Available online at www.sciencedirect.com ScienceDirect Energy Procedia 101 (2016) 18 – 25 71st Conference of the Italian Thermal Machines Engineering Association, ATI2016, 14-16 September 2016, Turin, Italy Impingement cooling experimental investigation using different heating elements Carlo Carcasci, Lorenzo Cocchi*, Bruno Facchini, Daniele Massini Department of Industrial Engineering Università di Firenze, Via Santa Marta 3, 50139, Firenze, Italy Abstract The aim of this work is to study heat transfer due to a single submerged impinging jet by using different experimental techniques, thus to provide cross-validation of the various employed methods In particular, steady-state technique is applied, and target wall temperature is evaluated thanks to thermochromic liquid crystals To allow heat transfer to take place, electrically heated film elements are applied to the jet target surface Both a smooth Inconel sheet (arithmetic average roughness 1.2 μm) and a fine stainless steel mesh (arithmetic average roughness 23 μm) have been used as heating elements, with the aim to evaluate their ability to provide a uniform and controllable heat flux condition, as well as to determine the effect of surface roughness on heat transfer Impingement hole characteristics include nozzle diameter Dj=12.5 mm, nozzle length to diameter ratio L/Dj=0.4 and nozzle-totarget surface distance to diameter ratio H/Dj=2 The performances of the different techniques are evaluated on a flat target surface, and the coolant is evacuated from the impingement cavity by two extraction holes with a diameter of 12.5 mm each, located on the target surface itself The Inconel heating element is cut in correspondence of such holes, in order to allow coolant extraction; on the contrary, the permeability of stainless steel mesh allows to lay it above the holes without cutting it, thus providing a more uniform heat flux Jet Reynolds number up to 40000 are investigated, and results are presented in terms of Nusselt number distributions and area averaged values Results obtained with the two heating elements show that surface roughness has a significant effect on heat transfer, the steel mesh coated surface providing up to 15.6% higher area averaged Nusselt values with respect to the Inconel coated one The application of a thin non conducting film over the mesh reduces surface roughness and causes Inconel and mesh to provide coherent results, thus providing a validation for the second method © Published by Elsevier Ltd This © 2016 2016The TheAuthors Authors Published by Elsevier Ltd is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the Scientific Committee of ATI 2016 Peer-review under responsibility of the Scientific Committee of ATI 2016 Keywords: heat transfer; impingement cooling; steady-state method; surface roughness; thermochromic liquid crystal; mesh heater Corresponding author Tel.: +39 055 2758 713 E-mail address: lorenzo.cocchi@htc.de.unifi.it 1876-6102 © 2016 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the Scientific Committee of ATI 2016 doi:10.1016/j.egypro.2016.11.003 19 Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 Introduction Jet impingement is a widely applied technique in many heating or cooling processes, since it produces outstanding heat transfer rates with respect to other techniques [1] The impact of a fluid jet onto a solid surface is associated with complex thermal and fluid-dynamic phenomena, strongly dependent upon system geometry, fluid properties and jet and target wall characteristics As a consequence, many works were dedicated to the study of such technique, like the ones of Gardon and Akfirat [2] and Martin [3] Experimental investigation of jet impingement can be performed through different techniques, both to perform flow visualization (many of which are applied and compared in a work of Carcasci [4]) and to study heat transfer, which in turn include both transient [5] and steady-state [6,7] techniques The implementation of steady-state technique requires a constant heat flux to be applied to the heat transfer surface, which is often obtained through an electrically supplied film heater or thin metal sheet It is evident that a similar technique is particularly challenging to apply on complex geometries, especially when the heat transfer surface houses coolant extraction holes or slots In this case, the heating element continuity is interrupted by the extraction orifices and the generated heat flux becomes strongly nonuniform, requiring elaborate FEM postprocessing procedures to be applied (examples of which are reported in [8] and [9]) Such