded spindle used in heavy-duty lifting systems provide significant added value. The whirling process enables the production of a greater variety of surface textures, resulting in a 25.5% reduction in the coefficient of friction. This reduction in the coefficient of friction leads to a proportional decrease in the overall system’s energy consumption, considering the specific bearing of the lifting system, as the efficiency is significantly influenced by the frictional contact between the spindle and nut. Furthermore, it has been shown that the original surface of a threaded spindle textured by the whirling process remains largely intact after an experimentally emulated service life of 10 years. This indicates that the proportion of solid-state friction, where the roughness peaks of the spindle and nut are in direct contact, could be reduced. With the surface textures presented, the adhesive wear on the thread flank in test series 4 was reduced to 10.1%, compared to 36.3% for a rolled threaded spindle.

Although the entire 10-year service life of a threaded spindle was emulated in the test, the service life of the threaded nut is not presented. While only one part of the spindle is subjected to tribological contact, the internal thread of the nut undergoes continuous loading, resulting in a significantly longer exposure to tribological interaction. Future investigations could focus more on this aspect of the friction system, particularly considering the challenge of measuring nut wear. Another positive outcome of a threaded spindle with a pronounced surface texture could be the reduction of wear on the counter-friction body, in this case, the nut.

References

1. Langsdorff W (1969) Gewindefertigung und Herstellung von Schnecken, 6th edn. Springer, Berlin, Heidelberg
2. Gereke-Bornemann H, von Soden M (2010) Wie man den Stickslip-Effekt bei Gewindespindeln vermeidet
3. Stribeck R (1902) Die wesentlichen Eigenschaften der Gleit- und Rollenlager
4. Wang X, Kato K, Adachi K et al (2003) Loads carrying capacity map for the surface texture design of SiC thrust bearing sliding in water. Tribol Int 36(3):189–197. https://doi.org/10.1016/S0301-679X(02)00145-7
5. Touche T, Cayer-Barrioz J, Mazuyer D (2016) Friction of textured surfaces in EHL and mixed lubrication: effect of the groove topography. Tribol Lett 63:25. https://doi.org/10.1007/s11249-016-0713-8
6. Flores G, Abeln T, Klink U (2007) Functionally adapted final machining for cylinder bores made of cast iron. MTZ Worldw 68:6–9. https://doi.org/10.1007/BF03226811
7. Buckley DH (1981) Surface effects in adhesion, friction, wear and lubrication. Elsevier, Amsterdam
8. Etsion I (2004) Improving tribological performance of mechanical components by laser surface texturing. Tribol Lett 17:733–737. https://doi.org/10.1007/s11249-004-8081-1
9. Hamilton DB, Walowit JA, Allen CM (1966) A theory of lubrication by microirregularities. ASME J Basic Eng 88(1):177–185. https://doi.org/10.1115/1.3645799
10. Brizmer V, Kligerman Y, Etsion I (2003) A laser surface textured parallel thrust bearing. Tribol Transact 46(3):397–403. https://doi.org/10.1080/10402000308982643
11. Kovalchenko A, Ajayi O, Erdemir A, Fenske G, Etsion I (2005) The effect of laser surface texturing on transitions in lubrication regimes during unidirectional sliding contact. Tribol Int 38(3):219–225. https://doi.org/10.1016/j.triboint.2004.08.004
12. Denkena B, Köhler J, Kästner J, Göttsching T, Dinkelacker F, Ulmer H (2013) Efficient machining of microdimples for friction reduction. ASME J Micro Nano Manuf 1(1):11003. https://doi.org/10.1115/1.4023757
13. Stender W (1954) Schälen von Gewindespindeln: Die zeitgemäße Lösung eines alten Problems. Werkstatttechnik Maschinenbau 44:531–538
14. Matsumura T, Serizawa M, Ogawa T, SasakiM(2015) Surface dimple machining in whirling. JManuf Syst 37(2):487–493. https://doi.org/10.1016/j.jmsy.2014.07.008
15. Denkena B, Bergmann B, Keitel M, Wege C, Poll G, Kelley J, Pape F (2023) Process strategies for milling of dimples on tapered roller bearings. Prod Eng Res Devel 17:893–905. https://doi.org/10.1007/s11740-023-01208-416.
16. Steinert P (2017) Fertigung und Bewertung deterministischer Oberflächenmikrostrukturen zur Beeinflussung des tribologischen Verhaltens von Stahl-Bronze-Gleitpaarungen. Technischen Universität Chemnitz
17. Greenwood JA, Tripp JH (1967) The elastic contact of rough spheres. ASME J Appl Mech 89:153
18. Haessig DA, Friedland B (1990) On the modeling and simulation of friction. American Control Conference, San Diego
19. Aurégan G, Fridrici V, Kapsa P, Rodrigues F (2015) Experimental simulation of rolling—sliding contact for application to planetary roller screw mechanism. Wear 332–333:1176–1184. https://doi.org/10.1016/j.wear.2015.01.047
20. Denkena B, Böß V, Nespor D, Gilge P, Hohenstein S, Seume J (2015) Prediction of the 3D surface topography after ball end milling and its influence on aerodynamics. Procedia CIRP 31:221–227. https://doi.org/10.1016/j.procir.2015.03.049
21. Böge G, Böge W (2021) Statik starrer Körper in der Ebene. In: Böge A, Böge W(eds) Handbuch Maschinenbau. Springer Vieweg, Wiesbaden
22. Blau PJ (2006) On the nature of running-in. Tribol Int 38(11):1007–1012. https://doi.org/10.1016/j.triboint.2005.07.020

Acknowledgements

The authors thank the Federal Ministry for Economic Affairs and Climate Action (BMWK) for the funding and the project partner Bornemann Gewindetechnik GmbH & Co. KG for the constructive and close cooperation.

Funding

The investigations were funded by the Federal Ministry for Economic Affairs and Climate Action (BMWK) as part of the Central Innovation.

Author Contribution

B. Denkena reviewed and edited the manuscript together with B. Bergmann. C. Wege developed the concept of this work, conducted the experiments, analyzed the data and wrote the manuscript. M. von Soden and H. Gereke-Bornemann manufactured the tools and provided the experimental setup.

Funding

Open Access funding enabled and organized by Projekt DEAL.

Conflict of Interest

B. Denkena, B. Bergmann, C. Wege, M. von Soden and H. Gereke-Bornemann declare that they have no competing interests.

Open Access

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Publishers Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Springer Nature, 24.05.2024