Research Papers: Gas Turbines: Aircraft Engine

J. Eng. Gas Turbines Power. 2014;136(8):081201-081201-10. doi:10.1115/1.4026810.

Multiple volcanoes erupt yearly propelling volcanic ash into the atmosphere and creating an aviation hazard. The plinian eruption type is most likely to create a significant aviation hazard. Plinian eruptions can eject large quantities of fine ash up to an altitude of 50,000 m (164,000 ft). While large airborne particles rapidly fall, smaller particles at reduced concentrations drift for days to weeks as they gradually descend and deposit on the ground. Very small particles, less than 1 μm, can remain aloft for years. An average of three aircraft encounters with volcanic ash was reported every year between 1973 and 2003. Of these, eight resulted in some loss of engine power, including a complete shutdown of all four engines on a Boeing 747. However, no crashes have been attributed to volcanic ash. The major forms of engine damage caused by volcanic ash are: (1) deposition of ash on turbine nozzles and blades due to glassification (2) erosion of compressor and turbine blades (3) carbon deposits on fuel nozzles. The combination of these effects can push the engine to surge and flame out. If a flame out occurs, engine restart may be possible. Less serious engine damage can also occur. In most cases the major damage will require an engine overhaul long before the minor damage becomes an operational issue, but under some conditions no sign of volcanic ash is evident and the turbine cooling system blockage could go unnoticed until an engine inspection is performed. Several organizations provide aircrew procedures to respond to encounters with a volcanic ash cloud. If a volcanic ash encounter is suspected, then an engine inspection, including borescope, should be performed with particular attention given to the turbine cooling system.

Commentary by Dr. Valentin Fuster

Research Papers: Gas Turbines: Combustion, Fuels, and Emissions

J. Eng. Gas Turbines Power. 2014;136(8):081501-081501-10. doi:10.1115/1.4026809.

Nonuniform combustor outlet flows have been demonstrated to have significant impact on the first and second stage turbine aerothermal performance. Rich-burn combustors, which generally have pronounced temperature profiles and weak swirl profiles, primarily affect the heat load in the vane but both the heat load and aerodynamics of the rotor. Lean burn combustors, in contrast, generally have a strong swirl profile which has an additional significant impact on the vane aerodynamics which should be accounted for in the design process. There has been a move towards lean burn combustor designs to reduce NOx emissions. There is also increasing interest in fully integrated design processes which consider the impact of the combustor flow on the design of the high pressure vane and rotor aerodynamics and cooling. There are a number of current large research projects in scaled (low temperature and pressure) turbine facilities which aim to provide validation data and physical understanding to support this design philosophy. There is a small body of literature devoted to rich burn combustor simulator design but no open literature on the topic of lean burn simulator design. The particular problem is that in nonreacting, highly swirling and diffusing flows, vortex instability in the form of a precessing vortex core or vortex breakdown is unlikely to be well matched to the reacting case. In reacting combustors the flow is stabilized by heat release, but in low temperature simulators other methods for stabilizing the flow must be employed. Unsteady Reynolds-averaged Navier–Stokes and large eddy simulation have shown promise in modeling swirling flows with unstable features. These design issues form the subject of this paper.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081502-081502-7. doi:10.1115/1.4026655.

Premixed compression ignition (CI) combustion has attracted increasing research effort recently due to its potential to achieve both high thermal efficiency and low emissions. Dual-fuel strategies for enabling premixed CI have been a focus using a low-reactivity fumigant and direct diesel injection to control ignition. Alternative fuels like hydrogen and ethanol have been used as fumigants in the past but typically with diesel injection systems that did not allow the same degree of control or mixing enabled by modern common rail systems. In this work, we experimentally investigated hydrogen, ethanol, and gasoline as fumigants and examined three levels of fumigant energy fraction (FEF) using gasoline over a large, direct diesel injection timing range with a single-cylinder diesel engine. It was found that the operable diesel injection timing range at constant FEF was dependent on the fumigant's propensity for autoignition. Peak indicated gross cycle efficiency occurred with advanced diesel injection timing and aligned well with combustion phasing near top dead center (TDC), as we found in an earlier work. The use of hydrogen as a fumigant resulted in very low hydrocarbon (HC) emissions compared with ethanol and gasoline, establishing that they mainly result from incomplete combustion of the fumigated fuel. Hydrogen emissions were independent of diesel injection timing, and HC emissions were strongly linked to combustion phasing, giving further indication that squish and crevice flows are responsible for partially burned species from fumigation combustion.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081503-081503-13. doi:10.1115/1.4026803.

