Evaluation of the Friction Coefficient for TRIP1000 Steel under Different Conditions of Lubrication, Contact Pressure, Sliding Speed and Working Temperature

Abstract

The use of ultra-high-strength steels (UHSS) has been growing in recent years, mainly in the automotive industry. Since these steels have high strength and hardness, more applied stresses are required to deform them, probably also impacting friction behaviour. In this article, a variation in the process parameters commonly observed in sheet-metal forming, such as contact pressure, sliding speed, lubrication and working temperature was performed. The material used was TRIP1000. These process parameters were varied, aiming to investigate the friction-coefficient behaviour; however, it was observed that there were no significant variations, indicating that the steel hardness may have contributed to this. Another finding is that, even if the lubricant did not change the average value of the friction coefficient, it contributed to a more stable process, favouring the absence of premature wear of the tools.

1. Introduction

Friction and wear in forming have a great importance for improving the quality of the products made by this process. Many works have been published on this subject in recent years and several advances have been observed, such as better lubricants and more suitable finishes [1]. However, when a new material appears, there is a need to carry out a survey of mechanical properties and performance under various aspects, such as the tribological behaviour at each stage of the manufacturing process. From there, the material becomes known, and its behaviour determined so that it can be used in simulation software with a minimum of errors, as shown in the work of [2].

It is known that the coefficient of friction in sheet-metal forming is influenced by several factors, including surface finishing, sliding speed, contact pressure between sheet and tools, working temperature, type of lubrication, the hardness of the material and tools, surface treatments and chemical composition of the sheet [3]. Trzepiecinski and Lemu [4] reported the main parameters that can influence the friction behaviour along with the types of tests that are used for sheet-metal forming. Other authors carried out specific studies, as in the work of Fratinei et al. [5], where several materials such as aluminium, stainless steel, copper and brass were tested under different conditions, for example with and without coating, using grease or Teflon as lubricant and even in a dry condition. The results reveal that friction varies substantially for each situation and better in-depth studies are necessary for each material and boundary conditions. Additionally, there are works are more focused on the analysis of the types of surface that generate less friction [6,7], where the best surface preparations for lubricated and dry-forming processes were found. There are also efforts to understand how the performance of lubricants will occur over the lifetime of the tools [1], where lubricants were tested for 1200 strokes of parts during punch cutting and those more suitable for mass production were selected.

The friction condition regarding position and process parameters is also important to improve manufacturing quality and several studies of it have been conducted [8]. In this work, friction was measured under several contact pressures and the results analysed in finite element modelling to evaluate its impact on spring-back behaviour. The analyses revealed that a variable friction improves the spring-back behaviour due to frictional shear stress close to the punch corner radius (responsible for the spring-back effect) for constant friction to be almost 3.3 times greater than its for variable friction model. Regarding the temperature, at first, it may seem that the temperature change should not be present in the cold sheet-forming processes, but in the work of Kim and Alta [9], a numerical simulation was performed, and they observed that the temperature can reach 85 °C in the sheet during forming. Thus, in the work of [10], the temperature and the sliding speed were measured through pin-on-disk test and implemented in a simulation software. A biaxial stretch forming test was performed and the simulated and tested results were found to be in good agreement. The evaluating and determination of ‘the friction coefficient in sheet-metal forming is very important to avoid manufacturing failures, as high values can damage the surface and finishing of the workpiece. In this regard, the work presented by Olguner and Bozdana [11] showed that, using numerical simulation, a friction coefficient of 0.2 would already be enough to cause sheet-forming failure during the manufacturing process, which could be improved by changing or adapting the lubricant material and conditions. There are also some works focused on new simulation models that can predict friction conditions using software like TriboForm^®, in which the main process variables (contact pressure, velocity and deformation) are accounted for with numerical calculations in the software. This was already reported in [2,12], and the results revealed that the addition of variable friction in the simulation can improves the predictability and reliability of the results. Other works [13,14,15,16,17] have made efforts to evaluate all possible friction variables such as the topological profile of the surfaces in contact, the lubricant viscosity, the variation with pressure, temperature, sliding speed, level of deformations and amount of lubricant applied in each region of the blank so that it can be implemented in finite element software in order to make the simulation more and more accurate.

