- Email: [email protected]
Experimental and numerical analysis of friction in high aspect ratio combined forward-backward extrusion with retreat and advance pulse ram motion on a servo press
Journal of Materials Processing Technology 214 (2014) 936–944
Contents lists available at ScienceDirect
Journal of Materials Processing Technology journal homepage: www.elsevier.com/locate/jmatprotec
Experimental and numerical analysis of friction in high aspect ratio combined forward-backward extrusion with retreat and advance pulse ram motion on a servo press Ryo Matsumoto ∗ , Kazunori Hayashi, Hiroshi Utsunomiya Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita 565-0871, Japan
a r t i c l e
i n f o
Article history: Received 27 August 2013 Received in revised form 11 November 2013 Accepted 16 November 2013 Available online 25 November 2013 Keywords: Forging Extrusion Friction Wear Servo press
a b s t r a c t A method for maintaining lubrication in the backward extrusion of deep holes for lightweight structural components is proposed utilizing a servo press and a punch with an internal channel for liquid lubricant supply. In this forming method, the punch is pushed into the specimen with a servo press in a manner that combines pulsed and stepwise modes. Sufficient liquid lubricant is periodically supplied to the deformation zone through the internal channel upon the retreat of the punch. This forming method with pulse punch ram motion was tested in combined forward-backward extrusion process with a high aspect ratio (height/diameter) in this study. The material flow of the aluminum specimen during the extrusion with pulse punch ram motion was investigated to determine the coefficient of shear friction at the specimen–punch interface. The punch wear was assessed by a finite element analysis of the material flow of the specimen during the extrusion with pulse punch ram motion. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The programming of the ram speed and motion of servo presses with a servomotor through CNC control has led to new forming processes. Osakada et al. (2011) have reviewed servo press designs and their major applications in sheet metal forming and bulk metal forming processes. For example, Palaniswamy and Altan (2003) developed heating and stamping processes for Mg sheet with a servo press. Maeno et al. (2011) reduced the friction in cold plate forging with a servo press by implementing load pulsation. Groche and Moller (2012) investigated the friction in deep-drawing processes with a servo press that utilizes forming speed control. Terano et al. (2013) investigated the shape accuracy of the products forged on a servo press under several press ram motions. For the fabrication of lightweight components such as hollow components, we have proposed an extrusion method for forming deep holes with a servo press that utilizes a punch with an internal channel for the supply of liquid lubricant (Matsumoto et al., 2011). The concept of this forming method was derived from the machining of deep holes with tools that have internal channels for lubricant. In machining, an internal channel for lubricant in a drill makes it possible to cut deep holes by supplying lubricant to
∗ Corresponding author. Tel.: +81 6 6879 7500; fax: +81 6 6879 7500. E-mail address: [email protected] (R. Matsumoto). 0924-0136/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jmatprotec.2013.11.017
the cutting part (Weinert et al., 2004). It has been demonstrated that this forming method prevents galling in the backward extrusion of holes with an aspect ratio (height/diameter) of six when the appropriate punch ram motions are applied. In addition, it has been confirmed that this forming method provides formed holes with high shape accuracy (Matsumoto et al., 2013). However, the friction and punch wear in this forming method with pulse punch ram motion have not previously been investigated. It is difficult to directly measure the friction at the specimen–punch interface during forming with pulse punch ram motion. Sagisaka and Nakamura (2007) developed a testing method for determining the friction at the specimen–punch interface during combined forward-backward extrusion with aspect ratio of one. In this method, the friction was estimated from the material flow of the specimen in the forward and backward extruded parts. Murai et al. (2009) investigated the material flows of specimens during combined forward-backward extrusions with aspect ratios in the range of 0.4–2.0. The material flow and friction in combined forward-backward extrusion with aspect ratios greater than two have rarely been investigated. In this study, the forming method with pulse punch ram motion is applied to combined forward-backward extrusion with a high aspect ratio. The relationship between the punch motion and the material flow of the aluminum specimen is investigated in extrusion with pulse punch ram motion. The friction and punch wear are determined by analyzing the material flow of the specimen during
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
937
extrusion with pulse punch ram motion by both experiment with a servo press and the finite element analysis. 2. Extrusion with pulse punch ram motion 2.1. Combined forward-backward extrusion method The extrusion method for reducing the friction over the punch surface is shown in Fig. 1 (Matsumoto et al., 2011). The punch with an internal channel for lubricant flow is pushed into the specimen in a manner that combines pulsed and stepwise modes and assists the supply of liquid lubricant from the punch nose. The punch is connected to a lubricant tank, and the lubricant is supplied to the internal channel from the tank. During forming with a manner that combines pulsed and stepwise modes, the internal pressure in the cavity formed in the previous forming steps is depressurized by the retreat of the punch, and the lubricant is sucked into the cavity through the internal channel (Fig. 1(b)). In this method, no pump and/or check valve for the prevention of backward flow is used in the supply of the lubricant from the punch nose. The lubricant is sucked into the deformation zone because of the change in the internal pressure in the cavity. After the retreat of the punch, the punch is advanced again to continue the forming of the hole (Fig. 1(c)). Each advance of the punch can be carried out without seizure because sufficient lubricant is supplied to the forming zone during the retreat of the punch. To describe the punch motion, the following parameters are defined: ntotal : total number of forming steps sai : advance stroke in the ith forming step (i = 1 to ntotal ) sri : retreat stroke in the ith forming step (i = 1 to ntotal ) sfi : forming stroke in the ith forming step (=s − sri ) (i = 1 to ntotal ) aintotal stotal : total forming stroke of the punch (= s ) i=1 fi In this study, sai , sri , and sfi were set as constant. Thus, sai , sri , and sfi can be written as sa , sr , and sf , respectively. 2.2. Experimental conditions The tool arrangement for the forming method is shown in Fig. 2. The punch with an internal channel for lubricant flow is connected to the lubricant tank by a tube. No equipment such as a pump or a valve to prevent the backflow of the lubricant is used. Schematic illustrations of the punch with an internal channel for lubricant
Fig. 2. Schematic illustration of the tool arrangement for combined forwardbackward extrusion (DC : inner diameter of the container).
supply and the counter punch are shown in Fig. 3. The punch diameter was DP = 6.0 mm, and the diameters of the internal channel were DI = 1.5 mm at the inside of the channel and 0.5 mm at the output of the channel. A counter punch with diameter DCP = 4.5 mm was prepared to examine the material flow of the specimen in the forward and backward extruded parts. There is no internal channel for lubricant in the counter punch. The inner diameter of the container was DC = 9 mm. The extrusion ratios for the forward and backward parts were 1.80 and 1.33, respectively. The materials used for the punches and container were high speed tool steel (HRC63–65) and matrix high speed tool steel (Hitachi Metals, Ltd., YXR3, HRC59–62), respectively. The punches and container surfaces were polished to a mirror finish with Ra = 0.02–0.04 m. The initial dimensions of the specimen was 8.9 mm in diameter and L0 = 30 mm in height. The specimen material was an AA6061-T6 aluminum alloy. Mineral oil with a kinematic viscosity of 32 mm2 /s (at 40 ◦ C) was used as the lubricant. The tools were installed on a 450 kN servo press (Komatsu Industrial Corp., H1F45). The servo press was driven by an AC servomotor through a mechanical link (0–70 spm). The punch position–time and speed–position diagrams for the retreat and advance pulse ram motion are shown in Fig. 4. The total step number (ntotal ) was limited to less than five because of the press specifications. The forming stroke in every forming step was set in the range sf = 6–24 mm (sf /DP = 1.0–4.0), and the total forming stroke of the punch was fixed at stotal = 24 mm (stotal /DP = 4.0). The retreat stroke of the punch in every forming step was fixed at sr = 6 mm (sr /DP = 1.0) because it was confirmed that sufficient lubricant (approximately 18 mm3 , nominal thickness: 110 m) enters the forming zone during the retreat action of the punch when sr /DP ≥ 0.5 (Matsumoto et al., 2013). The average forming speed range was vavg = 20–80 mm/s.
Fig. 1. Retreat and advance pulse ram motion of a punch with an internal channel for pulsating lubricant supply during extrusion (sai : advance stroke of punch in the ith forming step; sri : retreat stroke of punch in the ith forming step; sfi : forming stroke of punch in the ith forming step, i = 1 to ntotal ).