The first vane leading edge film cooling is challenging because of the highest thermal load and the complex flow interaction between the hot mainstream gas and the coolant flow. This interaction varies significantly from the stagnation region to the regions of high curvature and acceleration further downstream. Additionally, in industrial gas turbines with multiple combustor chambers around the annulus the first vane leading edges may also be exposed to large wake disturbances shed from the upstream combustor walls. The influence of these vortical structures on the first vane leading edge film cooling is numerically analyzed in this paper. In order to assess the capabilities of the flow solver TBLOCK to simulate these complex interactions an experimental test case is modeled numerically. The test case is available in the open literature and consists of a cylindrical leading edge and two rows of film cooling holes representative of industrial practice. A LES turbulence modeling strategy with the WALE subgrid scale (SGS) model is applied and compared against experimental results. Based on this validation it is decided to analyze also the wake–leading edge interaction, dominated by large scale unsteady vortical structures, using the same WALE subgrid scale LES model. The initial flow domain with the cylindrical leading edge and cooling holes is extended to incorporate the effect of the combustor wall, which is modeled as a flat plate with a square trailing edge. The location and the size of the plate are scaled to be representative of industrial practice: the plate is located upstream from the leading edge at a distance twice the leading edge diameter, and the thickness of the plate is one half of the leading edge diameter. Two different clockwise positions of the vertical combustor wall model were investigated and compared with the datum configuration: the former where the axis of the plate and the leading edge are aligned (central wake location), the latter with the combustor wall circumferentially shifted up by a quarter of the leading edge diameter (circumferentially shifted wake location). Numerical predictions show that the shed vortices from the combustor wall trailing edge have a highly detrimental effect on the leading edge film cooling by periodically removing the coolant flow from the leading edge surface. This results in an increased unsteady thermal load. These negative effects are less significant in the case of circumferentially shifted wake, due to the combined action of both shed vortices.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081504-081504-8. doi:10.1115/1.4026806.

The laminar flamelet model (LFM) (Peters, 1986, “Laminar Diffusion Flamelet Models in Non-Premixed Combustion,” Prog. Energy Combust. Sci., 10, pp. 319–339; Peters, “Laminar Flamelet Concepts in Turbulent Combustion,” Proc. Combust. Inst., 21, pp. 1231–1250) represents the turbulent flame brush using statistical averaging of laminar flamelets whose structure is not affected by turbulence. The chemical nonequilibrium effects considered in this model are due to local turbulent straining only. In contrast, the flamelet-generated manifold (FGM) (van Oijen and de Goey, 2000, “Modeling of Premixed Laminar Flames Using Flamelet-Generated Manifolds,” Combust. Sci. Technol., 161, pp. 113–137) model considers that the scalar evolution; the realized trajectories on the thermochemical manifold in a turbulent flame are approximated by the scalar evolution similar to that in a laminar flame. This model does not involve any assumption on flame structure. Therefore, it can be successfully used to model ignition, slow chemistry, and quenching effects far away from the equilibrium. In FGM, 1D premixed flamelets are solved in reaction-progress space rather than physical space. This helps better solution convergence for the flamelets over the entire mixture fraction range, especially with large kinetic mechanisms at the flammability limits (ANSYS FLUENT 14.5 Theory Guide Help Document, http://www.ansys.com). In the present work, a systematic comparative study of the FGM model with the LFM for four different turbulent diffusion/premixed flames is presented. The first flame considered in this work is methane-air flame with dilution air at the downstream. The second and third flames considered are jet flames in a coaxial flow of hot combustion products from a lean premixed flame called Cabra lifted H2 and CH4 flames (Cabra, et al., 2002, “Simultaneous Laser Raman-Rayleigh-LIF Measurements and Numerical Modeling Results of a Lifted Turbulent H2/N2 Jet Flame in a Vitiated Coflow,” Proc. Combust. Inst., 29(2), pp. 1881–1888; Lifted CH4/Air Jet Flame in a Vitiated Coflow, http://www.me.berkeley.edu/cal/vcb/data/VCMAData.html) where the reacting flow associated with the central jet exhibits similar chemical kinetics, heat transfer, and molecular transport as recirculation burners without the complex recirculating fluid mechanics. The fourth flame considered is a Sandia flame D (Barlow et al., 2005, “Piloted Methane/Air Jet Flames: Scalar Structure and Transport Effects,” Combust. Flame, 143, pp. 433–449), a piloted methane-air jet flame. It is observed that the simulation results predicted by the FGM model are more physical and accurate compared to the LFM in all the flames presented in this work. The autoignition-controlled flame lift-off is also captured well in the cases of lifted flames using the FGM model.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081505-081505-11. doi:10.1115/1.4026812.

