- Email: [email protected]
Influence of pressure and surface roughness on the heat transfer efficiency during water spray quenching of 6082 aluminum alloy
Journal of Materials Processing Technology 214 (2014) 2877–2883
Contents lists available at ScienceDirect
Journal of Materials Processing Technology journal homepage: www.elsevier.com/locate/jmatprotec
Influence of pressure and surface roughness on the heat transfer efficiency during water spray quenching of 6082 aluminum alloy Rong Xu a,b , Luoxing Li a,b,∗ , Liqiang Zhang a,c , Biwu Zhu a,d , Xiao Liu a,d , Xiaobing Bu a,b a
State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, 410082 Hunan, PR China College of Mechanical and Vehicle Engineering, Hunan University, Changsha, 410082 Hunan, PR China College of Mechanical and Electrical Engineering, Central South University of Forestry and Technology, Changsha, 410004 Hunan, PR China d College of Electromechanical Engineering, Hunan University Science and Technology, Xiangtan, 411002 Hunan, PR China b c
a r t i c l e
i n f o
Article history: Received 14 March 2014 Received in revised form 28 June 2014 Accepted 30 June 2014 Available online 8 July 2014 Keywords: Interfacial heat transfer Spray quenching Spray pressure Surface roughness Aluminum alloy Nucleation site density
a b s t r a c t The heat flux (q) and heat transfer coefficient (h) at the interface between hot aluminum surface and spray water were determined by using an inverse heat conduction method. Good agreements between numerically calculated temperatures with the inverse identified h and experimentally measurements demonstrate that the method is valid for solving the q and h of spray quenching process. The estimated heat flux consists of three main stages of transition boiling, nucleate boiling and single-phase cooling. The results show that both the heat flux and heat transfer coefficient increase with the increasing of spray pressure. When the surface temperature is lower than 170 ◦ C, the q, h and the maximum heat transfer coefficient (hmax ) decrease and then increase as surface roughness increases. However, when the surface temperature is higher than 170 ◦ C, the influence of surface is insignificant. This phenomenon may be attributed to the variation of nucleation site density with surface roughness. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Spray quenching is an effective heat treatment method to improve the material performance due to its ability of high heat removal rate. The final mechanical properties of metal workpiece highly depend on the heat transfer efficiency at the metalquenchant interface during spray quenching process. There are many factors that can influence the heat transfer efficiency, such as the properties of metal workpiece and quenchant, nozzle type, spray pressure, spray distance, spray angle, surface roughness. Therefore, quantitative characterization of the heat transfer coefficient in the simple mathematical model is still very difficult. The previous investigation were mainly focused on a few parameters that can be isolated and easily changed in practice, such as spray pressure and surface roughness. Xu and Mohamed (2006) developed an iterative and sequential inverse heat transfer analysis procedure to obtain the heat fluxes at the stagnation points of stationary hot steel plates cooled by impingement water jet and studied the effects of water flow rate on
∗ Corresponding author at: State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, 410082 Hunan, PR China. Tel.: +86 73188821950; fax: +86 73188821950. E-mail address: [email protected] (L. Li). http://dx.doi.org/10.1016/j.jmatprotec.2014.06.027 0924-0136/© 2014 Elsevier B.V. All rights reserved.
heat transfer at different water temperature levels. They concluded that the water flow rate generally has little effect on the heat fluxes of the stagnation points at all water temperature levers. Wang et al. (2012) investigated the heat transfer phenomena of stationary hot steel plate under multiple top circular jets and also found that the cooling water flow rate in the range of 15–35 L/min has no effect on h within 70 mm distance from stagnation line. Mozumder et al. (2006) studied the quenching of hot cylindrical blocks made of copper, brass and steel with initial block temperature 250–400 ◦ C by a subcooled water jet. They suggested that jet velocity was one of predominant parameters and the maximum heat flux (qmax ) increased gradually with increasing jet velocity. Mascarenhas and Mudawar (2010) examined the quenching a solid alloy cylinder using a fullcone plain-orifice spray nozzles with a cone angle of 45◦ and found that the increasing of nozzle pressure drop or the decreasing of orifice-to-surface distance could cause the transitions to the initiation of lower temperature boiling regimes to occur at higher surface temperatures and hasten the exit from the poor film boiling regime to the more efficient transition boiling regime. There are also several studies involving the effect of surface roughness on the interfacial heat transfer over the past two decades. Prabhu and Fernandes (2007) prepared three identical stainless steel specimens with three different surface roughness (Ra = 1.0, 3.0 m and a groove) at the bottom surface to study the effect of surface roughness on heat transfer rates in various
2878
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
foil, to ensure the water flow only contacts the bottom end of the specimen. All the experimental conditions employed in the present study are displayed in Table 1. The temperature of the water spray (Tw ) is 8 ◦ C. Spray pressure (P) is the water pressure at the outlet of the nozzle. The value of the spray pressure was calculated by using the measured total flow rate and nozzle diameter and varies from 0.6 to 3.1 kPa. The distance from the outlet of the nozzle to the specimen was about 80 mm. 2.2. Specimen preparation and data acquisition
Fig. 1. Schematic diagram of quenching experimental set-up.
