Experimental investigation on the local heat transfer with a circular jet impinging on a metal foamed flat plate

https://doi.org/10.1016/j.ijheatmasstransfer.2020.120405Get rights and content

Highlights

  • Air jet is made to impinge on metal foamed flat plate.

  • Effect of Re, z/d and PPI on Nu is investigated for air jet.

  • Metal foamed flat plate enhance heat transfer performance compared to flat plate.

  • 10 PPI metal foam perform better than 20 PPI and 40 PPI metal foam.

  • Region wise correlations for Nu are developed for metal foamed flat plate.

Abstract

The present study is concerned with an experimental investigation of the local heat transfer coefficient for an air jet impinging on a metal foamed flat plate having different pore densities using a thin metal foil technique. The augmentation of heat transfer caused by the metal foams in comparison with heat transfer on a flat plate is quantified. The experiments are performed to measure the temperature distribution of the metal foamed flat plate with the help of an infrared themal camera. Alunumium metal foams having 8 mm thickness and pore densities of 10, 20 and 40 pores per inch (PPI) are studied. The porosity and thickness of the each aluminum metal foam is 92 %. The range of the Reynolds number (based on the nozzle diameter) covered in the study is 10000 to 25000 and the nozzle to plate spacing varied from 2 to 10 nozzle diameter. The local Nusselt number is highest at the stagnation region, and it decreases in the radial direction away from the stagnation point. Metal foamed flat plate enhances the heat transfer performance compared to a smooth flat plate irrespective of metal foam pore density (pores per inch). The stagnation Nusselt number and average Nusselt number for the metal foamed flat plate is consistently higher compared to the smooth flat plate. A metal foamed flat plate with a pore density of 10 PPI results in higher augmentation of the stagnation and average Nusselt number compared to metal foamed flat plates with a pore density of 20 and 40 PPI. Metal foamed flat plate shows little decrement in the stagnation Nusselt number with an increase in the nozzle to plate spacing irrespective of the pore density. The average Nusselt number seems to be insensitive to the nozzle to plate spacing. Region wise semi-empirical correlations of the local Nusselt number based on the Reynolds number (Re), non-dimensional radial distance (r/d) and non-diamnesional nozzle to plate spacing (r/d) are proposed.

Introduction

Jet impinging cooling with a suitable fluid medium is widely preferred where high rates of heat transfer are desired. It gives better local heat transfer performance as compared to parallel flow cooling, such as duct flow. When jet impinges on a heated plate, a thin boundary layer formed over it results in a higher convective heat transfer coefficient [1]. Due to this advantage, the jet impingement cooling technique is widely used in the cooling of gas turbine components, freezing of tissue in cryosurgery , drying of textiles products and paper, electronics cooling, glass tempering, metals cooling, etc. There have been many researchers reported the investigations of the heat transfer characteristic of impinging jets in the past decades. Jambunathan et al. [2] reviewed the heat transfer characteristics of a single circular jet impinging on a flat surface. It depends on the parameters such as Reynolds number, nozzle to plate spacing, radial distance, Prandtl number, the turbulent intensity at a nozzle exit, confinement of jet, and nozzle geometries. The heat transfer characteristics of an air jet impingement for a lower nozzle to plate spacing (z/d < 1) is investigated by Lytle and Webb [3]. They used a thermal imaging technique to measure the heat transfer and use a laser Doppler velocimetry for flow structure and pressure measurements. They reported that in the acceleration range of nozzle to plate spacing (z/d < 0.25), maximum Nusselt number shifts from the stagnation point to the point of the secondary peak. This effect is more marked at higher Reynolds numbers.

Katti and Prabhu [4] experimentally investigated the effect of the nozzle to plate spacing (0.5 < z/d < 8) and Reynolds number on the local heat transfer distribution for impingement of a submerged circular air-jet on a smooth flat surface using a thermal imaging technique. They identified three regions, namely stagnation region, transition region, and wall jet region on the impingement surface based on the flow characteristics and heat transfer distribution. They reported semi-empirical correlations of the local Nusselt numbers for each region. Gulati et al. [5] experimentally investigated the influence of the shape of the nozzle (circular, square and rectangular) on the local heat transfer distribution for submerged air jet impingement on a smooth flat surface. The Nusselt number distribution along the horizontal axis for a rectangular jet is higher in the stagnation region than those of circular and square jets up to non-dimensional nozzle to plate spacing of 6. The average Nusselt number is almost insensitive to the shape of the nozzle.

