0

Coal, Biomass and Alternative Fuels

1999;():V002T01A001. doi:10.1115/99-GT-081.
FREE TO VIEW

Co-utilization of solid fuels with combustion turbines (CTs) can be accomplished by blending feedstocks in indirectly heated gasifier/liquifiers (IHG/L), Blending could improve the quality and yields of gaseous and liquid fuel outputs and, with gas clean-up (GCU) or liquid distillation, serve valuable societal functions. Gas yields vs. temperature, time and other variables have been obtained using batch fed and continuous fed laboratory scale IHG/Ls using particle sizes suitable for commercial systems. The rates of solids to gas/liquid conversion are mainly dependent upon the heat transfer and diffusion coefficients of the biomass rather than the rate constants for chemical decomposition. Phenomenological engineering formulas are developed for use with biomass that can be treated as a blend of hemicellulose, cellulose and lignin. The main immediate goals are to develop: 1) laboratory scale process development units (PDUs) that can help anticipate results with commercial IHG/L systems, 2) predictive models for estimating gaseous, liquid and char yields when various blends of biomass and other domestic fuels are processed in IHG/Ls and 3) applications of IHG-GCU-CT systems that could drive technology development during periods of low oil and natural gas prices.

Topics: Feedstock
Commentary by Dr. Valentin Fuster
1999;():V002T01A002. doi:10.1115/99-GT-191.
FREE TO VIEW

An experimental study on co-gasification of coal and biomass blends in an oxygen-containing atmosphere has been carried out in a pressurized fluidized-bed gasifier. Several different biomass materials including wood and energy crops were used in the study, whereas two coals ranked of bituminite from Poland and UK were used in the investigation. The gasifier used was a Laboratory Development Unit (LDU) with an inner diameter of 144 mm. The operation temperature was 900 °C, and the pressure was 0.4 MPa. The research was part of the European Commission’s APAS and JOULE III clean coal technology programs.

The study was focused on possible synergistic effects in the thermochemical treatment of the fuel blends. The char formed was examined. The tar produced in the process was analyzed. The environmentally concerned nitrogen compounds emitted from the process were detected. An unexpected result was that the blends of the fuels and their char formed in situ exhibited higher gasification reaction rate under the studied conditions. The yield of char diminished and consequently the gas production increased. Furthermore, both the formation of tar and nitrogen compounds seemed also affected synergistically in co-gasification process of the fuel blends. The yields of tar and ammonia were lower than expected.

Commentary by Dr. Valentin Fuster
1999;():V002T01A003. doi:10.1115/99-GT-192.
FREE TO VIEW

Gas turbine based power and cogeneration schemes are likely to become more favored as turbine efficiencies improve, but the economics of local power generation may depend on the use of low cost fuels other than natural gas. Opportunities may arise in the application of gas turbines in the pulp and paper industry and the wider use of biomass derived fuels in general.

These fuels, as produced, typically contain inorganic impurities originating from ash forming substances and other minor constituents of the feedstock. Also, depending on the biomass treatment process, they contain varying amounts of complex organic derivatives, commonly referred to as tars, and some simpler condensable vapors.

The Department of Energy is sponsoring work aimed at providing realistic data on low level constituents and impurities in gas derived by indirect gasification of wood, some of which may have disproportionately severe effects on turbine operation, durability, and emissions performance. It is planned to sample gas from both laboratory scale (up to 20 tons/day) and pilot scale (200 tons/day) installations and to assess the effectiveness of wet scrubbing procedures and catalytic reforming of condensables in cleaning up the gases. This paper discusses the rationale for this work, experimental approach, and analytic procedures that will be used. The work will include the operation of a small (220-kWe) gas turbine to provide direct information on the impact of using the final biomass derived gas delivered by the system.

The laboratory scale work is currently under way, with a planned completion date in mid 2000. The second phase is dependent on arrangements for integration of the R&D effort with the operation of the pilot plant.

Commentary by Dr. Valentin Fuster
1999;():V002T01A004. doi:10.1115/99-GT-193.
FREE TO VIEW

Aircraft gas turbines, steam turbines, coal utilization turbines and many other systems operating under particulate flows are exposed to erosion wear and performance deterioration. This paper describes an experimental technique to evaluate particle rebound characteristics on turbomachinery leading edge geometry using laser velocimetry and high-speed photography. The experimental work was conducted with silica sand particle sizes of 1000–1500 microns and velocity of 15 and 30 m/s. Stainless Steel was the target material. Mach Number and Reynolds Number of the flow were 0.09 and 105 (Based on inlet velocity and test section width) respectively. Wollensax Fastax high-speed camera was used. A camera speed of 2500 frames per second was used. The data were analyzed using x-v, image-processing software. Velocity restitution ratio at various impact locations was calculated. The measured velocity restitution ratio could be used in the numerical simulation to predict particle trajectories, subsequently could be used to identify the critical zones and to predict the erosion rates.

Commentary by Dr. Valentin Fuster
1999;():V002T01A005. doi:10.1115/99-GT-267.
FREE TO VIEW

Several advanced, coal- and biomass-based combustion turbine power generation technologies are currently under development and demonstration. A key technology component in these power generation systems is the hot gas filter. These power generation technologies must utilize highly reliable and efficient hot gas filter systems to protect the turbine and to meet environmental constraints if their full thermal efficiency and cost potential is to be realized. Siemens Westinghouse Power Corporation (SWPC) has developed a hot gas filter system to near-commercial status for large-scale power generation applications. This paper reviews recent progress made by SWPC in hot gas filter test development programs and in major demonstration programs. Two advanced hot gas filter concepts, the “Inverted Candle” and the “Sheet Filter”, having the potential for superior reliability are also described.

Commentary by Dr. Valentin Fuster
1999;():V002T01A006. doi:10.1115/99-GT-268.
FREE TO VIEW

Iron aluminides have shown good to excellent high-temperature corrosion resistance in sulfur-bearing environments and thus have potential for use as the material of construction for metallic filters used to clean fossil-fuel-derived gases prior to their introduction into gas turbine systems. Consequently, a background for consideration of such alloys for filter applications is given in terms of a brief summary of the physical metallurgy and relevant high-temperature corrosion behavior of iron aluminides. In addition, preliminary characterization results on iron-aluminide filter elements exposed in test beds that simulate environments associated with advanced coal-fired energy production are presented. Although good corrosion resistance was found, there were minor to moderate strength reductions that did not necessarily scale with time. Little degradation in ductility was observed.

Topics: Filters , Iron
Commentary by Dr. Valentin Fuster
1999;():V002T01A007. doi:10.1115/99-GT-293.
FREE TO VIEW

The use of biologically derived oils and fuels has received increasing visibility in recent years. A combination of fuel availability, refinery capacity and environmental responsibility has resulted in interest in their use in turbine engine fuels.

Such a soy methyl ester (SME) is being evaluated as a possible extender and/or additive for aviation turbine fuel per ASTM D-1655. Laboratory testing indicates additive levels of up to 2% by volume can be used while still meeting ASTM D-1655. Engine testing performed at 20% blending levels have demonstrated potential fuel consumption improvements as well as reduction in NOx emissions. The final blend levels have not yet been determined.

The use of SME even at low levels could provided performance benefits. Because of the oil nature of SME, a small addition could result in significant increases in lubricity. The use of the higher flashpoint SME could result in an upward shift in flashpoint with little or no effect to other physical properties. With increased visibility to work place considerations, the potential for “odor” abatement is also of interest.

Topics: Fuels , Turbines , Biofuel , Aviation
Commentary by Dr. Valentin Fuster
1999;():V002T01A008. doi:10.1115/99-GT-294.
FREE TO VIEW

Detailed chemical kinetic modeling has been used to study the reduction of nitrogen oxides at gas turbine (GT) combustor conditions. A gas from gasification of wood with air has been used as the fuel. An air-staged combustion technique has been adapted.

In our previous study a simple plug flow model was used to study the effects of pressure and temperature among others process variables. The air-fuel mixing was assumed perfect and instantaneous. Results showed the NOx reduction mainly affected by both pressure and temperature.

