International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 1 Assessment of the Baffle Effects on the Mixed Convection in Open Cavity Nabil Jamil Yasin 1, Kadhum Audaa Jehhef 2, Asraa Mahdi Shaker 3 1,3 Engineering Technical College-Baghdad, Middle Technical University, Baghdad, Iraq, email:nabilyasin@toc.edu.iq 2 Department of Mechanics, Institute of Technology, Middle Technical University, Iraq, jehhefk@gmail.com Abstract-- Several experimental and numerical studies were carried out to determine the effects of a vertical unheated baffle on the mixed convection heat transfer process in a square crosssectional bottom-grooved cavity heated from three sides. The mixed convection heat transfer and fluid flow within the cavity were evaluated by the buoyancy parameter, Reynolds number, Grashof number, and Richardson number. In the numerical part, the validated CFD model has been used to solve the governing continuity equations for the temperature and velocity distribution. The wall temperature profile and the Nusselt numbers were investigated and presented in this study. The experimental results show that the maximum temperature values increased by increasing the baffle height and the maximum Nusselt numbers were found at the full length of the baffle. The comparison between the numerical and experimental study was done and shown a good agreement with a maximum deviation of ±17%. Index Term Heat transfer, Mixed convection, Grooved cavity, Baffle. I. INTRODUCTION The method of using baffles for the heat transfer enhancement is widely used in engineering applications such as compact heat exchangers, air-cooled solar collectors, and electronic packages. A numerical study presented by [1] for the mixed convection heat transfer in open-ended enclosures for three different flow attack angles. Another numerical study of mixed square cavity inside inclined heater was presented by [2]. The effects of the horizontal baffle on the heat transfer characteristics of pulsating opposing mixed convection in a parallel vertical open channel investigated by [3]. However, [4] presented a study on the opposing mixed convection in a differentially heated partitioned enclosure. The influence of baffles on natural convection in trapezoidal cavity was discussed by [5]. Their results showed that the heat transfer in case of opposing mixed convection is more than that in natural convection in a centrally partitioned enclosure when the partition height exceeds one-third of the enclosure height. The Laminar periodic flow and heat transfer in a two dimensional horizontal channel with isothermal walls and with staggered diamond-shaped baffles presented by [6]. A numerical study [7] on a laminar mixed-convection heat transfer to air using two identical protruding heat sources was presented adopting twodimensional horizontal channel simulate electronic components. A numerical study on the steady air flow free convection in a partially open square 2-D cavity carried out by [8] with internal heat source with an adiabatic condition given to the bottom and top walls while the vertical walls were maintained at different constant temperatures. Their results showed that the thermal and fluid dynamics of the fluid are highly influenced by the presence of the heat source by the opening size and by the temperature difference between the vertical walls. Another study, [9], discussed the thermal transport and fluid flow in a three-dimensional horizontal channel through an open-ended cavity by a comprehensive numerically for Reynolds number between 100 < Re < 1500 and Richardson number between 0.01 < Ri < 10. They found that at both low Ri and Re, the flow becomes steady and the diffusion dominates the heat transfer mechanism. The mixed convection in an open cavity with a heated wall bounded by a horizontally unheated plate was studied by [10]. Their results showed that the larger the distance between the vertical walls of the cavity the lower the surface temperature. Moreover, [11] presented an experimental and numerical study on the mixed convection. Their results showed that the effect of the orientation of the heater in a horizontal and inclined position where the inclined position offers lower heat transfer rate compared to horizontal one. A number of numerical investigations were carried out [12] on mixed convection heat transfer in a two-dimensional channel with open cavity. The technique requires generating a forced flow assisting and opposing the motion by the natural convection inside the cavity. The temperature was assumed uniformly and alternatively associated on the left surface and on the right side of the cavity while the other surfaces were considered