Tim Roediger

Time-Resolved Heat Transfer Measurements on the Tip Wall of a Ribbed Channel Using a Novel Heat Flux Sensor—Part I: Sensor and Benchmarks

A novel heat flux sensor was tested that allows for time-resolved heat flux measurements in internal ribbed channels related to the study of passages in gas turbine blades. The working principle of the atomic layer thermopile (ALTP) sensor is based on a thermoelectric field created by a temperature gradient over an yttrium-barium-copper-oxide (YBCO) crystal (the transverse Seebeck effect). The sensors very fast frequency response allows for highly time-resolved heat flux measurements up to the 1 MHz range. This paper explains the design and working principle of the sensor, as well as the benchmark-ing of the sensor for several flow conditions. For internal cooling passages, this novel sensor allows for highly accurate, time-resolved measurements of heat transfer coefficients, leading to a greater understanding of the influence of fluctuations in temperature fields.

Novel Sensor for Fast Heat Flux Measurements

§§ *** A new category of heat flux sensor is presented with a working principle based on the so-called Transverse Seebeck Effect. The capability of this isothermal thin film sensor, also called Atomic Layer Thermopile (ALTP), below 1 μm in thickness, a minimum active surface area of 0.4 x 2 mm is demonstrated in reference to the state of the art technology of several conventional gauges including heat flux microsensors. Because of small size and fast frequency response high spatial and temporal resolution up to almost 1 MHz is attained. The output of the signal is directly proportional to heat flux density and has a linear characteristic for heat flux rates from mW/cm² to MW/m². The threshold limit is about 20 kW/cm² with a duration of 1 ms. A short description about structure and working principle of the sensor is given. Steady and unsteady response, first for calibration purposes, were determined experimentally. Transient response was measured to step changes of imposed radiative flux and a convective heat flux step caused by a traveling shock wave in a shock tube. The ALTP gauge recorded a complete heat flux response with a captured heat flux peak of 1.2 MW/m² in less than 1 μs after a shock passing time of 0.3 μs over the sensor, demonstrating that the ALTP sensor has a frequency response covering DC to 1 MHz. Theoretical prediction of the frequency response is in good agreement with experimental results. Besides the highly time resolved convective heat step rise in the very beginning, the ALTP also detects the resulting wall heat flux assigned to the transitional process in the transient unsteady boundary layer (BL) development downstream of the traveling shock. After the initial laminar BL state during the first 11 μs, indicated by a temporal heat flux decay, following the inverse of square root of time, according Mirels’ theory, a transitional stretched region of about 50 μs is captured with a constant heat flux history before a sudden increase according to the turbulent state is initiated. The heat flux density in the turbulent BL state is in good agreement with values measured simultaneous by conventional sensors. Another example to demonstrate temporal and especially spatial resolution of the ALTP, BL instability studies have been performed in a steady hypersonic conical BL at M = 6. By means of ALTP single point measurements in the cone surface a Second Mode (SM) could be detected between 220-370 kHz, depending on free stream Reynolds number. A second peak in the spectra of the ALTP signal at 430-730 kHz revealed a first harmonic of this SM instability. Comparatively performed mass flux density fluctuation measurement by a hot wire probe in the BL confirmed the SM but not the first harmonic, because of limited bandwidth of the CCA system. On the other hand while the First instability Mode (FM) at lower frequency of about 100 kHz was captured by the hot wire, this FM was not detected as a momentary heat flux density “foot print” by the ALTP. The reason of this fact can be explained by theoretical studies with the linear stability theory, showing that the decisive parameter for the wall heat flux the temperature gradient close to the wall for the first mode is only about 1/10 compared with that of the SM.

Hypersonic Instability Waves Measured Using Fast-Response Heat-Flux Gauges

DOI: 10.2514/1.37026Instability and transition were measured on a 7-degree half-angle sharp cone at zero angle of attack. Surface-mountedheat-fluxgaugeswitha1-MHzfrequencyresponseweremountedinastreamwisearray.Experimentswerecarried out under noisy and quiet Mach-6 flow. Second-mode instability waves and their first harmonics weredetectedundernoisyflowforstagnationpressuresrangingfrom3.9to5.8bar.Underquietflow,however,thesecondmodecould onlybedetectedat8.6bar dueto themuchloweramplitude ofthefluctuations. Theamplification rateswereingoodagreementwithlinearstabilitytheoryfornoisyflow.Underquietflow,maximumgrowthratescouldnotbe determined due to the low signal-to-noise ratio.

Using anisotropic heat flux sensors in aerodynamic experiments

We present the results of comparative measurements of the heat flux to a flat plate in a supersonic flow at a Mach number of M = 6, which were performed using the two following anisotropic heat sensors with different thicknesses of sensor elements: (i) Atomic Layer Thermo Pile (ALTP, Fortech GmbH, Germany) with a thickness of ∼0.5 × 10−6 m and (ii) gradient heat flux sensor (GHFS, St. Petersburg State Polytechnic University, Russia) with a thickness of ∼2 × 10−4 m. The ALTP sensor can be used for directly measuring heat fluxes in processes with a characteristic time above 10−6 s. A method for mathematically processing the GHFS response signal is proposed that allows heat flux oscillations to be revealed in gasdynamic process with a characteristic time on the order of 10−4 s.