938
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Fig. 3. Schematic illustrations and photographs of (a) punch with an internal channel for lubricant supply and (b) counter punch (DP : punch diameter; DI : diameter of the internal channel; DCP : counter punch diameter).
Distance from bottom dead center /mm
sf = 24mm, ntotal = 1 (vavg = 80mm/s) = 2 (v = 40mm/s) s = 12mm, n
f total avg 40 sf = 8mm, ntotal = 3 (vavg = 27mm/s) sf = 6mm, ntotal = 4 (vavg = 20mm/s) 6 35 sf = 24mm, ntotal = 1 (vavg = 20mm/s) 30 5 Top of specimen 25 4 20 3 15 2 10 1 5 0 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time /s
Distance from bottom dead center/Punch diameter
To investigate the influence of varying the total forming duration on the forming characteristics of the combined forward-backward extrusion, a punch ram motion with sf = 24 mm, ntotal = 1, and vavg = 20 mm/s was tested. The total forming duration of the punch ram motion was the same as that of a pulse punch ram motion with sf = 6 mm, ntotal = 4, and vavg = 20 mm/s.
Distance from bottom dead center/Punch diameter 6 5 4 3 2 1 0 400 sf = 24mm, ntotal = 1 (vavg = 80mm/s) 350 sf = 12mm, ntotal = 2 (vavg = 40mm/s) 300 sf = 8mm, ntotal = 3 (vavg = 27mm/s) sf = 6mm, ntotal = 4 (vavg = 20mm/s) 250 sf = 24mm, ntotal = 1 (vavg = 20mm/s) 200 150 Top of 100 specimen 50 0 –50 –100 40 35 30 25 20 15 10 5 0 Distance from bottom dead center /mm
Ram speed /mm·s
–1
(a) Punch position–time.
(b) Punch speed–position. Fig. 4. (a) Punch position–time and (b) speed–position diagrams for the retreat and advance pulse ram motion on a servo press (total forming stroke: stotal = 24 mm; vavg : average forming speed) (Komatsu Industrial Corp., H1F45).
Fig. 5 shows the initial and formed specimen shapes and a photograph of the extruded specimen. The forward and backward extruded lengths are symbolized as LF and LB , respectively. A boss was formed at the center of the bottom of the formed hole by the internal channel of the punch. It has been confirmed that the boss height increases with increases in the total punch stroke, irrespective of the punch ram motion (Matsumoto et al., 2013).
2.3. Finite element analysis conditions To analyze the material flow of the specimen during combined forward-backward extrusion with pulse punch motion, a finite element analysis was carried out by employing a commercial elastic-plastic finite element analysis code (Simufact Engineering GmbH, simufact.forming ver.11). In this simulation, the elasticplastic deformation and temperature change in the aluminum specimen were calculated with two-dimensional axisymmetric analysis, as shown in Fig. 6. The dies were assumed to be rigid bodies. The dimensions, geometries, and temperatures of the specimen and the dies used in the finite element simulation were identical to the experimental values. The punch ram motions shown in Fig. 4 were employed. Fig. 7 shows the flow stress curves of the AA6061-T6 aluminum alloy employed at several temperatures. The flow stress was measured at various temperatures by the upsettability test (Osakada et al., 1981). Since the friction conditions at the specimen–punch interface are strongly affected by the amount of lubricant supplied during extrusion, the frictional condition at the specimen–punch interface was assumed to be specified by the coefficient of shear friction mP = 0–1.0, and the coefficients of shear friction at the specimen–counter punch and specimen–container interfaces were assumed to be fixed at 0.2. The change in the friction conditions during extrusion was not considered. The sliding velocity dependency on the coefficient of shear friction was not also considered. The heat transfer coefficients for the specimen–die contact interfaces and the free surfaces of the specimen were determined with heating and cooling tests to be 10,000 W m−2 K−1 and 20 W m−2 K−1 , respectively. In the heating and cooling tests, the temperature change of the specimen surface with contacting the
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
939
Fig. 5. Cross-sectional views of initial and extruded aluminum specimens: (a) schematic illustration and (b) photograph.
die or air was measured by thermocouple. The measured temperature change was compared with the finite element analysis results under various heat transfer coefficients conditions. The specific heat capacity and the thermal conductivity of the specimen were assumed to be constant (no considering with temperature dependency) from the literature (Japan Aluminium Association, 2001). The finite element analysis conditions are summarized in Table 1. 3. Experimental results 3.1. Extrusion with no pulse punch ram motion Fig. 8 shows the surface roughness of the sidewall part of the backward extruded hole obtained with combined
forward-backward extrusion with no pulse punch motion (conventional forming method). The lubricant was applied to the top surface of the specimen before the forming. The surface roughness was measured in the circumferential direction of the formed hole by using a stylus type surface roughness tester. When the surface roughness was Ra > 0.4 m, the galling was assumed to occur in this study from comparing the surface roughness and visual inspection of the formed hole. Galling was observed in forming with sf /DP ≥ 3.0, galling occurred particularly in the early stages of the forming with sf /DP = 4.0. The maximum forming stroke of the punch for preventing galling in the forming with no pulse punch motion was found to be sf /DP = 2.0. Fig. 9 shows the forward and backward extruded lengths of the specimen obtained with combined forward-backward extrusion with no pulse punch motion with stotal /DP = 4.0. The backward extruded length increased slightly with increases in the average forming speed, while the forward extruded length decreased slightly with increases in the average forming speed.
Flow stress /MPa
500
400
20°C
300
100°C 200°C
200 300°C 400°C
100
500°C 0 0.0
0.2
0.4
0.6
0.8
1.0
1.2
Average equivalent strain Fig. 6. Two-dimensional axisymmetric analysis model of combined forwardbackward extrusion with pulse punch motion for finite element analysis.
Fig. 7. Flow stress curves for the AA6061-T6 aluminum alloy obtained with upsettability tests.
940
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Distance from top of hole of specimen z /mm
Table 1 Finite element analysis conditions for combined forward-backward extrusion with pulse punch motion. Quadrilateral (0.2 mm × 0.2 mm) 69.7 0.3 896
Young’s modulus/GPa Poisson’s ratio Specific heat capacity/J kg−1 K−1 Density/kg m−3 Thermal conductivity/W m−1 K−1 Coefficient of shear friction
Heat transfer coefficient
2700 167
Specimen–punch interface, mP Specimen–counter punch interface Specimen–container interface
0–1.0
Specimen–die interface/W m−2 K−1 Free surface of specimen/W m−2 K−1
10,000
0.2
Surface roughness Ra / µm
0
5
10
15
20
25
4.0 3.5
30
35
1.5
Galling
0.5 0.0 2
3
4
5
6
Distance from top of hole of specimen/ Punch diameter z/DP Fig. 8. Variation with total forming stroke of the surface roughness of backward extruded holes obtained with combined forward-backward extrusion with no pulse punch motion.
1.4
1.2
1.2
LB/L0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
LF/L0
0.2
0.0 0
20
40
60
1 (vavg = 2 (vavg = 3 (vavg = 4 (vavg =
30
35
80mm/s) 40mm/s) 27mm/s) 20mm/s)
3.0 2.5 2.0 1.5 Galling
1.0
1
2
3
4
5
6
3.2. Extrusion with pulse punch ram motion Fig. 10 shows the surface roughness of the sidewall of the backward extruded hole obtained with combined forward-backward extrusion with stotal /DP = 4.0 with pulse punch motion. Galling of the hole surface is caused by the sliding of the punch during the advance and/or retreat of the punch. Since the surface roughness with pulse punch motions of sf /DP ≤ 2.0 was lower than that with a no pulse punch motion of sf /DP = 4.0, the pulse punch motion is confirmed to be effective to reduce the surface roughness. Galling occurred in forming with a pulse punch motion of sf /DP ≥ 2.0. The appropriate pulse punch motion forms a hole with a smooth surface (no galling). The maximum forming stroke of the punch for preventing galling at each forming step in the forming with pulse punch motion was found to be sf /DP = 1.3. Fig. 11 shows the nominal forming pressure–stroke diagram for the forming with stotal /DP = 4.0 with pulse ram motion. The nominal forming pressure was calculated that the forming load was divided by the cross-sectional area of the punch ((DP /2)2 ). Although sufficient lubricant was periodically supplied into the formed hole cavity before every forming step in the forming with pulse punch
80
Average forming speed vavg /mm·s
0.0 100 –1
Fig. 9. Variation with forming speed on the material flow of the specimen obtained with combined forward-backward extrusion with no pulse punch motion (stotal /DP = 4.0, ntotal = 1).