adiabatic. The simulation was carried out for Reynolds numbers in the range of 10 < Re < 1000 and the range of Richardson number was 0.1 < Ri < 1.7 10 7. Their results showed that the enhancement of heat transfer rate is generated principally by the increasing Re and the opposing configuration is thermally more efficient with respect to the assisting one. Another numerical study was performed with unsteady mixed convection in a two dimensional open-ended cavity with different aspect ratios was presented by [13]. In this study, a uniform temperature was set to the left and the right sides of the cavity while the other surfaces are adiabatic. The simulation is performed for a wide range of Reynolds numbers Re between 100 and 1000 and Richardson numbers Ri between 0.132 and 6.5 10 2, and at various cavity aspect ratios L/D of 0.5 to 4.0 and H/D at 0.1. It was noted that to achieve greater heat dissipation, it is better to have a heated surface aiding type configuration, which is in the same direction of air inlet into the channel. Also, the temperature profiles is, surely, that for which further decreases the L/D ratio, that is, more it tightens the cavity, the greater the temperature along the wall and inside the cavity. Two recent publications where the effect of two baffles in a square cavity was presented experimentally and numerically [14] while the other study discussed the behavior of an adiabatic square baffle
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 2 using nanofluid and considered buoyancy technique [15]. The present work focused on the effect of baffle height with an angle of 90 o perpendicular to the air flow on the mixed convection in a cavity with a square groove. II. EXPERIMENTAL APPARATUS The schematic diagram of the fabricated experimental rig is shown in Fig. 1. The air flow was generated by using a centrifugal fan with impeller diameter of 100 mm rotated at 1300 rpm. The fan is capable to provide air with a speed up to 6.6 m/s. The entrance section of the employed duct was fabricated from a 2 mm-stainless sheet material in order to achieve the fully developed air flow. This duct was connected with the test section by a contraction part with a square screen grid to assess the smoothness of air flow. The test section of the air passage was constructed of 5 mm-aluminum sheet thickness in a 25 cm-square cross-section. The cavity walls are perfectly insulated to ensure no heat is flowing across them. Also, the baffle was made from a 6 mm-aluminum plate and placed vertically downward from the top wall perpendicular to the air stream. The baffle position was at the center of the groove. The baffle is mounted on the top wall of the cavity and is equipped with a simple mechanism to adjust its vertical position at h = 0, 10, 12.5, 15, and 25 cm. Three uniform 900 W-heaters were inserted at the vertical left, right and bottom walls of the bottom groove to provide a constant heat flux surface. The heater was provided by a Variac-HSN0101-4 A max and voltage between 0 and 220 V. A Digital Clamp Meter (RE266) was used to measure the electricity-powered to the heater with an error deviation of ±0.5% in voltage and ±%1 in current. The test section and U-cavity for duct were well insulated using a glass wool insulation to reduce the heat loss as much as possible. Nine type-k thermocouples were used to measure the inlet, outlet air temperatures, surface temperature, and the air temperature at the midline of the test section in the top cavity. Moreover, a temperature recorder which has 12-channel equipped with a data logger uses SD memory card was used to save data which can be a loaded to an excel software. The experimental uncertainties have been determined using the standard error analysis method [16]. According to the measurement instruments, the maximum measurement errors of the flow rate were found to be ±6.0%, the air properties (density and viscosity) ±1.3%, Re ±3.8%, and Nusselt Number (Nu) ±4.2%. Fig. 1. Schematic diagram of the experimental setup. A. Data Reduction The heat (Q heater ) and heat the convection (Q conv ) delivered by the heaters to the air that flows in the duct can be described by Equation (1) and (2), respectively;