quenchants during end quenching. The results indicated that the specimens were found to be very sensitive to surface roughness and the interfacial heat transfer increased with increasing the surface roughness during quenching in water and brine. Luke (1997) investigated the effect of the surface roughness (11.3, 1.00 and 0.16 m) on the heat transfer coefficient by experiments with propane boiling on copper and steel tubes and observed that increase of surface roughness could increase the nucleation sites which resulted in the increase of heat transfer. Park et al. (2004) found that the wettability in rough surface with a roughness of 6.97 m was more than that of smooth surface with a roughness of 0.39 m. The enhancement of wettability on the heated surface directly improved the heat transfer rate. Pais et al. (1992) studied the effects of surface roughness (values ranged 0.3–22.0 m) on heat transfer during spray cooling and found that the 0.3 m rough surface achieved higher heat fluxes over the temperature domain sampled. According to the above literature, research on the effects of spray pressure and surface roughness on the heat transfer is still not systematic, different researchers obtained inconsistent results under different experimental conditions. The present study is aimed to investigate the effects of spray pressure and surface roughness on the heat flux and heat transfer coefficient at the interface between a heated aluminum surface and spray cooling water, to simulate the quenching process of aluminum alloys. The q and h are determined by inverse heat conduction method. 2. Experimental apparatus and procedure 2.1. Experimental set-up The schematic diagram of quenching experimental set-up is shown in Fig. 1. It consists of water spray quenching system (left dash box in Fig. 1), specimen and data acquisition system (right dash box in Fig. 1). A tank containing quenchant was placed in the bottom of the set-up. Circulating cooling water was used as quenching medium. A pump located in the tank was used to pump the water through a nozzle of diameter 10 mm. A flowmeter and a control valve were mounted in the middle of the pump and the nozzle in order to adjust the spray water flow as needed. Quench specimen was placed directly above the nozzle. All specimens were instrumented with three K-type thermocouples. The side face of the specimen was wrapped with an asbestos layer tighten by aluminum
Cylindrical specimens were used to study the interaction between water and hot aluminum alloy specimens during quenching and particularly to determine q and h in different quenching conditions. The chemical composition of the cylindrical specimens is illustrated in Table 2. Aiming to measure the evolution of temperature, three K-type thermocouples labeled by T1, T2 and T3 with the diameter of 0.2 mm were centrally positioned in the lateral surface of each specimen. The distances between the surface and the three thermocouples are 3 mm, 10 mm and 20 mm, respectively. The dimensions of specimens and the location of thermocouples are shown in Fig. 2. Three specimens with different milled surface roughness (Ra = 0.6, 7.6, 40.2 m) at the quench end were prepared to study the effect of surface roughness on the interfacial heat transfer coefficients. The specimens were heated to 520 ◦ C and soaked for 3 hours in the electric resistance furnace, and then quickly transferred to the operating position within 10 s. NI USB 9213 data acquisition instrument (National Instruments, Inc., Austin, Texas, US) was used to collect temperature data at a rate of 100 Hz. This is a 16 channel, 24 bit data acquisition instrument with a sampling rate of 1200 samples/sec. It has built-in cold junction compensation. The uncertainty of the temperature measurement is less than 0.25 ◦ C for K-type thermocouple, while the uncertainty of the placement of the thermocouples is estimated to be ±0.2 mm. 3. Results and discussion 3.1. Inverse identification of the heat flux As shown in Fig. 2, the length of the specimen is far greater than its diameter. Thus the quenching process can be approximately considered as one-dimensional heat transfer in order to simplify the data processing. The measured temperature data at the location T1 in Fig. 2 is used as the known temperature history for estimating the interfacial heat flux and the metal surface temperature at the quenching interface using an inverse heat conduction method as described in Zhang and Li (2013). The one dimensional transient heat conduction equation is defined as follows: Cp
∂T =k ∂t
2
∂ T ∂x2
(1)
where , Cp and k are the density, specific heat and thermal conductivity of the specimen. T is the temperature of the specimen, t is the time. The surface heat flux density is estimated from the T1 temperature by minimizing the objective function: 2 1 i+R (Tn − Yni+R ) M M
F(qi ) =
(2)
n=1
where M is the number of the heat fluxes to be predicted, Tni+R and Yni+R are the calculated and measured temperature at time i + R, R is the number of future time steps to compensate for the heat difference and is a discrete time interval. The objective function is minimized through iterations. At each iteration step,
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2879
Table 1 Experimental conditions of the present study. Specimen material
Water temperature, Tw [◦ C]
Spray pressure, P [103 Pa]
Spray distance, d [mm]
Surface roughness, Ra [m]
6082
8
0.6, 1.0, 1.9, 3.1
80
0.6, 7.6, 40.2
Table 2 The chemical composition of 6082 T6 aluminum alloy.
6082
Si
Fe
Cu
Mn
Mg
Cr
Zn
Ti
Al
0.7–1.3
≤0.50
≤0.10
0.40–1.0
0.6–1.2
≤0.25
≤0.20
≤0.20
Bal.
the heat flux is increased by q. This procedure is continued until the ratio (q/q) becomes less than 0.001, and then an estimate of qi can be obtained. At this point, the time is increased by , qi is used as the initial heat flux guess for the next time step and the process is repeated for qi+1 , yielding the full history of q(t). Then, the interfacial heat transfer coefficient can be determined by the following equation: h=
q Ts − Tw
(3)
where Ts is the surface temperature of the specimen, Tw is the temperature of spray water. This procedure simultaneously yields the temperature of the specimen surface in contact with the quenchant. It is worth mentioning that specimens’ thermal properties such as density, specific heat and conductivity are expressed as a function of temperature. These functions used in the inverse method can improve the calculation accuracy. Generally, the boiling curve during bath quenching is usually divided into four phases: film boiling regime, transition boiling regime, nucleate boiling regime and single-phase regime as shown in Fig. 3. Mascarenhas and Mudawar (2010) indicated that a large number of bubbles are produced in the quenching surface and gather into a vapor film in the film boiling regime. It should be note that the thin vapor film could form beneath individual droplets upon impact with the surface in the case of spray quenching. The minimum heat flux point (the Leidenfrost point) marks the temperature at which the insulating vapor layer begins to break up. The critical heat flux (CHF) point marks the temperature at which the entire surface becomes available to liquid wetting. In the ensuing nucleate boiling regime, vigorous boiling and full liquid contact result in very high heat removal rates. Eventually, the boiling completely subsides and enters the final single-phase cooling regime. When the surface roughness is 0.6 m and the spray pressure is 3.1 kPa, the variation of calculated interfacial heat flux at 6082 aluminum alloy surface with surface temperature during water spray quenching are shown in Fig. 4(a). It can be seen that the heat flux at the early stage rapidly increases with decreasing surface temperature and reaches the maximum heat flux of 4.4 MW/m2 at surface
temperature of 303 ◦ C. When the surface temperature is less than 303 ◦ C, the heat flux shows the characteristic of nucleate boiling regime, in which heat flux rapidly decreases with decreasing surface temperature. When the surface temperature is less than 110 ◦ C, the boiling completely subsides and enters the single-phase cooling regime. It also can be found from Fig. 4(a) that the film boiling regime and the minimum heat flux (Leidenfrost point) are not observed. For the experiments in this paper, the relative lower initial surface temperature and the relative larger subcooling (i.e. relative lower water temperature) is one of the main reasons for the absence of film boiling. At the beginning of spray, the surface within the spray range is immediately wetted by water and then the surface temperature drops abruptly due to the relative greater subcooling. The surface temperature drop (Fig. 4(b)), reduces the capacity of heating the water to the vaporization temperature, and results in the formation of bubbles in limited local areas. Hence, the fresh water can touch partially with the high temperature surface, leading to the transition boiling status. Xu and Mohamed (2006) also found that there is no film boiling regime at the beginning when the water temperature is lower than 60 ◦ C. Liu and Wang (2001) also indicated that no film boiling occurs on stagnation point when the water with a temperature of 20 ◦ C hits the test plate whose temperature is lower than 920 ◦ C. The effect of gravity on heat transfer caused by the upward spray is another main reason for the absence of film boiling. As pointed out by Hsieh and Yao (2006), the gravity can lower the ability of the liquid film attachment to the heated surface, which will shorten the contact time between water and surface. The shorter contact time will hinder the formation and the growth of bubbles or accelerate the bursting of bubbles which results in the absence of film boiling. 3.2. Verification identified the interfacial heat flux In order to further confirm the validity of the inverse method for calculating the q and h in the present experiments, the estimated q values are used to calculate the temperature evolutions at the location T2 and T3 with the same boundary condition. The
Fig. 2. Schematic diagram of the specimen dimensions and thermocouples layout (unit: mm).