Punch stroke s /mm 0
5
10
15
20
25
30
2.0
Forming pressure /GPa
1.6
1.4
Extruded length of backward part/ Initial specimen length LB/L0
Extruded length of forward part/ Initial specimen length LF/L0
1.6
0.2
4.0, ntotal = 2.0, ntotal = 1.3, ntotal = 1.0, ntotal =
25
Distance from top of hole of specimen/ Punch diameter z/DP
These effects arise mainly because the specimen temperature around the punch corner is raised by heat generation due to plastic deformation, and the specimen tends to be extruded backward.
1.0
3.5
= = = =
20
Fig. 10. Variation with punch motion of the surface roughness of the backward extruded hole in combined forward-backward extrusion with pulse punch motion (stotal /DP = 4.0).
2.0
1
4.0
15
20
2.5
0
sf/DP sf/DP sf/DP sf/DP
0
3.0
1.0
10
0.0
sf/DP = 4.0, ntotal = 1 sf/DP = 3.0, ntotal = 1 sf/DP = 2.0, ntotal = 1 sf/DP = 1.0, ntotal = 1
4.5
5
0.5
0.2
Distance from top of hole of specimen z /mm 5.0
0
4.5
Surface roughness Ra / µm
Initial element shape
Specimen (aluminum alloy)
5.0
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s)
1.5
1.0
0.5
0.0 0
1
2
3
4
5
Punch stroke/Punch diameter s/DP Fig. 11. Forming pressure–stroke diagram for the combined forward-backward extrusion of an AA6061 aluminum specimen with pulse punch motion (stotal /DP = 4.0).
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Extruded length of forward part/ Initial specimen length LF/L0
0
5
10
15
20
25
1.6 LF
1.4
LB
1.2
30 1.6 1.4
with lubrication without lubrication
1.2
1.0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0 0
1
2
3
4
Extruded length of backward part/ Initial specimen length LB/L0
Forming stroke in each forming step sf /mm
5
Forming stroke in each forming step/ Punch diameter sf/DP Fig. 12. Influence of punch motion on the material flow of an AA6061 aluminum specimen during combined forward-backward extrusion with pulse punch motion (stotal /DP = 4.0).
motion, the forming load was not sensitive to the lubricating conditions because the area of contact between the sidewall of the formed hole and the straight part of the punch (1.0 mm in length, as shown in Fig. 3(a)) was small. It is well-known that the material flow of a specimen in combined forward-backward extrusion is strongly affected by the friction at the specimen–punch interface (Sagisaka and Nakamura, 2007). Low friction at the specimen–punch interface means that the specimen extrudes backward, whereas high friction at the specimen–punch interface means that the specimen extrudes forward. Fig. 12 shows the material flows of the specimen during combined forward-backward extrusion of stotal /DP = 4.0 with pulse punch motion with and without lubricant supply to the internal channel of the punch. When the lubricant is not supplied to
941
the internal channel of the punch, the backward extruded length tends to be shorter due to the high friction and galling at the specimen–punch interface than that obtained with lubrication. Furthermore, the backward extruded length decrease with decreases in the forming stroke (sf ) under dry forming condition. The difference between the backward extruded lengths with no pulse punch motion with vavg = 20 mm/s and 80 mm/s is small (Fig. 9), while a larger difference is evident between the backward extruded lengths with the punch motions of (sf /DP = 1.0, ntotal = 4) and (sf /DP = 4.0, ntotal = 1), as shown in Fig. 12. Here, the average forming speeds of the punch motions (sf /DP = 1.0, ntotal = 4) and (sf /DP = 4.0, ntotal = 1) were vavg = 20 mm/s and 80 mm/s, respectively. Since the total sliding distance of the specimen–punch interface is long in the pulse punch motion, high friction and heavy galling are considered to be caused, especially in the dry pulse forming with sf /DP < 2.0. On the other hand, when the lubricant was supplied to the internal channel of the punch, the specimen tended to be extruded backward in the forming with sf /DP < 2.0. This result means that the lubricant is periodically supplied to the forming zone by the retreat action of the punch and that the extrusion is successfully conducted in a state of good lubrication state. The above experimental results confirm that the friction at the specimen–punch interface is effectively reduced in the forming with pulse punch motion and lubricant supply. 4. Finite element analysis results The calculated temperature distributions of the aluminum specimen after forming are shown in Fig. 13. The temperature is raised by heat generation due to plastic deformation and the specimen is cooled down mainly through heat transfer at the specimen–die contact. The temperature distribution is affected by the punch motion and the average forming speed (the total forming duration). The maximum temperature was appeared around punch corner or
Fig. 13. Temperature distributions for the aluminum specimen after combined forward-backward extrusion with/without pulse punch motion (stotal /DP = 4.0, mP = 0.2) (finite element simulation).
942
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Punch stroke s /mm 5
10
15
20
25
30
Specimen temperature /°C
400
sf/DP sf/DP sf/DP sf/DP sf/DP
350 300 250
4.0, ntotal = 2.0, ntotal = 1.3, ntotal = 1.0, ntotal = 4.0, ntotal =
= = = = =
1 (vavg = 2 (vavg = 3 (vavg = 4 (vavg = 1 (vavg =
80mm/s) 40mm/s) 27mm/s) 20mm/s) 20mm/s)
200 150
Maximum
100
Extruded length of forward part/ Initial specimen length LF/L0
0
0.7
0.5 0.4 0.3 0.2 0.1 0.0 0.0
Minimum
50
0.6
0 0
1
2
3
4
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s)
0.2
0.4
0.6
0.8
1.0
Coefficient of shear friction at specimen–punch interface mP
5
Punch stroke/Punch diameter s/DP
(a) Forward part.
at backward extruded part, while the minimum temperature was appeared at forward extruded part. Fig. 14 shows the changes in the maximum and minimum calculated temperatures of the aluminum specimen during forming. In the pulse punch motions, forming is carried out with interruptions because the punch is pushed into the specimen in a manner that combines pulsed and stepwise modes. Due to these interruptions, the maximum temperature drops periodically during the punch retreat in every forming step, and thus increases in the specimen temperature are prevented. Fig. 15 shows the changes in the forward and backward extruded lengths of the specimen during combined forward-backward extrusion. Although the material flow of the specimen changes somewhat as a result of the punch motion under the same coefficient of shear friction, the specimen is simultaneously extruded to forward and backward parts in all punch motions. The temperature distribution is strongly affected by the punch motion, as shown in Fig. 14, whereas the changes in the material flow are small under the same frictional condition (mP = 0.2). Thus, the friction at the specimen–punch interface is the dominant factor in the material flow of the specimen during the combined forward-backward extrusion with pulse punch motion. Fig. 16 shows the relationship between the extruded length of the specimen and the coefficient of shear friction at the specimen–punch interface for combined forward-backward extrusion with stotal /DP = 4.0. As mentioned in Section 3.2, the backward
1.4
Extruded length of backward part/ Initial specimen length LB/L0
Fig. 14. Changes in the maximum and minimum temperatures of the aluminum specimen during combined forward-backward extrusion with pulse punch motion (mP = 0.2) (finite element simulation).
1.3 1.2
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s)
1.1 1.0 0.9 0.8 0.7 0.0
0.2
0.4
0.6
0.8
1.0
Coefficient of shear friction at specimen–punch interface mP
(b) Backward part. Fig. 16. Relationship between the formed specimen shape and the coefficient of shear friction at the specimen–punch interface (stotal /DP = 4.0) (finite element simulation): (a) forward part and (b) backward part.
extruded length increases with decreases in the friction, whereas the forward extruded length increases with increases in the friction.