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 3 Q heater = IV (1) Q conv = ha(t w T avg ) (2) where T w is the wall temperature and T avg is the average temperature of the test section measured according to Equation (3); T avg = T 4+T 5 +T 6 +T 7 +T 8 +T 9 6. (3) By using energy balance with assuming no heat losses, the heat transfer of convection equals to the heat absorbed by air as follow according to equation (4); Q conv = Q air = m c pa (T out T in ).. (4) where m is the mas flow rate defined by the density (ρ), velocity (u), and the cross-sectional area (A) and given by Equation (5); m = ρua.. (5) The other important parameter is the Nusselt number (Nu) whose average is given in Equation (6) and defined in terms of heat transfer coefficient (h), the hydraulic diameter (d h ), thermal conductivity (k); Nu = hd h k. (6) In addition, Grashof number (Gr) and Reynolds number (Re) are also essential for the analysis. Grashof number (Gr) is defined in terms of the acceleration due to gravity (g), the coefficient of thermal expansion (β = 1/T), wall temperature (T w ), average temperature (T avg ), vertical length (H), and kinematic viscosity (ν) and is given in Equation (7); Gr = gβ(t w T avg )H 3 ν 2.. (7) The other number is Re number which is defined in terms of the density of the fluid (ρ), velocity of the fluid (u), hydraulic diameter (d h ), and the viscosity (μ), and represented as in Equation (8); Re = ρud h μ. (8) It is important to note that all parameter such density and viscosity are taken at the average temperature. III. NUMERICAL SOLUTION A. Geometry Specification The geometry considered in this study is a 2-D duct-cavity configuration, the horizontal square duct of H = 25 cm and exit length of 2H, the square cavity depth is equal to H as described in Fig. (2). Air is introduced to the duct at a uniform velocity (u i ), and at ambient temperature (T i ). The air flow is assumed to be laminar, incompressible and the Prandtl number of Pr = 0.71 with negligible viscous dissipation. The width and height of the heat source is H. The baffle is placed in on the upper duct walls. The boundary conditions were employed as the fluid velocity at the surface of the solid is zero in all directions, the heat flux is assumed to be constant along the right and left walls the confining wall is assumed to be adiabatic, signifying a zero temperature gradient normal to the partition surface, the air inlet with constant velocity u x = U i = 0.02, 0.03. 0.05, and 0.07, temperature and T = T i = 298 K, while the outlet vent, specify a zero pressure differential normal to the open surface. At the outlet section, the flow and temperature fields are assumed fully developed. Outflow boundary condition has been implemented for the outlet section. This boundary condition implies zero normal gradients for all flow variables except pressure.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 4 Fig. 2. Geometric dimensions of the duct and the baffle employed in the present numerical simulation. B. The governing equations The equations governing the buoyancy-induced regime inside an enclosure resemble the same basic equations of the conservation of mass, momentum, and energy typical equations of fluid motion. The scenario of free convection heat transfer requires at least one fluid parameter, density, to vary within the domain of interest. The equation of the buoyancy-induced flows inside a vertical truncated cone presented by the following set of equations [17]: dρ dt d(ρv) dt d(ρt ) dt + (ρ v ) = 0. (9) + (ρv v ) = p + τ + ρg (10) + (ρcpv T ) = [k. T ] (11) The stress tensor in the momentum form, τ, is described by: τ = μ [( v + v T ) 2 v I]] (12) 3 the above Equations could be reduced to simpler forms [17]: (ρv ) = 0... (13) (ρv v ) = p + μ( 2 v ) + ρg.. (14) (ρcpv T ) = (k T ) (15) The fluid density can be simplified at all points within the employed domain, by employing the Oberbeck-Boussinesq approximation to linearize the temperature dependency of density. The buoyancy force is redefined as [17]: ρg = g [ρ ref ρ ref β(t T f )] = ρ ref g [1 β(t T f )]... (16) where β represent a fluid compressibility found via: β = 1 V V P... (17) The air is considered as an ideal gas at nominal atmospheric pressures and temperatures. The compressibility of the air can be found by: β = 1 T f... (18)
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 5 A necessary simplification was made to the mass and momentum equations present in Equations (10) and (11) allows for the development of the final forms for incompressible flow: (v ) = 0. (19) v. (. v ) = 1 ρ ref P + v ref ( 2. v ) + g [1 β(t T f )].. (20) C. Mesh and Grid Independence The ANSYS-FLUENT-v18.5 meshing software was used with advanced by CAD/geometry reading, for generating the mesh of duct-cavity in the Design Modular. The global number of grid points was 131351 with a global number of elements of 129900 quadrilateral cells as shown in Fig. 3. Fig. 3. Computational domain and the grid model. To make sure that the results are due to the boundary conditions and physics used, the mesh independence was studied and the results for the average Nusselt number is presented in Fig. 4. Fig. 4. Variation of Nusselt number of air flow with total node number. IV. RESULTS AND DISCUSSION
