Impingement cooling is one of the powerful cooling methods in high-temperature devices. For the gas turbine applications, impingement cooling is commonly applied in the transition piece of a combustor and in the leading edge, suction and pressure sides of a turbine blade/vane. In the suction side and pressure side, impingement cooling is applied as a form of an array jet. However, due to the small gap between the jet hole and target surface, the wall jet faces a crossflow inside of the gap. This crossflow has an adverse effect on the jets and deteriorates the heat/mass transfer performance. Therefore, several studies have been conducted to minimized the crossflow effect. The present study also investigated the effect of crossflow reduction in the gap by having a castellated hole plate. The heat/mass transfer was measured using the naphthalene sublimation method. Heat/mass transfer data are compared among three different cases. One is the baseline case which is simple array impinging jets. Others are the castellated cases with and without rib structures on the target wall. Jet-to-jet spacing (s/d) and jet-to-target spacing (z/d) are selected as geometrical variables. Also, the experiments were conducted for the Reynolds numbers (based on jet hole diameter) of 5,000, 15,000 and 30,000. The baseline case was named as B case, the castellated case without rib structure as C case and with rib structure as CR case. Both C and CR cases showed the crossflow reduction effect and resulted high and similar Nusselt number values.

Thermal design of dual circulating fluidized bed reactors for carbon dioxide (CO 2 ) capture was carried out. To handle large heat duties for regeneration, a thin rectangular reactor was proposed. For feasible thermal design, the effect of varying reactor thickness on the gas-solid flow and heat transfer of the thin rectangular fluidized bed was investigated. Reactor thickness of 10, 30, and 60 mm was tested. Numerical simulations were conducted to analyze the pressure difference, solid particle hold-up distribution, particle velocity, granular temperature, and heat transfer in detail. According to our results, when the reactor is between 10 mm and 30 mm thick, a large solid hold-up occurs adjacent to the narrow wall. This causes a large pressure difference due to the wall effect. Furthermore, the particle velocities were analyzed to evaluate that there is the two-dimensional (2D) particle mixing behaviors. On the other hand, in the case of reactors with a thickness of 60 mm, tuning flows occur adjacent to the narrow wall. This reduced the pressure difference and the three-dimensional (3D) particle mixing behaviors. This difference in particle behavior affected heat transfer. In the case of reactor thicknesses between 10 mm and 30 mm, the heat transfer increased with the reactor thickness. In particular, the heat transfer at the narrow wall of the reactor with a thickness of 10 mm was extremely low due to the low particle mixing. On the other hand, there was more heat transfer with a thickness at the 60 mm wall, despite the low solid hold-up.

Different types of vortices, such as horseshoe vortex, passage vortex, corner vortex, cause high heat transfer distributions and complex heat transfer characteristics at the endwall of turbine blades. In addition, the endwall heat transfer is also affected when the main flow is highly turbulent and wakes are generated by the trailing edge of the vane. Detailed heat transfer measurements are necessary to protect the blades under harsh and complex flow conditions. Therefore, this study investigated the heat transfer characteristics on the blade endwall under flow conditions that simulate high turbulence intensity of the main flow and the generation of wakes by the trailing edge of the vane. The endwall heat transfer was measured using the naphthalene sublimation method. A turbulence generating grid was installed in a linear cascade to simulate the main flow with high turbulence intensity and a wake generator with a rod bundle was used to simulate the wakes generated by the trailing edge of the vane. In the case of high turbulence intensity without wakes, the main flow with high turbulence intensity (Turbulence intensity, T.I = 7.5%) had little impact on the effect of the horseshoe vortex and passage vortex on the heat transfer characteristics. However, increasing turbulence caused the endwall heat transfer to decrease near the pressure side of the blade and increase near the suction side of the blade. On the other hand, the wakes resulted in heat transfer characteristics similar to high turbulence intensity, but decreased heat transfer by horseshoe vortex and passage vortex. The endwall heat transfer distributions were similar regardless of the turbulence intensity (T.I = 1.2%, 7.5%) in the cases with wakes (Rod passing Strouhal number, S = 0.3). This means that the flow condition of S = 0.3 has a more significant influence on the endwall heat transfer than that of T.I = 7.5%.