This paper investigates the effect of a cetane improver on the autoignition characteristics of Sasol IPK in the combustion chamber of the ignition quality tester (IQT). The fuel tested was Sasol IPK with a derived cetane number (DCN) of 31, treated with different percentages of Lubrizol 8090 cetane improver ranging from 0.1 to 0.4%. Tests were conducted under steady state conditions at a constant charging pressure of 21 bar. The charge air temperature before fuel injection varied from 778 to 848 K. Accordingly, all the tests were conducted under a constant charge density. The rate of heat release was calculated and analyzed in detail, particularly during the autoignition period. In addition, the physical and chemical delay periods were determined by comparing the results of two tests. The first was conducted with fuel injection into air according to ASTM standards where combustion occurred. In the second test, the fuel was injected into the chamber charged with nitrogen. The physical delay is defined as the period of time from start of injection (SOI) to point of inflection (POI), and the chemical delay is defined as the period of time from POI to start of combustion (SOC). Both the physical and chemical delay periods were determined under different charge temperatures. The cetane improver was found to have an effect only on the chemical ID period. In addition, the effect of the cetane improver on the apparent activation energy of the global combustion reactions was determined. The results showed a linear drop in the apparent activation energy with the increase in the percentage of the cetane improver. Moreover, the low temperature (LT) regimes were investigated and found to be presented in base fuel, as well as cetane improver treated fuels.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081506-081506-11. doi:10.1115/1.4026861.

The current work involves the validation of presumed shape multi-environment Eulerian probability density function (PDF) transport method (MEPDF) using direct quadrature method of moments (DQMOM)-interaction by exchange with mean (IEM) approach for modeling turbulence chemistry interactions in nonpremixed combustion problems. The joint composition PDF is represented as a collection of finite number of Delta functions. The PDF shape is resolved by solving the governing transport equations for probability of occurrence of each environment and probability-weighted mass fraction of species and enthalpy in Eulerian frame for each environment. A generic implementation of the MEPDF approach is carried out for an arbitrary number of environments. In the current work, the MEPDF approach is used for a series of problems to validate each component of MEPDF approach in an isolated manner as well as their combined effect. First of all, a nonreactive turbulent mixing problem with two different Reynolds numbers (Re = 7000 and 11,900) is used for validation of the mixing and correction terms appear in the MEPDF approach. The second problem studied is a diffusion flame with infinitely fast chemistry having an analytical solution. The reaction component is validated by considering a 1D premixed laminar flame. In order to validate the combined effect of mixing and turbulence chemistry interactions, two different turbulent nonpremixed problems using global one-step chemistry are used. The first reactive problem used is H2 combustion (DLR Flame H3), while the second reactive validation case is a pilot-stabilized CH4 flame. The current predictions for all validation problems are compared with experimental data or published results. The study is further extended by modeling a turbulent nonpremixed H2 combustion using finite-rate chemistry effects and radiative heat transfer. The current model predictions for different flame lengths as well as minor species are compared with experimental data. The current model gave excellent predictions of minor species like OH. The differences in the current predictions with experimental data are discussed.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):081507-081507-13. doi:10.1115/1.4026862.