Another popular cooling technique is the heat transfer augmentation using a metal foam. A metal foam is a cellular structure consisting of solid metal with gas-filled pores comprising a large portion of the volume. The pores are interconnected in an open-cell metal foam. These metal foams are ultra-light because of their low relative density and high porosity (~ 90 %). Metal foam has special characteristics like low density, high surface area to volume ratio, lightweight, good impact energy absorption [6], high heat dissipation capacity, and excellent fluid mixing due to tortuous flow paths [7]. The application of metal foams for enhancement of heat transfer has started to receive a reasonable amount of attention from researchers in the recent past. Kim et al. [8] experimentally examined the flow and convective heat transfer characteristics of aluminum foam heat sinks having different pore densities. They reported that an aluminium foam used as a heat sink is 28 % more efficient than a conventional parallel-plate heat sink of the same size.

The combination of the metal foams and the jet impingement cooling gives an attractive solution for enhancing the convective heat transfer. Recently, the combined effect of the jet impingement and the open-cell metal foams on the heat transfer performance is considered by many researchers to evaluate the effect of various aspects of porous properties such as foam height, porosity, pore density (or pore size) and thermal conductivity. Summary of the experimental investigations on the heat transfer of metal foams impinged by circular jets, as reported in the literature, is given in Table 1.

Yakkatelli et al. [9] investigated a flow visualization study for a single round jet impinging on aluminium metal foam by using smoke-wire flow visualization to study the fluid dynamics behavior as a function of foam porosity, nozzle to plate spacing, and Reynolds number. They identified recirculation regions and flow deflection patterns. They have found that the permeability of the porous media significantly affected the penetration of the impinging flow into the porous media. The flow deflection from the impact surface of the foam increases as the permeability of the foam decreases, and it approaches the fluid dynamics behavior of impact on a solid cube. Using high permeability foam with a turbulent jet and a higher nozzle to surface spacing would result in better flow to penetration into the foam.

Chan [10] experimentally investigated the heat transfer characteristics of an aluminum foam heat sink subject to air jet impingement under fixed pumping power condition. They reported that there is no significant effect of impinging distance on the Nusselt number under a fixed Reynolds number. However, under fixed pumping power condition, the Nusselt number decreases with decreasing impinging distance. A nigligible effect of the shape of the nozzle on the heat transfer performance is observed . They also reported that a metal foam heat sink would give better heat transfer performance compared to a pin fin heat sink with the same porosity for low Reynolds number.

Buonomo et al. [11] showed that the metal foam improves the heat transfer performance for impinging jet on the heated wall with the penalty of increase in the friction factor. For a given heat input, metal foam having a thickness of 20 mm has a higher heat transfer performance than a metal foam having 10 mm and 40 mm thickness at lower Reynolds number. However, at higher Reynolds number, metal foam with 10 mm thickness has a higher heat transfer performance compared to 20 mm and 40 mm thick metal foam.

Hsieh et al. [12] experimentally investigated the effects of porosity and pore density of a metal foam on the heat-transfer characteristics of jet impingement on an aluminium metal foam. They reported that the Nusselt number increases with the increase of the porosity and the pore density of the metal foam. They observed the phenomenon of non-local thermal equilibrium (LTNE) between the solid- and gas-phases and it may inhance with an increase in the porosity and pore density of the metal foam.

The effect of metal foam height on the heat-transfer characteristics for jet impingement on an aluminium metal foam is experimentally investigated by Shih et al. [13]. They reported that the decrease in the height of aluminium foam would first results in an increase and then decrease the cooling performance of the aluminium foam under jet impingement condition.

Kuang et al. [14] experimentally investigated that the influence of the jet structures and their variation with the nozzle to plate spacing for circular jet impingement on a circular block of an open-cell copper foam. They used semi turbulent and fully turbulent jets. They have obseved an appearance of the laminar lenght in the case of the semi trubulent jet. However, it is on obseved in the case of fully turbulent jet. The length of the undisturbed flow region is the sum of the laminar length and the potential core length. . In the case of semi turbulent jet, the overall heat transfer remains almost unchanged within the undisturbed region. However, outside the undisturbed region, the overall heat transfer begins to increase, peaking at a certain downstream location followed by a monotonic decrease. In the case of fully turbulent jet, the overall heat transfer increases and peaks at a certain downstream location, followed by a monotonic decrease. Turbulent jets with z/d spacing larger than the length of the potential core can be used to improve flow penetration into the foam and gives better heat transfer performance.