Film cooling effectiveness and heat transfer were measured in squealer tip configurations on the blade tip surface. Three different shelf squealer tip geometries were studied: conventional, vertical, and inclined. The experiment was carried out in a wind tunnel with an inlet mainstream Reynolds number, based on the axial chord length of the blade, of 140,000. The experiments were conducted in five blades in linear cascade with an averaged turbulence intensity of 8.5%. The film cooling effectiveness and heat transfer coefficient on the tip surface were obtained using the transient IR thermography technique. For the pressure side film cooling holes, averaging blowing ratios (M) of 1.0 and 2.0 were set. The results showed the film cooling effectiveness distributions on the tip surface. Owing to the mainstream, the cooling effect appeared after x / C x = 0.15 and the film cooling effectiveness tended to increase toward downstream of the trailing edge. Additionally, the heat transfer distributions were investigated regarding the film cooling holes. In the presence of film cooling holes, the heat transfer distribution had more uniformity than without them on the pressure side. As the blowing ratio increased from 1 to 2, the heat transfer was decreased on the tip surface. The heat transfer ratio represented the change of heat transfer distribution with and without film cooling holes. Those of results were compared in three squealer tip geometries. The overall area-averaged net heat flux reduction (NHFR) levels on the tip surface were enhanced as the blowing ratio increased. The NHFR of the shelf squealer tip configurations was better than that with the conventional squealer tip.

An experimental study has been conducted to investigate the heat-transfer characteristics of blade tips and shrouds with and without unsteady wakes. Depending on the presence of unsteady wakes, the local heat/mass-transfer coefficients of the tip and shroud were measured using the naphthalene sublimation method. Wakes from unsteady blades were modeled as wakes generated from moving cylindrical rod bundles. Test conditions were set to the Reynolds number of 100,000, based on an inlet velocity of 11.4 m/s and the axial chord length. The Strouhal number was varied from 0 to 0.22. For St = 0, high heat/mass-transfer coefficients appeared in regions where various flow patterns, such as flow reattachment, swirling flow, and vortexes, occurred. For St = 0.22, the heat/mass-transfer distributions of the tip and shroud were changed due to the unsteady wakes. Unsteady wakes made high turbulence intensity of leakage flow and flow patterns such as flow reattachment, swirling flow, and tip leakage vortex in the tip and shroud were changed and dispersed. There were also variations in the pitch-wise averaged Sherwood number of the blade tip and shroud on the presence of the unsteady wakes due to vortex shedding and dispersed flow patterns. Thus, considering the effects of unsteady wakes on the heat transfer of the blade tip and shroud, proper cooling designs for blade tips and shrouds should be investigated and adopted for effective cooling of gas turbine blades.

Multi-layered impingement/effusion cooling is an advanced cooling configuration that combines impingement jet cooling, pin cooling, and effusion cooling. The arrangement of the pins is a critical design factor because of the complex heat transfer in the internal structure. Therefore, it is important to measure the local heat transfer at all internal surfaces as a function of the pin spacing. In this study, a naphthalene sublimation method was employed to measure the details of the heat/mass transfer at the internal surfaces, including the injection plate, effusion plates, and the pins. An staggered array of holes was formed at the injection plate and effusion plates where the ratio of the height to the diameter of the pins, h/d , was fixed at 0.25. The ratio of the pin spacing to the diameter, s p /d , was varied in the range 1.5≤ s p /d ≤6, and the Reynolds number based on the hole diameter was 3000. As a result, a vortex ring formed near the pin, leading to re-impingement flows in the narrow channel. The jet flow impinged strongly on the pin, resulting in a large heat transfer region at each surface. The total average Sherwood number with s p /d =1.5 was larger than that with s p /d =6 by a factor of 1.5.