approach tends to emphasize and amplify even small disturbances or measurement noise, often requiring some kind of filtering to be applied; moreover, the measurement confidence in proximity of extraction holes is strongly compromised Such issues could be partly addressed if a permeable heating medium was applied, which could be laid on the extraction holes without the need to cut it: a uniform heat flux could thus be obtained, and thus a significantly simpler postprocessing, at the expense of a flow alteration in proximity of the extraction holes In the present work, a candidate for such permeable heating medium has been identified in a fine stainless steel mesh, composed of 30 μm diameter wires and with a 69 μm pitch The performances of such heating element are compared with the ones of a smooth Inconel sheet under a single impinging jet A primary concern about the mesh is that surface roughness could influence heat transfer, thus providing results which are not suitable for a smooth wall Literature on roughness effect on heat transfer is limited with respect to other subjects, even more if impingement is considered Gabour and Lienhardt [10] studied the effect of surface roughness on the stagnation point heat transfer of an unsubmerged liquid jet The authors found up to 50% increase in heat transfer for the higher jet Reynolds numbers and roughness values, and also showed the presence of a critical jet Reynolds number under which the rough surface acts as a smooth one More recently, Beitelmal et al [11] evaluated area averaged heat transfer performances of a jet impinging onto a flat and roughened (through 0.5 mm height protrusions) surfaces, finding up to 6% heat transfer enhancement on the rough surface Better cooling performances for rough surfaces were also found by Sesen et al [12] (up to 22.4%) for a single phase liquid jet impinging onto a nanostructured plate A numerical investigation of Xu et al [13] confirmed the overall heat transfer enhancement, but also found local reductions in case the fluid gets trapped into the roughness elements All of these investigations show that a general theory on roughness effect on impingement is not easy to develop, and that a case-by-case evaluation is needed: thus, the effect of the adoption of a fine mesh as a heating element cannot be derived from precedent studies In this context, the present work also provides a contribution on the subject of impingement-roughness interaction Nomenclature A area [m2] D diameter [m] nozzle-to-target distance [m] H h convective heat transfer coefficient [W/m2K] k thermal conductivity [W/mK] nozzle length [m] L m mass flow rate [kg/s] μ dynamic viscosity [Pa s] Nu Nusselt number [-] q heat flux [W/m2] R roughness [μm] Re Reynolds number [-] r radial coordinate [m] T temperature x horizontal coordinate y vertical coordinate Subscripts a arithmetic average ad adiabatic conv convective ex external in internal jet j joule generated by Joule effect loss not exchanged by convection maximum t w wall [K] [m] [m] 20 Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 Experimental analysis 2.1 Experimental apparatus 100 mm 50 mm Inconel mesh Double-side tape pe Bus bar PC Target plate IInconel/mesh onel/mesh Air out CCD camera S Spacer DSA Led array DC power supply p T Foam I,V mesh Impingement hole(s) T AIR OUT Target plate p T Model Plenum 55 mm Inconel Impingementt plate p T Baffle AIR IN Orifice Fan 55 mm I,V Air from centrifugal fan Impingement plate (a) (b) (c) Figure 1: test rig (a), model (b) and impingement and target plates (c) schemes Measurements were performed in the Heat Transfer and Combustion Laboratory of the Department of Industrial Engineering of the University of Florence (DIEF) The test rig (depicted in Figure 1) consists of an open-loop wind tunnel, and is designed to replicate and record the thermal and fluid-dynamic phenomena involved by a jet impingement cooling system The main element of the rig is the model, which defines the actual cooling geometry and in which heat transfer phenomenon takes place The model is entirely made of transparent PMMA, thus granting a complete optical access to the inner geometry, and is composed by four main elements: a feeding plenum, an impingement plate, a spacer and a target plate The feeding plenum provides air at uniform total pressure to the cooling geometry, thus ensuring a controllable upstream condition for the impingement system Pressurized coolant air is provided to the plenum by means of an inverter controlled centrifugal fan The plenum inner volume is 220x220x120 mm3 A square impingement plate encloses the upper