Based on this, measuring friction under conditions of process variables is very important for the correct determination of input data for simulation software as well as for understanding the processing of the work material. Therefore, this article aimed to determine the friction coefficient with variations in contact pressures between sheet and tool, sliding speed, type of lubrication (with oil and grease) and system temperature. The goal is to understand better the friction behaviour for a possible candidate material to integrate the next generation of automotive platforms, TRIP1000 steel.

2. Materials and Methods

2.1. Sheet Material

The material investigated in this work is a TRIP1000 sheet metal with 1.5 mm thickness, supplied by the steel making company USIMINAS (Usinas Siderurgicas de Minas Gerais S.A, Belo Horizonte, MG, Brazil). The composition of the material is shown in

Table 1. Chemical Composition of the TRIP1000 steel specified.

Element	C	Mn	Si	Al	Ti + Nb	P	S
Content (wt%)	<0.30	<2.50	<2.00	<2.00	<0.10	<0.04	<0.01

and the mechanical properties obtained through the uniaxial tensile test is shown in

Table 2. Mechanical properties measured in the tensile test.

Rolling Direction	Yield Strength (MPa)	UTS (MPa)	Elunifor. (%)	Eltotal (%)	n6%-ue	r10%
90°	746	1064	14.2	18.3	0.167	1.047
45°	726	1054	15.2	19.3	0.176	0.840
0°	724	1056	16.1	20.6	0.180	0.802

. The sheet manufacturing process went through hot rolling, cold rolling and final annealing.

As a preliminary work, the microhardness of the sheet was also measured in a Shimadzu HMV-2T E microhardness tester (SHIMADZU DO BRASIL, Barueri, SP, Brazil) with a load of 1 kgf (9.8 N) in ten (10) randomly positioned indentations for each sample. The average hardness obtained from the sheet was 302.50 ± 10.17 HV (≈30 HRC).

2.2. Pin-on-Disk Test

The tests to determine the friction coefficient were performed in a Bruker® UNT-Tribolab tribometer (Bruker Corporation, Billerica, MA, USA). In the pin-on-disk tests, the disk was manufactured with the sheet material following two geometries, one for cold tests with a diameter of 69.85 mm and another for hot tests with a diameter of 50.80 mm, as shown in Figure 1a. A straight steel pin 5 mm in diameter (Figure 1b) with 47 HRC hardness was used in the tribological system. In order to ensure no contamination that could influence the results, before starting a new test, 1200-grit sandpaper was applied to the tip of the pin to remove any supposedly adhered material coming from preview tests. The tests were conducted with forces applied to the pin of 100, 300, 500, 700 and 900 N [8], which corresponds to a pressure of approximately 5, 15, 25, 35 and 45 MPa, respectively. The speeds applied to the pin (converted from disk rotations) were 0.5, 5 and 10 mm/s, based on the work of [10]. The tests were divided in three categories of lubrication conditions: without lubricant, using a machine oil (highly refined, additive, extreme pressure mineral oil) and using a grease (lithium grease, NLGI grade 2, containing oxidation and corrosion inhibitors). The test temperature was also set in different conditions: room temperature, 28 °C for lubricant applications, at 60 °C and at 100 °C without the addition of lubrication [10]. Three valid tests were performed for each condition and the lubricants were added so that they formed a layer about 0.5 mm thick. The temperature was applied in the tribological tests as a function of the mechanical work that generates the heating during the forming processes. All these values were based on articles previously mentioned.