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 6 The present study considers the effect of the baffle of a height ratio (h = h H) which varies in the range of 0.2 to 1.0 measured from the top wall and heater position on the performance of the mixed convection heat transfer inside a cavity equipped with an under the groove. The working fluid inside the cavity was dry air with Pr = 0.7, the Grashof number range of 1.2 10 7 < Gr < 8.3 10 7, Reynolds number (Re) in a range given by 250 < Re < 1400, and the Richardson number given by 1.0 < Ri < 700. The experimental results presented using the temperature distribution in the upper opening of the heated groove measured by seven thermocouples and the heater wall Nusselt numbers. The numerical results is given by the temperature and velocity contours of the cavity two-dimensional domain. constant heat flux with a variable fan speed of the case of without baffle insertion inside the cavity. The results for the air temperature distribution in the cavity are plotted against the dimensional axial distance with a variation of the applied constant heat fluxes and shown in Fig. 5 and 6 for Re = 800 and 1400, respectively. It is shown that for all cases, the variation of the temperature along the top opening of the under groove of the cavity will slightly change with time after (50 min) of the test operation and the differences between the values of temperature with time will be small and at this point the case can be treated as steady-state case. The comparison between the two figures shows that the level of the temperature values decreased as the Reynolds number increased from 800 to 1400. A. Effect of Heater Position Experiments were carried out by mounting the heater at three different positions in the cavity (left, right and bottom) using a (a) Left heater (b) Right heater (c) Bottom heater Fig. 5. Temperature variation along the mid line of the cavity at h = 0, q = 500 W, Re = 800. (a) Left heater (b) Right heater (c) Bottom heater Fig. 6. Temperature variation along the mid line of the cavity at h = 0, q = 500 W, Re = 1400. B. Effect of the Input Reynolds Number To study the effect of Reynolds number, Fig. 7 shows the variation of the air temperature values inside the cavity at h = 0 along the center line of the cavity at a constant heat flux of q = 300 W with different Reynolds number and by switching the location of the heater. The figures show that the air temperature inside the cavity decreases as Re increases due to increasing air velocity and increasing cooling effect. Fig. 7a shows the values when the heater is mounted at the left side of the cavity in the direction opposing the flow. It may be noticed that the value of temperature near that left heater is higher than the value remote from the heater. It is clear that the value of temperature near the right heater is higher than the value remote from the heater as indicated by Fig. 7b. It is clearly presented that the flow is considered as assisting flow. For the assisting
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 7 flow, the temperature gradient is larger close to the top corner, which is adjacent to the inlet cavity opening. Moreover, when the heater is located at the bottom wall of the cavity, the difference between the temperature values at a different position along the midline of the cavity seems to be uniform as shown in Fig. 7c. By comparing with the above figures, it has shown for assisting and opposing flow. (a) Left heater (b) Right heater (c) Bottom heater Fig. 7. Temperature with dimensional axial distance at Gr = 10 6. C. Effect of Applied Heat Flux The effect of variation in heat flux on air temperature distribution at constant Reynolds number of Re = 800 is shown in Fig. 8. The figure shows a similar trend to the air temperatures for two heat flux of 300 and 500 W. The air temperature increases when the heat flux increases and reaches maximum values near the left heater for opposing flow case while it reaches maximum values near the right heater for assisting flow case. Also, it shows that the air temperature values increase with increasing the heat flux at fixed value of Re for bottom heater (heating from below), but their maximum values are shown in the middle of the cavity and it can be shown that the region in front of the heater is higher temperatures from remote regions of the heater. (a) Left heater (b) Right heater (c) Bottom heater Fig. 8. Temperature variation with dimensional axial distance at Re = 800. D. Effect of the Baffle Height Fig. 9 represents the variation of the air temperature values inside the cavity along the center line of the cavity by using different baffle heights and by changing the location of the heater in the cavity. The results show that the air temperature inside the cavity are decreasing with increasing the baffle height ratio h = h H from 0, 0.2, 0.5, to 1.0.