part of the plenum and houses two impingement holes, both with a diameter ratio Dj=12.5 mm, whose position is reported in Figure Since the impingement plate thickness is mm, a nozzle length-to-diameter L/Dj=0.4 is obtained A 25 mm thick spacer separates the impingement plate from the target surface: this component is designed as a square PMMA frame, thus defining an impingement cavity which is laterally closed from all the four sides The spacer has not been replaced between the various tests, thus a fixed nozzleto-target distance of H/Dj=2 is obtained The 20 mm thick target plate, made of transparent PMMA, is parallel to the impingement plate and houses four extraction holes, two for each impingement jet, symmetrically located at the opposite side of the impingement cavity with respect to the cooling holes Each extraction hole is 12.5 mm in diameter A constant heat flux in the jet impact surface is required to perform a steady-state heat transfer evaluation: for this reason, the surface itself is coated with a flat heating element Given that the aim of this work is to compare the performances of different heating elements, target surface is divided into two regions, one for each impingement jet: one is covered with a smooth, 25.4 μm thick Inconel 600 sheet, and the other with a fine AISI 304 steel square mesh, composed of 30 μm diameter wires and with a mesh pitch of 69 μm The two heating elements are exposed to the impingement jets, and are fixed to the target plate thanks to a layer of 0.2 mm thick, high temperature resistant double side tape Each element is heated by Joule effect, which is obtained by connecting the opposite sides of each element to an independent DC supply through copper bus bars Since the surface roughness of the two elements is different, such parameter was evaluated thanks to a MahrPerthen Perthometer PRK surface profilometer Arithmetic average roughness Ra is 1.2 μm for the Inconel and 23 μm for the mesh, while maximum roughness Rt resulted 2.0 μm for the Inconel and 55 μm for the mesh: as a consequence, mesh roughness is an order of magnitude higher than the Inconel one Surface roughness can influence heat transfer, 55 mm p T Extraction hole(s) Data acquisition 170 mm PMMA TLC Black paint Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 since roughness elements can protrude from the laminar sublayer into the turbulent boundary layer core: in this case, laminar sublayer structure is essentially destroyed, and thermal resistance is reduced Each heated region is 100x170 mm2 in size, and houses two of the four extraction holes Since the Inconel is not a permeable medium, the metal sheet had to be cut in correspondence of the extraction holes: this generates a strongly nonuniform heat flux around the holes themselves In this case, however, jet impact location is sufficiently far from the holes to allow heat flux to be considered uniform, which was demonstrated through an electric FEM simulation of the Inconel sheet Heat transfer surface temperature is evaluated thanks to wide band thermochromic liquid crystals (TLCs) coating The adopted TLC formulation is the 30C20W provided by Hallcrest ltd, with a nominal activation temperature of 30°C and 20°C color play bandwidth TLCs were calibrated in order to ensure an accurate color-temperature response, which allowed to relate surface hue values to precise temperature values TLC coating is directly sprayed onto the inner surface of the PMMA target plate, and is then covered with a black coating to provide a nonreflecting backing: in such way, TLCs can be observed from the opposite side of the impingement jets through the transparent PMMA, and color response can thus be recorded from the outside of the model: this is achieved thanks to a CCD camera (Sony XCD-SX90CR), connected to a PC via IEEE 1394b interface Two 8W white led arrays, capable of 750-800 lumen, provide uniform illumination to the TLC coated surface To ensure the desired flow conditions to be replicated, impingement mass flow rate is measured upstream the feeding plenum through a calibrated orifice, according to the standard EN ISO 5167-1 Mass flow rate is controlled by varying fan speed Temperature and pressure measurements in definite locations are also required for the rig management Air temperature in different locations of the rig is recorded through several T type thermocouples (0.5 K accuracy), connected to a reference junction and then acquired by a data acquisition/switch unit (Agilent 34970A), while static pressure is measured by a Scanivalve DSA 3217 pressure scanner with 16 piezoresistive relative pressure