5. Discussions on the friction and punch wear 5.1. Coefficient of shear friction at the specimen-punch interface
Extruded length of forward part/ Initial specimen length LF/L0
0 1.6 1.4 1.2 1.0
5
10
30 1.6 sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) 1.4 sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) 1.2 sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s) 1.0
0.8
15
20
25
0.8
LB
0.6
0.6
0.4
0.4
LF
0.2
0.2
0.0
0.0 0
1
2
3
4
Extruded length of backward part/ Initial specimen length LB/L0
Punch stroke s /mm
5
Punch stroke/Punch diameter s/DP Fig. 15. Material flow of the specimen during combined forward-backward extrusion with pulse punch motion (mP = 0.2) (finite element simulation).
The nominal coefficient of shear friction at the specimen–punch interface (mP ) was determined by comparing the experimental extruded lengths (Fig. 12) and the results obtained with finite element analysis (Fig. 16). The determined coefficient of shear friction is shown in Fig. 17. The coefficients of shear friction determined from the forward and backward extruded lengths are almost the same. The coefficient of shear friction for forming without lubricant (dry condition) is higher than that for forming with lubricant. In forming without lubricant, the coefficient of shear friction for the pulse punch motion is higher than that without pulse punch motion because heavy galling was caused in the formed hole, as discussed in Section 3.2. In contrast, a low coefficient of shear friction arises in forming with the pulse punch motion and lubricant because good lubrication state is maintained during forming by the periodic supply of lubricant to the forming zone. Thus the forming method with pulse punch motion reduces the friction at the specimen–punch interface.
1.0
2.0 0
0.8
0.6
Ratio of punch wear
Determined Determined from LB from LF with lubrication without lubrication
0.4
0.2
943
mP = 0.4 mP = determine ed in Fig. 17 7
1.5 5
1.0 0
0.5 5
0.0 4
= 1 4 .0, to ta P l =
s n f /D
to ta P l =
s n f /D
= 4 1 .0,
Fig. 17. Relationship between determined coefficient of shear friction at the specimen–punch interface and punch ram motion during combined forwardbackward extrusion with pulse punch motion.
= 1 3 .3,
0.0 0 = 2 2 .0,
3
to ta P l =
2
s n f /D
1
Forming stroke in every forming step/ Punch diameter sf/DP
to ta P l =
0
s n f /D
Determined coefficient of shear friction at specimen–punch interface mP
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Fig. 19. Effect of punch ram motion and the coefficient of shear friction at the specimen–punch interface (mP ) on the punch wear during backward extrusion with pulse punch motion (finite element simulation).
5.2. Punch wear The punch wear caused during forming with pulse punch motion is discussed from the finite element analysis results. The punch wear (W) was calculated by using the Archard’s equation (Archard, J.F., 1953) as follows,
W=
K
Pv dt H
(1)
where K is the wear coefficient, P is the contacting pressure, v is the sliding velocity at the specimen–punch interface, H is the hardness of the punch material, and t is the time. K was assumed to be constant in this discussion because it was difficult to determine the value of K. The influence of the punch motion on the material flow of the specimen in the forward part should be removed to simplify this analysis of the relationship between punch motion and wear. Thus a counter punch with DCP = 9.0 mm was used in the finite element analysis for the calculation of the punch wear, and the punch wear was calculated for backward extrusion with pulse punch motion; the formed specimen shape was assumed to have the same shape in all punch motions.
The calculated distribution of the punch wear after one forming cycle is shown in Fig. 18. The maximum punch wear was caused around the punch corner in this extrusion ratio with all punch motions. Compared with the distribution of the punch wear in the forming with no pulse punch motion, larger punch wear was appeared at the bottom part of the punch in the forming with pulse punch motion. The calculated punch wear in one forming cycle is shown in Fig. 19. The ratio of punch wear for pulse punch motion to that for no pulse punch motion was calculated. Here, the punch wear for no pulse punch motion (sf /DP = 4.0, ntotal = 1, mP = 0.4) was set as W/K = 1.0. The punch wear with pulse punch motions in the calculation for mP = 0.4 was 1.1–1.6 times larger than that with no pulse punch motion. In the calculation of the coefficient of shear friction determined in Fig. 17, 1.1–1.6 times the punch wear was found to be also caused in the forming with pulse punch motions for maintaining good lubrication. The low friction at the specimen–punch interface provides significant sliding at the specimen–punch contact around the punch corner. As a result, although the pulse punch motion provides low friction, it does cause large punch wear.
Fig. 18. Calculated distribution of punch wear after backward extrusion with pulse punch motion (forming cycle: one time) (finite element simulation).
944
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
6. Conclusions A forming method with pulse punch motion was applied to combined forward-backward extrusion with a high aspect ratio. The coefficient of shear friction at the specimen–punch interface with pulse punch motion was determined from experimental results with a servo press and the finite element analysis results. The punch wear was discussed by using the finite element analysis. The following conclusions were obtained. (1) The proposed forming method with appropriate pulse punch motion reduces the friction between the specimen and the punch with an internal channel for lubricant to the coefficient of shear friction lower than 0.2 because sufficient liquid lubricant to prevent galling is periodically supplied to the deforming zone through the internal channel during the retreat action of the punch. (2) The punch wear in the forming method with pulse punch motion for maintaining good lubrication is 1.1–1.6 times larger than that in the conventional forming method with no pulse punch motion. Acknowledgements The authors would like to thank Nichidai Corporation for providing the dies used in this study. This study was financially supported in part by the Japan Ministry of Education, Culture, Sports, Science and Technology with a Grant-in-Aid for Young Scientists (B).
References Archard, J.F., 1953. Contact and rubbing of flat surfaces. Journal of Applied Physics 24 (8), 981–988. Groche, P., Moller, N., 2012. Tribological investigation of deep-drawing processes using servo press. In: Proceedings of the ASME 2012 International Manufacturing Science and Engineering Conference (MSEC2012), pp. 127–137. Japan Aluminium Association (Ed.), 2001. Aluminium Handbook, 26–29. (in Japanese). Maeno, T., Osakada, K., Mori, K., 2011. Reduction of friction in compression of plates by load pulsation. International Journal of Machine Tools and Manufacture 51 (7–8), 612–617. Matsumoto, R., Sawa, S., Utsunomiya, H., Osakada, K., 2011. Prevention of galling in forming of deep hole with retreat and advance pulse ram motion on servo press. CIRP Annals – Manufacturing Technology 60 (1), 315–318. Matsumoto, R., Jeon, J.Y., Utsunomiya, H., 2013. Shape accuracy in the forming of deep holes with retreat and advance pulse ram motion on a servo press. Journal of Materials Processing Technology 213 (5), 770–778. Murai, E., Ono, M., Tozawa, Y., Sawai, Y., Kojima, T., Hamaya, S., 2009. Effects of shape ratio of billet on extruded length and extrusion load in forward can-backward can combined extrusion forging – research on cold forward-backward combined extrusion forging of aluminum alloy III. Journal of the JSTP 50 (584), 847–851 (in Japanese). Osakada, K., Kawasaki, T., Mori, K., 1981. A method of determining flow stress under forming conditions. CIRP Annals – Manufacturing Technology 30 (1), 135–138. Osakada, K., Mori, K., Altan, T., Groche, P., 2011. Mechanical servo press technology for metal forming. CIRP Annals – Manufacturing Technology 60 (2), 651–672. Palaniswamy, H., Altan, T., 2003. Warm deep drawing of Mg alloy sheet. Stamping Journal 10/11, 34. Sagisaka, Y., Nakamura, T., 2007. Evaluation of friction characteristics at piercing punch by tribo-testing method based on double cup extrusion. In: Proceedings of the 3rd International Conference on Tribology in Manufacturing Processes, pp. 73–78. Terano, M., Guo, F., Yukawa, N., Ishikawa, T., 2013. Investigation of dimensional accuracy of cold backward extruded cup by using servo press. In: Proceedings of the 6th JSTP International Seminar on Precision Forging, pp. 137–140. Weinert, K., Inasaki, I., Sutherland, J.W., Wakabayashi, T., 2004. Dry machining and minimum quantity lubrication. CIRP Annals – Manufacturing Technology 53 (2), 511–537.