The determination of the friction coefficient was made directly by the results provided by the tribometer after the test, as shown in Figure 2, for example. The measurements were taken of the vertical and tangential forces on the pin and thus, the coefficient of friction was calculated by dividing these forces. The test was run for the same amount of time (60 s) for all specimens and, as mentioned, repeated three times for each condition. Consequently, the mean value and the standard deviation of each curve was taken based in the average value of these three measurements.

3. Results and Discussions

The tribological tests showing the friction coefficient for all three different lubricant conditions (dry, oil and grease) are shown in Figure 3. The first analyses show that the friction coefficient values do not have a significant change from its mean value (of about 0.11), but some differences can be observed. Comparing the lubricant condition, speed and pressure for the dry condition, it seems that the friction does have a higher variation with the change in speed and pressure and its value fluctuates around the mean value 0.11. The same occurred for the application of oil. However, for the grease, there was an increase in the coefficient of friction, mainly for a low speed, which agrees with the work of [2,10,18]. There was also a tendency to increase friction at low pressures, which is in agreement with the literature [2,8,12]. Figure 4 presents the average friction coefficient with respect to speed and pressure and provides a better view of the reported behaviour. In the case of grease, it is easier to observe the influence of speed on the friction coefficient which shows that a low speed has a higher coefficient of friction. This probably occurs because, in a lubrication regime, a transition from a mixed friction system to a hydrodynamic one is expected, and thus, the metal-to-metal contact is reduced, stabilizing at lower friction for higher speeds.

Figure 5 show the results of friction coefficient analyses for the dry condition in different temperature conditions (room temperature, 60 °C and 100 °C). It can be noticed that as soon as the temperature increases, the friction behaviour becomes irregular, showing no specific trend. One could say that for the 100 °C condition, there is weak trend to decrease the friction with increasing the pressure and velocity (Figure 6). Here, this behaviour may be connected with a pronounced oscillation in the friction coefficient (stick–slip effect) showing that temperature acts as a mechanism to trigger material adhesion and promote the galling effect. This was originally associated with formation and destruction of interfacial junctions on a microscopic scale. The motion resulting from stick–slip is sometimes referred to as jerky motion [19]. Galling is a form of damage and can be considered a type of adhesive wear which is caused by sliding of two solids and often includes plastic flow, material transfer or both [20]. The adhesive wear occurs when the tools slide against the workpiece under high pressure after the breakdown of the lubricant. Here, temperature can cause microwelds (adhesion) between the sliding surfaces and can damage both the tools and the workpiece surfaces [21].

In general, the results of the friction coefficient analyses did not show much variation in all conditions and its mean value ranged from 0.1 to 0.13, which is considered a low variation for cold-stamping applications and is in accordance with the results of other authors [5,6,8,10,11,22]. However, analysing Figure 3, it is noticed that the lubricant plays a fundamental role in the stabilization of friction, especially in high pressure and high speed conditions. Additionally, as the viscosity of the lubricant increases, the pressure and speed begin to present a behaviour more consistent with the literature, despite the values being quite low. Another important observation is that the temperature of 100 °C promoted an enormous dispersion of the results, causing quite significant variations regarding the main value. This occurs mainly because at that moment, friction mechanisms are activated due to material adhesion [21], which generates the oscillation in the coefficient values when particles are detached from the pin or sheet. Additionally, it can be noticed that the test pressure and speed did not cause significant variations in the value of the friction coefficient, which disagrees with some references in the literature [2,8,10,12,18]. The main hypothesis that explains this is the high hardness of the sheet (300 HV ≈ 30 HRC), which is close to that of the pin (474 HV ≈ 47 HRC). Works that carried out friction studies with high-strength steels are scarce and they are practically non-existent for this level of resistance. However, in the study by [23,24], the tribological response of some high-strength-steel sheet was evaluated. In the work done by [24], the friction behaviour in steels whose tensile strength varies from 305 MPa to 1079 MPa, with and without lubricants, was evaluated. The results show variations within the same average obtained by this study. In [23], the sheet finish was basically evaluated in relation to friction and contact pressure for two materials (SPFC 590 and SPFC 980). It was shown that the material with higher strength has a slightly lower friction and that the contact pressure, within the same range of this article, did not vary much. The friction-coefficient values were reported to stay between 0.1 and 0.2, which agrees with the values found in the present work. However, for this study, pressure and velocity acted to stabilize friction, making its role still important in the forming of high-strength steels. Figure 5 shows that the higher the temperature, the greater the problems related to tool wear, and this can be avoided with good lubrication, since in this condition the lubricant will also act as a heat sink.