sensors (7 Pa accuracy) 2.2 Experimental procedure and data postprocessing As mentioned above, Heat transfer tests were performed using a steady-state technique The desired flow conditions are set by controlling the fan speed, and are monitored through a custom built LabView program Subsequently, both the DC power supplies are turned on and electric power provided to both the heating elements Target surface heat flux is controlled in order to ensure the TLCs on the whole surface of interest to be colored, allowing a complete temperature distribution to be recorded In any case, temperature differences between surface and coolant are set not to be lower than 5°C, thus achieving a satisfactory accuracy in heat transfer evaluation Once a steady temperature distribution is reached, digital camera records one set of frames (one frame per second) in RAW format, while the LabView program acquires the other thermal, fluid-dynamic and electric data Raw data need to be postprocessed in order to determine heat transfer on the target surface, which is entirely performed by a Matlab macro Convective heat transfer coefficient h is defined as: qconv,in h (1) Tw Tad ,w where qconv,in is the convective heat flux on the inner side of the impingement target plate, Tw is the wall temperature and Tad,w is the adiabatic wall temperature Tw is obtained from the camera acquisitions: the RAW frames are ensemble averaged to reduce digital signal noise and then converted to a single RGB image from which the hue distribution is extracted TLC calibration allows to convert hue values into temperature distribution on the target wall In the present case, Tad,w was considered equal to the jet temperature Tj, which is evaluated as the arithmetic average of the two jet temperatures measured on the jet centerlines (in any condition, these two temperatures not differ for more than 0.2°C) The convective heat flux qconv,in is determined by performing an energy balance of the target surface: qconv,in q joule qloss (2) where qjoule is the electrically generated heat flux and qloss is the heat flux which is not exchanged by convection For the present case, the generated heat flux is calculated as qjoule=VI/A, where V and I are voltage and current applied to each heating element, and A is the element area exposed to the coolant flow For the Inconel, qjoule is corrected to consider the presence of the extraction holes, while this operation is not necessary for the mesh Since radiative heat 21 22 Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 fluxes can be neglected, given the low temperature differences, qloss is assumed equal to the conductive heat flux through the PMMA, which in turn is equal to the convective heat flux at the exterior side of the plate, qloss=hex(Tex-Tamb) External heat transfer coefficient hex was evaluated though correlations available in the open literature (Lloyd and Moran [15]), and being dependent from the external temperature Tex itself, an iterative calculation had to be implemented to determine both these values In any case, the dispersed heat flux never exceeds 4.5% of the internal convective heat flux The obtained h values are reformulated in a dimensionless form as Nusselt number: hD j Nu (3) k where k is air thermal conductivity, which is evaluated at the average value of air and wall temperatures Impingement flow conditions are defined through the jet Reynolds number: mD j (4) Re j Aj P where m is the overall impingement mass flow rate, Aj is the impingement hole cross section and μ is air dynamic viscosity Tests were contemporarily performed using both impingement holes, which were supposed to discharge the same mass flow rate given the symmetric geometry layout In this case, the two extraction holes from the Inconel side were covered with the same mesh used as the heating element, thus achieving similar pressure losses and avoiding flow unbalances This operation does not significantly alter heat transfer in the jet impingement region, since the extraction holes are sufficiently far from this zone An uncertainty analysis was performed according to the standard ANSI/ASME PTC 19.1 [16], based on the Kline and McClintock method [17] Maximum uncertainty on Rej value was found to be 1.34% for the Rej=10000 test A point by point implementation of the same method was applied to evaluate the uncertainty on local Nu values, thus obtaining error maps It was found local uncertainty to be up to 15% in the stagnation region, since there the lowest temperature differences are registered Even so, uncertainty on area averaged Nu value never exceeds 7.75% Uncertainty in