Effect of Production Rate on Lubrication Performance of Environmentally-Friendly Lubricant in Combined Forward-Can and Backward-Can Cold Extrusion Test of Aluminum Alloy

Abstract

The effect of production rate on the lubrication performance of an environmentally-friendly lubricant in the cold forging of aluminum alloy was investigated using the tribological test of combined forward-can and backward-can extrusion proposed by one of the present authors. In this test, the lubrication performance was examined by both finite element analysis and experimentally. A double-layer-type environmentally-friendly solid lubricant was used. The undercoat plays the role of mitigating pick up, whereas the overcoat reduces friction. Using finite element analysis, a calibration diagram of the relationship between the extruded geometry of the workpiece and the punch stroke was prepared at the production rate of 1 and 20 stroke per minutes. The Coulomb friction coefficient between the workpiece and the outer die was identified by plotting the punch stroke and the extruded height of the workpiece on the calibration diagram. Three different surfaces of annealed aluminum alloy A4032 formed by different treatment conditions were used. The first is a surface roughened by a lathe, the second is a surface roughened by a lathe and then alkali-etched, and the third is a surface roughened by a lathe and then shot blasted. The tests were performed by varying the punch stroke between the two production rates. As a result, the dependency of the friction coefficient on the production rate was reasonably clarified, and the friction coefficient value estimated was higher when the production rate was smaller. A significant influence of the surface treatment on the friction coefficient was not observed due to the moderate deformation resistance of workpiece material.

1. Introduction

A large number of cold forged parts made of aluminum alloy have been used mainly in the automotive industry. Furthermore, forging processes with less finishing procedures or no finishing steps at all, known as near net shape or net shape forging, are in demand.¹⁾ On the other hand, cold forging has problems with the high pressures that need to be applied on the workpiece and tool, and it will subject the metal to a harsh and severe forming process. Therefore, an appropriate lubricant is required for a sufficient deformation of the material, the reduction of forming force to prevent premature tool failure and the avoidance of direct contact between the workpiece and the tools during the various processes.²⁾

The aluminum fluoride conversion coating (Al–F) is well known lubricant for aluminum cold forging in Japan. It exhibits a superior anti-pickup property under the excessive surface expansion of the workpiece. However, it possesses high environmental risks due to the wastes generated during the coating process. As an alternative to the conversion coatings, dry-in-place type environmentally friendly lubricants have been developed. Their coating process is simple and generates little waste. Therefore, they are more advantageous from the perspective of environmental risks and costs.³⁾

On the other hand, it is necessary to evaluate whether the newly developed lubricants have sufficient lubrication performance for the increasingly complex forging conditions encountered in near net shape or net shape forging or not. Friction tests for cold forging must be able to reproduce the harsh lubrication conditions. In particular, backward extrusion type forging is one such rigorous process due to the larger surface expansion at the punch shoulder region.

One of the present authors proposed the combined forward-can and backward-can extrusion (WCL) test to evaluate the lubrication performance of lubricant in backward extrusions. This method enables us to evaluate the friction characteristic value of piercing punches by plotting the dimensions of the deformed workpiece on a theoretical calibration diagram, without a measurement of the punch load.^4,5)

Sagisaka et al. tried to reveal the impact of production rate on the friction coefficient by the WCL test. However, it was not possible to estimate a reasonable friction coefficient at the higher production rate because of the lack of precision on the calibration diagram using rigid-plastic finite element analysis.⁶⁾ The higher production rate is often selected in the actual mass production of cold forged parts, though the lower production rate is employed during the design phase of the forging process. Therefore, it is imperative that the WCL test appropriately reflects the change in the production rate.

In the present study, the WCL test was applied to reveal the influence of the production rate on the lubrication performance for the backward extrusion type cold forging of an aluminum alloy. A double-layer-type environmentally friendly lubricant (DLT) was employed as the lubricant, which is a type of solid lubricant that has been developed in recent years by the present authors. The experiments were conducted using two different production rates. Since a precise calibration diagram to estimate the reasonable friction coefficient is necessary, strain rate- and temperature-dependent stress-strain curves were employed for use in elastic-plastic finite element analysis.

2. Double-Layer-Type Environmentally-Friendly Solid Lubricant

Environmentally friendly solid lubricant films have been developed by the present authors as an alternative to the Al–F coating.