Sultan et al. [15] showed the enhancement of the convective heat transfer in graphite foam compared to the bare substrate under jet impingement condition. For airflow having velocity ranging from 7 to 11 m/s and foam thickness of 2 to 10 mm, the enhancement with graphite foam shows 30 to 70% compared to the bare plate surface.

Several studies are reported on improvements in the heat transfer performance with jet impingement on a metal foam by changing metal foam configurations. Shih et al. [16] presented the use of a flow restricting mask along with a metal foam. The use of the flow restricting mask guides the cooling air towards the heat-generating surface. This will reduce the convective resistance by increasing the velocity near the heat-generating surface, which enhances the cooling performance.

Kumar and Pattamatta [17] experimentally investigated the heat transfer performance of the different configurations of an aluminium metal foam with a laminar slot jet impingement. They have used liquid crystal thermography technique. They investigated the impingement of a slot jet over the (1) porous heat sink (2) porous obstacle and (3) through porous passage configuration. They reported that, jet impingement through porous passage performes better compared to other configurations studied because of the increase in the local jet velocities.

Feng et al. [18] experimentally investigated the heat transfer performance of the Metal foam (MF) heat sink and a Fin metal foam (FMF) heat sink made of an aluminium. Here, the finned metal foam heat sink developed by inserting a copper metal foam between two nearby fins. They reported that the heat transfer performance of FMF heat sinks could be 1.5–2.8 times that of the MF heat sinks with the same foam height under either a given flow rate or a given pumping power condition. For a given flow rate condition, the heat transfer performance of the MF heat sink decreases monotonously with the foam height. However, the heat transfer performance first increases and then slightly decreases with the height of the metal foam in the case of FMF heat sink. For a given pumping power condition, the heat transfer performance is insensitive to the foam height in case of the MF heat sink. On the other hand, the heat transfer performance increases with the increase in the metal foam hight in the case of FMF heat sinks. A bonding material in FMF enhances heat transfer performance. They suggested that the FMF heat sink is more suitable for larger foam height compared to the MF heat sink.

Similar study is reported by Wang et al. [19] for the finned copper metal foam heat sink subjected to the impingement cooling by a rectangular slot jet as well as an axial fan cooling method used in the CPUs. They reported that, for the same number of fins, the finned copper foams heat sink shows better heat transfer performance than a conventional heat sink under jet impingement condition with the penalty of increase in the pumping power. However, for the impingement by an axial fan, finned copper foams show poor thermal performance than that of the conventional heat sinks because of larger flow resistance. The thermal performance of finned copper foam heat sink is enhanced as the height of foam decreases. For 10 PPI finned copper foam, 30 mm and for 20 PPI and 30 PPI metal foam 15 mm is the optimal foam height compared to all the other sample thickness covered in their study.

Singh et al. [20] experimentally and numerically investigated the fluid flow and thermal characteristics in high porosity aluminum foams placed in a channel of square cross-section and subjected to an array of jet impingement. The parameters of the study are jet-to-jet spacing, metal foam pore density and Reynolds number (based on channel hydraulic diameter and channel inlet velocity). The major finding of their study is that the impingement configurations had a higher heat transfer rate in comparison with the baseline configuration (the metal foam in a channel flow, without jet plates). Also, for all the values of nozzle -to-jet spacings, the heat transfer rate increases with an increase in the pore density of the metal foam. The numerical computation shows a recirculating flow structure immediately downstream of the jet exits and the extent of flow penetrating in the foam. These flow patterns aid in understanding the role of the volume of metal foam in thermal transport and the convective heat transfer rate.

It is observed from the available literature that the open-cell metal foam as a heat sink enhances the heat transfer. Several research works are carried out to understand the heat transfer performance of the metal foam heat sink under the impinging condition. It is observed that all the studies report average Nusselt number instead of local Nusselt number as listed in Table 1 with the confined jet condition. Also, most of the reported Nusselt number is limited up to the stagnation region (r/d=1.0) only. Although, few of the studies report the local Nusselt number for higher r/d′s. The most important observation is that almost all of the studies use a thick conductive plate (copper or aluminium of approximately 3 mm or more) as the base plate on which the metal foam is placed, as given in Table 1. Temperature measurement on such a thick plate will only provide average temperature, not a local one due to the conduction within the plate. Along with that, most of the experimental studies reported use thermocouples on the base plate to measure the baseplate temperature. Hence, most of the reported Nusselt number of data is averaged in nature. There is a need to measure the local temperature distribution to identify the hot spots which are necessary for an engineering approach. In the present study, the local temperature distribution of the aluminium metal foamed flat plate under unconfined air jet impingement condition using thermal imaging technique is carried out to find the local heat transfer distribution. In this work, we refer to a heat transfer along the heated wall as “local heat transfer” as against pore-level phenomena.