We investigated the local heat and mass transfer on the endwall surface at different Strouhal numbers. The Strouhal number represents the interaction between the turbine stator and rotor. Thus, characteristics of flow and heat transfer on the endwall surface change for different Strouhal numbers. The experiment was performed in five-bladed linear cascade blades with moving cylindrical rods simulating unsteady wakes effects. The Reynolds number which was based on the blade cord length and inlet velocity was 100,000. The range of the Strouhal number varies from 0 to 0.22 to investigate the effect of unsteady wakes on the endwall surface. The local heat and mass transfer were measured on the endwall surface using the naphthalene sublimation method. The results showed that the local heat transfer characteristics changed depending on the Strouhal number. In the steady case, the occurrence of a horseshoe vortex, passage vortex and corner vortex caused non-uniform heat transfer on the endwall surface. However, in the unsteady case, the unsteady wake effect caused more uniform heat transfer on the endwall surface. As the Strouhal number increased, heat transfer increased and became more uniform compared with the steady case. Thus, an appropriate cooling system is necessary for stator and rotor endwalls in situations of real gas turbine operation.

Film cooling is a cooling method used to protect the hot components of a gas turbine from high temperature conditions. For this purpose, high and uniform film cooling effectiveness is required to protect the vanes/blades from excessive thermal stress. Backward injection is proposed as one of the methods for the improvement of film cooling effectiveness. In this study, experiments were performed to investigate the effect of backward injection on film cooling effectiveness, using pressure sensitive paint (PSP) method. Four experimental configurations were composed of forward and backward injection cylindrical holes. The cylindrical holes were aligned in two staggered rows with pitch ( p ) of 6 d and row spacing ( s ) of 3 d . The injection angles (α) of the cylindrical holes were 35° and 145° for forward and backward injection, respectively. The blowing ratios ( M ) ranged from 0.5 to 2.0 and the density ratio ( DR ) was about 1. The results indicate that backward injection enhanced not only film cooling effectiveness but also the lateral cooling uniformity. At a high blowing ratio, all configurations demonstrated higher film cooling effectiveness with backward injection than with only forward injection; thus, the dispersion of the backward injection jets enhanced the lateral coverage over wide areas. Configuration, in particular, arranged with forward injection in the first row and backward injection in the second row, obtained the highest film cooling effectiveness among the four cases studied, due to the dispersion of the backward injection jets and the coolant supply from the forward injection jets at a high blowing ratio.

We report an investigation of the total cooling effectiveness of a film cooled surface with staggered array impingement jet cooling using infra-red thermography. Heat transfer experiments were carried out using three film cooled test plates of different thermal conductivities: stainless steel (with a thermal conductivity, k = 13.4 W/mK), Corian ® ( k = 1 W/mK), and polycarbonate ( k = 0.2 W/mK). The effects of conduction through the test plates and convective heat transfer due to the arrayed impingement jets were analyzed. The inclination angle of the film cooling holes was 35° and that of the impingement jet holes was 90°. The film and impingement jet holes on each plate were arranged in a staggered pattern, and the film cooling holes and impingement jet holes were also positioned in a staggered pattern. The jet Reynolds number based on the hole diameter was Re jet = 3,000 and the equivalent blowing rate was M = 0.3. The ratio of the target surface height to the hole diameter was varied in the range 1 < H/d < 5. The diameter of both the film cooling holes and impingement jet holes was 5 mm. The total cooling effectiveness was investigated with and without the impingement jets. When the impingement jets were added to the internal cooling, the averaged total cooling effectiveness was enhanced about 8.4%. The stainless steel plate was found to exhibit better cooling performance with more uniform temperature distribution. The total cooling effectiveness was increased up to 0.87 in the stainless steel plate, and the maximum deviation of total cooling effectiveness in the stainless steel was reduced to 85% from that in polycarbonate plate along the lateral direction. The total cooling effectiveness was related to the Biot number of the film cooled plate, however, the effect of the H/d ratio was not significant.