The solid lubricant film must have anti-pick up properties and sufficient strength to prevent breakage of the film. Therefore, it needs to adhere firmly to the workpiece and needs a low shear resistance to improve the lubrication performance. Low shear resistance films have low strength, while highly adhesive films tend to stick to the die and increase friction. Thus, it is difficult to realize these contradictory requirements with a single-layer lubricating film.³⁾

To overcome this problem, the present authors have developed a double layer structured dry-in-place type lubricant (DLT) as shown in the Fig. 1.^3,8,9) The undercoat adheres to the workpiece material, making it easy for the lubricant to follow the surface expansion of workpiece, and further, it facilitates seizure resistance. The overcoat is made of metal soap that acts to reduce friction and prevents the destruction of the undercoat by responding to the shear deformation of the lubricating film. This structure is formed by repeating the process of dipping and drying. The undercoat is applied by first dipping and drying, following which step, the overcoat is applied. This simple coating process leads to waste- and cost-savings.

Fig. 1

Schematics of Double-Layer-Type Environmentally-friendly solid lubricant structure.

3. Combined Forward Can–Backward Can Type Extrusion Test

Figure 2 shows the principle of the combined forward-can and backward-can type extrusion test (WCL test).^8,9) Generally, in backward extrusion, material flow with a large surface expansion at the punch surface occurs. Therefore, the material deforms under the harsh lubrication conditions during this process. The WCL test can help in analyzing the lubrication performance of the upper and lower punches. Furthermore, the Coulomb friction coefficient of the punches μ_P can be estimated by using the calibration diagram prepared with the help of finite element analysis by assuming various μ_P values when calculating.

Fig. 2

Schematics of WCL test.

In the WCL test, however, not only μ_P but also the Coulomb friction coefficient of the straight die surface μ_D affect the changes in geometry of the workpiece. Therefore, it is necessary to identify μ_D prior to the WCL test. For this purpose, the test (WC test) that estimates μ_D was conducted as shown in Fig. 3.

Fig. 3

Schematics of WC test.

In the WC test, the material flow is more downward because of a lesser reduction in area on the side of the lower punch. However, as μ_D increases, the upward flow increases due to the increase of frictional resistance to the downward flow. The influence of punches on the WC test can be negligible due to the short bearing length. In that case, μ_D can be estimated by plotting the measured relationship between H_U and S_P on a calibration diagram showing μ_D calculated by FE analysis. After μ_D is determined, a calibration diagram showing the relationship between H_U and S_P for the WCL test was calculated. Next, μ_P was estimated by plotting the measured values of the workpiece on the diagram.

4. Calculation of Calibration Diagrams

The software used was Simufact Forming (MSC software). Table 1 shows the analysis parameters. The upper and lower punches were assumed to have the same friction coefficient because they have the same coating film to avoid pick-up. In the WC test, μ_P was set to 0.1, since the influence of punches on the WC test can be negligible due to the short bearing length, as previously mentioned.

Table 1 Simulation condition.

Aluminum alloy A4032 was used in the present experiments, which has the chemical composition Al–10.4Si–2.3Cu–0.35Mg–0.11Fe–0.03Mn–0.02Zn–0.01Ti. The material model was prepared using the software package JMatPro (Sante Software). The calculated material model was modified using the result of the compression test of the workpiece material at room temperature (20°C) and a small strain rate (order of 10⁻²/s), so that the calculated material data can yield the experimental result. The strain rate- and temperature-dependent equivalent stress–equivalent plastic strain relationship is shown in Fig. 4. Negligible or negative work hardening behavior (equivalent to softening) due to recrystallization is observed, which is salient under the smaller strain rate and higher temperature. Lee et al. conducted a compression test of the aluminum alloy A4032 at certain temperatures and strain rates to obtain the flow stress successfully with negligible work hardening due to recrystallization. These results are similar to those obtained in the present study. In addition, the flow stress curves in the present study are in the range of strains of 4.0. Such a range is deemed sufficient to complete the calculations required for the WCL test.

Fig. 4

Strain rate and temperature-dependent equivalent stress–equivalent plastic strain curves.