Spark assist (SA) has been demonstrated to extend the operating limits of homogeneous charge compression ignition (HCCI) modes of engine operation. This experimental investigation focuses on the effects of 100% indolene and 70% indolene/30% ethanol blends on the ignition and combustion properties during SA HCCI operation. The spark assist effects are compared to baseline HCCI operation for each blend by varying spark timing at different fuel/air equivalence ratios ranging from Φ = 0.4–0.5. High speed imaging is used to understand connections between spark initiated flame propagation and heat release rates. Ethanol generally improves engine performance with higher net indicated mean effective pressure (IMEPn) and higher stability compared to 100% indolene. SA advances phasing within a range of ∼5 crank angle degrees (CAD) at lower engine speeds (700 rpm) and ∼11 CAD at higher engine speeds (1200 rpm). SA does not affect heat release rates until immediately (within ∼5 CAD) prior to auto-ignition. Unlike previous SA HCCI studies of indolene fuel in the same engine, flames were not observed for all SA conditions.

Commentary by Dr. Valentin Fuster

Research Papers: Gas Turbines: Structures and Dynamics

J. Eng. Gas Turbines Power. 2014;136(8):082501-082501-6. doi:10.1115/1.4026905.

A hydrodynamic rolling hybrid bearing (HRHB) assembled coaxially by a rolling bearing and hydrodynamic bearing is developed to achieve two functions at low and high speeds. At low speeds, the rolling bearing of the HRHB can be utilized to avoid wear in the hydrodynamic bearing. While at high speeds, the rotor is entirely supported by the hydrodynamic bearing, keeping away the interference of the rolling bearing. However, because the HRHB is mounted coaxially by two bearings, a misalignment cannot be avoided. This can lead two unexpected consequences: either the malfunction of the rolling bearing at low speeds or the interference of rolling bearing applied on the rotor. In this paper, a computational method to calculate the maximum allowable misalignment based on the noninterference conditions of the locus of the shaft center and rolling bearing at high speeds is proposed, and the influence of the rotating speed, the spindle, and bearing structural parameters on the maximum allowable misalignment is also analyzed. The results show that the locus of the maximum allowable misalignment forms a circle along the circumferential direction. The noninterference conditions are satisfied when the maximum allowable misalignment is inside the circle.

Commentary by Dr. Valentin Fuster

Research Papers: Gas Turbines: Turbomachinery

J. Eng. Gas Turbines Power. 2014;136(8):082601-082601-10. doi:10.1115/1.4026621.

Adjustable inlet guide vanes (IGVs) and variable speed drivers are known as providing process compressors with an effective regulation all throughout the operating envelope of the machine. A large amount of work exists in literature reporting the successful control of multistage centrifugal compressors by means of IGVs or speed separately, while a few studies document the combined use of both devices and their effect on compressor performance. The present paper details the off-design behavior of a multistage centrifugal compressor equipped with both types of control. It is shown that classical IGVs' advantage in extending the operating envelope of a fixed speed multistage compressor tends to reduce when speed regulation is active too. In this sense, the average level of peripheral Mach numbers inside the compressor may be interpreted as a sort of threshold since it deeply affects the stage mismatching at off-design conditions. This consideration is corroborated by a number of applications in a wide range of design peripheral Mach numbers. Based on those cases, the paper reviews the general effectiveness of the combined regulation, thus outlining some general rules of thumb for the choice of the optimal control device for a multistage centrifugal compressor.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):082602-082602-8. doi:10.1115/1.4026610.