The present study investigates heat/mass transfer for flow through perforated plates for application to combustor wall and turbine blade film cooling. The experiments are conducted for hole length-to-diameter ratios of 0.68 to 1.5, for hole pitch-to-diameter ratios of 1.5 and 3.0, for gap distance between two parallel perforated plates of 0 to 3 hole diameters, and for Reynolds numbers of 60 to 13,700. Local heat/mass transfer coefficients near and inside the cooling holes are obtained using a naphthalene sublimation technique. Detailed knowledge of the local transfer coefficients is essential to analyze thermal stress in turbine components. The results indicate that the heat/mass transfer coefficients inside the hole surface vary significantly due to flow separation and reattachment. The transfer coefficient near the reattachment point is about four and half times that for a fully developed circular tube flow. The heat/mass transfer coefficient on the leeward surface has the same order as that on the windward surface because of a strong recirculation flow between neighboring jets from the array of holes. For flow through two in-line layers, the transfer coefficient affected by the gap spacing is approximately 100% higher on the windward surface of the second wall and is about 20% lower on the inside hole surface than that with a single layer. The transfer coefficient on the leeward surface is not affected by upstream flow conditions due to probably strong recirculation in the wake flow.

Effusion flow from a blade tip region, not only protect the blade tip directly against hot gas but also influence the cooling performance of the internal cooling passage. The present study investigated the heat/mass transfer of a rotating two-passage rectangular duct using three different types of effusion hole arrays on the internal tip surface. The duct inlet was 80 mm × 16 mm, and the hydraulic diameter was 26.67 mm. The Reynolds number, which is based on the hydraulic diameter, was 10,000. The mass flow rate of the effusion flow was 10% of the total inlet mass flow. The naphthalene sublimation method was used to determine the local heat/mass transfer coefficient. The results indicated that effusion flow enhanced the heat transfer on the tip surface: they decreased the recirculation flow and counter-rotating flow in the turning region of the internal passage. The effusion holes enhanced the local heat transfer around holes due to local secondary flows such as tripping flows. The rotating channel had a different heat transfer distribution compared to the stationary channel due to Coriolis and centrifugal forces.

This study was an experimental investigation of the effect of an intersecting rib on heat/mass transfer performance in rectangular channels with angled ribs and different aspect ratios. In a rib-roughened channel with angled ribs, heat/mass transfer performance deteriorates as the channel aspect ratio increases, since the vortices induced by angled ribs diminish with increasing aspect ratio. A longitudinal rib that bisects the angled ribs is suggested to overcome this disadvantage. The heat transfer performance of angled rib configurations with a 60° attack angle were tested with and without an intersecting rib using naphthalene sublimation method. The channel aspect ratio is varied from 1 to 4. When the intersecting rib was present, additional vortices were generated at every point of intersection with the angled ribs. Thus the heat/mass transfer performance was significantly enhanced for all channel aspect ratios when an intersecting rib was added to an ordinary angled rib configuration.

The next generation aircraft combustor liner will be operating in more severe conditions. This means that the current cooling system needs significant amounts of cooling air to maintain cooling intensity. The present study investigates experimentally the total cooling effectiveness of an integrated impingement/effusion cooling system (thin perforated laminate plate) and effusion cooling system (single plate) at the same blowing ratio of 0.2 to 1.2. The infrared thermography method was employed to evaluate total cooling effectiveness and to determine the fully developed region of cooling performance. The holes arrays on both plates are 13 × 13 and the centers formed a square pattern (i.e., an in-line array). The perforated laminate plate is constructed of three layers and pins that were installed between the layers. In order to avoid increasing the thickness and volume, the layer thickness-to-hole diameter ratio was 0.29, and the pin height-to-hole diameter ratio, which is equivalent to the gap between the plates, was 0.21. The single plate had the same total plate thickness-to-hole diameter, but was composed of only one layer. As a result, the total cooling effectiveness of the laminate plate is 47% ∼ 141% better than single plate depending on the blowing ratio. Also, a fully developed region appears on the 2nd or 3th row of holes.