2880
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
Fig. 3. (a) Boiling curve and, (b) corresponding surface temperature curve (Mascarenhas and Mudawar, 2010).
measured and calculated cooling curves for surface roughness of 0.6 m and spray pressure of 3.1 kPa are displayed in Fig. 5. The curves labeled T1, T2 and T3 were measured at different depths from the interface (3, 10 and 20 mm). The curves labeled C1, C2 and C3 were obtained by inverse heat conduction program at different distances from the surface (0, 10 and 20 mm). There are two stages in the cooling curves: rapid decline stage and slow decline stage. The cooling curves in the rapid decline stage (0–20 s) show that the surface temperature drops rapidly, while the internal temperature decreases relatively slowly. The surface temperature drops from 510 to 100 ◦ C within 8 s after the water spray first strikes. It takes about 13 s, 36 s and 72 s for the internal temperature T1, T2 and T3. This indicates that the axial temperature distribution of the specimen in the initial stage is highly nonlinear. After the initial stage, all cooling curves drop slowly. It can be obviously seen from Fig. 5 that the calculated temperatures C2 and C3 are in good agreement with the measured T2 and T3. It is demonstrated that the inverse method is very feasible and effective to determine the interfacial heat flux and the heat transfer coefficient during quenching of aluminum alloy in water. 3.3. Effect of spray pressure on q and h Estimated heat flux vs. surface temperature plots during water spray quenching of 6082 aluminum alloy with four different spray pressures are illustrated in Fig. 6. The surface roughness of specimens is 0.6 m. It can be seen from Fig. 6 that the heat flux increases with increasing spray pressure. The difference of maximum heat
flux when the spray pressure varies from 0.6 to 1.0 kPa is larger than that when the spray pressure varies from 1.0 to 3.1 kPa. The main reason is due to the surface tension of the cooling medium and the pressure of entrapped air in the micro cavities of specimen surface. With increasing spray pressure, the liquid is able to fill into microcavities deeply, and the liquid–solid contact heat transfer area will increase. At the same time, the liquid surface tension and the pressure of the air to be captured in the cavities, hindering the liquid filling into the microcavities, also increase with increasing spray pressure. When the spray pressure is low, each unit increment of spray pressure will lead to the increasing of the heat transfer area and heat flux due to the smaller liquid surface tension and entrapped air pressure. However, when the spray pressure is high, each unit increment of spray pressure produces less increase in the heat transfer area and heat flux due to the increasing liquid surface tension and entrapped air pressure. The result suggests that the effect of spray pressure on the maximum heat flux decreases as the spray pressure increases. The effect of spray pressure on the heat transfer coefficient during water spray quenching with a specimen roughness of 0.6 m is shown in Fig. 7. The four h curves have the similar trend. The values of h increase almost linearly with decreasing surface temperature at 500–180 ◦ C. In the surface temperature ranging from 180 to 135 ◦ C, all of interfacial heat transfer coefficients reach the maximum values (hmax ) due to both vigorous boiling and full liquid contact. After the maximum, h decreases rapidly with decreasing surface temperature. It can be seen clearly from Fig. 7 that the hmax increases with increasing spray pressure and the maximum hmax was obtained
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2881
Heat flux, q / MW/m
2
4.5
3.0
P=0.6kPa P=1.0kPa P=1.9kPa P=3.1kPa
1.5
0.0 100
200
300
Surface temperature, Ts
400
500
o
C
Fig. 6. The effect of spray pressure on the heat flux.
Fig. 4. (a) variation of calculated q at 6082 aluminum alloy surface with Ts during water quenching and (b) variation of calculated Ts at 6082 aluminum alloy surface with time during water quenching.
500 400
o
T/ C
T2 C2
300
Fig. 7. The effect of spray pressure on the heat transfer coefficient.
T1 Measured 3mm T2 Measured 10mm T3 Measured 20mm C1 Calculated 0mm C2 Calculated 10mm C3 Calculated 20mm
T3 C3
convection. The greater single-phase convection heat transfer is able to reduce the decline of q. As a result, the hmax value with higher spray pressure appears at lower surface temperature. 3.4. Effect of surface roughness on q and h
200 100 C1 0 0
T1 20
40
t/s
60
80
100
Fig. 5. The cooling curves for surface roughness of 0.6 m and spray pressure of 3.1 kPa.
at spray pressure of 3.1 kPa and surface temperature 135 ◦ C. Moreover, the hmax value with higher spray pressure appears at lower surface temperature. Higher spray pressure corresponds to greater total water flow volume. The increase of total water flow volume will not only increase the heat exchange of bubble evaporation, but also increase the heat exchange due to single-phase
The heat fluxes and heat transfer coefficients at 1.9 kPa with different surface roughnesses are shown in Figs. 8 and 9. It can be found that the surface roughness has a significant influence on q, h and hmax in the nucleate boiling regime (300–100 ◦ C), but its influence is insignificant in the transition boiling regime (500–300 ◦ C). When the surface temperature is less than 170 ◦ C, q, h and hmax first decrease and then increase as the surface roughness increases. Bernardin and Mudawar (1996) also found that not all boiling regimes are equally impacted by surface roughness and suggested that this phenomenon is mainly related with the ability of liquid access to the hot surface. During the nucleate boiling regime, direct access of the liquid to the surface renders this boiling regime most sensitive to surface roughness. However, liquid access is limited during the transient boiling regime due to an intermittent vapor blanket between the liquid and the surface. The effect of surface roughness is negligible because the thickness of the intermittent
2882
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
Fig. 8. The effect of surface roughness on the heat flux at pressure of 1.9 kPa.
and indicated that the secondary nucleation was a main heat transfer mechanism when the surface finish is smooth. The results of Pais et al. (1992) implied that departure frequency of bubbles increases with the decrease of the surface roughness. The greater departure frequency could produce more bubbles per unit time and increase nucleation site density. Thus, the decrease of surface roughness may result in the enhancement of heat transfer due to the increase of the nucleation site density. As can be seen from the above analysis, most researchers agreed that the heat transfer increased with increasing surface roughness. Only a few researchers (Pais et al., 1992) held the opposite view. It can be found that the common essence of the heat transfer enhancement resulted from roughness increase or decrease is the increase of the nucleation number per unit nominal area and unit time (i.e. the increase of nucleation site density). The real contact area increases as surface roughness increases and leads to the increase of nucleation site density. Reducing roughness increases the cavity number and secondary nucleation which result in the increase of nucleation site density. Therefore, it can be inferred that the influence of surface roughness on the nucleation site density during the nucleation boiling regime could reflect the essence of the influence of surface roughness on the heat transfer. In order to better understand the effect of surface roughness on the heat transfer, the relationship between surface roughness and the heat transfer should be described quantitatively. Abishek et al. (2013) indicated that the total heat transfer rate due to boiling of spray subcooled liquid is partitioned into three components including liquid phase convection (qC ), quenching (transient convection, qQ ) and evaporation (qE ). Because of the pretty high saturation temperature difference, evaporation heat flux (qE ) is dominant in the total heat transfer rate in the present study. Evaporation heat flux is defined as (Abishek et al., 2013) qE =
Fig. 9. The effect of surface roughness on the heat transfer coefficients at a spray pressure is 1.9 kPa.