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 8 (a) Re = 300 (b) Re = 1400 Fig. 9. Temperature with dimensional axial distance at Gr = 10 7 for the left heater. E. Average Nusselt number The variation of the average Nu with Re number for various heater positions with no baffle was presented Fig. 10 and 11. It shows that Nu increases by increasing the applied heat flux and, simultaneously, Nu increases also with increasing Re values. But the Nu values increased in right heater by 35% greater than the bottom heater case at q = 300 and Re = 800 due to the increased the heat convection in this position. However, the Nu increased in right heater by 46 % greater than the left heater case at the conditions. (a) Gr = 10 6 (b) Gr = 10 7 Fig. 10. Variation of Nusselt number with Reynolds number for the cavity h = 0 at constant heat flux. (a) Left heater (b) Right heater Fig. 11. Variation of Nusselt number with Reynolds number for the cavity at h = 0.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 9 The Richardson number represented the ratio of the natural effect and the forced convection heat transfer effect showing a good indication in the mixed convection cases. Fig. 12 shows the variation of Nu with Ri for cavity region in case of no baffle using constant heat flux of q = 300 and 500 W. Nu number increases as Ri decreases. Due to the increase of Re, the dominated natural convection effect was converting to the forced convection domination. The values of the Nusselt number become at its maximum values when using the left heater as a heat source. (a) Left heater (b) Right heater Fig. 12. Variation of Nusselt number with Richardson number for the cavity at h = 0. A comparison of variation of the average Nu with Re in the case of using h = 0.2 and 1.0 is presented in Fig.13. It shows a good enhancement in heat transfer rate for different positions of the heater. Nu values of the left heater are more than that of right and the bottom heater at the baffle height ratio of h = 0.2. But if the baffle height ratio increased to h = 1.0, where the baffle height is equal the duct height, Nu values of the bottom heater becomes more than that of the right heater. (a) h = 0.2 (b) h = 1.0 Fig. 13. Variation of Nusselt number and Reynolds number with different height baffle at Gr = 10 7. Moreover, the figure shows that Nu number increases by increasing Re. Also, Fig. 14 shows the variation of Nu with Ri at constant heat flux of q = 500 W in the case of using a different baffle heights.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 10 (a) h = 0.2 (b) h = 1.0 Fig. 14. Variation of the Nusselt number and Richardson number with different height baffle Gr = 10 7. Heat transfer enhancement parameter (η) is defined as the ratio of the average Nu with baffle to that without the baffle. Fig. 15 shows the variation of this parameter plots with baffle height ratio at different Ri and constant Re which helps to quantify the enhancement and determine the best height of the baffle. As observed from the figure, there is an appreciable heat transfer enhancement, with the use of the baffle and the enhancement parameter is increase with increasing the baffle height ratio. However, there is a significant increase in enhancement parameter when the heat source was in the right position of the cavity. At h = 1.0, there is appreciable enhancement exists for all values of Ri and at all positions of the heat source right, bottom and left. Fig. 15. Variation of heat transfer enhancement with baffle height ratio at Re = 250 and different Ri number. The numerical isotherm and velocity contours for three different location of the heater (left, bottom and right) at Re = 300 and q = 300 W was plotted in Fig. 16 and 17 in order to study the effect of heater position by using temperature visualization. It is noticed that when using the left heater as presented in Fig. 16a, the temperature contours stratified align the heater in the groove region and thermal boundary layer was very thinner without any cell circulation inside the grove area. But when applied the heater power on the right heater Fig. 16b, temperature contours spread inside the cavity region and thermal boundary layer becomes thicker than the first case, and the contours began to generate a large circulation cell inside the grooved region. For the bottom heater shown in Fig. 16c, the temperature contours were more stratified horizontally due to the conduction phenomenon of the air in case of the heated horizontal surfaces showing no circulation cell in the groove. The contours for velocity lines presented in Fig. 17 show that using the right heater will lead to generating more than one cell of vortices inside the cavity especially near the heated wall due to the convection current.