Topological Profile of Samples

In order to verify the effect of uncoated sheets and the lack of lubricant in the material surface, topological analyses before and after the tests were conducted. Figure 7 shows SEM images of the surface of the sheet before the tests. It is possible to verify that the surface of the sheet is quite irregular, presenting well-pronounced peaks and valleys, with the presence of some elements of dirt (circle in red) and the presence of voids (circle in green).

Aiming to identify the effect of each variable applied to the sheet surface, SEM micrographs were taken of some tested samples, which will be shown below. For the test without lubricant at room temperature (Figure 8), it was observed that the speed did not reduce the friction coefficient but made the stick–slip effect more stable. The pressure, however, slightly lessened friction but brought more galling to the surface. The high speed and pressure resulted in grinding off the micro bulges so that the contacting surfaces became smoother, mainly for high pressure, and led to small drop in friction coefficients. It could probably be attributed to the wear from the stage of softening, which is adhesive in nature with high inertia to restrain sliding of the tool pin against the softer steel disc [8]. The softening effect is common in sheet-metal forming and it can be seen in Figure 9, where the green line is the stress–strain curve and the red lines are the slope of the tangent lines. Here, the alpha angles decrease until they reach zero, decreasing the material’s resistance with increasing deformation.

The effects of increased contact pressure will probably occur naturally due to the high strength of the steel, which requires more effort to stamp. As a result, there may be a greater potential to generate heating due to friction, which may reach high values in localized regions. For tests done with a temperature of 60 °C (Figure 10), the pressure generated less stick–slip friction, but more galling on the surface could be observed. Generally, the speed stabilizes the stick–slip effect and reduces the friction coefficient slightly. The pressure combined with the speed generated even less stick–slip, probably due to the softening effect together with the adhesion and the surface flattening, but it was the one that generated the most galling. Figure 11 shows that according to [25], the effects of surface flattening and adhesion are antagonistic, but indeed at a certain level they can be found to be balanced when friction is minimal.

Figure 12 show the results for friction-coefficient tests at 100 °C. It can be seen that the speed was not able to decrease the stick–slip as the adhesion effects are kept very intense. In this case, the pressure ends up reducing the stick–slip and the friction. However, the galling effect ends up being much greater.

4. Conclusions

This study aimed to increase the knowledge about the tribological phenomena for TRIP1000 steel, so that it can have greater predictability in its use in final parts when a project is made accompanied by a specific numerical simulation. Thus, some conclusions can be listed below:

• Lubrication did not act to greatly decrease the friction coefficient, and, for the grease, there was an increase in the friction coefficient, mainly for a low speed.
• Likewise, speed and contact pressure did not act to greatly reduce friction, but some speeds and pressures helped to stabilize the friction stick–slip.
• The increase in temperature favoured the appearance of the stick–slip mechanism due to the increase in material adhesion, which can increase tool wear and galling. This highlights the need for a lubricant that, in addition to reducing tool wear, acts as a coolant and thus reduces material adhesion.
• Measuring the coefficient of friction with these process variables is very important so that the friction performance is understood and applied in numerical simulations of the final parts more accurately.

Some of these conclusions do not corroborate the information collected in the bibliography, as they showed that the mean numerical value of the friction coefficient did not significantly vary with the contact pressure, speed and lubrication. However, it should be noted that the material sheet used has a very high strength for sheet-metal forming application standards, which may have contributed to this phenomenon.