The physical properties of the material used in the present study were thermal conductivity: 160 W/(m·°C), heat capacity: 950 J/(kg·°C), density: 2700 kg/m³, and linear thermal expansion ratio: 2.1 × 10⁻⁵/°C.

The heat transfer coefficient between the workpiece and tools was assumed to be 2.0 × 10⁴ W/(m²·°C).

5. Experimental Conditions

The experimental conditions are shown in Table 2. In this study, workpieces were prepared by turning raw material into a shape that was 20 mm in diameter and 20 mm in length. After degreasing, some workpieces were alkali-etched or shot-blasted, in order to examine the effect of the surface treatment on the lubrication performance. Following this step, DLT lubricant was applied to all the workpieces. For the coating of die and punches, AlCrN film (ALCRONA PRO, Oerlikon Balzers) was applied for the sake of prevention of the pick-up of the workpiece material to die and punches.

Table 2 Experiment condition.

Two production rates of 1 and 20 strokes per minute (spm) were employed in the present experiments. Figure 5 shows the relationship between the punch velocity v and the punch position from the bottom dead center δ at 1 and 20 spm’s. The vs at 1 and 20 spm’s at a position 10 mm above the bottom dead center are around 5.25 and 105 mm/s, respectively. Sagisaka et al. referred to the impact of the velocity on the friction coefficient evaluation using the WCL and WC test.⁶⁾ They conducted the experiments using only a velocity of 3 mm/s due to the lack of material data with both strain rate and temperature dependencies of aluminum alloy A6061. Though it was calculated by the dedicated software, the strain rate- and temperature-dependent flow stress curves were employed to prepare more precise calibration diagrams in the present study.

Fig. 5

Relationship between punch velocity and punch position from bottom dead center at 1 and 20 spm’s.

6. Experimental Results and Discussion

Figures 6 and 7 show the calibration diagrams and test results of the WC test at 1 and 20 spm’s. The test results can be confirmed to be located between μ_D = 0.1 and 0.05. The relationship of μ_D to S_P is shown in Fig. 8. From the figure, μ_D can be estimated to be around 0.05. The μ_D used in the WCL test was determined to be $\overline{\mu _{D}} = 0.055$, which is an average value.

Fig. 6

Calibration diagrams of μ_D and experimental results of H_U vs. S_P at 1 spm.

Fig. 7

Calibration diagrams of μ_D and experimental results of H_U vs. S_P at 20 spm.

Fig. 8

Relationship between μ_D and S_P.

Figures 9 and 10 show the calibration diagrams of the WCL test and the test results at 1 and 20 spm’s. It can be confirmed that the experimental value of μ_P at 1 spm is located between 0.1 and 0.2, whereas it lies between 0.01 and 0.1 at 20 spm.

Fig. 9

Calibration diagrams of μ_P and experimental results of H_U vs. S_P at 1 spm.

Fig. 10

Calibration diagrams of μ_P and experimental results of H_U vs. S_P at 20 spm.

Figure 11 shows the change in μ_P to S_P. There are two tendencies observed from the figure, one of which is μ_P at 20 spm is smaller than that at 1 spm. The other tendency is that μ_P decreases as S_P increases in both 1 and 20 spm’s. It was assumed that the increases of production rate and S_P lead temperature rise during the process, which may induce not only the softening of the workpiece material but also the change in lubrication performance of DLT. Furthermore, the increase of S_P leads the increase of surface expansion, which makes the lubricant thinner to reduce the lubrication performance. The change in μ_P results from the combined effect of the temperature and surface expansion.

Fig. 11

Relationship between μ_P and S_P.

The effect of surface treatments on friction was not significant, except for the higher μ_P of alkali-etched workpiece at 1 and 20 spm’s when S_P are small. As stated previously by the present authors,¹⁰⁾ the surface topography is not a dominant factor when the deformation resistance of the workpiece material is moderate, as is the case with annealed aluminum alloy. This may also hold in the present study.

The effect of production rate on the temperature rise, which is a commonly known phenomenon in the cold forging process,¹¹⁾ in the present experiments was shown in Fig. 12 at the S_P of 15.5 mm. The temperature at 20 spm is higher than that at 1 spm. The temperature rise is mainly due to heat generation by plastic deformation and friction. The difference in the temperature rise results from the transient phenomenon of heat transfer from the workpiece to the tools.