roughness values can be estimated to be around 10%, based on measurement repetition Results Different heat transfer tests were performed by varying the impingement mass flow rate, and thus jet Reynolds numbers, ranging from around 10000 to around 40000 Geometric parameters were the same in all tests The obtained Nu distributions in the area surrounding the stagnation region are reported in Figure 2: only the top half of the jet impact area is represented, since the other half was disturbed by the vicinity of the bus bar, especially for the mesh covered surface In fact, the mesh exhibits a lower capability of redistributing heat and electric current with respect to the Inconel, and is thus more influenced by even small differenced in current supply at the bus bar This causes the distributions obtained with the mesh to be also slightly noisier than the Inconel ones An axisymmetric Nu distribution can be noticed for every condition, having its center in the jet stagnation point: this demonstrates the small interactions of the jet with the surrounding walls and with the adjacent jets near the impact location For both the Inconel and the mesh, a peak is present at the jet stagnation point: however, as Rej increases, a secondary high heat transfer ring is generated and becomes more and more evident This is a well known and documented phenomenon [2], which is enhanced for low H/Dj and high Rej values [1] Similar considerations can be performed by observing Nu trends extracted along a radial direction (x/Dj=0) and reported in Figure 3, which also provide a quantitative insight on the effect of Rej For both the Inconel and the mesh, Nu values increase as Rej increases in every radial position The analysis of Figure 3a shows that for the Inconel both the high heat transfer regions have a ring shape, since a local minimum is present in the jet stagnation point (r/Dj=0) and two distinct Nu peaks are visible The first peak is located at a distance r/Dj ranging from 0.3 to 0.6 as Rej increases, while the second peak, which emerges as Rej grows, is always located at an r/Dj of around 1.5 The shape of Nu distribution in the stagnation region seems to indicate that local heat transfer is enhanced by boundary layer acceleration, due to the mass flow adduction directly under the jet: this creates a laminar, yet thin boundary layer [1] Nu grows until this mass flow adduction is maintained, and then decreases supposedly when the wall jet starts to decelerate, which leads to the second local minimum (r/Dj=1.1) The second peak (r/Dj=1.5) can be interpreted as the 23 Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 Heat transfer surface Investigated region INCONEL MESH 100 80 60 40 40 1 20 -3 -2 -1 x/Dj [-] 4 20 0 -4 150 -3 -2 -1 x/Dj [-] 150 Nu [-] y/Dj [-] 100 Nu [-] 50 50 1 -3 -2 -1 x/Dj [-] -4 4 -3 -2 -1 x/Dj [-] 4 160 140 100 80 60 60 40 40 20 -3 -2 -1 x/Dj [-] 20 -4 -3 -2 -1 x/Dj [-] 160 140 140 120 120 y/Dj [-] 80 Nu [-] 100 100 80 60 60 40 40 20 -3 -2 -1 x/Dj [-] 20 0 -4 -3 -2 -1 x/Dj [-] 4 -3 -2 -1 x/Dj [-] y/Dj [-] 180 160 140 120 100 80 60 40 20 Nu [-] y/Dj [-] Rej=40000 Nu [-] -4 160 y/Dj [-] Rej=32500 -4 Nu [-] 80 120 y/Dj [-] Nu [-] 100 140 120 y/Dj [-] Rej=25000 160 -4 -3 -2 -1 x/Dj [-] 180 160 140 120 100 80 60 40 20 Nu [-] y/Dj [-] 100 -4 -4 Nu [-] y/Dj [-] 80 Nu [-] 60 -4 Rej=17500 100 y/Dj [-] Rej=10000 Figure 2: Nu distributions for different Rej values and heating elements; different scales are applied for different Rej values effect of boundary layer transition, induced by the shear layer vortices impinging in such region [1] For the investigated conditions, the first Nu peak is always higher than the second one The analysis of Figure 3b shows that Nu distributions obtained on the mesh heating element are similar in shape to the Inconel ones, even if a local minimum is not present at r/Dj=0 and the second peak is less evident It can also be noticed that Nu values are sensibly higher than the Inconel ones, up to 30% in the stagnation region for the Rej=40000 test Since every thermal, fluid-dynamic and geometric conditions were held symmetrical for the two heating elements, such discrepancy can only be attributed to the higher roughness of the mesh In fact, if roughness elements are high enough to protrude through the viscous sublayer, flow perturbation occur in the very nearest region to the wall, and 24 Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 250 250 Inconel Mesh 200 150 Nu [-] Nu [-] 200 100 Re Re10000 j=10000 150 Re Re17500 j=17500 Re Re25000 j=25000 100 Re32500 Re j=32500 50 Re40000 Re j=40000 50 0 r/Dj [-] r/Dj [-] (a) (b) Figure 3: radial (x/Dj=0) Nusselt number distributions for Inconel (a) and mesh (b) heating elements thermal resistance is reduced To verify if this is the case, boundary layer thickness in the stagnation region has been evaluated, based on the relations reported by Martin [3], and compared to Inconel and mesh surface roughness In this case, stagnation dynamic boundary layer thickness was found to range from 251 μm (for the Rej=10000 test) to 125 μm (for the Rej=40000 test), which may correspond to thermal boundary layers of similar magnitude, given the fluid properties Considering these values, it is unlikely Inconel roughness elements to protrude into the thermal sublayer, which allows to assimilate the Inconel heating element to a smooth surface In contrast, mesh roughness is likely of the same order of magnitude of the thermal sublayer thickness, which causes the wall flow to act, if not as fully rough, at least as transitionally rough [10] To validate this hypothesis, an additional test has been performed by covering half of the mesh with a smooth polypropylene adhesive tape, with the aim to reduce surface roughness: indeed, measurements showed tape covered mesh to have Ra=7 μm and Rt=10 μm In this case, the additional thermal resistance and heat flux reduction due to the tape were taken into account by knowing its thickness and thermal conductivity Radial Nu distributions for such test, which was performed at Rej=30000, are reported in Figure and show tape covered mesh to provide heat transfer results comparable with the Inconel ones, taking into account experimental uncertainties This result also demonstrates that the fine mesh can be used as a heating element for heat transfer steady-state tests on smooth surfaces, providing that it is covered by a smooth, nonconductive layer Figure reports Nu values averaged over a circular area with a radius of r/Dj=4 It can be observed relative difference between Inconel and mesh to increase as Rej increases, rising from 9.6% at Rej=10000 to 15.9% at Rej=40000: in fact, boundary layer thickness decreases with an increase in Rej value [3], and so the roughness effect is expected to be stronger Measured values are also compared with Martin correlation [3] for single round nozzles over a flat plate, showing satisfactory agreement for the Inconel values (maximum discrepancy of 7.5% for Rej=10000 test) 250 200 180 160 140 120 Tape covered mesh Inconel 200 Mesh Mesh Inconel Martin correlation [3] 100 Nu [-] Nu [-] 150 80 Nu = 0.1089 Rej0.6609 R² = 0.9987 60 Nu = 0.1576 Rej0.6118 R² = 0.9939 100 40 50 0 r/Dj [-] Figure 4: Nu radial trends for different surfaces, Rej=30000 20 5000 20000 10000 30000 40000 50000 Rej [-] Figure 5: area averaged Nu values Carlo Carcasci et al / Energy Procedia 101 (2016) 18 – 25 Conclusions In the present work, steady-state technique has been applied to evaluate heat transfer distribution under an impinging jet, using two different heating elements: a smooth Inconel sheet (Ra=1.2 μm) and a fine stainless steel mesh (Ra=23 μm) Wide band TLCs have been used to evaluate impingement target surface temperature A fixed cooling geometry has been investigated (Dj=12.5 mm, L/Dj=0.4, H/Dj=2) for jet Reynolds numbers ranging from 10000 to 40000 Nusselt number distributions show a typical heat transfer pattern for jet impingement, with a circular peak in the jet stagnation region and a surrounding ring-shaped high heat transfer zone Nu values increase in every location as Rej increases, as well as the relative intensity of the ring-shaped region with respect to the central peak Differences have emerged in heat transfer for the two heating elements: a local minimum directly in the jet stagnation point and a more definite ring-shaped region have been recorded for the Inconel, while the mesh presents significantly higher Nu values (with increases up to 30% for local values and up to 15.9% for area-averaged values) Moreover, heat transfer differences between Inconel and mesh seem to increase with Rej value This discrepancy has been interpreted as the effect of surface roughness, which enhances heat transfer through the disruption of thermal sublayer As a consequence, stainless steel mesh can be used as heating element to investigate heat transfer on smooth surfaces only if the resulting surface roughness is reduced, e.g through the application of a smooth, nonconducting film over the mesh itself, as demonstrated by a dedicated test Despite this, Inconel substitution with the mesh can be interesting for complex geometries involving extraction holes or slots, since the permeable mesh can be laid upon such orifices