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 11 (a) Left eater (b) Right heater (c) Bottom heater Fig. 16. Isothermal lines for three heater positions of without baffle case h = 0 at Re = 300 and q = 300. (a) Left heatere (b) Right heater (c) Bottom heater Fig. 17. Velocity lines for three heater positions of without baffle case h = 0 at Re = 300 and q = 300. Fig. 18 shows contours of temperatures streamlines without baffle case at Re = 200 and 1400. It is observed that when Re increases, the temperatures lines induce inward to the duct by the air flow, and a circulating cell zone encompassing above the entire square groove of the cavity is formed and the velocity increased along the heated vertical wall due to increasing the convection currents. (a) Re = 300 (b) Re = 1400 Fig. 18. Isothermal lines for Left heater positions of without baffle case of h = 0 at Gr = 10 6. Increasing the power of the heater and the heat flux giving the value of Grashof number as Gr = 10 6 and 10 7 leads to increasing the temperatures along the right heater in the grooved cavity as shown in Fig. 19. When using right heater, a second circulating cell inside the grooved region will be generated.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 12 (c) Gr = 10 6 (d) Gr = 10 7 Fig. 19. Isotherms lines and velocity lines for right heater positions of without baffle case of h = 0 at Re = 1400. The isotherm contours and velocity lines three different baffle heights ratio h = h H of 0.2, 0.5 and 1.0 for left and right heater positions at Re = 300 presented in Fig. 20 and 21. The figures show that in the square cavity a distributed isotherm contours exist only for the right position at Re=300 but for left heater condition, a high valued contours are limited at the regions near the hot wall while low valued contours dominate along the other regions of the cavity, which is a result of the flow influence on circulating cell of cavity. (a) h = 0.2 (b) h = 0.5 (c) h = 1.0 Fig. 20. Velocity lines generated by the left heater using three different baffle heights at Re = 300. (a) h = 0.2 (b) h = 0.5 (c) h = 1.0 Fig. 21. Isotherms lines generated by the left heater using three different baffle heights at Re = 300. The presence of the baffle which acts as an obstruction to the airflow shows a distributed contours at Re = 300. The isotherm contours for all baffle heights show low valued contours extending in the left end of the channel passage while the high valued contours mainly were occurred align with the regions near the heated wall. Fig. 22 and 23 show contours of velocity at Re = 300 for two positions of the heater (left and right) position. It is observed that for low baffle height, h =
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 13 h H of 0.2, a circulating cell encompassing nearly the entire duct and extension to the cavity is formed, but in condition of the right heater operation, a circulating cell encompassing nearly the entire square cavity is formed, but in bottom heater condition will more than cell the entire square cavity is formed. Thus, the height of the baffle plays an important role in the shape of the circulating cell formed in the square cavity and its bottom groove as well as the one behind the baffle. With the increase in baffle height ratio to h = h H of 1.0 as shown in Fig. 23c, the flow is forced to enter and disturb the circulating cell formed in the square cavity. (a) h = 0.2 (b) h = 0.5 (c) h = 1.0 Fig. 22. Velocity lines generated by the right heater using three different baffle heights at Re = 300. (a) h = 0.2 (b) h = 0.5 (c) h = 1.0 Fig. 23: Isothermal lines generated by the right heater using three different baffle heights at Re = 300 F. Comparison Experimental and Numerical Results Fig. 24 shows a comparison between the present experimental and numerical study which shows a good agreement between them subjected to a maximum deviation of ±17% between the value of Nusselt number then case without the baffle. Also, a deviation of ± 6.7% for h = 0.5 and ± 6.1% for h = 1.0, were observed due to the assumption and the boundary condition used here. (a) h = 0 (b) h = 0.5 (c) h = 1.0 Fig. 24. Average Nusselt number vs. Reynolds number for present experimental and numerical results.