Fig. 12

Distribution of temperature at 1 and 20 spm’s.

Figure 13 shows the surface of the workpiece formed at 1 and 20 spm’s, where S_P was set to 17.5 mm and the surface was shot blasted. At 20 spm, the metal surface was exposed because of the breakage of the lubricating film. At 1 spm, though the overcoat is broken, the undercoat still remains and follows the surface expansion of the workpiece. The reason for this is that the surface expansion at the die surface at 20 spm is larger than that at 1 spm.

Fig. 13

Surface of workpiece at 1 and 20 spm’s.

By the analysis, the surface area expansion ratio at the region in contact with the die surface at 20 spm was found to be about 50 times larger than that at 1 spm. In the WCL test, though S_P is the same, the increase of the forming speed promotes the surface expansion at the die surface. As a result, the lubricant film becomes thinner and there is a possibility of film breakage that leads to pick-up. In the present experiments, it is considered that the only the impact of temperature on lubrication performance emerged, because, no pick-up occurred thanks to the use of coating film of AlCrN.

Finally, let us note that the friction coefficient estimated is but overall (average) value on entire contact surface between the workpiece and tools, not the relationship between the specific strain rate and friction coefficient, because the strain rate is widely distributed in the workpiece. Another method of experiment is necessary to clarify the relationship between strain rate and friction coefficient.

7. Conclusions

The effect of the production rate on the lubrication performance of the environmentally-friendly solid lubricant for the cold forging of aluminum alloys was investigated using the combined forward-can and backward-can extrusion test. The conclusions obtained were summarized as follows.

1. (1)    The Coulomb friction coefficient of the punches at 20 spm is identified to be approximately the value of the Coulomb friction coefficient of the die.
2. (2)    The Coulomb friction coefficient of the punches at 20 spm is lower than that at 1 spm. This is because that the temperature rises at 20 spm is greater than that at 1 spm.
3. (3)    In the combined forward-can and backward-can extrusion test, the workpiece surface expansion on the die side is larger at 20 spm than at 1 spm. Therefore, the lubricant could not follow the deformation of the material and caused the film to crack at 20 spm. On the other hand, at 1 spm, only the overcoat was broken, while the undercoat remained intact.

REFERENCES

• 1)   R.  Lorenz,  H.  Hagenah and  M.  Merklein: Mater. Sci. Forum 918 (2018) 65–70.
• 2)   T.  Schrader,  M.  Shirgaokar and  T.  Altan: J. Mater. Process. Technol. 189 (2007) 36–44.
• 3)   Y.  Sagisaka,  K.  Hayakawa,  T.  Nakamura,  I.  Ishibashi and  T.  Nakakura: Adv. Mater. Res. 966–967 (2014) 301–310.
• 4)   T.  Nakamura,  S.  Tanaka,  K.  Hayakawa and  I.  Takahashi: Lubr. Eng. 59 (2003) 12–17.
• 5)  Y. Sagisaka and T. Nakamura: Proc. 3rd Int. Conf. Tribo. Manuf. Process., (2007) pp. 73–78.
• 6)   Y.  Sagisaka,  T.  Nakamura,  S.  Tanaka and  K.  Hayakawa: J. Jpn. Soc. Technol. Plast. 48 (2007) 56–60.
• 7)   S.W.  Lee,  J.M.  Lee and  M.S.  Joun: Tribol. Int. 141 (2020) 105855.
• 8)   Y.  Sagisaka,  I.  Ishibashi,  T.  Nakamura,  M.  Sekizawa,  Y.  Sumioka and  M.  Kawano: J. Mater. Process. Technol. 212 (2012) 1869–1874.
• 9)   Y.  Sagisaka,  K.  Hayakawa,  T.  Nakamura and  I.  Ishibashi: J. Manuf. Process. 15 (2013) 96–101.
• 10)   Y.  Sagisaka,  I.  Ishibashi,  T.  Nakakura,  T.  Nakamura and  K.  Hayakawa: J. Jpn. Soc. Technol. Plast. 57 (2016) 473–478.
• 11)   T.  Ishikawa,  T.  Ishiguro,  N.  Yukawa and  T.  Goto: CIRP Ann. 63 (2014) 289–292.