The aim of the present work is to establish the effect of air-fuel mixing delay on NOx predictions and to extrapolate indications options for GT. To model the mixing delay, a varying number of air sub-streams are mixed with the fuel gas during different time periods. Alternatively, a combination of a perfectly mixed zone followed by a plug flow zone is illustrated.

Results by any air-fuel mixing model show similar affect of process variables on NOx reduction. When a mixing delay is assumed instead of the instantaneous mixing the NOx reduction is enhanced, and only with delayed mixing NOx are affected by CH4.

Lower temperature and higher pressure in the GT-combustor can enhance the NOx reduction. Also air staging is an effective option: a 3 stages combustor designed for low mixing speed appear competitive compared to more complicate combustors. The fewer hydrocarbons in the gasification gas the high NOx reduction.

Commentary by Dr. Valentin Fuster
1999;():V002T01A009. doi:10.1115/99-GT-320.
FREE TO VIEW

Advanced, coal-fueled, power generation systems utilizing pressurized fluidized bed combustion (PFBC) and integrated gasification combined cycle (IGCC) technologies are currently being developed for high-efficiency, low emissions, and low-cost power generation. In spite of the advantages of these promising technologies, the severe operating environment often leads to material degradation and loss of performance in the barrier filters used for particle entrapment. To address this problem, LoTEC Inc., and Oak Ridge National Laboratory are jointly designing and developing a monolithic cross-flow ceramic hot-gas filter. The filter concept involves a truly monolithic cross-flow design that is resistant to delamination, can be easily fabricated, and offers flexibility of geometry and material make-up.

During Phase I of the program, a thermo-mechanical analysis was performed to determine how a cross-flow filter would respond both thermally and mechanically to a series of thermal and mechanical loads. The cross-flow filter mold was designed accordingly, and the materials selection was narrowed down to Ca0.5Sr0.5Zr4P6O24 (CS-50) and 2Al2O3−3SiO2 (mullite). A fabrication process was developed using gelcasting technology and monolithic cross-flow filters were fabricated. The program focuses on obtaining optimum filter permeability and testing the corrosion resistance of the candidate materials.

Commentary by Dr. Valentin Fuster
1999;():V002T01A010. doi:10.1115/99-GT-321.
FREE TO VIEW

Oxygen production rates of 10,000 to 20,000 tons per day from large, cryogenic air separation units are being studied by many alternative fuel project developers. These projects utilize oxygen to partially oxidize hydrocarbon materials, producing a clean synthesis gas that can be used as a fuel or for conversion into valuable chemical products. Specific market applications include natural gas or waste material conversion processes and multi-train integrated gasification combined cycle facilities. In an effort to reduce specific facility cost project developers increase facility output to obtain economies of scale, resulting in large oxygen requirements for the partial oxidation step. One of the challenges to provide cost effective oxygen is the economic supply of large quantities of compressed air for use in the cryogenic air separation process. To date, gas turbines have found limited application for use in air separation facilities due to their relatively high capital cost compared to traditional electric motor drives. The need for large, single train air separation units to support alternative fuel projects creates opportunities for the use of gas turbines. This paper explores the use of commercially available equipment, configured to integrate with air separation processes, to improve the economics of oxygen production. Long term developmental equipment configurations are presented to further improve the economics of these facilities.

Commentary by Dr. Valentin Fuster
1999;():V002T01A011. doi:10.1115/99-GT-322.
FREE TO VIEW

The present paper reports on a demonstration project supported by the THERMIE program of the European Commission and by the VLIET program of the Flemish Government. A CHP gas turbine plant fueled by product gas from a biomass fluidized bed gasifier has been constructed. The demonstration scale is 500 kWe for production of power and heat for the university campus district heating. At the present stage 150 kW power has been delivered to the grid. Problems encountered and results achieved during the first startups of the power plant will be discussed. In the future some natural gas topping combustion will be included to overcome the temperature limitation of materials used in the metallic high temperature air heater. Water injection in this air heater will be included to enhance power output and to allow flexible power to heat ratios. The target commercial scale is 2 to 5 Mwe using atmospheric gasification and external firing through a high temperature metallic heater.

Commentary by Dr. Valentin Fuster
1999;():V002T01A012. doi:10.1115/99-GT-343.
FREE TO VIEW

The Power Systems Development Facility (PSDF) is a Department of Energy (DOE) sponsored engineering scale demonstration of two advanced coal-fired power systems. Particulate cleanup is achieved by several High Temperature, High Pressure (HTHP) gas filtration systems. The PSDF was designed at sufficient scale so that advanced power systems and components could be tested in an integrated fashion to provide confidence and data for commercial scale-up. This paper provides an operations summary of a Siemens-Westinghouse Particulate Control Device (PCD) filtering combustion gas from a Kellogg Brown & Root (KBR) transport reactor located at the PSDF.

The transport reactor is an advanced circulating fluidized bed reactor designed to operate as either a combustor or a gasifier. Particulate cleanup is achieved by using one of two PCDs, located downstream of the transport reactor. As of the end of 1998, the transport reactor has operated on coal as a combustor for over 3500 hours. To date, filter elements from 3M, Blasch, Coors, Allied Signal (DuPont), IF&P, McDermott, Pall, Schumacher and Specific Surface have been tested up to 1400°F in the Siemens-Westinghouse PCD.

The PSDF has a unique capability for the collection of samples of suspended dust entering and exiting the PCD with Southern Research Institute’s (SRI) in-situ particulate sampling systems. These systems have operated successfully and have proven to be invaluable assets. Isokinetic samples using a batch sampler, a cascade impactor and a cyclone manifold have provided valuable data to support the operation of the transport reactor and the PCD. Southern Research Institute has also supported the PSDF by conducting filter element material testing.

Commentary by Dr. Valentin Fuster
1999;():V002T01A013. doi:10.1115/99-GT-353.
FREE TO VIEW

In order to reduce the production of green house gases, the combustion of biomass has been gaining importance in electricity generation. Especially the direct combustion of biomass in gas turbines of a few MW output would offer a very attractive option because of low investment costs and high operational flexibility. Therefore, since 1991 the Institute of Thermal Turbomachines and Powerplants at the Vienna University of Technology has been working on realising a wood particle fired gas turbine with direct combustion. With reference to earlier studies (c.f. Hamrick (1991), Fredriksson and Kallner (1993)), it had been concluded that the design and the operating characteristics of the fuel feed system would strongly influence the combustion and so would be a very important part of the whole facility.

Following an overview of the planned gas turbine test facility including the combustion chamber and the recently installed pneumatic fuel feed system, the paper will deal with three basic requirements of fuel feeding in the case of a directly fired gas turbine: feeding against back pressure, continuous fuel flow rate and a low conveying air ratio, which is the ratio of fuel conveying air to total combustion air of the combustion chamber. While the first two requirements, i.e. feeding against back pressure and continuous feeding, are briefly considered, the minimisation of the conveying air ratio is discussed in detail. For instance, important parameters affecting the conveying air ratio are fuel moisture, combustion air ratio and, in particular, techniques. Following theoretical estimation of the conveying air ratio, results of fuel feeding measurements are presented and conclusions drawn with respect to system integration.

Commentary by Dr. Valentin Fuster
1999;():V002T01A014. doi:10.1115/99-GT-354.
FREE TO VIEW

Ceramic candle filters have been developed for cleaning high-temperature high-pressure (HTHP) gas streams. They meet environmental and economical considerations in Combined cycle power plant, where gas turbine blades can be protected from the erosion that occurs due to using HTHP exhaust from the fluidized bed. Ceramic candle filters are the most promising hot gas filtration technology, which has demonstrated high collection efficiencies at high-temperature high-pressure conditions.

