Front. Mech. Eng. China 2007, 2(3): 283–287 DOI 10.1007/s11465-007-0049-z

RESEARCH ARTICLE

GUO Zhipeng, XIONG Shoumei, CHO SangHyun, CHOI JeongKil

Interfacial heat transfer coefficient between metal and die during high pressure die casting process of aluminum alloy

© Higher Education Press and Springer-Verlag 2007

Abstract The present work focused on the determination of the interfacial heat transfer coefficient (IHTC) between metal and die during the high pressure die casting (HPDC) process. Experiments were carried out on an aluminum alloy, ADC12Z, using “step shape” casting—so-called because of its shape. The IHTC was successfully determined by solving one of the inverse heat problems using the nonlinear estimation method first used by Beck. The calculation results indicated that the IHTC immediately increased after liquid metal was brought into the cavity by the plunger and decreased as the solidification process of the liquid metal proceeded. The liquid metal eventually solidified completely, a condition when the IHTC tended to be stable. Casting thickness played an important role in affecting the IHTC between the metal and die not only in terms of its value but also in terms of its change tendency. Also, under the test conditions, different change tendencies of the metal solid fraction were found between castings with different thicknesses and the die. Keywords high pressure die casting, Aluminum alloy, ADC12Z, interfacial heat transfer coefficient

1

Introduction

The high pressure die casting (HPDC) process is one of the most increasingly popular and efficient methods for the production of complex shape castings of magnesium and aluminum alloys in manufacturing. Because of the excellent properties of die castings, more and more die casting products Translated from Acta Metallurgica Sinica, 2007, 43(1): 103–106 [译自: 金属学报] GUO Zhipeng, XIONG Shoumei ( ) Key Laboratory for Advanced Manufacturing by Materials Processing Technology, Department of Mechanical Engineering, Tsinghua University, Beijing 100084, China E-mail: [email protected] CHO SangHyun, CHOI JeongKil Advanced Material R&D Center, Korean Institute of Industrial Technology, 994-32, DongchunDong, YonsuKu, Inchon, Republic of Korea

are used in automotive, aerospace, electronics, and other industries [1−2]. As more commercial software for numerical simulation, such as Magma, AnyCasting and Procast, are used during the research and design process, precise simulation of filling and solidification during the foundry process has become possible. It is well known that the simulation of these processes is precise as long as boundary conditions and initial conditions are accurate [3]. One of the boundary conditions, namely the interfacial heat transfer coefficient (IHTC), is important during the solidification process because it quantifies thermal resistance at the metal-mold interface [4−6]. The determination of the IHTC is a difficult task because it changes when the temperature and the process parameters are varied. As a result, most current software assume the IHTC to be constant, which under some conditions leads to serious mistakes in calculation [7]. Researches have already been conducted in determining IHTC. However, most studies have only focused on the sand or permanent mold casting processes; there is very limited knowledge about interfacial heat transfer during HPDC solidification. In this work, the IHTC at the metal-die interface in HPDC was determined by solving one of the inverse thermal problems on the basis of the temperature profiles obtained at different locations inside the die. “Step shape” casting, a special casting, was used during the experiment with ADC12Z as the casting alloy. The influence of casting thickness on the IHTC was examined and studied in detail to obtain a clear understanding of heat transfer behavior during the HPDC process.

2

Experiment

The casting geometry shown in Fig. 1 indicates five steps with different thicknesses. Two thermocouples (TA and TC) were situated in the die at specific locations corresponding to steps 2−4. The distances from the temperature measurement points TA2, TA3 and TA4 to the cavity surface were 7, 7, and 6.8 mm, respectively. The distances from the temperature measurement points TC2, TC3 and TC4 to the cavity surface were 4, 3.8 and 3.4 mm, respectively.

284

Fig. 1 “Step shape” casting and adjustment of temperature sensors

A TOYO 650 t cold chamber die casting machine was used in the study. The temperatures were recorded by an integrated measurement and control (IMC) data acquisition system at 50 Hz. The casting material was ADC12Z and the die material was H13.