Time-Resolved Heat Transfer Measurements on the Tip Wall of a Ribbed Channel Using a Novel Heat Flux Sensor—Part II: Heat Transfer Results

Measurements using a novel heat flux sensor were performed in an internal ribbed channel representing the internal cooling passages of a gas turbine blade. These measurements allowed for the characterization of heat transfer turbulence levels and unsteadiness not previously available for internal cooling channels. In the study of heat transfer, often the fluctuations can be equally as important as the mean values for understanding the heat loads in a system. In this study, comparisons are made between the time-averaged values obtained using this sensor and detailed surface measurements using the transient thermal liquid crystal technique. The time-averaged heat flux sensor and transient TLC results showed very good agreement, validating both methods. Time-resolved measurements were also corroborated with hot film measurements at the wall at the location of the sensor to better clarify the influence of unsteadiness in the velocity field at the wall on fluctuations in the heat flux. These measurements resulted in turbulence intensities of the velocity and heat flux of 20%. The velocity and heat flux integral length scales were about 60% and 35% of the channel width, respectively, resulting in a turbulent Prandtl number of 1.7 at the wall.

Time-Resolved Heat Transfer Measurements on the Tip Wall of a Ribbed Channel Using a Novel Heat Flux Sensor: Part I — Sensor and Benchmarks

A novel heat flux sensor was tested which allows for time-resolved heat flux measurements in internal ribbed channels related to the study of passages in gas turbine blades. The working principle of the Atomic Layer Thermopile (ALTP) sensor is based on a thermoelectric field created by a temperature gradient over an YBCO crystal (the transverse Seebeck effect). The sensors very fast frequency response allows for highly time-resolved heat flux measurements up to the 1 MHz range. This paper explains the design and working principle of the sensor, as well as the benchmarking of the sensor for several flow conditions. For internal cooling passages, this novel sensor allows for highly accurate, time-resolved measurements of heat transfer coefficients, leading to a greater understanding of the influence of fluctuations in temperature fields.Copyright

A novel fast-response heat-flux sensor for measuring transition to turbulence in the boundary layer behind a moving shock wave

To study the local heat transfer rate to the shock tube wall and to observe the transition mechanisms from laminar to turbulent in the shock-induced flow, an investigation is performed by use of a new fast response heat flux gauge. The initial shock tube conditions establish a unit Reynolds number range of more than one order of magnitude, reaching from 0.5 × 106 < Reunit m < 11 × 106. The new sensor possesses a high spatial and temporal resolution with a time constant down to 1 μ s. The temporal resolution allows the detection of the boundary layer transition not only by a rise of the mean heat flux density but also by the detection of the increase of the heat flux fluctuation level.

Time-Resolved Heat Transfer Measurements on the Tip Wall of a Ribbed Channel Using a Novel Heat Flux Sensor: Part II — Heat Transfer Results

Measurements using a novel heat flux sensor were performed in an internal ribbed channel representing the internal cooling passages of a gas turbine blade. These measurements allowed for the characterization of heat transfer turbulence levels and unsteadiness not previously available for internal cooling channels. In the study of heat transfer, often the fluctuations can be equally as important as the mean values for understanding the heat loads in a system. In this study comparisons are made between the time-averaged values obtained using this sensor and detailed surface measurements using the transient thermal liquid crystal technique. The time-averaged heat flux sensor and transient TLC results showed very good agreement, validating both methods. Time-resolved measurements were also corroborated with hot film measurements at the wall at the location of the sensor to better clarify the influence of unsteadiness in the velocity field at the wall on fluctuations in the heat flux. These measurements resulted in turbulence intensities of the velocity and heat flux of about 20%. The velocity and heat flux integral length scales were about 60% and 35% of the channel width respectively, resulting in a turbulent Prandtl number of about 1.7 at the wall.Copyright

Hypersonic instability waves measured on a circular cone at M=12 using fast-response surface heat-flux and pressure gauges

Conical boundary layers are prevalent on many hypersonic vehicles and stability measurements on sharp cones have been conducted since the 1970’s. Schneider [1] gives a review of experimental and numerical studies on the subject and provides an overview of instability mechanisms on circular cones including a comprehensive list of references. Up to now, experimental stability investigations of a conical boundary layer (BL) at free stream Mach numbers larger than M=8 are very sparse. The following experiments present spectral wave amplitude distributions at M=12 measured by a staggered array of surface-mounted, fast-response pressure and ALTP heat-flux gauges. Both gauges cover a frequency range up to 1MHz and allow highly time-resolved experimental studies where conventional measurement techniques like hot-wires cannot be used due to their limited durability, limited overheat ratio and temporal resolution. The data quality allows the calculation of amplificationrates, which are compared with predictions from LST computations (2D) and amplification rates obtained with the same model in two different Mach-6 facilities.