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:04 14 V. CONCLUSIONS An experimental and numerical investigation on mixed convection heat transfer in a partially open cavity with a square groove at the bottom wall and a vertical baffle with a variable height attached at the top wall of the cavity were considered in this study. The bottom groove equipped with three heaters at left, bottom and right position with a variable applied heat flux. The baffle height to cavity height ratio h = h H can be varied from 0, 0.2, 0.5, to 1.0. The experimental results presented in term of the top groove side temperatures profile and the heater wall Nusselt numbers according to the flow parameter of 300 < Re < 1000 and 1.0 < Ri < 700. The results indicate that the temperature profiles were significantly affected by the Re and Ri values, and it increases in the region near the mainstream. Also, the temperature profile increases by increasing the baffle height ratio h. The heated wall Nusselt numbers increased by increasing the h in the investigated range of Richardson numbers. The numerical flow visualization had pointed out that the air flow at Re = 300, there were two nearly distinct fluid motions: a parallel forced flow in the channel and a recirculation flow inside the cavity groove. REFERENCES [1] K. Khanafer, K. Vafai, L. and L. Marilyn, (2002). Mixed convection heat transfer in two-dimensional open-ended enclosures, International Journal of Heat and Mass Transfer 45, pp.5171 5190. [2] N. Hamici and D. Sadaoui, (2017). Numerical study of Mixed convection and thermal Radiation in a Square Cavity with an inside Inclined Heater, 23 ème Congrès Français de Mécanique, 28 Août au 1er Septembre 2017. [3] T. Chang, S. Yann-Huei, (2005). Flow pulsation and baffles effects on the opposing mixed convection in a vertical channel, International Journal of Heat and Mass Transfer 48, pp.4190 4204. [4] S. K. Mahapatra and A. Sarkar, (2007). Numerical simulation of opposing mixed convection in differentially heated square enclosure with partition, International Journal of Thermal Sciences 46, pp.970 979. [5] E. Fontana, A. da Silva, V. C. mariani and F. Marcondes, (2010). The influence of Baffles on the natural Convection in Trapezoidal cavities, Numerical heat transfer, Part A, 58, 125-145. [6] S. Sripattanapipat and P. Promvonge, (2009). Numerical analysis of laminar heat transfer in a channel with diamondshaped baffles, International Communications in Heat and Mass Transfer 36, pp.32 38. [7] A. Hamouche and R. Bessaïh, (2009). Mixed convection air cooling of protruding heat sources mounted in a horizontal channel, International Communications in Heat and Mass Transfer 36, pp.841 849. [8] E. Fontana, A. da Silva, V. C. Mariani, (2011). Natural convection in a partially open square cavity with internal heat source: An analysis of the opening mass flow, International Journal of Heat and Mass Transfer 54, pp.1369 1386. [9] Y. Stiriba, J. A. Ferré, F. X. Grau, (2013). Heat transfer and fluid flow characteristics of laminar flow past an open cavity with heating from below, International Communications in Heat and Mass Transfer 43, pp. 8 15. [10] O. Manca, S. Nardini, and K. Vafai, (2006). Experimental Investigation of Mixed Convection in a Channel with an Open Cavity, Experimental Heat Transfer, 19:53 68, 2006. [11] T. V. Radhakrishnan, A. K. Verma, C. Balaji, and S. P. Venkateshan, (2007). An experimental and numerical investigation of mixed convection from a heat generating element in a ventilated cavity, Experimental Thermal and Fluid Science 32 (2007) 502 520. [12] A. Carozza, O. Manca, and S. Nardini, (2014). Numerical Investigation on Heat Transfer Enhancement due to Assisting and Opposing Mixed Convection in an Open Ended Cavity, 2nd International Conference on Emerging Trends in Engineering and Technology (ICETET'2014), May 30-31, 2014 London (UK). [13] A. Carozza, (2018). Numerical Study on Mixed Convection in Ventilated Cavities with Different Aspect Ratios, Fluids 3, 11. [14] G. Nardini, M. Paroncini and R. Vitali, Natural convection in a square cavity with two baffles on the vertical walls: experimental and numerical investigation, International Journal of Mechanics, 9, 120-127. [15] M. Mahmoodi and M. H. Esfe, (2015).Buoyancy driven heat transfer of a nanofluid in a differentially heated square cavity under effect of an adiabatic square baffle, Journal of Heat and Mass Transfer Research 1, 1-13. [16] J. P. Holman, (2012). Experimental methods for engineers, 8 th edition, Department of Mechanical Engineering Southern Methodist University. [17] A. A. Ganguli, A. B. Pandit and J. B. Joshi, (2009). CFD simulation of heat transfer in a two-dimensional vertical enclosure, Chemical Engineering Research and Design, 87, pp. 711 727.