vapor blanket during transient boiling regime is greater than any surface roughness. Therefore, the transient boiling regime is not sensitive to the microsurface geometry. In Figs. 8 and 9, q, h and hmax first decrease and then increase as the surface roughness increases when the surface temperature is less than 170 ◦ C. Conventional wisdom would lead one to expect lager heat transfer area and hence higher heat transfer coefficient with increasing surface roughness – that is, rougher surface give rise to higher heat transfer coefficient. For example, Prabhu and Fernandes (2007) indicated that a maximum peak heat flux was obtained with rough surface for quenching in aqueous quenchants. They suggested that the increase of surface roughness leads to the increase of real surface area and aqueous quenchants due to lower viscosity is able to penetrate easily into the cavities on the rough surface resulting in larger heat transfer contact area. Heat transfer significantly increases with increasing surface roughness, attributing to the increase of heat transfer contact area. Conversely, Pais et al. (1992) studied the effects of surface roughness (22, 14 and 0.3 m) on heat transfer when using spray cooling and found that the 0.3 m rough surface achieved higher heat fluxes over the temperature domain sampled. They attributed the heat transfer enhancement in smoother surface to the increase of the nucleation site density of bubble during the nucleate boiling regime
3 d fNw v L 6 bw
(4)
where dbw , f, Nw , v and L are bubble departure diameter, bubble departure frequency, nucleation site density, density of vapor and latent heat of vaporization. The heat flux obviously increases monotonically with increasing nucleation site density when other parameters remain constant. Therefore, Eq. (4) confirms that the relationship between surface roughness and nucleation site density could reflect the relationship between the heat transfer and surface roughness. The nucleation site density can be determined by the correlation of Benjamin and Balakrishnan (1997) N = 218.8(Pr )1.63 A
1
−0.4 (T )
3
(5)
The N/A of Eq. (5) is equal to Nw of Eq. (4). Where Pr is the Prandlt number and is defined as Pr =
Cp k
(6)
where Cp , and k are the specific heat, dynamic viscosity and thermal conductivity of the liquid. The surface–liquid interaction parameter is defined by
=
kw w Cpw kl l Cpl
0.5
(7)
is given by Benjamin and Balakrishnan (1997) = 14.5 − 4.5
R P a
R P 2
+ 0.4
a
(8)
where Ra , P and are arithmetic average roughness, external pressure and surface tension. By substituting the experimental parameters into Eqs. (5)–(8), the variation of nucleation site density with the surface roughness can be determined as illustrated
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2883
cooling (below 100 ◦ C). The film boiling and the minimum heat flux (Leidenfrost point) are not detected in all heat flux curves. (3) Higher spray pressure tend to produce higher interfacial heat flux and heat transfer coefficient. The qmax and hmax increase with increasing spray pressure. (4) In the transition boiling regime (Ts > 300 ◦ C), surface roughness has little effect on the q and h. However, its influence is obvious in the nucleate boiling regime (100 ◦ C < Ts < 300 ◦ C). When the Ts is less than 170 ◦ C, the q, h and hmax decrease and then increase as the surface roughness increases. Acknowledgements
Fig. 10. The effect of surface roughness on nucleation density and hmax .
in Fig. 10. The square points represent the maximum heat transfer coefficients with different surface roughnesses calculated by inverse method. It can be seen that the curve first decreases and then increases as the surface roughness increases, consisting with the trend of maximum heat transfer coefficient. Why are there minimum values? Benjamin and Balakrishnan (1997) suggested that increasing surface roughness decreases the available superheat before the minimum, leading to lower nucleation site density. Beyond the minimum, one or more individual bubbles are probably nucleated in the same cavity, so nucleation site density starts increasing again. It should be noted that there is a difference between the two minimum values. The difference may be attributing to the calculation and measurement error. In addition, the hmax of the roughest surface (40.2 m) appears in the single-phase cooling regime instead of the nucleate boiling regime, suggesting that the increase of surface roughness leads to the increase of heat transfer in both nucleate boiling regime and single-phase cooling regime. 4. Conclusions (1) The numerically calculated temperatures with the inverse identified heat transfer coefficient are in good agreement with those measured experimentally, indicating that the inverse method was a feasible and effective tool for determination of the interfacial heat flux and heat transfer coefficient during spray quenching of aluminum alloy in water. (2) In the water spray quenching of aluminum alloy, heat transfer consists of three main stages including transition boiling (500–300 ◦ C), nucleate boiling (300–100 ◦ C) and single-phase
The authors gratefully acknowledge research support from the National Key Project of Science and Technology “High-grade CNC machine tools and basic manufacturing equipment” (No: 2014ZX04002071) and the Changsha science and technology project (No. K1112067-11). References Abishek, S., Narayanaswamy, R., Narayanan, V., 2013. Effect of heater size and Reynolds number on the partitioning of surface heat flux in subcooled jet impingement boiling. Int. J. Heat Mass Transfer 59, 247–261. Benjamin, R.J., Balakrishnan, A.R., 1997. Nucleation site density in pool boiling of saturated pure liquids: effect of surface microroughness and surface and liquid physical properties. Exp. Therm. Fluid Sci. 15, 32–42. Bernardin, J.D., Mudawar, I., 1996. Experimental and statistical investigation of changes in surface roughness associated with spray quenching. Int. J. Heat Mass Transfer 39 (10), 2023–2037. Hsieh, C.C., Yao, S.C., 2006. Evaporative heat transfer characteristics of a water spray on micro-structured silicon surfaces. Int. J. Heat Mass Transfer 49, 962–974. Liu, Z.H., Wang, J., 2001. Study on film boiling heat transfer for water jet impinging on high temperature flat plate. Int. J. Heat Mass Transfer 44, 2475–2481. Luke, A., 1997. Pool boiling heat transfer from horizontal tubes with different surface roughness. Int. J. Refrig. 20 (8), 561–574. Mascarenhas, N., Mudawar, I., 2010. Analytical and computational methodology for modeling spray quenching of solid alloy cylinders. Int. J. Heat Mass Transfer 53, 5871–5883. Mozumder, A.K., Monde, M., Woodfield, P.L., Islam, M.A., 2006. Maximum heat flux in relation to quenching of a high temperature surface with liquid jet impingement. Int. J. Heat Mass Transfer 49, 2877–2888. Pais, M., Chow, L., Mahefkey, E., 1992. Surface roughness and its effects on the heat transfer mechanism of spray cooling. J. Heat Transf. Trans. ASME 114 (1), 211–219. Park, C.W., Cho, H.C., Kang, Y.T., 2004. The effect of heat transfer additive and surface roughness of micro-scale hatched tubes on absorption performance. Int. J. Refrig. 270, 264–270. Prabhu, K.N., Fernandes, P., 2007. Effect of surface roughness on metal/quenchant interfacial heat transfer and evolution of microstructure. Mater. Des. 28, 544–550. Wang, H.M., Yu, W., Cai, Q.W., 2012. Experimental study of heat transfer coefficient on hot steel plate during water jet impingement cooling. J. Mater. Process. Technol. 212, 1825–1831. Xu, F.C., Mohamed, S.G., 2006. Heat transfer behavior in the impingement zone under circular water jet. Int. J. Heat Mass Transfer 49, 3785–3799. Zhang, L.Q., Li, L.X., 2013. Determination of heat transfer coefficients at metal/chill interface in the casting solidification process. Heat Mass Transfer 49, 1071–1080.