This paper reports a computational fluid dynamics (CFD) investigation of a candle filter. Constant filtration velocity boundary models have been used to investigate the filter in cross flow conditions using the CFD code FLUENT. Different approach (inlet) velocity to filter face velocity ratios and different face velocities (ranging from 2 to 5 cm/s) are used in the CFD calculation. Particles in the diameter range 1 to 100 microns are tracked through the domain. The radius of convergence (or the critical trajectory) is compared and plotted as a function of many parameters. The deposition process and the factors that affect the build up of the filter cake have also been studied.

Commentary by Dr. Valentin Fuster
1999;():V002T01A015. doi:10.1115/99-GT-355.
FREE TO VIEW

Gas turbines fired on syngas may show thermo-acoustic combustion instabilities. The theory on these instabilities is well developed. From this theory it can be shown that the acoustic system of a combustion installation can be described as a control loop with a set of transfer functions. The transfer function of the flame plays a decisive role in the occurrence of combustion instabilities. It is however very difficult to predict this flame transfer function analytically. In this paper a numerical method will be presented to calculate the flame transfer function from time-dependent combustion calculations. Also an experimental method will be discussed to determine this flame transfer function. Experiments have been performed in a 25 kW atmospheric test rig. Also calculations have been done for this situation. The agreement between the measurements and CFD calculations is good, especially for the phase at higher frequencies. This opens the way to apply CFD-modeling for acoustics in a real gas turbine situation.

Commentary by Dr. Valentin Fuster
1999;():V002T01A016. doi:10.1115/99-GT-397.
FREE TO VIEW

Imatran Voima (TVO) is the biggest energy producer in Finland. IVO builds, owns and operates several biomass combustion power plants and carries out active R&D work on further development of small-scale biomass and peat-fired power plants. Both conventional power plants based on fluidized bed boilers and future power plants based on gasification are also developed by IVO.

Typically, biomass fuels have high moisture content. High moisture content of fuel has, however, a negative effect on power generation cycles because of higher combustion losses.

IVO has researched and developed a steam drying system for small-scale gasification power plants for the sole domestic fuels peat and biomass. This is described and the experience gained in testing it is briefly reviewed.

The main focus of this paper is the thermal analysis of pressurized steam drying for biomass gasification power plants. Special emphasis is given to the thermal analysis of fuel moisture in alternative power plant processes. Fuel moisture removal is analyzed as a separate process.

Topics: Fuels , Gas turbines
Commentary by Dr. Valentin Fuster
1999;():V002T01A017. doi:10.1115/99-GT-398.
FREE TO VIEW

The development of integrated, coal-gasification combined cycle (IGCC) systems provides cost-effective and environmentally sound options for meeting future coal-utilizing power generation needs in the world. The Japanese government and the Electric Power Industries in Japan promoted research and development of an IGCC system. We have being working on developing a low-NOx combustion technology used in gas turbine combustors for IGCC.

Each gaseous fuel produced from some raw materials contained CO and H2 as the main combustible components, and a small amount of CH4. Compositions and calorific values of gasified coal fuels varied widely depending on raw materials and gasifier types. Gaseous fuel, produced in various gasifiers, has a calorific value of 4–13MJ/m 3, which is about one-tenth to one-third that of natural gas. The flame temperatures of fuels increase as the fuel calorific value rises. When the fuel calorific value rises 8MJ/m 3 or higher, the flame temperature is higher than that of natural gas, and so NOx production from nitrogen fixation is expected to increase significantly. Also, some gasified coal fuels contain fuel nitrogen, such as ammonia, if the hot/dry type gas cleaning system is employed. These factors affect the combustion characteristics of the gasified coal fuel.

In this paper, we clarified the influence of gasified coal fuel properties on NOx and CO emissions through experiments using a small diffusion burner and through numerical analysis based on reaction kinetics. The main results were as follows:

(1) NH3 conversion to NOx increases with increasing CH4 concentration in gaseous fuel.

(2) If gaseous fuel contains CH4, there will be some specific equivalence ratio in the primary combustion zone for the minimum NH3 conversion to NOx in the two-staged combustion.

(3) Its specific equivalence ratio in the primary combustion zone increases with decreasing CH4 concentration in gaseous fuel.

(4) If the fuel contains a small percent of CH4, there is no influence of the CO/H2 molar ratio in the fuel on the conversion rate of NH3 to NOx, while there is an influence in the case where fuel contains no CH4. The conversion rate increases with rises in the CO/H2 molar ratio.

(5) As the pressure increases, the conversion rate of NH3 to NOx slightly decreases and the CO emission declines significantly.

Topics: Combustion , Fuels , Coal
Commentary by Dr. Valentin Fuster

Combustion and Fuels

1999;():V002T02A001. doi:10.1115/99-GT-003.
FREE TO VIEW

There has been increased demand in recent years for gas turbines that operate in a lean, premixed (LP) mode of combustion in an effort to meet stringent emissions goals. Unfortunately, detrimental combustion instabilities are often excited within the combustor when it operates under lean conditions, degrading performance and reducing combustor life. To eliminate the onset of these instabilities and develop effective approaches for their control, the mechanisms responsible for their occurrence must be understood.

This paper describes the results of an investigation of the mechanisms responsible for these instabilities and approaches for their control. These studies found that combustors operating in a LP mode of combustion are highly sensitive to variations in the equivalence ratio (ϕ) of the mixture that enters the combustor. Furthermore, it was found that such ϕ variations can be induced by interactions of the pressure and flow oscillations with the reactant supply rates. The ϕ perturbations formed in the inlet duct (near the fuel injector) are convected by the mean flow to the combustor where they produce large amplitude heat release oscillations that drive combustor pressure oscillations. It is shown that the dominant characteristic time associated with this mechanism is the convective time from the point of formation of the reactive mixture at the fuel injector to the point where it is consumed at the flame. Instabilities occur when the ratio of this convective time and the period of the oscillations equals a specific constant, whose magnitude depends upon the combustor design. Significantly, these predictions are in good agreement with available experimental data, strongly suggesting that the proposed mechanism properly accounts for the essential physics of the problem. The predictions of this study also indicate, however, that simple design changes (i.e., passive control approaches) may not, in general, provide a viable means for controlling these instabilities, due to the multiple number of modes that may be excited by the combustion process. This conclusion indicates that active control strategies may be necessary for controlling these instabilities.

Commentary by Dr. Valentin Fuster
1999;():V002T02A002. doi:10.1115/99-GT-008.
FREE TO VIEW

An experimental and modeling study has been performed jointly by UTRC and DOE-FETC to determine the effect of humidity in the combustion air on emissions and stability limits of gas turbine premixed flames. This study focuses on developing gas turbine combustor design criteria for the Humid Air Turbine (HAT) cycle. The experiments were conducted at different moisture levels (0%, 5%, 10% and 15% by mass in the air), at a total pressure of 200 psi, pilot levels (0%, 1%, 3% and 5% total fuel), and equivalence ratio (0.4 to 0.8 depending on the moisture levels). The moisture levels were achieved by injecting steam into dry air well upstream of the fuel-air premixing nozzle. Computations were made for comparison to the experiments using GRI Mech 2.11 kinetics and thermodynamic database for modeling the flame chemistry. A Perfectly Stirred Reactor (PSR) network code was used to create a network of PSRs to simulate the flame. Excellent agreement between the measured and modeled NOx (5–10%) was obtained. Trends of added moisture reducing NOx and the effects of equivalence ratio and piloting level were well predicted. The CO predictions were higher by about 30–50%. The CO discrepancies are attributed to in-probe oxidation. The agreement between the data and model predictions over a wide range of conditions indicate the consistency and reliability of the measured data and usefulness of the modeling approach. An analysis of NOx formation revealed that at constant equilibrium temperature, Teq, the presence of steam leads to lower O-atom concentration which reduces “Zeldovich and N2O” NOx while higher OH-atom concentration reduces “Fenimore” NOx.