Contents lists available at ScienceDirect
Journal of Materials Processing Technology journal homepage: www.elsevier.com/locate/jmatprotec
Experimental and numerical analysis of friction in high aspect ratio combined forward-backward extrusion with retreat and advance pulse ram motion on a servo press Ryo Matsumoto ∗ , Kazunori Hayashi, Hiroshi Utsunomiya Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita 565-0871, Japan
a r t i c l e
i n f o
Article history: Received 27 August 2013 Received in revised form 11 November 2013 Accepted 16 November 2013 Available online 25 November 2013 Keywords: Forging Extrusion Friction Wear Servo press
a b s t r a c t A method for maintaining lubrication in the backward extrusion of deep holes for lightweight structural components is proposed utilizing a servo press and a punch with an internal channel for liquid lubricant supply. In this forming method, the punch is pushed into the specimen with a servo press in a manner that combines pulsed and stepwise modes. Sufficient liquid lubricant is periodically supplied to the deformation zone through the internal channel upon the retreat of the punch. This forming method with pulse punch ram motion was tested in combined forward-backward extrusion process with a high aspect ratio (height/diameter) in this study. The material flow of the aluminum specimen during the extrusion with pulse punch ram motion was investigated to determine the coefficient of shear friction at the specimen–punch interface. The punch wear was assessed by a finite element analysis of the material flow of the specimen during the extrusion with pulse punch ram motion. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The programming of the ram speed and motion of servo presses with a servomotor through CNC control has led to new forming processes. Osakada et al. (2011) have reviewed servo press designs and their major applications in sheet metal forming and bulk metal forming processes. For example, Palaniswamy and Altan (2003) developed heating and stamping processes for Mg sheet with a servo press. Maeno et al. (2011) reduced the friction in cold plate forging with a servo press by implementing load pulsation. Groche and Moller (2012) investigated the friction in deep-drawing processes with a servo press that utilizes forming speed control. Terano et al. (2013) investigated the shape accuracy of the products forged on a servo press under several press ram motions. For the fabrication of lightweight components such as hollow components, we have proposed an extrusion method for forming deep holes with a servo press that utilizes a punch with an internal channel for the supply of liquid lubricant (Matsumoto et al., 2011). The concept of this forming method was derived from the machining of deep holes with tools that have internal channels for lubricant. In machining, an internal channel for lubricant in a drill makes it possible to cut deep holes by supplying lubricant to
∗ Corresponding author. Tel.: +81 6 6879 7500; fax: +81 6 6879 7500. E-mail address: [email protected] (R. Matsumoto). 0924-0136/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jmatprotec.2013.11.017
the cutting part (Weinert et al., 2004). It has been demonstrated that this forming method prevents galling in the backward extrusion of holes with an aspect ratio (height/diameter) of six when the appropriate punch ram motions are applied. In addition, it has been confirmed that this forming method provides formed holes with high shape accuracy (Matsumoto et al., 2013). However, the friction and punch wear in this forming method with pulse punch ram motion have not previously been investigated. It is difficult to directly measure the friction at the specimen–punch interface during forming with pulse punch ram motion. Sagisaka and Nakamura (2007) developed a testing method for determining the friction at the specimen–punch interface during combined forward-backward extrusion with aspect ratio of one. In this method, the friction was estimated from the material flow of the specimen in the forward and backward extruded parts. Murai et al. (2009) investigated the material flows of specimens during combined forward-backward extrusions with aspect ratios in the range of 0.4–2.0. The material flow and friction in combined forward-backward extrusion with aspect ratios greater than two have rarely been investigated. In this study, the forming method with pulse punch ram motion is applied to combined forward-backward extrusion with a high aspect ratio. The relationship between the punch motion and the material flow of the aluminum specimen is investigated in extrusion with pulse punch ram motion. The friction and punch wear are determined by analyzing the material flow of the specimen during
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
937
extrusion with pulse punch ram motion by both experiment with a servo press and the finite element analysis. 2. Extrusion with pulse punch ram motion 2.1. Combined forward-backward extrusion method The extrusion method for reducing the friction over the punch surface is shown in Fig. 1 (Matsumoto et al., 2011). The punch with an internal channel for lubricant flow is pushed into the specimen in a manner that combines pulsed and stepwise modes and assists the supply of liquid lubricant from the punch nose. The punch is connected to a lubricant tank, and the lubricant is supplied to the internal channel from the tank. During forming with a manner that combines pulsed and stepwise modes, the internal pressure in the cavity formed in the previous forming steps is depressurized by the retreat of the punch, and the lubricant is sucked into the cavity through the internal channel (Fig. 1(b)). In this method, no pump and/or check valve for the prevention of backward flow is used in the supply of the lubricant from the punch nose. The lubricant is sucked into the deformation zone because of the change in the internal pressure in the cavity. After the retreat of the punch, the punch is advanced again to continue the forming of the hole (Fig. 1(c)). Each advance of the punch can be carried out without seizure because sufficient lubricant is supplied to the forming zone during the retreat of the punch. To describe the punch motion, the following parameters are defined: ntotal : total number of forming steps sai : advance stroke in the ith forming step (i = 1 to ntotal ) sri : retreat stroke in the ith forming step (i = 1 to ntotal ) sfi : forming stroke in the ith forming step (=s − sri ) (i = 1 to ntotal ) aintotal stotal : total forming stroke of the punch (= s ) i=1 fi In this study, sai , sri , and sfi were set as constant. Thus, sai , sri , and sfi can be written as sa , sr , and sf , respectively. 2.2. Experimental conditions The tool arrangement for the forming method is shown in Fig. 2. The punch with an internal channel for lubricant flow is connected to the lubricant tank by a tube. No equipment such as a pump or a valve to prevent the backflow of the lubricant is used. Schematic illustrations of the punch with an internal channel for lubricant
Fig. 2. Schematic illustration of the tool arrangement for combined forwardbackward extrusion (DC : inner diameter of the container).
supply and the counter punch are shown in Fig. 3. The punch diameter was DP = 6.0 mm, and the diameters of the internal channel were DI = 1.5 mm at the inside of the channel and 0.5 mm at the output of the channel. A counter punch with diameter DCP = 4.5 mm was prepared to examine the material flow of the specimen in the forward and backward extruded parts. There is no internal channel for lubricant in the counter punch. The inner diameter of the container was DC = 9 mm. The extrusion ratios for the forward and backward parts were 1.80 and 1.33, respectively. The materials used for the punches and container were high speed tool steel (HRC63–65) and matrix high speed tool steel (Hitachi Metals, Ltd., YXR3, HRC59–62), respectively. The punches and container surfaces were polished to a mirror finish with Ra = 0.02–0.04 m. The initial dimensions of the specimen was 8.9 mm in diameter and L0 = 30 mm in height. The specimen material was an AA6061-T6 aluminum alloy. Mineral oil with a kinematic viscosity of 32 mm2 /s (at 40 ◦ C) was used as the lubricant. The tools were installed on a 450 kN servo press (Komatsu Industrial Corp., H1F45). The servo press was driven by an AC servomotor through a mechanical link (0–70 spm). The punch position–time and speed–position diagrams for the retreat and advance pulse ram motion are shown in Fig. 4. The total step number (ntotal ) was limited to less than five because of the press specifications. The forming stroke in every forming step was set in the range sf = 6–24 mm (sf /DP = 1.0–4.0), and the total forming stroke of the punch was fixed at stotal = 24 mm (stotal /DP = 4.0). The retreat stroke of the punch in every forming step was fixed at sr = 6 mm (sr /DP = 1.0) because it was confirmed that sufficient lubricant (approximately 18 mm3 , nominal thickness: 110 m) enters the forming zone during the retreat action of the punch when sr /DP ≥ 0.5 (Matsumoto et al., 2013). The average forming speed range was vavg = 20–80 mm/s.
Fig. 1. Retreat and advance pulse ram motion of a punch with an internal channel for pulsating lubricant supply during extrusion (sai : advance stroke of punch in the ith forming step; sri : retreat stroke of punch in the ith forming step; sfi : forming stroke of punch in the ith forming step, i = 1 to ntotal ).
938
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Fig. 3. Schematic illustrations and photographs of (a) punch with an internal channel for lubricant supply and (b) counter punch (DP : punch diameter; DI : diameter of the internal channel; DCP : counter punch diameter).