Contents lists available at ScienceDirect
Journal of Materials Processing Technology journal homepage: www.elsevier.com/locate/jmatprotec
Influence of pressure and surface roughness on the heat transfer efficiency during water spray quenching of 6082 aluminum alloy Rong Xu a,b , Luoxing Li a,b,∗ , Liqiang Zhang a,c , Biwu Zhu a,d , Xiao Liu a,d , Xiaobing Bu a,b a
State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, 410082 Hunan, PR China College of Mechanical and Vehicle Engineering, Hunan University, Changsha, 410082 Hunan, PR China College of Mechanical and Electrical Engineering, Central South University of Forestry and Technology, Changsha, 410004 Hunan, PR China d College of Electromechanical Engineering, Hunan University Science and Technology, Xiangtan, 411002 Hunan, PR China b c
a r t i c l e
i n f o
Article history: Received 14 March 2014 Received in revised form 28 June 2014 Accepted 30 June 2014 Available online 8 July 2014 Keywords: Interfacial heat transfer Spray quenching Spray pressure Surface roughness Aluminum alloy Nucleation site density
a b s t r a c t The heat flux (q) and heat transfer coefficient (h) at the interface between hot aluminum surface and spray water were determined by using an inverse heat conduction method. Good agreements between numerically calculated temperatures with the inverse identified h and experimentally measurements demonstrate that the method is valid for solving the q and h of spray quenching process. The estimated heat flux consists of three main stages of transition boiling, nucleate boiling and single-phase cooling. The results show that both the heat flux and heat transfer coefficient increase with the increasing of spray pressure. When the surface temperature is lower than 170 ◦ C, the q, h and the maximum heat transfer coefficient (hmax ) decrease and then increase as surface roughness increases. However, when the surface temperature is higher than 170 ◦ C, the influence of surface is insignificant. This phenomenon may be attributed to the variation of nucleation site density with surface roughness. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Spray quenching is an effective heat treatment method to improve the material performance due to its ability of high heat removal rate. The final mechanical properties of metal workpiece highly depend on the heat transfer efficiency at the metalquenchant interface during spray quenching process. There are many factors that can influence the heat transfer efficiency, such as the properties of metal workpiece and quenchant, nozzle type, spray pressure, spray distance, spray angle, surface roughness. Therefore, quantitative characterization of the heat transfer coefficient in the simple mathematical model is still very difficult. The previous investigation were mainly focused on a few parameters that can be isolated and easily changed in practice, such as spray pressure and surface roughness. Xu and Mohamed (2006) developed an iterative and sequential inverse heat transfer analysis procedure to obtain the heat fluxes at the stagnation points of stationary hot steel plates cooled by impingement water jet and studied the effects of water flow rate on
∗ Corresponding author at: State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, 410082 Hunan, PR China. Tel.: +86 73188821950; fax: +86 73188821950. E-mail address: [email protected] (L. Li). http://dx.doi.org/10.1016/j.jmatprotec.2014.06.027 0924-0136/© 2014 Elsevier B.V. All rights reserved.
heat transfer at different water temperature levels. They concluded that the water flow rate generally has little effect on the heat fluxes of the stagnation points at all water temperature levers. Wang et al. (2012) investigated the heat transfer phenomena of stationary hot steel plate under multiple top circular jets and also found that the cooling water flow rate in the range of 15–35 L/min has no effect on h within 70 mm distance from stagnation line. Mozumder et al. (2006) studied the quenching of hot cylindrical blocks made of copper, brass and steel with initial block temperature 250–400 ◦ C by a subcooled water jet. They suggested that jet velocity was one of predominant parameters and the maximum heat flux (qmax ) increased gradually with increasing jet velocity. Mascarenhas and Mudawar (2010) examined the quenching a solid alloy cylinder using a fullcone plain-orifice spray nozzles with a cone angle of 45◦ and found that the increasing of nozzle pressure drop or the decreasing of orifice-to-surface distance could cause the transitions to the initiation of lower temperature boiling regimes to occur at higher surface temperatures and hasten the exit from the poor film boiling regime to the more efficient transition boiling regime. There are also several studies involving the effect of surface roughness on the interfacial heat transfer over the past two decades. Prabhu and Fernandes (2007) prepared three identical stainless steel specimens with three different surface roughness (Ra = 1.0, 3.0 m and a groove) at the bottom surface to study the effect of surface roughness on heat transfer rates in various
2878
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
foil, to ensure the water flow only contacts the bottom end of the specimen. All the experimental conditions employed in the present study are displayed in Table 1. The temperature of the water spray (Tw ) is 8 ◦ C. Spray pressure (P) is the water pressure at the outlet of the nozzle. The value of the spray pressure was calculated by using the measured total flow rate and nozzle diameter and varies from 0.6 to 3.1 kPa. The distance from the outlet of the nozzle to the specimen was about 80 mm. 2.2. Specimen preparation and data acquisition
Fig. 1. Schematic diagram of quenching experimental set-up.
quenchants during end quenching. The results indicated that the specimens were found to be very sensitive to surface roughness and the interfacial heat transfer increased with increasing the surface roughness during quenching in water and brine. Luke (1997) investigated the effect of the surface roughness (11.3, 1.00 and 0.16 m) on the heat transfer coefficient by experiments with propane boiling on copper and steel tubes and observed that increase of surface roughness could increase the nucleation sites which resulted in the increase of heat transfer. Park et al. (2004) found that the wettability in rough surface with a roughness of 6.97 m was more than that of smooth surface with a roughness of 0.39 m. The enhancement of wettability on the heated surface directly improved the heat transfer rate. Pais et al. (1992) studied the effects of surface roughness (values ranged 0.3–22.0 m) on heat transfer during spray cooling and found that the 0.3 m rough surface achieved higher heat fluxes over the temperature domain sampled. According to the above literature, research on the effects of spray pressure and surface roughness on the heat transfer is still not systematic, different researchers obtained inconsistent results under different experimental conditions. The present study is aimed to investigate the effects of spray pressure and surface roughness on the heat flux and heat transfer coefficient at the interface between a heated aluminum surface and spray cooling water, to simulate the quenching process of aluminum alloys. The q and h are determined by inverse heat conduction method. 2. Experimental apparatus and procedure 2.1. Experimental set-up The schematic diagram of quenching experimental set-up is shown in Fig. 1. It consists of water spray quenching system (left dash box in Fig. 1), specimen and data acquisition system (right dash box in Fig. 1). A tank containing quenchant was placed in the bottom of the set-up. Circulating cooling water was used as quenching medium. A pump located in the tank was used to pump the water through a nozzle of diameter 10 mm. A flowmeter and a control valve were mounted in the middle of the pump and the nozzle in order to adjust the spray water flow as needed. Quench specimen was placed directly above the nozzle. All specimens were instrumented with three K-type thermocouples. The side face of the specimen was wrapped with an asbestos layer tighten by aluminum