3

Mathematical model

The IHTC was determined using the nonlinear estimation method first used by Beck et al. [8]. This method greatly improves computation stability by applying future time steps. The nonlinear estimation method involves the minimization of the function F(q), which is defined as J

I

(

F (h) = ∑ ∑ h mj ,i -h cj ,i j =1 i = 0

)

2

(1)

4

Results and discussion

A TOYO 650 t cold chamber die casting machine was used, nearly 24 shots were performed, and the first 19 served to preheat the dies to a thermal equilibrium condition. The melt temperature was increased to 680°C, while the ram low speed and high speed were set to 0.4 and 2 m/s, respectively. The casting pressure after the filling process was set to 24 MPa. The die-holding time was 13 s and the die casting cyclic time was 80 s. Figure 2 shows the temperature readings from thermocouples TA2, TC2, TA3, TC3, TA4 and TC4 during the last five cycles. The temperature profiles show that cyclic temperature change corresponded to the cyclic die casting shots. The die temperature rose quickly after each shot until reaching its maximum value, after which the temperature dropped until

where h mj ,i and h cj ,i are the measured and estimated temperatures of the jth temperature-measurement point in the ith time step respectively. h is the heat transfer coefficient. By mathematical deduction, the IHTC can be determined by the following equations

∑ ∑ (h

m j ,l + i

J

I

J

Dhl +1 =

)

I

- h cj , l + i wj , l + i

j =1 i = 0

∑∑w

(2) 2 j ,l + i

j =1 i = 0

hl +1 = hl +Dhl +1 wj ,i =

yh cj ,i yhl

(3) (4)

This calculation procedure is continued until Dhl +1 / hl +1
(5)

Fig. 2 Measured die internal temperature profiles during die casting experiments

285 the next shot. The temperature of the die farther away from the cavity surface was much lower than that closer to the cavity surface. For example, the temperature profile of TA2 was located below that of TC2 in each cycle, which can be clearly seen in Fig. 2. Also, with the same distance from the cavity surface, the die temperature at step 4 was higher than that at step 2 or step 3. For example, TC4 had a much higher temperature during these cycles than TC2 and TC3, even though they were of the same distance from the cavity surface. Step 2 is taken as an example. The IHTC can be determined from the thermocouple measurements at TA2 and TC2. Figure 3 shows the calculated IHTC between step 2 and the die during the last five shots. Similar to the temperature profiles, the IHTC profile explicitly displays the response of the procedures to the operations applied. The IHTC rose immediately when the shot was performed until reaching its maximum value of about 5.20 kW/(m2K) and then fell as the solidification process proceeded. It is also noteworthy that the IHTC was negative before the shot was carried out, which implies the direction of the heat flux before the shot was from the die to the surrounding air.

Fig. 4 IHTC, Die surface temperature (DST) and measured temperatures (a) Calculated IHTC during cycle 2 between step 2 and die; (b) DST and measured temperatures during cycle 2 of step 2

Fig. 3 Calculated IHTC during cycles 1 to 5 between step 2 and the die

The second shot (cycle 2) of the last 5 shots is a good example of the direction of heat loss. Small fluctuations in the IHTC could be found when the die was opened, as shown in Fig. 4(a). The spraying process in the later part of the cycle dramatically changed the heat transfer process. During the spraying process the dies lost heat to the surrounding air, leading to a high heat transfer coefficient of nearly −1.40 kW/(m2K) at its minimum value, which ultimately resulted in a drop in die surface temperature of as much as 20°C, as shown in Fig. 4(b). Figure 5 shows the die surface temperatures and the metal surface temperatures in the second shot of steps 2−4. Figure 5(a) shows that the DST increased with increasing thickness of the casting. Additionally, the response was much

faster when the casting step was thicker. As the shot was performed, the DST of step 4 increased at a much faster speed than those of the other two steps. Similar characteristics were found in the DST profiles of steps 2 and 3 in that both increased when the shot was performed until reaching their maximum values and dropped as solidification proceeded, while the DST profile of step 4 changed in its own particular way. The DST of step 4 rose as the shot was performed until approaching its first peak value, and then dropped. After a small decrease in its value, the DST of step 4 continued to increase until reaching its maximum value. As shown in Fig. 5(b), the MST of step 4 decreased more slowly than those of the other two steps when the MST reached a point between the liquidus and solidus temperatures, resulting in an obvious temperature plateau. Additionally, the MST was higher in a thicker casting step when the die was opened.