In this paper, a dynamic neural network (DNN) based on robust identification scheme is presented to determine compressor surge point accurately using sensor fault detection (FD). The main innovation of this paper is to present different and complementary technique for surge suppressing studies within sensor FD. The proposed method aims to utilize the embedded analytical redundancies for sensor FD, even in the presence of uncertainty in the compressor and sensor noise. The robust dynamic neural network is developed to learn the input–output map of the compressor for residual generation and the required data is obtained from the compressor Moore–Greitzer simulated model. Generally, the main drawback of DNN method is the lack of systematic law for selecting of initial Hurwitz matrix. Therefore, the subspace identification method is proposed for selecting this matrix. A number of simulation studies are carried out to demonstrate the advantages, capabilities, and performance of our proposed FD scheme and a worthwhile direction for future research is also presented.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):082603-082603-15. doi:10.1115/1.4026845.

Hot and harsh environments, sometimes experienced by gas turbine airfoils, can create undesirable effects such as clogging of the cooling holes. Clogging of the cooling holes along the trailing edge of an airfoil on the tip side and its effects on the heat transfer coefficients in the cooling cavity around the clogged holes is the main focus of this investigation. Local and average heat transfer coefficients were measured in a test section simulating a rib-roughened trailing edge cooling cavity of a turbine airfoil. The rig was made up of two adjacent channels, each with a trapezoidal cross sectional area. The first channel supplied the cooling air to the trailing-edge channel through a row of racetrack-shaped slots on the partition wall between the two channels. Eleven crossover jets, issued from these slots entered the trailing-edge channel, impinged on eleven radial ribs and exited from a second row of race-track shaped slots on the opposite wall that simulated the cooling holes along the trailing edge of the airfoil. Tests were run for the baseline case with all exit holes open and for cases in which 2, 3, and 4 exit holes on the airfoil tip side were clogged. All tests were run for two crossover jet angles. The first set of tests were run for zero angle between the jet axis and the trailing-edge channel centerline. The jets were then tilted towards the ribs by five degrees. Results of the two set of tests for a range of jet Reynolds number from 10,000 to 35,000 were compared. The numerical models contained the entire trailing-edge and supply channels with all slots and ribs to simulate exactly the tested geometries. They were meshed with all-hexa structured mesh of high near-wall concentration. A pressure-correction based, multiblock, multigrid, unstructured/adaptive commercial software was used in this investigation. The realizable k-ε turbulence model in combination with enhanced wall treatment approach for the near wall regions were used for turbulence closure. Boundary conditions identical to those of the experiments were applied and several turbulence model results were compared. The numerical analyses also provided the share of each crossover and each exit hole from the total flow for different geometries. The major conclusions of this study were: (a) clogging of the exit holes near the airfoil tip alters the distribution of the coolant mass flow rate through the crossover holes and changes the flow structure. Depending on the number of clogged exit holes (from 3 to 6, out of 12), the tip-end crossover hole experienced from 35% to 49% reductions in its mass flow rate while the root-end crossover hole, under the same conditions, experienced an increase of the same magnitude in its mass flow rate. (b) Up to 64% reduction in heat transfer coefficients on the tip-end surface areas around the clogged holes were observed which might have devastating effects on the airfoil life. At the same time, a gain in heat transfer coefficient of up 40% was observed around the root-end due to increased crossover flows. (c) Numerical heat transfer results with the use of the realizable k-ε turbulence model in combination with enhanced wall treatment approach for the near wall regions were generally in a reasonable agreement with the test results. The overall difference between the CFD and test results was about 10%.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):082604-082604-14. doi:10.1115/1.4026804.

Transonic flows through axial, multistage, transcritical organic rankine cycle (ORC) turbines are investigated by using a numerical solver including advanced multiparameter equations of state and a high-order discretization scheme. The working fluids in use are the refrigerants R134a and R245fa, classified as dense gases due to their complex molecules and relatively high molecular weight. Both inviscid and viscous numerical simulations are carried out to quantify the impact of dense gas effects and viscous effects on turbine performance. Both supercritical and subcritical inlet conditions are studied for the considered working fluids. In the former case, flow across the turbine is transcritical, since turbine output pressure is subcritical. Numerical results show that, due to dense gas effects characterizing the flow at supercritical inlet conditions, supercritical ORC turbines enable, for a given pressure ratio, a higher isentropic efficiency than subcritical turbines using the same working fluid. Moreover, for the selected operating conditions, R134a provides a better performance than R245fa.