Topics: Modeling , Flames
Commentary by Dr. Valentin Fuster
1999;():V002T02A003. doi:10.1115/99-GT-009.
FREE TO VIEW

Thermoacoustic resonance is a difficult technical problem that is experienced by almost all lean-premixed combustors. The Industrial Trent combustor is a novel dry-low-emissions (DLE) combustor design, which incorporates three stages of lean premixed fuel injection in series. The three stages in series allow independent control of two stages — the third stage receives the balance of fuel to maintain the desired power level — at all power conditions. Thus, primary zone and secondary zone temperatures can be independently controlled. This paper examines how the flexibility offered by a 3-stage lean premixed combustion system permits the implementation of a successful combustion noise avoidance strategy at all power conditions and at all ambient conditions. This is because at a given engine condition (power level and day temperature) a characteristic “noise map” can be generated on the engine, independently of the engine running condition. The variable distribution of heat release along the length of the combustor provides an effective mechanism to control the amplitude of longitudinal resonance modes of the combustor. This approach has allowed the Industrial Trent combustion engineers to thoroughly “map out” all longitudinal combustor acoustic modes and design a fuel schedule that can navigate around regions of combustor thermoacoustic resonance. Noise mapping results are presented in detail, together with the development of noise prediction methods (frequency and amplitude) that have allowed the noise characteristics of the engine to be established over the entire operating envelope of the engine.

Commentary by Dr. Valentin Fuster
1999;():V002T02A004. doi:10.1115/99-GT-052.
FREE TO VIEW

A Solar fuel injector that provides lean premixed combustion conditions has been studied in a combined experimental and numerical investigation. Lean premixed conditions can be accompanied by excessive combustion driven pressure oscillations which must be eliminated before the release of a final combustor design. In order to eliminate the pressure oscillations the location of fuel injection was parametrically evaluated to determine a stable configuration. It was observed that small axial changes in the position of the fuel spokes within the premix duct of the fuel injector had a significant positive effect on decoupling the excitation of the natural acoustic modes of the combustion system.

In order to further understand the phenomenon, a time-accurate 2D CFD analysis was performed. 2D analysis was first calibrated using 3D steady-state CFD computations of the premixer in order to model the radial distribution of velocities in the pre mixer caused by non-uniform inlet conditions and swirling flow. 2D time-accurate calculations were then performed on the baseline configuration. The calculations captured the coupling of heat release with the combustor acoustics, which resulted in excessive pressure oscillations. When the axial location of the fuel injection was moved, the CFD analysis accurately captured the fuel time lag to the flame-front, and qualitatively matched the experimental findings.

Commentary by Dr. Valentin Fuster
1999;():V002T02A005. doi:10.1115/99-GT-053.
FREE TO VIEW

Through a series of computational studies, carbon monoxide has been identified as an important promoter of NO oxidation to NO2 in combustion turbine exhaust gas at intermediate temperatures (450 to 750°C). NO2 formation is accompanied by enhanced CO burnout at these temperatures. Perfectly stirred reactor and plug flow reactor calculations indicate that concentrations of CO as low as 50 ppmv in exhaust gas containing 25 ppmv NO can result in the conversion of 50% of the NO to NO2 in less than 1 second. NO2 concentrations as low as 15 ppmv can result in visible, yellow-brown plumes from large diameter exhaust stacks. If NO2 plumes are to be prevented, then designers of gas turbines and heat recovery steam generators need to be aware of the relationships between time, temperature, and composition which cause NO2 to form in exhaust gas.

Reaction path analysis indicates that the mutually promoted oxidation of CO and NO occurs through a self-propagating, three-step chain reaction mechanism. CO is oxidized by OH, Display Formula

(R2)
CO+OHCO2+H
while NO is oxidized by HO2:Display Formula
(R23)
NO+HO2NO2+OH
In a narrow temperature range, the H-atom produced by R2 can react with O2 in a three body reaction to yield the hydroperoxy radical needed in R23:Display Formula
(R9)
H+O2+MHO2+M
where M is any third body. The observed net reaction isDisplay Formula
CO+O2+NOCO2+NO2
which occurs stoichiometrically at temperatures below about 550°C. As the temperature increases, additional reaction pathways become available for H, HO2, and OH which remove these radicals from the chain and eventually completely decouple the oxidation of CO from NO.

An abbreviated set of elementary chemical reactions, including 15 species and 33 reactions, has been developed to model CO-enhanced oxidation of NO to NO2. This reaction set was derived from a larger reaction set with more than 50 species and 230 elementary chemical reactions, and was validated by comparison of PSR and PFR calculations using the two sets.

Commentary by Dr. Valentin Fuster
1999;():V002T02A006. doi:10.1115/99-GT-054.
FREE TO VIEW

An advanced thermal management analysis tool, named Advanced Thermal Hydraulic Energy Network Analyzer (ATHENA), has been used to simulate a fuel system for gas turbine engines. The ATHENA tool was modified to account for JP-8/dodecane fuel properties. The JP-8/dodecane fuel thermodynamic properties were obtained from the SUPERTRAP property program. A series of tests of a fuel system simulator located at the Air Force Research Laboratory (AFRL)/Wright Patterson Air Force Base were conducted to characterize the steady state and dynamic behavior of the fuel system. Temperature, pressures and fuel flows for various fuel pump speeds, pressure rise and flow control valve stem positions (orifice areas), heat loads and engine fuel flows were measured. The predicted results were compared to the measured data and found to be in excellent agreement. This demonstrates the capability of the ATHENA tool to reproduce the experimental data and, consequently, its validity as an analysis tool that can be used to carry out analysis and design of fuel systems for advanced gas turbine engines. However, some key components in the fuel system simulator such as control components, which regulate the engine fuel flow based on predetermined parameters such as fan speed, compressor inlet and exit pressures and temperatures, combustor pressure, turbine temperature and power demand, were not simulated in the present investigation due to their complex interactions with other components functions. Efforts are currently underway to simulate the operation of the fuel system components with control as the engine fuel flow and power demands are varied.

Commentary by Dr. Valentin Fuster
1999;():V002T02A007. doi:10.1115/99-GT-055.
FREE TO VIEW

Liquid-phase oxidation of two jet fuels classified as hydro-treated and straight-run (without dissolved metals) is studied at 185°C by measuring the depletion of dissolved oxygen in an isothermal, high-pressure flowing system. The goal is to simulate some of the effects of fuel recirculation by studying changes in oxidation that occur during the course of reaction. Recirculation of fuel is used in modem aircraft to optimize the ability of the fuel to dissipate heat from component systems. Following extended oxidation (prestressing), hydrotreated fuel oxidizes more rapidly than in its neat form resulting, in part, from formation of catalytic oxidation products such as hydroperoxides and depletion of synthetic primary antioxidants. In contrast, the straight-run fuel oxidizes more slowly following prestressing; this effect is ascribed to the formation of efficient secondary antioxidants and to depletion of radical initiators. Results are discussed in terms of changes in the distribution of primary and secondary antioxidants and their effect on subsequent oxidation and the thermal stability of recirculated fuel.

Topics: Fuels , oxidation , Aviation
Commentary by Dr. Valentin Fuster
1999;():V002T02A008. doi:10.1115/99-GT-056.
FREE TO VIEW

The effectiveness of hydrogen donor compounds as additives to reduce pyrolytic deposition in JP-8+100 at high temperatures was assessed. Decalin and 1,2,3,4 tetrahydroquinoline (THQ) were added to JP-8+100 at 0.5% (decalin only), 1.0 and 2.5% w/w concentrations and tested in a flow reactor at a fuel exit temperature of 600°C at 5.2 MPa. Measurements of carbon deposits along the tube and gas chromatography/mass spectrometry (GC/MS) analysis of the stressed and unstressed liquid fuel were used to assess effectiveness of the additive, and the degree of fuel decomposition. Additionally, liquid-to-gas conversion was determined, and the composition of the gas was determined via GC. Experimental results show significant reductions in pyrolytic deposition in JP-8+100 with the additives relative to the baseline fuel. Tests with decalin showed negligible effects on thermal oxidative deposits, while THQ produced significant increases in thermal oxidative deposits. The effects of the additives on fuel thermal decomposition and conversion rates are also discussed.