Distance from bottom dead center /mm
sf = 24mm, ntotal = 1 (vavg = 80mm/s) = 2 (v = 40mm/s) s = 12mm, n
f total avg 40 sf = 8mm, ntotal = 3 (vavg = 27mm/s) sf = 6mm, ntotal = 4 (vavg = 20mm/s) 6 35 sf = 24mm, ntotal = 1 (vavg = 20mm/s) 30 5 Top of specimen 25 4 20 3 15 2 10 1 5 0 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time /s
Distance from bottom dead center/Punch diameter
To investigate the influence of varying the total forming duration on the forming characteristics of the combined forward-backward extrusion, a punch ram motion with sf = 24 mm, ntotal = 1, and vavg = 20 mm/s was tested. The total forming duration of the punch ram motion was the same as that of a pulse punch ram motion with sf = 6 mm, ntotal = 4, and vavg = 20 mm/s.
Distance from bottom dead center/Punch diameter 6 5 4 3 2 1 0 400 sf = 24mm, ntotal = 1 (vavg = 80mm/s) 350 sf = 12mm, ntotal = 2 (vavg = 40mm/s) 300 sf = 8mm, ntotal = 3 (vavg = 27mm/s) sf = 6mm, ntotal = 4 (vavg = 20mm/s) 250 sf = 24mm, ntotal = 1 (vavg = 20mm/s) 200 150 Top of 100 specimen 50 0 –50 –100 40 35 30 25 20 15 10 5 0 Distance from bottom dead center /mm
Ram speed /mm·s
–1
(a) Punch position–time.
(b) Punch speed–position. Fig. 4. (a) Punch position–time and (b) speed–position diagrams for the retreat and advance pulse ram motion on a servo press (total forming stroke: stotal = 24 mm; vavg : average forming speed) (Komatsu Industrial Corp., H1F45).
Fig. 5 shows the initial and formed specimen shapes and a photograph of the extruded specimen. The forward and backward extruded lengths are symbolized as LF and LB , respectively. A boss was formed at the center of the bottom of the formed hole by the internal channel of the punch. It has been confirmed that the boss height increases with increases in the total punch stroke, irrespective of the punch ram motion (Matsumoto et al., 2013).
2.3. Finite element analysis conditions To analyze the material flow of the specimen during combined forward-backward extrusion with pulse punch motion, a finite element analysis was carried out by employing a commercial elastic-plastic finite element analysis code (Simufact Engineering GmbH, simufact.forming ver.11). In this simulation, the elasticplastic deformation and temperature change in the aluminum specimen were calculated with two-dimensional axisymmetric analysis, as shown in Fig. 6. The dies were assumed to be rigid bodies. The dimensions, geometries, and temperatures of the specimen and the dies used in the finite element simulation were identical to the experimental values. The punch ram motions shown in Fig. 4 were employed. Fig. 7 shows the flow stress curves of the AA6061-T6 aluminum alloy employed at several temperatures. The flow stress was measured at various temperatures by the upsettability test (Osakada et al., 1981). Since the friction conditions at the specimen–punch interface are strongly affected by the amount of lubricant supplied during extrusion, the frictional condition at the specimen–punch interface was assumed to be specified by the coefficient of shear friction mP = 0–1.0, and the coefficients of shear friction at the specimen–counter punch and specimen–container interfaces were assumed to be fixed at 0.2. The change in the friction conditions during extrusion was not considered. The sliding velocity dependency on the coefficient of shear friction was not also considered. The heat transfer coefficients for the specimen–die contact interfaces and the free surfaces of the specimen were determined with heating and cooling tests to be 10,000 W m−2 K−1 and 20 W m−2 K−1 , respectively. In the heating and cooling tests, the temperature change of the specimen surface with contacting the
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
939
Fig. 5. Cross-sectional views of initial and extruded aluminum specimens: (a) schematic illustration and (b) photograph.
die or air was measured by thermocouple. The measured temperature change was compared with the finite element analysis results under various heat transfer coefficients conditions. The specific heat capacity and the thermal conductivity of the specimen were assumed to be constant (no considering with temperature dependency) from the literature (Japan Aluminium Association, 2001). The finite element analysis conditions are summarized in Table 1. 3. Experimental results 3.1. Extrusion with no pulse punch ram motion Fig. 8 shows the surface roughness of the sidewall part of the backward extruded hole obtained with combined
forward-backward extrusion with no pulse punch motion (conventional forming method). The lubricant was applied to the top surface of the specimen before the forming. The surface roughness was measured in the circumferential direction of the formed hole by using a stylus type surface roughness tester. When the surface roughness was Ra > 0.4 m, the galling was assumed to occur in this study from comparing the surface roughness and visual inspection of the formed hole. Galling was observed in forming with sf /DP ≥ 3.0, galling occurred particularly in the early stages of the forming with sf /DP = 4.0. The maximum forming stroke of the punch for preventing galling in the forming with no pulse punch motion was found to be sf /DP = 2.0. Fig. 9 shows the forward and backward extruded lengths of the specimen obtained with combined forward-backward extrusion with no pulse punch motion with stotal /DP = 4.0. The backward extruded length increased slightly with increases in the average forming speed, while the forward extruded length decreased slightly with increases in the average forming speed.
Flow stress /MPa
500
400
20°C
300
100°C 200°C
200 300°C 400°C
100
500°C 0 0.0
0.2
0.4
0.6
0.8
1.0
1.2
Average equivalent strain Fig. 6. Two-dimensional axisymmetric analysis model of combined forwardbackward extrusion with pulse punch motion for finite element analysis.
Fig. 7. Flow stress curves for the AA6061-T6 aluminum alloy obtained with upsettability tests.
940
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Distance from top of hole of specimen z /mm
Table 1 Finite element analysis conditions for combined forward-backward extrusion with pulse punch motion. Quadrilateral (0.2 mm × 0.2 mm) 69.7 0.3 896
Young’s modulus/GPa Poisson’s ratio Specific heat capacity/J kg−1 K−1 Density/kg m−3 Thermal conductivity/W m−1 K−1 Coefficient of shear friction
Heat transfer coefficient
2700 167
Specimen–punch interface, mP Specimen–counter punch interface Specimen–container interface
0–1.0
Specimen–die interface/W m−2 K−1 Free surface of specimen/W m−2 K−1
10,000
0.2
Surface roughness Ra / µm
0
5
10
15
20
25
4.0 3.5
30
35
1.5
Galling
0.5 0.0 2
3
4
5
6
Distance from top of hole of specimen/ Punch diameter z/DP Fig. 8. Variation with total forming stroke of the surface roughness of backward extruded holes obtained with combined forward-backward extrusion with no pulse punch motion.
1.4
1.2
1.2
LB/L0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
LF/L0
0.2
0.0 0
20
40
60
1 (vavg = 2 (vavg = 3 (vavg = 4 (vavg =
30
35
80mm/s) 40mm/s) 27mm/s) 20mm/s)
3.0 2.5 2.0 1.5 Galling
1.0
1
2
3
4
5
6
3.2. Extrusion with pulse punch ram motion Fig. 10 shows the surface roughness of the sidewall of the backward extruded hole obtained with combined forward-backward extrusion with stotal /DP = 4.0 with pulse punch motion. Galling of the hole surface is caused by the sliding of the punch during the advance and/or retreat of the punch. Since the surface roughness with pulse punch motions of sf /DP ≤ 2.0 was lower than that with a no pulse punch motion of sf /DP = 4.0, the pulse punch motion is confirmed to be effective to reduce the surface roughness. Galling occurred in forming with a pulse punch motion of sf /DP ≥ 2.0. The appropriate pulse punch motion forms a hole with a smooth surface (no galling). The maximum forming stroke of the punch for preventing galling at each forming step in the forming with pulse punch motion was found to be sf /DP = 1.3. Fig. 11 shows the nominal forming pressure–stroke diagram for the forming with stotal /DP = 4.0 with pulse ram motion. The nominal forming pressure was calculated that the forming load was divided by the cross-sectional area of the punch ((DP /2)2 ). Although sufficient lubricant was periodically supplied into the formed hole cavity before every forming step in the forming with pulse punch
80
Average forming speed vavg /mm·s
0.0 100 –1
Fig. 9. Variation with forming speed on the material flow of the specimen obtained with combined forward-backward extrusion with no pulse punch motion (stotal /DP = 4.0, ntotal = 1).