Commentary by Dr. Valentin Fuster
J. Eng. Gas Turbines Power. 2014;136(8):082605-082605-11. doi:10.1115/1.4026805.

This paper presents an investigation on the hot streak migration across rotor blade tip clearance in a high pressure gas turbine with different tip clearance heights. The blade geometry is taken from the first stage of GE-E3 turbine engine. Three tip clearances, 1.0%, 1.5%, and 2.5% of the blade span with a flat tip were investigated, respectively, and the uniform and nonuniform inlet temperature profiles were taken as the inlet boundary conditions. A new method for heat transfer coefficient calculation recommended by Maffulli and He has been adopted. By solving the unsteady compressible Reynolds-averaged Navier–Stokes equations, the time dependent solutions were obtained. The results indicate that the large tip clearance intensifies the leakage flow, increases the hot streak migration rate, and aggravates the heat transfer environment on the blade tip. However, the reverse secondary flow dominated by the relative motion of casing is insensitive to the change of tip clearance height. Attributed to the high-speed rotation of rotor blade and the low pressure difference between both sides of blade, a reverse leakage flow zone emerges over blade tip near trailing edge. Because it is possible for heat transfer coefficient distributions to be greatly different from heat flux distributions, it becomes of great concern to combine both of them in consideration of hot streak migration. To eliminate the effects of blade profile variation due to twist along the blade span on the aerothermal performance in tip clearance, the tested rotor (straight) blade and the original rotor (twisted) blade of GE-E3 first stage with the same tip profile are compared in this paper.

Commentary by Dr. Valentin Fuster

Research Papers: Nuclear Power

J. Eng. Gas Turbines Power. 2014;136(8):082901-082901-7. doi:10.1115/1.4026811.

The exposure routes by which the dispersion of radionuclides from the accident at the Fukushima atomic power plant were estimated, and the risk was evaluated based on the overall exposure routes, which include the ingestion of food, ingestion and inhalation of soil, and external air dose. This study shows that the air dose from this disaster should be less than 0.2 μSv/h to control the radiation dose with the consumption of food being less than 1 mSv/yr. However, to maintain the lifetime dose under 100 mSv, several mSv/yr is sufficient, considering radioactive decay and dilution by advection and diffusion.

Commentary by Dr. Valentin Fuster

Technical Brief

J. Eng. Gas Turbines Power. 2014;136(8):084501-084501-7. doi:10.1115/1.4026943.

The bypass dual throat nozzle (BDTN) is a new kind of fluidic vectoring nozzle. A bypass is set between the upstream convergent section and upstream minimum area based on the conventional dual throat nozzle (DTN). The BDTN shows a minimum or even no penalty on the nozzle's thrust performance, while it would be able to produce steady and efficient vectoring deflection similar to the conventional DTN. A BDTN model has been designed and subjected to experimental and computational study. The main results show that: (1) BDTN does not consume any secondary injection from the other part of the engine, while it can produce steady and efficient vectoring deflection. (2) Under the same condition, it can provide the maximum thrust vectoring efficiency of all the known fluidic thrust vectoring concepts reported in the literature. (3) The thrust vector angle is also greater than that of the conventional DTN that has been reported up to now. Especially, under NPR = 10, the thrust vector angle of BDTN can reach 21.3 deg. (4) For a wide NPR range from 2 to 10, the BDTN generates the best thrust vectoring performance under NPR = 4. Above all, the BDTN is well suited to produce vectored thrust for nozzles.

Commentary by Dr. Valentin Fuster


J. Eng. Gas Turbines Power. 2014;136(8):087001-087001-1. doi:10.1115/1.4026906.

Faculty of Aerospace Engineering, Technion, Israel Institute of Technology, Technion, Haifa 3200003, Israel

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In