Cylindrical specimens were used to study the interaction between water and hot aluminum alloy specimens during quenching and particularly to determine q and h in different quenching conditions. The chemical composition of the cylindrical specimens is illustrated in Table 2. Aiming to measure the evolution of temperature, three K-type thermocouples labeled by T1, T2 and T3 with the diameter of 0.2 mm were centrally positioned in the lateral surface of each specimen. The distances between the surface and the three thermocouples are 3 mm, 10 mm and 20 mm, respectively. The dimensions of specimens and the location of thermocouples are shown in Fig. 2. Three specimens with different milled surface roughness (Ra = 0.6, 7.6, 40.2 m) at the quench end were prepared to study the effect of surface roughness on the interfacial heat transfer coefficients. The specimens were heated to 520 ◦ C and soaked for 3 hours in the electric resistance furnace, and then quickly transferred to the operating position within 10 s. NI USB 9213 data acquisition instrument (National Instruments, Inc., Austin, Texas, US) was used to collect temperature data at a rate of 100 Hz. This is a 16 channel, 24 bit data acquisition instrument with a sampling rate of 1200 samples/sec. It has built-in cold junction compensation. The uncertainty of the temperature measurement is less than 0.25 ◦ C for K-type thermocouple, while the uncertainty of the placement of the thermocouples is estimated to be ±0.2 mm. 3. Results and discussion 3.1. Inverse identification of the heat flux As shown in Fig. 2, the length of the specimen is far greater than its diameter. Thus the quenching process can be approximately considered as one-dimensional heat transfer in order to simplify the data processing. The measured temperature data at the location T1 in Fig. 2 is used as the known temperature history for estimating the interfacial heat flux and the metal surface temperature at the quenching interface using an inverse heat conduction method as described in Zhang and Li (2013). The one dimensional transient heat conduction equation is defined as follows: Cp
∂T =k ∂t
2
∂ T ∂x2
(1)
where , Cp and k are the density, specific heat and thermal conductivity of the specimen. T is the temperature of the specimen, t is the time. The surface heat flux density is estimated from the T1 temperature by minimizing the objective function: 2 1 i+R (Tn − Yni+R ) M M
F(qi ) =
(2)
n=1
where M is the number of the heat fluxes to be predicted, Tni+R and Yni+R are the calculated and measured temperature at time i + R, R is the number of future time steps to compensate for the heat difference and is a discrete time interval. The objective function is minimized through iterations. At each iteration step,
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2879
Table 1 Experimental conditions of the present study. Specimen material
Water temperature, Tw [◦ C]
Spray pressure, P [103 Pa]
Spray distance, d [mm]
Surface roughness, Ra [m]
6082
8
0.6, 1.0, 1.9, 3.1
80
0.6, 7.6, 40.2
Table 2 The chemical composition of 6082 T6 aluminum alloy.
6082
Si
Fe
Cu
Mn
Mg
Cr
Zn
Ti
Al
0.7–1.3
≤0.50
≤0.10
0.40–1.0
0.6–1.2
≤0.25
≤0.20
≤0.20
Bal.
the heat flux is increased by q. This procedure is continued until the ratio (q/q) becomes less than 0.001, and then an estimate of qi can be obtained. At this point, the time is increased by , qi is used as the initial heat flux guess for the next time step and the process is repeated for qi+1 , yielding the full history of q(t). Then, the interfacial heat transfer coefficient can be determined by the following equation: h=
q Ts − Tw
(3)
where Ts is the surface temperature of the specimen, Tw is the temperature of spray water. This procedure simultaneously yields the temperature of the specimen surface in contact with the quenchant. It is worth mentioning that specimens’ thermal properties such as density, specific heat and conductivity are expressed as a function of temperature. These functions used in the inverse method can improve the calculation accuracy. Generally, the boiling curve during bath quenching is usually divided into four phases: film boiling regime, transition boiling regime, nucleate boiling regime and single-phase regime as shown in Fig. 3. Mascarenhas and Mudawar (2010) indicated that a large number of bubbles are produced in the quenching surface and gather into a vapor film in the film boiling regime. It should be note that the thin vapor film could form beneath individual droplets upon impact with the surface in the case of spray quenching. The minimum heat flux point (the Leidenfrost point) marks the temperature at which the insulating vapor layer begins to break up. The critical heat flux (CHF) point marks the temperature at which the entire surface becomes available to liquid wetting. In the ensuing nucleate boiling regime, vigorous boiling and full liquid contact result in very high heat removal rates. Eventually, the boiling completely subsides and enters the final single-phase cooling regime. When the surface roughness is 0.6 m and the spray pressure is 3.1 kPa, the variation of calculated interfacial heat flux at 6082 aluminum alloy surface with surface temperature during water spray quenching are shown in Fig. 4(a). It can be seen that the heat flux at the early stage rapidly increases with decreasing surface temperature and reaches the maximum heat flux of 4.4 MW/m2 at surface
temperature of 303 ◦ C. When the surface temperature is less than 303 ◦ C, the heat flux shows the characteristic of nucleate boiling regime, in which heat flux rapidly decreases with decreasing surface temperature. When the surface temperature is less than 110 ◦ C, the boiling completely subsides and enters the single-phase cooling regime. It also can be found from Fig. 4(a) that the film boiling regime and the minimum heat flux (Leidenfrost point) are not observed. For the experiments in this paper, the relative lower initial surface temperature and the relative larger subcooling (i.e. relative lower water temperature) is one of the main reasons for the absence of film boiling. At the beginning of spray, the surface within the spray range is immediately wetted by water and then the surface temperature drops abruptly due to the relative greater subcooling. The surface temperature drop (Fig. 4(b)), reduces the capacity of heating the water to the vaporization temperature, and results in the formation of bubbles in limited local areas. Hence, the fresh water can touch partially with the high temperature surface, leading to the transition boiling status. Xu and Mohamed (2006) also found that there is no film boiling regime at the beginning when the water temperature is lower than 60 ◦ C. Liu and Wang (2001) also indicated that no film boiling occurs on stagnation point when the water with a temperature of 20 ◦ C hits the test plate whose temperature is lower than 920 ◦ C. The effect of gravity on heat transfer caused by the upward spray is another main reason for the absence of film boiling. As pointed out by Hsieh and Yao (2006), the gravity can lower the ability of the liquid film attachment to the heated surface, which will shorten the contact time between water and surface. The shorter contact time will hinder the formation and the growth of bubbles or accelerate the bursting of bubbles which results in the absence of film boiling. 3.2. Verification identified the interfacial heat flux In order to further confirm the validity of the inverse method for calculating the q and h in the present experiments, the estimated q values are used to calculate the temperature evolutions at the location T2 and T3 with the same boundary condition. The
Fig. 2. Schematic diagram of the specimen dimensions and thermocouples layout (unit: mm).