The objectives of the present study are to

  • Conduct an experimental investigation to measure the local heat transfer distribution along the radial direction by measuring local temperature using an infrared camera.

  • Investigate the effect of various parameters such as Reynolds number, pore density, and nozzle to plate spacing on the local Nusselt number distribution.

  • Region-wise correlation for the local Nusselt number for the metal foamed flat plate with having different pore density is obtained as a function of Reynolds number, nozzle to plate spacing, and non-dimensional redial distance of a metal foam.

Section snippets

Experimental set-up

A schematic layout of the experimental set-up used for a circular air jet impingement heat transfer study is shown in Fig. 1. Air is supplied to a circular nozzle from an air compressor through a calibrated orifice flow meter. Air coming from an air filter is compressed in the compressor (ELGI make) and stored in an air receiver. A needle valve is attached downstream of an air receiver. It ensures not to allow the flow downstream until the required gauge pressure of 4 bar is reached in an air

Data reduction

The Reynolds number of a jet is calculated byRe=ρvdμ=4m˙πμdwhere ρ is the density of the jet fluid in kg/m3, v is the average velocity of the jet fluid in m/s, d is the diameter of the nozzle in m and μ is the dynamic viscosity of the jet fluid in kg/ms and m˙ is the mass flow rate in kg/s. The properties of the fluid are evaluated at the jet temperature Tj in°C.

The temperature of the targeted surface is found by averaging five IR images. The electric heat flux is supplied to the plate for

Results and discussion

Experiments are performed to measure the local heat transfer distribution on the metal foamed flat plate impinged by an air jet. An air jet is provided by a circular tube of diameter 7.06 mm. Aluminium metal foam is of 8 mm thickness with a porosity of 92%. The local temperature distribution is measured using a thermal imaging technique. The parameters varied are nozzle to plate spacing (z/d=2,4,6,8and10), Reynolds number (Re=10000,15000,20000and25000), and pore density

Correlations

Katti and Prabhu [4] identified three distinct regions on the smooth flat plate with the help of the local Nusselt number distribution pattern for the case of air jet impingement on the smooth flat surface. These regions are (1) stagnation region (0 < r/d < 1.0) (2) transition region (1.0 < r/d < 2.5) and (3) turbulent wall jet region (r/d > 2.5). They suggest a region-wise correlation for the local Nusselt number. In the present study, three different regions are identified on the metal foamed

Conclusions

The local heat transfer characteristics of a circular air jet impingement on an aluminium metal foamed flat plate are experimentally investigated using a thin metal foil technique. The range of the Reynolds number covered in the study is 10000, 15000, 20000, and 25000. The diameter of the circular nozzle is 7.06 mm. The metal foams having a thickness of 8 mm, the porosity of 0.92 and pore density of 10 PPI, 20 PPI, and 40 PPI is used in the present study. The distance between the nozzle and

Contribution of the present work

The present work presents a comprehensive literature review on a circular air jet impingement on the metal foamed heat sink. Present work represents a technique to measure the local wall temperature of a metal foamed flat plate under jet impingement condition using a thermal imaging technique. The local Nusselt number distribution presented in the present study for a metal foamed flat plate would serve as benchmark data for validating numerical results. By taking a suitable jet temperature, the

CRediT authorship contribution statement

Ketan Yogi: Data curation, Formal analysis, Investigation, Writing - original draft. Mayur Manik Godase: Data curation, Formal analysis, Investigation, Writing - original draft. Mikhil Shetty: Data curation, Formal analysis, Investigation, Writing - original draft. Shankar Krishnan: Conceptualization, Visualization, Supervision, Writing - original draft. S.V. Prabhu: Conceptualization, Visualization, Supervision, Writing - original draft.

Declaration of Competing Interest

We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property.

Acknowledgments

Authors acknowledge the efforts put in by Mr. Rahul Shirsat in building the experimental setup and fixing the mechanical problems during the experiments.

References (24)

Cited by (25)

  • Forced convection of nanofluids in metal foam: An essential review

    2023, International Journal of Thermal Sciences
View all citing articles on Scopus
View full text