without obstructing them, with negligible distortion of the generated heat flux This may result in a simpler postprocessing with respect to the Inconel, and in a greater freedom in the definition of power supply configuration The investigation also highlights that significant heat transfer enhancements can be obtained by increasing surface roughness: this is an aspect to be held in consideration for the design of impingement-based cooling systems References [1] Zuckerman, N.; Lior, N Jet impingement heat transfer: physics, correlations, and numerical modeling Advances in heat transfer, 2006, 39.06: 565-631 [2] Gardon, R.; Akfirat, J C The role of turbulence in determining the heat-transfer characteristics of impinging jets International Journal of Heat and Mass Transfer, 1965, 8.10: 1261-1272 [3] Martin, H Heat and mass transfer between impinging gas jets and solid surfaces In: In: Advances in heat transfer Volume 13 New York, Academic Press, Inc., 1977, p 1-60 1977 p 1-60 [4] Carcasci, C An experimental investigation on air impinging jets using visualisation methods International journal of thermal sciences, 1999, 38.9: 808-818 [5] Andrei, L., Carcasci, C., Da Soghe, R., Facchini, B., Maiuolo, F.,Tarchi, L., Zecchi, S Heat transfer measurements in a leading edge geometry with racetrack holes and film cooling extraction Journal of Turbomachinery, Vol 135, Issue 3, 25 March 2013, Article number 031020 [6] Beniaiche, A., Ghenaiet, A., Carcasci, C., Facchini, B., Heat transfer investigation in new cooling schemes of a stationary blade trailing edge Applied Thermal Engineering, Vol 87, July 2015, Article number 6592, Pages 816-825 [7] Beniaiche, A., Ghenaiet, A., Carcasci, C., Pievarolli, M., Facchini, B., Experimental study of a cooling scheme for a turbine blade trailing edge, Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, Vol 229, Issue 8, December 2015, Pages 832-848 [8] Facchini, B., Maiuolo, F., Tarchi, L., and Coutandin, D Combined effect of slot injection, effusion array and dilution hole on the heat transfer coefficient of a real combustor liner: Part 1—Experimental analysis In: ASME Turbo Expo 2010: Power for Land, Sea, and Air American Society of Mechanical Engineers, 2010 p 753-762 [9] Massini, D., Burberi, E., Carcasci, C., Cocchi, L., Facchini, B., Armellini, A., Casarsa, L., Furlani, L., Effect of rotation on a gas turbine blade internal cooling system: experimental investigation, Proceedings of ASME Turbo Expo (GT2016-57954), 2016 [10] Gabour, L A.; Lienhard, J H Wall roughness effects on stagnation-point heat transfer beneath an impinging liquid jet Journal of heat transfer, 1994, 116.1: 8187 [11] Beitelmal, A.H.; Saad, M.A.; Patel, C.D Effects of surface roughness on the average heat transfer of an impinging air jet International Communications in Heat and Mass Transfer, 2000, 27.1: 1-12 [12] Sesen, M., Demir, E., Izci, T., Khudhayer, W., Karabacak, T., and Kosar, A Submerged jet impingement cooling using nanostructured plates International Journal of Heat and Mass Transfer, 2013, 59: 414-422 [13] Xu, P., Sasmito, A P., Qiu, S., Mujumdar, A S., Xu, L., & Geng, L Heat transfer and entropy generation in air jet impingement on a model rough surface International Communications in Heat and Mass Transfer 72 (2016): 48-56 [14] Chan, T L.; Ashforth-Frost, S.; Jambunathan, K Calibrating for viewing angle effect during heat transfer measurements on a curved surface International Journal of Heat and Mass Transfer, 2001, 44.12: 2209-2223 [15] Lloyd, J R.; Moran, W R Natural convection adjacent to horizontal surface of various planforms Journal of Heat Transfer, 1974, 96.4: 443-447 [16] ANSI/ASME PTC 19.1-1985, Instruments and Apparatus, Part 1, Measurement Uncertainty, ANSI, New York, 1985, 64 [17] Kline, S J., Mcclintock, F A., Describing uncertainties in single-sample experiments Mechanical engineering, 1953, 75.1: 3-8 [18] Viskanta, R Heat transfer to impinging isothermal gas and flame jets Experimental thermal and fluid science, 1993, 6.2: 111-134 25 ... with a flat heating element Given that the aim of this work is to compare the performances of different heating elements, target surface is divided into two regions, one for each impingement. .. Rej=17500 100 y/Dj [-] Rej=10000 Figure 2: Nu distributions for different Rej values and heating elements; different scales are applied for different Rej values effect of boundary layer transition,... an impinging jet, using two different heating elements: a smooth Inconel sheet (Ra=1.2 μm) and a fine stainless steel mesh (Ra=23 μm) Wide band TLCs have been used to evaluate impingement target