2880
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
Fig. 3. (a) Boiling curve and, (b) corresponding surface temperature curve (Mascarenhas and Mudawar, 2010).
measured and calculated cooling curves for surface roughness of 0.6 m and spray pressure of 3.1 kPa are displayed in Fig. 5. The curves labeled T1, T2 and T3 were measured at different depths from the interface (3, 10 and 20 mm). The curves labeled C1, C2 and C3 were obtained by inverse heat conduction program at different distances from the surface (0, 10 and 20 mm). There are two stages in the cooling curves: rapid decline stage and slow decline stage. The cooling curves in the rapid decline stage (0–20 s) show that the surface temperature drops rapidly, while the internal temperature decreases relatively slowly. The surface temperature drops from 510 to 100 ◦ C within 8 s after the water spray first strikes. It takes about 13 s, 36 s and 72 s for the internal temperature T1, T2 and T3. This indicates that the axial temperature distribution of the specimen in the initial stage is highly nonlinear. After the initial stage, all cooling curves drop slowly. It can be obviously seen from Fig. 5 that the calculated temperatures C2 and C3 are in good agreement with the measured T2 and T3. It is demonstrated that the inverse method is very feasible and effective to determine the interfacial heat flux and the heat transfer coefficient during quenching of aluminum alloy in water. 3.3. Effect of spray pressure on q and h Estimated heat flux vs. surface temperature plots during water spray quenching of 6082 aluminum alloy with four different spray pressures are illustrated in Fig. 6. The surface roughness of specimens is 0.6 m. It can be seen from Fig. 6 that the heat flux increases with increasing spray pressure. The difference of maximum heat
flux when the spray pressure varies from 0.6 to 1.0 kPa is larger than that when the spray pressure varies from 1.0 to 3.1 kPa. The main reason is due to the surface tension of the cooling medium and the pressure of entrapped air in the micro cavities of specimen surface. With increasing spray pressure, the liquid is able to fill into microcavities deeply, and the liquid–solid contact heat transfer area will increase. At the same time, the liquid surface tension and the pressure of the air to be captured in the cavities, hindering the liquid filling into the microcavities, also increase with increasing spray pressure. When the spray pressure is low, each unit increment of spray pressure will lead to the increasing of the heat transfer area and heat flux due to the smaller liquid surface tension and entrapped air pressure. However, when the spray pressure is high, each unit increment of spray pressure produces less increase in the heat transfer area and heat flux due to the increasing liquid surface tension and entrapped air pressure. The result suggests that the effect of spray pressure on the maximum heat flux decreases as the spray pressure increases. The effect of spray pressure on the heat transfer coefficient during water spray quenching with a specimen roughness of 0.6 m is shown in Fig. 7. The four h curves have the similar trend. The values of h increase almost linearly with decreasing surface temperature at 500–180 ◦ C. In the surface temperature ranging from 180 to 135 ◦ C, all of interfacial heat transfer coefficients reach the maximum values (hmax ) due to both vigorous boiling and full liquid contact. After the maximum, h decreases rapidly with decreasing surface temperature. It can be seen clearly from Fig. 7 that the hmax increases with increasing spray pressure and the maximum hmax was obtained
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2881
Heat flux, q / MW/m
2
4.5
3.0
P=0.6kPa P=1.0kPa P=1.9kPa P=3.1kPa
1.5
0.0 100
200
300
Surface temperature, Ts
400
500
o
C
Fig. 6. The effect of spray pressure on the heat flux.
Fig. 4. (a) variation of calculated q at 6082 aluminum alloy surface with Ts during water quenching and (b) variation of calculated Ts at 6082 aluminum alloy surface with time during water quenching.
500 400
o
T/ C
T2 C2
300
Fig. 7. The effect of spray pressure on the heat transfer coefficient.
T1 Measured 3mm T2 Measured 10mm T3 Measured 20mm C1 Calculated 0mm C2 Calculated 10mm C3 Calculated 20mm
T3 C3
convection. The greater single-phase convection heat transfer is able to reduce the decline of q. As a result, the hmax value with higher spray pressure appears at lower surface temperature. 3.4. Effect of surface roughness on q and h
200 100 C1 0 0
T1 20
40
t/s
60
80
100
Fig. 5. The cooling curves for surface roughness of 0.6 m and spray pressure of 3.1 kPa.
at spray pressure of 3.1 kPa and surface temperature 135 ◦ C. Moreover, the hmax value with higher spray pressure appears at lower surface temperature. Higher spray pressure corresponds to greater total water flow volume. The increase of total water flow volume will not only increase the heat exchange of bubble evaporation, but also increase the heat exchange due to single-phase
The heat fluxes and heat transfer coefficients at 1.9 kPa with different surface roughnesses are shown in Figs. 8 and 9. It can be found that the surface roughness has a significant influence on q, h and hmax in the nucleate boiling regime (300–100 ◦ C), but its influence is insignificant in the transition boiling regime (500–300 ◦ C). When the surface temperature is less than 170 ◦ C, q, h and hmax first decrease and then increase as the surface roughness increases. Bernardin and Mudawar (1996) also found that not all boiling regimes are equally impacted by surface roughness and suggested that this phenomenon is mainly related with the ability of liquid access to the hot surface. During the nucleate boiling regime, direct access of the liquid to the surface renders this boiling regime most sensitive to surface roughness. However, liquid access is limited during the transient boiling regime due to an intermittent vapor blanket between the liquid and the surface. The effect of surface roughness is negligible because the thickness of the intermittent
2882
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
Fig. 8. The effect of surface roughness on the heat flux at pressure of 1.9 kPa.
and indicated that the secondary nucleation was a main heat transfer mechanism when the surface finish is smooth. The results of Pais et al. (1992) implied that departure frequency of bubbles increases with the decrease of the surface roughness. The greater departure frequency could produce more bubbles per unit time and increase nucleation site density. Thus, the decrease of surface roughness may result in the enhancement of heat transfer due to the increase of the nucleation site density. As can be seen from the above analysis, most researchers agreed that the heat transfer increased with increasing surface roughness. Only a few researchers (Pais et al., 1992) held the opposite view. It can be found that the common essence of the heat transfer enhancement resulted from roughness increase or decrease is the increase of the nucleation number per unit nominal area and unit time (i.e. the increase of nucleation site density). The real contact area increases as surface roughness increases and leads to the increase of nucleation site density. Reducing roughness increases the cavity number and secondary nucleation which result in the increase of nucleation site density. Therefore, it can be inferred that the influence of surface roughness on the nucleation site density during the nucleation boiling regime could reflect the essence of the influence of surface roughness on the heat transfer. In order to better understand the effect of surface roughness on the heat transfer, the relationship between surface roughness and the heat transfer should be described quantitatively. Abishek et al. (2013) indicated that the total heat transfer rate due to boiling of spray subcooled liquid is partitioned into three components including liquid phase convection (qC ), quenching (transient convection, qQ ) and evaporation (qE ). Because of the pretty high saturation temperature difference, evaporation heat flux (qE ) is dominant in the total heat transfer rate in the present study. Evaporation heat flux is defined as (Abishek et al., 2013) qE =
Fig. 9. The effect of surface roughness on the heat transfer coefficients at a spray pressure is 1.9 kPa.