Commentary by Dr. Valentin Fuster
1999;():V002T02A009. doi:10.1115/99-GT-057.
FREE TO VIEW

This paper discusses some of the advanced concepts and research and development associated with implementing catalytic combustion to achieve ultra-low-NOx emissions in the next generation of land-based gas turbine engines. In particular, the paper presents current development status and design challenges being addressed by Siemens Westinghouse Power Corp. for large industrial engines (> 200 MW) and by Solar Turbines for smaller engines (< 20 MW) as part of the U.S. Department of Energy’s (DOE) Advanced Turbine Systems (ATS) program. Operational issues in implementing catalytic combustion and the current needs for research in catalyst durability and operability are also discussed. This paper indicates how recent advances in reactor design and catalytic coatings have made catalytic combustion a viable technology for advanced turbine engines and how further research and development may improve catalytic combustion systems to better meet the durability and operability challenges presented by the high-efficiency, ultra-low emissions ATS program goals.

Topics: Combustion , Turbines
Commentary by Dr. Valentin Fuster
1999;():V002T02A010. doi:10.1115/99-GT-058.
FREE TO VIEW

This paper describes a reduced NOx diffusion flame combustor that has been developed for the MS5002 gas turbine. Laboratory tests have shown that when firing with natural gas, without water or steam injection, NOx emissions from the new combustor are about 40% lower than NOx emissions from the standard MS5002 combustor. CO emissions are virtually unchanged at base load, but increase at part load conditions. The laboratory results were confirmed in 1997 by a commercial demonstration test at a British Petroleum site in Prudhoe Bay, Alaska.

The standard MS5002 gas turbine is equipped with a conventional, swirl stabilized diffusion flame combustion system. The twelve standard combustors in an MS5002 turbine are cylindrical cans, approximately 27 cm (10.5 inches) in diameter and 112 cm (44 inches) long. A small, annular, vortex generator surrounds the single fuel nozzle that is centered at the inlet to each can. The walls of the cans are louvered for cooling, and contain an array of mixing and dilution holes that provide the air needed to complete combustion and dilute the burned gas to the desired turbine inlet temperature.

The new, reduced NOx emissions combustor (referred to as a “lean head end”, or LHE, combustor) retains all of the key features of the conventional combustor: the only significant difference is the arrangement of the mixing and dilution holes in the cylindrical combustor can. By optimizing the number, diameter, and location of these holes, NOx emissions were substantially reduced. The materials of construction, fuel nozzle, and total combustor air flow were unchanged.

The differences in NOx emissions between the standard and LHE combustors, as well as the variations in NOx emissions with firing temperature, were well correlated using turbulent flame length arguments. Details of this correlation are also presented.

Commentary by Dr. Valentin Fuster
1999;():V002T02A011. doi:10.1115/99-GT-059.
FREE TO VIEW

On of the major considerations in the development of advanced gas turbine engines are increased thrust to weight ratio and reduced development and operating costs. Improvements in engine thrust require an increase in combustion chamber heat release and inlet pressures. However, increasing the amount of heat release will also result in an increase in the radiative heat flux to the combustion chamber walls, proving detrimental to the operational life time of the combustor. To maximise combustor life, different cooling devices can be incorporated into the combustion chamber design. The effectiveness with which these devices are implemented is important and in the absence of a reliable predictive numerical tools, is difficult to quantify without undertaking expensive and timely testing. A computer analysis tool, based on a network model approach, has previously been developed to analyse airflow distributions in complex combustor geometries. A recent variant of this model has incorporated the Discrete Transfer radiation model, along with other convective and conductive sub-models, to account for heat transfer. These models have been validated against thermocouple measurements of wall temperature obtained in a sectored research combustor. The results of this comparison indicate that, whilst the model is capable of predicting the trends in wall temperature, it is currently unable to reproduce the magnitude of wall temperature with a greater accuracy than 80 K. However, the versatility of the discrete transfer model suggests that further improvements in accuracy are possible.

Commentary by Dr. Valentin Fuster
1999;():V002T02A012. doi:10.1115/99-GT-107.
FREE TO VIEW

An engine trend toward maximum use of fuel as a heat sink has created interest in thermal stability information on aviation fuels for high temperature applications. Testing of the fuels has been conducted in the Extended Duration Thermal Stability Test System. This system was established for evaluating the thermal stability of JP-8+100 fuel. The specific fuels evaluated were JP-8, JP-8+100, JPTS and JP-7. The testing has been conducted at temperature and residence times that represent future aircraft/engine fuel systems. Special thermal stability tests with copper doped fuels and at low oxygen content have also been conducted. Carbon deposition information of these fuels that occur in bulk fuel and on hot surfaces is covered in this paper. The description of the test system and its operating characteristics and the test results are also included.

Commentary by Dr. Valentin Fuster
1999;():V002T02A013. doi:10.1115/99-GT-108.
FREE TO VIEW

This paper reports results of experimental and numerical investigations on ethane-air two-stage combustion in a counterflow burner where the fuel stream, which is partially premixed with air for equivalence ratios from 1.6 to 3.0, flows against a pure air stream. Similar to methane, the two-stage ethane combustion exhibits a green fuel-rich premixed flame and a blue diffusion flame. Flame structures, including concentration profiles of stable intermediate species such as C2H4, C2H2 and CH4, are measured by a gas chromatography and are calculated by numerical integrations of the conservation equations employing an updated elementary chemical-kinetic data base. The implications of the results from these experimental measurements and numerical predictions are summarized, the flame chemistry of ethane two-stage combustion at different degrees of premixing (or equivalence ratio) is discussed, and the relationship between NOx formation and the degree of premixing is established. The present work helps to increase understanding of flame chemistry of hydrocarbon fuels, identify important reactions for pollutant formation and suggest means to reduce emissions.

Topics: Flames
Commentary by Dr. Valentin Fuster
1999;():V002T02A014. doi:10.1115/99-GT-109.
FREE TO VIEW

This paper reports the effect of changing the location of axial swirl vanes on premix combustion dynamics. Tests are conducted in a specially designed single-injector combustor operating at a pressure of 7.5 atmospheres and an inlet air temperature of 588K (600F). All of the tests are conducted using natural gas as the fuel. The air velocity and equivalence ratio are varied over an operating map for four different axial swirl vane positions in the premix nozzle. In contrast to earlier studies reported from this combustor, the fuel injection location is fixed. The results confirm the importance of the convective fuel time lag for the different swirl vane locations, but also show that changing the vane location at a fixed time lag can significantly affect the magnitude of the combustion oscillations.

Commentary by Dr. Valentin Fuster
1999;():V002T02A015. doi:10.1115/99-GT-110.
FREE TO VIEW

Measurements of vibration and combustion chamber dynamic pressures have been taken on a number of 150MW industrial gas turbines operating on pre-mixed natural gas, both during long periods of base-load operation and during short duration load-swings. The data has been analysed in terms of the frequency and bandwidth of the principle peak in the vibration and pressure spectra as a function of load and other operating parameters. It is observed that bandwidth, which is a measure of the damping of the resonant mode of the combustion chamber’s acoustic resonance, decreases towards zero as the machines approach their combustion stability limits.

A theoretical model of the thermoacoustic behaviour of the combustion system has been developed to see to what extent the observed behaviour on the operational machines can be explained in terms of an acoustic model of the ductwork and a flame characterised simply by a time-delay. This time delay is obtained from the frequency response function of the flame in response to unsteady perturbations in inlet velocity and is calculated using computational fluid dynamics.

The model has also been used to illustrate the importance of fuel supply system design in controlling combustion stability. It is shown that stability can be a strong function of the acoustic impedance of the fuel supply and that this can lead to enhanced or reduced stability depending on the flame characteristics.

Commentary by Dr. Valentin Fuster
1999;():V002T02A016. doi:10.1115/99-GT-111.
FREE TO VIEW

Self-induced combustion driven oscillations are a crucial challenge in the design of advanced gas turbine combustors. Lean premixed combustion, typically used in modern gas turbines, has a pronounced tendency to produce these instabilities.