Punch stroke s /mm 0
5
10
15
20
25
30
2.0
Forming pressure /GPa
1.6
1.4
Extruded length of backward part/ Initial specimen length LB/L0
Extruded length of forward part/ Initial specimen length LF/L0
1.6
0.2
4.0, ntotal = 2.0, ntotal = 1.3, ntotal = 1.0, ntotal =
25
Distance from top of hole of specimen/ Punch diameter z/DP
These effects arise mainly because the specimen temperature around the punch corner is raised by heat generation due to plastic deformation, and the specimen tends to be extruded backward.
1.0
3.5
= = = =
20
Fig. 10. Variation with punch motion of the surface roughness of the backward extruded hole in combined forward-backward extrusion with pulse punch motion (stotal /DP = 4.0).
2.0
1
4.0
15
20
2.5
0
sf/DP sf/DP sf/DP sf/DP
0
3.0
1.0
10
0.0
sf/DP = 4.0, ntotal = 1 sf/DP = 3.0, ntotal = 1 sf/DP = 2.0, ntotal = 1 sf/DP = 1.0, ntotal = 1
4.5
5
0.5
0.2
Distance from top of hole of specimen z /mm 5.0
0
4.5
Surface roughness Ra / µm
Initial element shape
Specimen (aluminum alloy)
5.0
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s)
1.5
1.0
0.5
0.0 0
1
2
3
4
5
Punch stroke/Punch diameter s/DP Fig. 11. Forming pressure–stroke diagram for the combined forward-backward extrusion of an AA6061 aluminum specimen with pulse punch motion (stotal /DP = 4.0).
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Extruded length of forward part/ Initial specimen length LF/L0
0
5
10
15
20
25
1.6 LF
1.4
LB
1.2
30 1.6 1.4
with lubrication without lubrication
1.2
1.0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0 0
1
2
3
4
Extruded length of backward part/ Initial specimen length LB/L0
Forming stroke in each forming step sf /mm
5
Forming stroke in each forming step/ Punch diameter sf/DP Fig. 12. Influence of punch motion on the material flow of an AA6061 aluminum specimen during combined forward-backward extrusion with pulse punch motion (stotal /DP = 4.0).
motion, the forming load was not sensitive to the lubricating conditions because the area of contact between the sidewall of the formed hole and the straight part of the punch (1.0 mm in length, as shown in Fig. 3(a)) was small. It is well-known that the material flow of a specimen in combined forward-backward extrusion is strongly affected by the friction at the specimen–punch interface (Sagisaka and Nakamura, 2007). Low friction at the specimen–punch interface means that the specimen extrudes backward, whereas high friction at the specimen–punch interface means that the specimen extrudes forward. Fig. 12 shows the material flows of the specimen during combined forward-backward extrusion of stotal /DP = 4.0 with pulse punch motion with and without lubricant supply to the internal channel of the punch. When the lubricant is not supplied to
941
the internal channel of the punch, the backward extruded length tends to be shorter due to the high friction and galling at the specimen–punch interface than that obtained with lubrication. Furthermore, the backward extruded length decrease with decreases in the forming stroke (sf ) under dry forming condition. The difference between the backward extruded lengths with no pulse punch motion with vavg = 20 mm/s and 80 mm/s is small (Fig. 9), while a larger difference is evident between the backward extruded lengths with the punch motions of (sf /DP = 1.0, ntotal = 4) and (sf /DP = 4.0, ntotal = 1), as shown in Fig. 12. Here, the average forming speeds of the punch motions (sf /DP = 1.0, ntotal = 4) and (sf /DP = 4.0, ntotal = 1) were vavg = 20 mm/s and 80 mm/s, respectively. Since the total sliding distance of the specimen–punch interface is long in the pulse punch motion, high friction and heavy galling are considered to be caused, especially in the dry pulse forming with sf /DP < 2.0. On the other hand, when the lubricant was supplied to the internal channel of the punch, the specimen tended to be extruded backward in the forming with sf /DP < 2.0. This result means that the lubricant is periodically supplied to the forming zone by the retreat action of the punch and that the extrusion is successfully conducted in a state of good lubrication state. The above experimental results confirm that the friction at the specimen–punch interface is effectively reduced in the forming with pulse punch motion and lubricant supply. 4. Finite element analysis results The calculated temperature distributions of the aluminum specimen after forming are shown in Fig. 13. The temperature is raised by heat generation due to plastic deformation and the specimen is cooled down mainly through heat transfer at the specimen–die contact. The temperature distribution is affected by the punch motion and the average forming speed (the total forming duration). The maximum temperature was appeared around punch corner or
Fig. 13. Temperature distributions for the aluminum specimen after combined forward-backward extrusion with/without pulse punch motion (stotal /DP = 4.0, mP = 0.2) (finite element simulation).
942
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Punch stroke s /mm 5
10
15
20
25
30
Specimen temperature /°C
400
sf/DP sf/DP sf/DP sf/DP sf/DP
350 300 250
4.0, ntotal = 2.0, ntotal = 1.3, ntotal = 1.0, ntotal = 4.0, ntotal =
= = = = =
1 (vavg = 2 (vavg = 3 (vavg = 4 (vavg = 1 (vavg =
80mm/s) 40mm/s) 27mm/s) 20mm/s) 20mm/s)
200 150
Maximum
100
Extruded length of forward part/ Initial specimen length LF/L0
0
0.7
0.5 0.4 0.3 0.2 0.1 0.0 0.0
Minimum
50
0.6
0 0
1
2
3
4
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s)
0.2
0.4
0.6
0.8
1.0
Coefficient of shear friction at specimen–punch interface mP
5
Punch stroke/Punch diameter s/DP
(a) Forward part.
at backward extruded part, while the minimum temperature was appeared at forward extruded part. Fig. 14 shows the changes in the maximum and minimum calculated temperatures of the aluminum specimen during forming. In the pulse punch motions, forming is carried out with interruptions because the punch is pushed into the specimen in a manner that combines pulsed and stepwise modes. Due to these interruptions, the maximum temperature drops periodically during the punch retreat in every forming step, and thus increases in the specimen temperature are prevented. Fig. 15 shows the changes in the forward and backward extruded lengths of the specimen during combined forward-backward extrusion. Although the material flow of the specimen changes somewhat as a result of the punch motion under the same coefficient of shear friction, the specimen is simultaneously extruded to forward and backward parts in all punch motions. The temperature distribution is strongly affected by the punch motion, as shown in Fig. 14, whereas the changes in the material flow are small under the same frictional condition (mP = 0.2). Thus, the friction at the specimen–punch interface is the dominant factor in the material flow of the specimen during the combined forward-backward extrusion with pulse punch motion. Fig. 16 shows the relationship between the extruded length of the specimen and the coefficient of shear friction at the specimen–punch interface for combined forward-backward extrusion with stotal /DP = 4.0. As mentioned in Section 3.2, the backward
1.4
Extruded length of backward part/ Initial specimen length LB/L0
Fig. 14. Changes in the maximum and minimum temperatures of the aluminum specimen during combined forward-backward extrusion with pulse punch motion (mP = 0.2) (finite element simulation).
1.3 1.2
sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s)
1.1 1.0 0.9 0.8 0.7 0.0
0.2
0.4
0.6
0.8
1.0
Coefficient of shear friction at specimen–punch interface mP
(b) Backward part. Fig. 16. Relationship between the formed specimen shape and the coefficient of shear friction at the specimen–punch interface (stotal /DP = 4.0) (finite element simulation): (a) forward part and (b) backward part.
extruded length increases with decreases in the friction, whereas the forward extruded length increases with increases in the friction.