vapor blanket during transient boiling regime is greater than any surface roughness. Therefore, the transient boiling regime is not sensitive to the microsurface geometry. In Figs. 8 and 9, q, h and hmax first decrease and then increase as the surface roughness increases when the surface temperature is less than 170 ◦ C. Conventional wisdom would lead one to expect lager heat transfer area and hence higher heat transfer coefficient with increasing surface roughness – that is, rougher surface give rise to higher heat transfer coefficient. For example, Prabhu and Fernandes (2007) indicated that a maximum peak heat flux was obtained with rough surface for quenching in aqueous quenchants. They suggested that the increase of surface roughness leads to the increase of real surface area and aqueous quenchants due to lower viscosity is able to penetrate easily into the cavities on the rough surface resulting in larger heat transfer contact area. Heat transfer significantly increases with increasing surface roughness, attributing to the increase of heat transfer contact area. Conversely, Pais et al. (1992) studied the effects of surface roughness (22, 14 and 0.3 m) on heat transfer when using spray cooling and found that the 0.3 m rough surface achieved higher heat fluxes over the temperature domain sampled. They attributed the heat transfer enhancement in smoother surface to the increase of the nucleation site density of bubble during the nucleate boiling regime
3 d fNw v L 6 bw
(4)
where dbw , f, Nw , v and L are bubble departure diameter, bubble departure frequency, nucleation site density, density of vapor and latent heat of vaporization. The heat flux obviously increases monotonically with increasing nucleation site density when other parameters remain constant. Therefore, Eq. (4) confirms that the relationship between surface roughness and nucleation site density could reflect the relationship between the heat transfer and surface roughness. The nucleation site density can be determined by the correlation of Benjamin and Balakrishnan (1997) N = 218.8(Pr )1.63 A
1
−0.4 (T )
3
(5)
The N/A of Eq. (5) is equal to Nw of Eq. (4). Where Pr is the Prandlt number and is defined as Pr =
Cp k
(6)
where Cp , and k are the specific heat, dynamic viscosity and thermal conductivity of the liquid. The surface–liquid interaction parameter is defined by
=
kw w Cpw kl l Cpl
0.5
(7)
is given by Benjamin and Balakrishnan (1997) = 14.5 − 4.5
R P a
R P 2
+ 0.4
a
(8)
where Ra , P and are arithmetic average roughness, external pressure and surface tension. By substituting the experimental parameters into Eqs. (5)–(8), the variation of nucleation site density with the surface roughness can be determined as illustrated
R. Xu et al. / Journal of Materials Processing Technology 214 (2014) 2877–2883
2883
cooling (below 100 ◦ C). The film boiling and the minimum heat flux (Leidenfrost point) are not detected in all heat flux curves. (3) Higher spray pressure tend to produce higher interfacial heat flux and heat transfer coefficient. The qmax and hmax increase with increasing spray pressure. (4) In the transition boiling regime (Ts > 300 ◦ C), surface roughness has little effect on the q and h. However, its influence is obvious in the nucleate boiling regime (100 ◦ C < Ts < 300 ◦ C). When the Ts is less than 170 ◦ C, the q, h and hmax decrease and then increase as the surface roughness increases. Acknowledgements
Fig. 10. The effect of surface roughness on nucleation density and hmax .
in Fig. 10. The square points represent the maximum heat transfer coefficients with different surface roughnesses calculated by inverse method. It can be seen that the curve first decreases and then increases as the surface roughness increases, consisting with the trend of maximum heat transfer coefficient. Why are there minimum values? Benjamin and Balakrishnan (1997) suggested that increasing surface roughness decreases the available superheat before the minimum, leading to lower nucleation site density. Beyond the minimum, one or more individual bubbles are probably nucleated in the same cavity, so nucleation site density starts increasing again. It should be noted that there is a difference between the two minimum values. The difference may be attributing to the calculation and measurement error. In addition, the hmax of the roughest surface (40.2 m) appears in the single-phase cooling regime instead of the nucleate boiling regime, suggesting that the increase of surface roughness leads to the increase of heat transfer in both nucleate boiling regime and single-phase cooling regime. 4. Conclusions (1) The numerically calculated temperatures with the inverse identified heat transfer coefficient are in good agreement with those measured experimentally, indicating that the inverse method was a feasible and effective tool for determination of the interfacial heat flux and heat transfer coefficient during spray quenching of aluminum alloy in water. (2) In the water spray quenching of aluminum alloy, heat transfer consists of three main stages including transition boiling (500–300 ◦ C), nucleate boiling (300–100 ◦ C) and single-phase
The authors gratefully acknowledge research support from the National Key Project of Science and Technology “High-grade CNC machine tools and basic manufacturing equipment” (No: 2014ZX04002071) and the Changsha science and technology project (No. K1112067-11). References Abishek, S., Narayanaswamy, R., Narayanan, V., 2013. Effect of heater size and Reynolds number on the partitioning of surface heat flux in subcooled jet impingement boiling. Int. J. Heat Mass Transfer 59, 247–261. Benjamin, R.J., Balakrishnan, A.R., 1997. Nucleation site density in pool boiling of saturated pure liquids: effect of surface microroughness and surface and liquid physical properties. Exp. Therm. Fluid Sci. 15, 32–42. Bernardin, J.D., Mudawar, I., 1996. Experimental and statistical investigation of changes in surface roughness associated with spray quenching. Int. J. Heat Mass Transfer 39 (10), 2023–2037. Hsieh, C.C., Yao, S.C., 2006. Evaporative heat transfer characteristics of a water spray on micro-structured silicon surfaces. Int. J. Heat Mass Transfer 49, 962–974. Liu, Z.H., Wang, J., 2001. Study on film boiling heat transfer for water jet impinging on high temperature flat plate. Int. J. Heat Mass Transfer 44, 2475–2481. Luke, A., 1997. Pool boiling heat transfer from horizontal tubes with different surface roughness. Int. J. Refrig. 20 (8), 561–574. Mascarenhas, N., Mudawar, I., 2010. Analytical and computational methodology for modeling spray quenching of solid alloy cylinders. Int. J. Heat Mass Transfer 53, 5871–5883. Mozumder, A.K., Monde, M., Woodfield, P.L., Islam, M.A., 2006. Maximum heat flux in relation to quenching of a high temperature surface with liquid jet impingement. Int. J. Heat Mass Transfer 49, 2877–2888. Pais, M., Chow, L., Mahefkey, E., 1992. Surface roughness and its effects on the heat transfer mechanism of spray cooling. J. Heat Transf. Trans. ASME 114 (1), 211–219. Park, C.W., Cho, H.C., Kang, Y.T., 2004. The effect of heat transfer additive and surface roughness of micro-scale hatched tubes on absorption performance. Int. J. Refrig. 270, 264–270. Prabhu, K.N., Fernandes, P., 2007. Effect of surface roughness on metal/quenchant interfacial heat transfer and evolution of microstructure. Mater. Des. 28, 544–550. Wang, H.M., Yu, W., Cai, Q.W., 2012. Experimental study of heat transfer coefficient on hot steel plate during water jet impingement cooling. J. Mater. Process. Technol. 212, 1825–1831. Xu, F.C., Mohamed, S.G., 2006. Heat transfer behavior in the impingement zone under circular water jet. Int. J. Heat Mass Transfer 49, 3785–3799. Zhang, L.Q., Li, L.X., 2013. Determination of heat transfer coefficients at metal/chill interface in the casting solidification process. Heat Mass Transfer 49, 1071–1080.