Thus, the prediction of these thermoacoustic instabilities in the design phase of an engine becomes more and more important. A method based on linear acoustic four-pole elements to predict the instabilities of the ring combustor of the Siemens 3A-series gas turbines will be presented in this paper. The complex network includes the entire system starting from both compressor outlet and fuel supply system and ending at the turbine inlet.

Most of the transfer elements can be described by analytical data. Nevertheless, the most important elements, “flame” and “combustion chamber”, have to be investigated more in detail due to their complex 3D acoustics.

For the turbulent, premixed and swirled flame, a numerical simulation of the transient behavior after a sudden jump in mass flow at the inlet (step-function approach) is used to obtain the flame frequency response for axial direction as well as circumferential direction. This method has been verified for numerous different flame types (Krüger et al. (1998), Bohn et al. (1997), Bohn et al. (1996)). The four-pole element of the annular combustor is derived by an eigenfrequency analysis of the chamber, including a numerical predicted temperature and flow distribution.

The results show the principle possibilities of the instability analysis described. The frequencies predicted correspond well with experience from engine test fields. The importance of several elements for self-induced combustion driven oscillations is pointed out clearly.

Topics: Gas turbines , Flames
Commentary by Dr. Valentin Fuster
1999;():V002T02A017. doi:10.1115/99-GT-112.
FREE TO VIEW

The drive for lower emissions has forced combustor designers to consider lean premixed combustion systems. Unfortunately, premixed combustion systems are particularly susceptible to instabilities, raising large periodic fluctuations in heat release and pressure, that may cause structural damage. A reliable computational tool for predicting the onset of these oscillations would be extremely useful during the design process.

The work contained in this paper utilises computational fluid dynamics to model a simple premixed combustor, consisting of a bluff-body stabilised flame burning within a cylindrical duct. State of the art models are used to represent the combustion heat release and the turbulent transport within the combustor. Both forced oscillations and a nearly self-excited condition are modelled and compared with experiment.

Commentary by Dr. Valentin Fuster
1999;():V002T02A018. doi:10.1115/99-GT-113.
FREE TO VIEW

To get a better understanding of the formation of thermoacoustic oscillations in an annular gasturbine combustor, an analysis of the acoustic eigenmodes has been conducted using the Finite Element (FE) method. The influence of different boundary conditions and a space dependent velocity of sound has been investigated. The boundary conditions actually define the eigenfrequency spectrum. Hence, it is crucial to know e.g. the burner impedance. In case of the combustion system without significant mixing air addition considered in this paper, the space dependence of the velocity of sound is of minor importance for the eigenfrequency spectrum leading to a maximum deviation of only 5% in the eigenvalues.

It is demonstrated that the efficiency of the numerical eigenvalue analysis can be improved by making use of symmetry, by splitting the problem into several steps with alternate boundaries conditions, and by choosing the shift frequency ωs in the range of frequencies one is interested in.

Commentary by Dr. Valentin Fuster
1999;():V002T02A019. doi:10.1115/99-GT-114.
FREE TO VIEW

Rotational CARS (Coherent anti-Stokes Raman Spectroscopy) was used to measure temperature and relative oxygen concentrations in the exhaust gas of a catalytic combustor. This laser technique has the general advantages of such techniques, i.e. possibility of performing in-situ measurements with high spatial and temporal resolution, and non-intrusiveness. Profiles of temperature and oxygen concentrations are presented for a full-load condition, and the measured temperatures are compared with calculated values. Of great importance for the results is the accuracy of the technique, and a sensitivity analysis is performed to test the temperature and oxygen concentration dependence on uncertainties in experimental parameters. It was shown that the accuracy of temperature and oxygen concentration could be improved by fitting the nonresonant susceptibility in the evaluation procedure. The measurements were performed as a project in the European gas turbine program AGATA with the aim to develop a catalytic combustor with ceramic structural components and producing low emissions of pollutants.

Commentary by Dr. Valentin Fuster
1999;():V002T02A020. doi:10.1115/99-GT-115.
FREE TO VIEW

A screening level study has been carried out to examine the potential of using H2-enriched natural gas to improve the combustion performance of gas turbines. H2 has wider flammability limits and a higher flame speed than methane. Many previous studies have shown that when H2 is added to fuel, more efficient combustion and lower emissions will result. However, to date no commercial attempt has been made to improve the combustion performance of a natural gas-fired gas turbine by supplementing the fuel with H2. Four potential options for supplementing natural gas with H2 have been analyzed. Three of these options use the exhaust heat of the gas turbine either directly or indirectly to partially reform methane. The fourth option uses liquid H2 supplied from an industrial gas producer.

Commentary by Dr. Valentin Fuster
1999;():V002T02A021. doi:10.1115/99-GT-116.
FREE TO VIEW

For an aircraft gas turbine engine, the ignition performance is usually expressed in terms of the range of flight conditions over which stable combustion can be established. At present, the size of an aircraft’s gas turbine combustor is governed predominately by its altitude relight performance and is principally derived from existing empirical design rules. These indicate the volume required to give adequate primary zone airodynamic loading to achieve the desired relight performance. A possible means of improving altitude relight performance is to relocate the point of ignition away from the combustor wall to a more favourable location within the combustion chamber. One way of achieving this is through the use of laser ignition. Reported in the paper is an initial programme of work in which the possibility of using laser ignition has been investigated in a gas turbine research combustor operating at several inlet conditions. Comparative results show that, when sited at the same location, laser ignition gave no noticeable improvement in ignition performance when compared to a standard surface discharge igniter (SDI). However, by using laser ignition to locate the ignition site away from the combustor wall, the range of combustor mass flow and AFR at which ≥75% ignition probability could be achieved was increased by approximately 33%.

In conjunction with the experimental study, an ignition probability model, based on the local magnitude of a Karlovitz stretch factor, has been developed to identify suitable regions within the combustor in which to apply laser ignition. The Karlovitz parameter gives an indication as to whether or not a flame kernel will propagate successfully and has been used in conjunction with flow field and scalar distributions generated by Computational Fluid Dynamics (CFD) to yield a 2D map of ignition probability. However, the sensitivity shown by the model to the accuracy of the CFD predictions meant that reliable estimates of optimum ignition sites could only be obtained using experimental fuel and turbulence intensity distributions.

Commentary by Dr. Valentin Fuster
1999;():V002T02A022. doi:10.1115/99-GT-117.
FREE TO VIEW

The effect of synchronized liquid fuel injection into combustion zones within excited air vortices was studied using Planar Laser Induced Fluorescence (PLIF) imaging of OH. A small-scale model of a swirl-stabilized combustor was investigated. Both the liquid fuel and air streams were forced at a normalized frequency (St) of approximately 6. Air vortices were produced by acoustically forcing the air stream. The liquid fuel was modulated at the same frequency and injected at different phase angles relative to the air vortex formation. Air was forced at two levels, and both non-swirling and swirling flow situations (Swirl number of 0.3) were considered. Instantaneous, and phase-averaged PLIF images of OH, and time-averaged temperature measurements were obtained for different phase differences between the two forcing signals.

The PLIF images indicated that with forcing there was an increase in the radial distribution of the reaction zones and a decrease in the height of the flame above the nozzle exit for both non-swirling and swirling conditions. Swirl also produced better mixing along the radial direction for low forcing amplitudes. The phase difference between the two forcing signals produced a significant change in the reaction zones particularly at a low forcing level and with swirl. The PLIF images indicated that the inner region of the flame was the primary reaction zone. When intense combustion occurred at the flame center, high vaporization rate prevented fuel droplets from reaching the external shear layer. The effect of the outer shear layer on combustion was enhanced by swirl, as the swirl increased the droplet dispersion away from the axis.