5. Discussions on the friction and punch wear 5.1. Coefficient of shear friction at the specimen-punch interface
Extruded length of forward part/ Initial specimen length LF/L0
0 1.6 1.4 1.2 1.0
5
10
30 1.6 sf/DP = 4.0, ntotal = 1 (vavg = 80mm/s) sf/DP = 2.0, ntotal = 2 (vavg = 40mm/s) 1.4 sf/DP = 1.3, ntotal = 3 (vavg = 27mm/s) sf/DP = 1.0, ntotal = 4 (vavg = 20mm/s) 1.2 sf/DP = 4.0, ntotal = 1 (vavg = 20mm/s) 1.0
0.8
15
20
25
0.8
LB
0.6
0.6
0.4
0.4
LF
0.2
0.2
0.0
0.0 0
1
2
3
4
Extruded length of backward part/ Initial specimen length LB/L0
Punch stroke s /mm
5
Punch stroke/Punch diameter s/DP Fig. 15. Material flow of the specimen during combined forward-backward extrusion with pulse punch motion (mP = 0.2) (finite element simulation).
The nominal coefficient of shear friction at the specimen–punch interface (mP ) was determined by comparing the experimental extruded lengths (Fig. 12) and the results obtained with finite element analysis (Fig. 16). The determined coefficient of shear friction is shown in Fig. 17. The coefficients of shear friction determined from the forward and backward extruded lengths are almost the same. The coefficient of shear friction for forming without lubricant (dry condition) is higher than that for forming with lubricant. In forming without lubricant, the coefficient of shear friction for the pulse punch motion is higher than that without pulse punch motion because heavy galling was caused in the formed hole, as discussed in Section 3.2. In contrast, a low coefficient of shear friction arises in forming with the pulse punch motion and lubricant because good lubrication state is maintained during forming by the periodic supply of lubricant to the forming zone. Thus the forming method with pulse punch motion reduces the friction at the specimen–punch interface.
1.0
2.0 0
0.8
0.6
Ratio of punch wear
Determined Determined from LB from LF with lubrication without lubrication
0.4
0.2
943
mP = 0.4 mP = determine ed in Fig. 17 7
1.5 5
1.0 0
0.5 5
0.0 4
= 1 4 .0, to ta P l =
s n f /D
to ta P l =
s n f /D
= 4 1 .0,
Fig. 17. Relationship between determined coefficient of shear friction at the specimen–punch interface and punch ram motion during combined forwardbackward extrusion with pulse punch motion.
= 1 3 .3,
0.0 0 = 2 2 .0,
3
to ta P l =
2
s n f /D
1
Forming stroke in every forming step/ Punch diameter sf/DP
to ta P l =
0
s n f /D
Determined coefficient of shear friction at specimen–punch interface mP
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
Fig. 19. Effect of punch ram motion and the coefficient of shear friction at the specimen–punch interface (mP ) on the punch wear during backward extrusion with pulse punch motion (finite element simulation).
5.2. Punch wear The punch wear caused during forming with pulse punch motion is discussed from the finite element analysis results. The punch wear (W) was calculated by using the Archard’s equation (Archard, J.F., 1953) as follows,
W=
K
Pv dt H
(1)
where K is the wear coefficient, P is the contacting pressure, v is the sliding velocity at the specimen–punch interface, H is the hardness of the punch material, and t is the time. K was assumed to be constant in this discussion because it was difficult to determine the value of K. The influence of the punch motion on the material flow of the specimen in the forward part should be removed to simplify this analysis of the relationship between punch motion and wear. Thus a counter punch with DCP = 9.0 mm was used in the finite element analysis for the calculation of the punch wear, and the punch wear was calculated for backward extrusion with pulse punch motion; the formed specimen shape was assumed to have the same shape in all punch motions.
The calculated distribution of the punch wear after one forming cycle is shown in Fig. 18. The maximum punch wear was caused around the punch corner in this extrusion ratio with all punch motions. Compared with the distribution of the punch wear in the forming with no pulse punch motion, larger punch wear was appeared at the bottom part of the punch in the forming with pulse punch motion. The calculated punch wear in one forming cycle is shown in Fig. 19. The ratio of punch wear for pulse punch motion to that for no pulse punch motion was calculated. Here, the punch wear for no pulse punch motion (sf /DP = 4.0, ntotal = 1, mP = 0.4) was set as W/K = 1.0. The punch wear with pulse punch motions in the calculation for mP = 0.4 was 1.1–1.6 times larger than that with no pulse punch motion. In the calculation of the coefficient of shear friction determined in Fig. 17, 1.1–1.6 times the punch wear was found to be also caused in the forming with pulse punch motions for maintaining good lubrication. The low friction at the specimen–punch interface provides significant sliding at the specimen–punch contact around the punch corner. As a result, although the pulse punch motion provides low friction, it does cause large punch wear.
Fig. 18. Calculated distribution of punch wear after backward extrusion with pulse punch motion (forming cycle: one time) (finite element simulation).
944
R. Matsumoto et al. / Journal of Materials Processing Technology 214 (2014) 936–944
6. Conclusions A forming method with pulse punch motion was applied to combined forward-backward extrusion with a high aspect ratio. The coefficient of shear friction at the specimen–punch interface with pulse punch motion was determined from experimental results with a servo press and the finite element analysis results. The punch wear was discussed by using the finite element analysis. The following conclusions were obtained. (1) The proposed forming method with appropriate pulse punch motion reduces the friction between the specimen and the punch with an internal channel for lubricant to the coefficient of shear friction lower than 0.2 because sufficient liquid lubricant to prevent galling is periodically supplied to the deforming zone through the internal channel during the retreat action of the punch. (2) The punch wear in the forming method with pulse punch motion for maintaining good lubrication is 1.1–1.6 times larger than that in the conventional forming method with no pulse punch motion. Acknowledgements The authors would like to thank Nichidai Corporation for providing the dies used in this study. This study was financially supported in part by the Japan Ministry of Education, Culture, Sports, Science and Technology with a Grant-in-Aid for Young Scientists (B).
References Archard, J.F., 1953. Contact and rubbing of flat surfaces. Journal of Applied Physics 24 (8), 981–988. Groche, P., Moller, N., 2012. Tribological investigation of deep-drawing processes using servo press. In: Proceedings of the ASME 2012 International Manufacturing Science and Engineering Conference (MSEC2012), pp. 127–137. Japan Aluminium Association (Ed.), 2001. Aluminium Handbook, 26–29. (in Japanese). Maeno, T., Osakada, K., Mori, K., 2011. Reduction of friction in compression of plates by load pulsation. International Journal of Machine Tools and Manufacture 51 (7–8), 612–617. Matsumoto, R., Sawa, S., Utsunomiya, H., Osakada, K., 2011. Prevention of galling in forming of deep hole with retreat and advance pulse ram motion on servo press. CIRP Annals – Manufacturing Technology 60 (1), 315–318. Matsumoto, R., Jeon, J.Y., Utsunomiya, H., 2013. Shape accuracy in the forming of deep holes with retreat and advance pulse ram motion on a servo press. Journal of Materials Processing Technology 213 (5), 770–778. Murai, E., Ono, M., Tozawa, Y., Sawai, Y., Kojima, T., Hamaya, S., 2009. Effects of shape ratio of billet on extruded length and extrusion load in forward can-backward can combined extrusion forging – research on cold forward-backward combined extrusion forging of aluminum alloy III. Journal of the JSTP 50 (584), 847–851 (in Japanese). Osakada, K., Kawasaki, T., Mori, K., 1981. A method of determining flow stress under forming conditions. CIRP Annals – Manufacturing Technology 30 (1), 135–138. Osakada, K., Mori, K., Altan, T., Groche, P., 2011. Mechanical servo press technology for metal forming. CIRP Annals – Manufacturing Technology 60 (2), 651–672. Palaniswamy, H., Altan, T., 2003. Warm deep drawing of Mg alloy sheet. Stamping Journal 10/11, 34. Sagisaka, Y., Nakamura, T., 2007. Evaluation of friction characteristics at piercing punch by tribo-testing method based on double cup extrusion. In: Proceedings of the 3rd International Conference on Tribology in Manufacturing Processes, pp. 73–78. Terano, M., Guo, F., Yukawa, N., Ishikawa, T., 2013. Investigation of dimensional accuracy of cold backward extruded cup by using servo press. In: Proceedings of the 6th JSTP International Seminar on Precision Forging, pp. 137–140. Weinert, K., Inasaki, I., Sutherland, J.W., Wakabayashi, T., 2004. Dry machining and minimum quantity lubrication. CIRP Annals – Manufacturing Technology 53 (2), 511–537.