The present study is conducted to investigate the local heat/mass transfer characteristics for flow through perforated plates. A naphthalene sublimation method is employed to determine the local heat/mass transfer coefficients on the effusion plate. Two parallel perforated plates are arranged in two different configurations: staggered and shifted in one direction. The experiments are conducted for hole pitch-to-diameter ratios of 6.0, for gap distance between the perforated plates of 0.33 to 10 hole diameters, and for Reynolds numbers of 5,000 to 12,000. The result shows that the high transfer region is formed at stagnation region and at the mid-line of the adjacent impinging jets due to secondary vortices and flow acceleration to the effusion hole. For flows through the perforated plates, the mass transfer rates on the surface of the effusion plate are about six to ten times higher than for effusion cooling alone (single perforated plate). In general, higher heat/mass transfer is obtained with smaller gap distance between two perforated plates.

An experimental investigation is conducted to improve a slot film cooling system used for the cooling of a gas turbine combustor liner. The tangential slots are constructed of discrete holes with different injection types which are the parallel, vertical and combined to the slot lip. The investigation is focused on the coolant supply systems of normal-, inline-, and counter-flow paths to the mainstream flow direction. A naphthalene sublimation technique has been employed to measure the local heat/mass transfer coefficients in a slot wall with various injection types and coolant feeding directions. A numerical simulation is also conducted to help understand the flow patterns inside the slot for different injection types. The velocity distributions at the exit of slot lip for the parallel and vertical injection types are fairly uniform with mild periodical patterns with respect to the injection hole positions. However, the combined injection type increases the nonuniformity of flow distribution with the period equaling twice that of hole-to-hole pitch due to splitting and merging of the ejected flows. The dimensionless temperature distributions at the slot exits differ little with blowing rates, injection types and secondary flow conditions. In the results of heat/mass transfer measurements, the best cooling performance inside the slot is obtained with the vertical injection type among the three different injection types due to the effects of jet impingement. The lateral distributions of heat/mass transfer coefficients with the inline- and counter-flow paths are more uniform than the normal-flow path. The average heat/mass transfer coefficients with the injection holes are about 2∼5 times higher than that of a smooth two-dimensional slot path.

This study investigates the local heat/mass transfer characteristics on the stationary shroud with blade tip clearances for flat tip geometry. A large scale linear cascade is used and the relative motion between the blade and shroud is neglected in this study. A naphthalene sublimation method is employed to determine the detailed local heat/mass transfer coefficients on the shroud surface. The geometry of blade tip used in this study is flat and the tip clearance varies from 0.66% to 2.85% of the blade chord length. The flow enters the gap between the blade tip and shroud at the pressure side due to the pressure difference. Therefore, the heat/mass transfer characteristics on the shroud are changed significantly from those for no tip clearance. High heat/mass transfer region is observed along the pressure side of blade due to the entrance effect and the acceleration of the tip gap flow. Complex heat transfer patterns on the shroud are observed in the region where the blade tip and shroud are overlapped due to the flow separation and reattachment. Then, the heat/mass transfer coefficients on the shroud increase along the suction side of blade because tip leakage vortices are generated with interacting the main flow. The experimental results show that the heat/mass transfer characteristics are changed significantly with the gap distance between the tip of turbine blade and the shroud. However, the turbulence intensity of incoming flow has little influence on the heat/mass transfer coefficients on the shroud with tip clearance.

An experimental study has been conducted to investigate the heat/mass transfer characteristics within film cooling holes of square and rectangular cross-section. The experiments for this study have been performed using a naphthalene sublimation method, and the flow field has been analyzed by numerical calculation using a commercial code (FLUENT). The rectangular cross-section has the aspect ratio of 2 and the same hydraulic diameter as the square cross-section. A duct flow enters into a film cooling hole in a cross-direction. For the film cooling hole with square cross-section, it is observed that the reattachment of separated flow and the vortices within the hole enhance considerably the heat/mass transfer around the hole entrance region. The heat/mass transfer on the leading edge side of hole exit region increases as the blowing rates decrease because the mainflow induces a secondary vortex. Heat/mass transfer patterns within the film cooling hole are changed slightly with the various Reynolds numbers. For the film cooling hole with rectangular cross-section, overall heat/mass transfer characteristics are similar with those for the square cross-section. However, heat/mass transfer on the leading edge side of hole entrance region has two peak regions due to split flow reattachment, and heat/mass transfer on the leading edge side of hole exit region is less sensitive to the blowing ratios than the square cross-section case.