Commentary by Dr. Valentin Fuster
1999;():V002T02A023. doi:10.1115/99-GT-118.
FREE TO VIEW

Unstable thermoacoustic modes were investigated and controlled in an experimental low-emission swirl stabilized combustor, in which the acoustic boundary conditions were modified to obtain combustion instability. Several axisymmetric and helical unstable modes were identified for fully premixed conditions. The combustion structure associated with the different unstable modes was visualized by phase locked images of OH chemiluminescence. The axisymmetric mode showed large variation of the heat release during one cycle, while the helical mode showed variation in the radial location of maximal heat release. The helical and axisymmetric unstable modes were associated with flow instabilities related to the recirculating flow in the wakelike legion on the combustor axis and shear layer instabilities at the sudden expansion (dump plane), respectively. A closed loop active control system was employed to suppress the thermoacoustic pressure oscillations and to reduce undesired emissions of pollutants during premixed combustion. Microphone and OH emission detection sensors were utilized to monitor the combustion process and provide input to the control system. High frequency valves were employed to modulate the fuel injection. The specific design of the investigated experimental burner allowed testing the effect of different modulated fuel injection concepts on the different combustion instability modes. Symmetric and antisymmetric fuel injection schemes were tested. Suppression levels of up to 12 dB in the pressure oscillations were observed. In some cases a concomitant reductions of NOx and CO emissions were obtained, however, in other instances increased emissions were recorded at reduced pressure oscillations. The effect of the various pulsed fuel injection methods on the combustion structure was investigated.

Commentary by Dr. Valentin Fuster
1999;():V002T02A024. doi:10.1115/99-GT-132.
FREE TO VIEW

A combined analytical/experimental investigation of the thermoacoustic properties of a gas turbine burner with a premixed, turbulent, swirl-stabilized flame is presented. In an enclosed flame, an interaction occurs between acoustic fluctuations and non-steady heat release, which may lead to thermoacoustic instabilities. This interaction may be characterized by the transfer matrix of the burner with flame. The transfer matrix describes the coupling between fluctuations of acoustic pressure and velocity on both sides of burner and flame, incorporating also the effects of heat release fluctuations on the acoustic quantities. The transfer matrix has been modeled and validated with experimental results. For the burner, an analytical model is proposed, which is based on the Bernoulli equation for instationary flow through compact elements. The model is based on the Rankine-Hugoniot relations across a thin heat source. The fundamental assumption underlying the model is that acoustic fluctuations cause modulations of fuel concentrations at the fuel injector, which result, after a certain time lag, in a fluctuating heat release rate at the flame. The oscillating heat release couples with pressure and velocity fluctuations in the combustion chamber, thereby creating a feedback loop between combustor acoustics and flame dynamics which may result in self-excited combustion instability.

The transfer matrix of the burner with flame has been determined experimentally in an atmospheric combustion test facility. The test rig was equipped with loudspeakers and microphones in order to measure the response to an acoustical excitation. Our new flame model shows to be in agreement with the measured results.

Topics: Modeling , Flames
Commentary by Dr. Valentin Fuster
1999;():V002T02A025. doi:10.1115/99-GT-133.
FREE TO VIEW

An experimental method to determine the thermoacoustic properties of a gas turbine combustor using a lean-premixed low emission swirl stabilized burner is presented. To model thermoacoustic oscillations, a combustion system can be described as a network of acoustic elements, representing for example fuel and air supply, burner and flame, combustor, cooling channels, suitable terminations, etc. For most of these elements, simple analytical models provide an adequate description of their thermoacoustic properties. However, the complex response of burner and flame (involving a three-dimensional flow field, recirculation zones, flow instabilities and heat release) to acoustic perturbations has — at least in a first step — to be determined by experiment. In our approach, we describe the burner as an active acoustical two-port, where the state variables pressure and velocity at the inlet and the outlet of the two port are coupled via a four element transfer matrix. This approach is similar to the “black box” theory in communication engineering. To determine all four transfer matrix coefficients, two test states, which are independent in the state vectors, have to be created. This is achieved by using acoustic excitation by loudspeakers upstream and downstream of the burner, respectively. In addition, the burner might act as an acoustic source, emitting acoustic waves due to an unsteady combustion process. The source characteristics were determined by using a third test state, which again must be independent from the two other state vectors. In application to a full size gas turbine burner, the method’s accuracy was tested in a first step without combustion and the results were compared to an analytical model for the burner’s acoustic properties. Then the method was used to determine the burner transfer matrix with combustion. An experimental swirl stabilized premixed gas-turbine burner was used for this purpose. The treatment of burners as acoustic two-ports with feedback including a source term and the experimental determination of the burner transfer matrix is novel.

Topics: Flames
Commentary by Dr. Valentin Fuster
1999;():V002T02A026. doi:10.1115/99-GT-134.
FREE TO VIEW

A detailed thermal analysis involving both measurements and calculations has been carried out in order to determine the wall heal load and to optimize the amount of cooling air for an annular combustor. In calculations, the convective wall heat flux has been detemined by application of a 3D Navier-Stokes Code. Furthermore, the radiation exchange between the hot combustion gases and the liner has been calculated using a multidimensional spectral approach.

Although a quite high thermal power density is found within the combustion chamber the wall heat load is at a low level. Values are well below 80 kW/m2, due to the application of ceramic tiles which have a low thermal conductivity. The wall heat load is dominated by radiation emitted in the lower gas radiation bands (λ < 2.9 μm). The convective wall heat flux is nearly balanced out by the sealing air which is discharged through gaps between the ceramic tiles. The cooling effect of the sealing air, however, is strongly influenced by the 3D near wall flow field in the combustion chamber.

Commentary by Dr. Valentin Fuster
1999;():V002T02A027. doi:10.1115/99-GT-135.
FREE TO VIEW

For lean premixed combustion the NOx emission can be reduced by optimizing the degree of fuel-air mixedness. Since both temporal and spatial mixture variations are of importance, the time resolved planar laser technique of acetone tracer-LIF (laser-induced fluorescence) is used to characterize the mixing quality in an one-to-one scale segment of a Siemens ring shaped gas turbine combustor. Variations of the combustor geometry and of additional mixing devices have been tested, showing the potential to increase the mixing quality. Subsequent tests in a fired atmospheric test rig confirm the influence of the mixing quality, leading to up to 30% further reduction of the NOx emissions.

Commentary by Dr. Valentin Fuster
1999;():V002T02A028. doi:10.1115/99-GT-136.
FREE TO VIEW

An endothermic fuel pyrolytic reactor may favor the formation of benzene and polynuclear aromatic hydrocarbon (PAH) soot precursors which degrade fuel system performance. Both experimental and computational studies are in progress which verify the formation of multi-ring aromatic compounds in endothermic thermal cracking processes. Classic gas phase pyrolytic soot mechanisms may be effective for the prediction of soot precursor concentrations. Hence, a need exists to examine thermal molecular growth processes in a versatile experimental apparatus which can help guide modeling efforts and increase basic understanding (and mitigation) of growth mechanisms. In the present work, cyclic formation experimental data are presented for three mixtures: n-decane, 90 volume % n-decane/10 volume % toluene, and 80 volume % n-decane/20 volume % toluene pyrolysis in a System for Thermal Diagnostic Studies (STDS). The data are qualitatively and quantitatively compared to computations at temperatures, pressures and residence times commensurate with those encountered in endothermic fuel reactor systems. These computations are an attempt to assess the applicability of gas-phase kinetic mechanisms to predict the chemistry of thermal decomposition in the fuel system. Preliminary results show the prominence of molecular growth reactions similar to gas phase pyrolysis mechanisms, starting with hydrocarbon unsaturation, cyclization, and finally, benzene, toluene and heavier PAH formation. Experimental results to date show general qualitative agreement to computational models. However, it is clear that molecular growth mechanisms found to be unimportant at the high temperatures generally associated with combustion can play a significant role in fuel systems.