286

Fig. 5 DST and metal surface temperature (MST) (a) DST during cycle 2 of steps 2−4; (b) MST during cycle 2 of steps 2−4

Figure 6 shows the metal solid fraction (MSF), the metal surface cell solid fraction (MSSF), and the IHTC of the different steps. The MSSF of step 4 changed differently compared to those of the other two steps. The MSSF of step 4 increased quickly when the shot was performed until the MST of step 4 reached a temperature between the liquidus and solidus temperature when the MSSF rose more slowly. The MSF of step 4, on the other hand, grew more slowly during the whole process compared with the MSSF. Additionally, as shown in Fig. 6(a), the increase in the casting step thickness greatly increased solidification time. Figure 6(b) shows the IHTC profiles between steps 2−4 and the die. Similar characteristics can be found in the change tendencies of the IHTC of steps 2 and 3 in that they both increased quickly right after the die casting shot until reaching their maximum value and then tended to be stable until opening of the die. The IHTC of step 4, on the other hand, exhibited different characteristics. Two peak values could be observed on the IHTC profile of step 4 during the whole solidification process, which means that the IHTC increased when the shot was performed until reaching its first peak value followed by a decrease and then it increased again

Fig. 6 (a) MSF and MSSF; (b) metal-die IHTC during cycle 2 of steps 2−4

until reaching a second peak value. In addition, the thicker the casting step was, the higher the IHTC was when the die was opened.

5

Conclusions

1) The IHTC exhibits different change tendencies due to the performance of different die casting operations: the IHTC increases quickly right after the die casting shot until reaching its maximum value and then tends to be stable until opening of the die. Fluctuations can be found in the IHTC profile when the die is opened. The spraying process greatly enhances the IHTC in the inverse direction. 2) The casting thickness has a great influence on the IHTC. The thicker the casting step is, the higher the IHTC. The IHTC profile for a casting cycle also varies with different casting thicknesses. 3) The higher the casting step is, the faster the MSSF will increase corresponding to the shot process. The MSSF changes more slowly during the later period of solidification. Additionally, the thicker the casting step is, the longer the solidification process will take.

287 Acknowledgements This research work was financially supported by Tsinghua-Toyo R&D Center of Magnesium and Aluminum Alloys Processing Technology (0503J18) and the National Basic Research Program of China (2005CB724105).

4.

References

5.

1. Griffiths W D. A model of the interfacial heat-transfer coefficient during unidirectional solidification of an aluminum alloy. Metallurgical and Materials Transactions B, 2000, 35B: 285−295 2. Santos C A, Quaresma J M V, Garcia A. Determination of transient interfacial heat transfer coefficients in chill mold castings. Journal of Alloys and Compounds, 2001, 319: 174−186 3. Michel F, Louchez P R, Samuel F H. Heat transfer coefficient during solidification of Al-Si alloys: Effects of mold temperature,

6. 7. 8.

coating type and thickness. AFS Transactions, 1995, 09: 275− 283 Prabhu K N, Kumar S A, Venkataraman N. Effect of coating/mold wall/casting thickness on heat transfer and solidification of Al-Cu-Si alloy (LM-21) in cast iron molds. AFS Transactions, 1994, 109: 827−832 Krishman M, Sharma D G R. Determination of heat transfer coefficient between casting and chill in unidirectional heat flow. AFS Transactions, 1994, 16: 769−774 Lau F, Lee W B, Xiong S M, et al. A study of the interfacial heat transfer between an iron casting and a metallic mould. Journal of Material Processing Technology, 1998, 79: 25−29 Hallam C P, Griffiths W D. A model of the interfacial heat-transfer coefficient for the aluminum gravity die-casting process. Metallurgical and Materials Transactions B, 2004, 35B: 721−733 Beck J V, Blackwell B, Clair C R S T. Inverse Heat Conduction: Ill-posed Problems. New York: Wiley, 1985, 145−162