Commentary by Dr. Valentin Fuster
1999;():V002T02A029. doi:10.1115/99-GT-137.
FREE TO VIEW

The adequacy and accuracy of the constant Schmidt number assumption in predicting turbulent scalar fields in jet-in-crossflows are assessed in the present work. A round jet injected into a confined crossflow in a rectangular tunnel has been simulated using the Reynolds-Averaged Navier-Stokes equations coupled with the standard k-ε turbulence model. A semi-analytical qualitative analysis was made to guide the selection of Schmidt number values. A series of parametric studies were performed, and Schmidt numbers ranging from 0.2 to 1.5 and jet-to-crossflow momentum flux ratios from 8 to 72 were tested. The principal observation is that the Schmidt number does not have an appreciable effect on the species penetration, but it does have a significant effect on species spreading rate in jet-in-crossflows, especially for the cases where the jet-to-crossflow momentum flux ratios are relatively small. A Schmidt number of 0.2 is recommended for best agreement with data. The limitations of the standard kε turbulence model and the constant Schmidt number assumption are discussed.

Topics: Scalars , Turbulence
Commentary by Dr. Valentin Fuster
1999;():V002T02A030. doi:10.1115/99-GT-215.
FREE TO VIEW

A system for the active control of combustion instabilities in liquid-fueled, lean, premixed combustors was demonstrated in a three-nozzle sector combustor, using full-scale engine hardware. Modulation of a portion of the premixed fuel flow led to a reduction of 6.5 dB (2.1X) in the amplitude of the dominant instability mode. Combustor emissions were not adversely affected by the control.

Commentary by Dr. Valentin Fuster
1999;():V002T02A031. doi:10.1115/99-GT-216.
FREE TO VIEW

The need for the development of jet fuels that are oxidatively and thermally stable to 480°C (900°F) has been previously detailed (Edwards and Liberio, 1996). Such a fuel is commonly referred to as JP-900. It is our belief that a new class of additives can be developed which, when added to JP-8, will provide both the oxidative and thermal stability requirements of JP-900.

Our most recent data is presented which explores the potential of 1,2,5-trimethylpyrrole (TMP) as a potential oxygen scavenger.

Topics: Jet fuels , Oxygen
Commentary by Dr. Valentin Fuster
1999;():V002T02A032. doi:10.1115/99-GT-217.
FREE TO VIEW

The development of a viable strategy for limiting coke deposition involves combining synergistic approaches for suppressing deposit buildup and reducing its impact on performance. Candidate approaches, including selection of favorable operating conditions (viz., pressure, temperature, heat flux, residence time and passage size) and coke-tolerant heat exchanger designs, were investigated to evaluate their effectiveness and provide a basis for combining them into a single design philosophy. These approaches were evaluated through testing of current jet fuels in single-tubes and segments of heat exchanger configurations at temperatures up to 1000 F, pressures up to 1200 psi and liquid hourly space velocities up to 40,000/h.

A key result of this work is the ranking of the importance of heat exchanger operating conditions on carbon deposition, with fuel temperature and those parameters that control species diffusion having the most pronounced impact. Residence time and pressure are of lesser importance. Alternative coke-tolerant heat exchanger designs featuring inter-channel communication were evaluated and ranked, with several of these concepts demonstrating improvement over continuous passages.

Topics: Fuels , Aircraft
Commentary by Dr. Valentin Fuster
1999;():V002T02A033. doi:10.1115/99-GT-218.
FREE TO VIEW

Present day land-based gas turbine combustors, operating on oil, must meet strict requirements for emissions (CO, unburned hydrocarbons, particulates, smoke and NOx) and burn stabily without pulsations over a wide range of operating conditions. In addition many engines, such as those produced by ABB, operate with both oil and natural gas fuels either together or independently. This paper concentrates on the development of an oil injection system which is optimised for ABB’s double cone burner (Figure 1) and which does not affect the operation of this burner on natural gas. The development procedure, which involved a coupling of numerical and experimental techniques, is described. The results of the application of this procedure indicate that a simple plain jet atomiser in conjunction with a small quantity of unswirling air admitted at the head of the burner is the best option for this burner.

Commentary by Dr. Valentin Fuster
1999;():V002T02A034. doi:10.1115/99-GT-236.
FREE TO VIEW

This paper describes the development of an ultra-low emission single-can combustor applicable to 200 kW to 3 MW size natural gas-driven gas turbines for cogeneration systems. The combustor, called a three-staged combustor, was designed by applying the theory of lean premixed staged combustion. The combustor comprises two sets of premixing injector tubes located around the combustor liner downstream of the premixing nozzle equipped with a pilot diffusion nozzle in the center. The combustor controls engine output solely by varying the fuel gas flow without the need for complex variable geometry, such as inlet guide vanes, for combustion airflow control. Reliability, response to load variation and retrofit capability have been greatly improved along with wide low-emission operating range. As the result of the atmospheric rig tests, the three-staged combustor has demonstrated superior performance of 3.5 ppm NOx (O2 = 15%) and 7 ppm CO (O2 = 15%) at full load. Assuming the relationship between NOx emission and pressure and taking into account sequential CO oxidation occurring in the scroll, the performance of the combustor at engine operation is expected to be less than 9 ppm NOx (O2 = 15%) and 50 ppm CO (O2 = 15%) emissions between 25% and 100% engine load. During the development, temperature distribution in the atmospheric combustion was measured in detail to gain comprehensive understanding of the low emissions combustion phenomena. The results of the measurement were compared with the theory of lean premixed staged combustion. Employing the concept of effective mixing ratio, the theory of lean premixed staged combustion has proved to be a powerful method to design a lean premixed staged combustor.

Commentary by Dr. Valentin Fuster
1999;():V002T02A035. doi:10.1115/99-GT-237.
FREE TO VIEW

This paper presents a technique for optimising the performance of a diffuser in an industrial gas turbine using validated CFD modelling. The combustor module of the Rolls Royce RB211-DLE industrial engine was modelled from diffuser inlet to combustor inlet, using a hybrid meshing procedure. A CFD model of the current RB211-DLE diffuser and casing was validated against perspex single sector rig data, including pressure probe measurements, oil dot flow tests and a sensitivity analysis. A three-dimensional design process was then undertaken to determine how the shape of the diffuser affects the loss through the system, and hence which type of diffuser would provide the best opportunity for maximising the engine performance. The best two general diffuser designs were optimised using an iterative two-dimensional design process. The performance of these optimised designs was then confirmed by full three-dimensional modelling. This work suggests that a significant improvement in sfc (based on a constant turbine temperature) would be achieved if the optimum diffuser design is installed into the RB211-DLE engine.

Commentary by Dr. Valentin Fuster
1999;():V002T02A036. doi:10.1115/99-GT-238.
FREE TO VIEW

Within a gas turbine engine the flow field issuing from the compression system is non-uniform containing, for example, circumferential and radial variations in the flow field due to wakes from the upstream compressor outlet guide vanes (OGVs). In addition, variations can arise due to the presence of radial load bearing struts within the pre-diffuser. This paper is concerned with the characterisation of this non-uniform flow field, prior to the combustion system, and the subsequent effect on the flame tube fuel injector flows and hence combustion processes.

A mainly experimental investigation has been undertaken using a fully annular test facility which incorporates a single stage axial flow compressor, diffuser and flame tube. Measurements have been made of the flow field, and its frequency content, within the dump cavity. Furthermore, the stagnation pressure presented to the core, outer and dome swirler passages of a fuel injector has been obtained for different circumferential positions of the upstream OGV/pre-diffuser assembly. These pressure variations, amounting to as much as 20% of the pressure drop across the fuel injector, also affect the flow field immediately downstream of the injector. In addition, general variations in pressure around the fuel injector have also been observed due to, for example, the fuel injector position relative to pre-diffuser exit and the flame tube cowl.

Commentary by Dr. Valentin Fuster
1999;():V002T02A037. doi:10.1115/99-GT-239.
FREE TO VIEW

This paper describes the design and initial testing of a second generation, lean-premixed combustor for a 6100 horsepower industrial gas turbine. The full scale, prototype combustor liner employed augmented backside cooling (ABC) as a means of reducing NOx and CO emissions. A thermal barrier coating (TBC) was applied on the liner hot side to reduce thermal flux from the flame zone. The goal of the effort was to demonstrate that the avoidance of film-cooling for the combustor liner would allow emissions reductions in a lean-premixed combustion system.

Testing of the combustor was conducted in both low and high pressure environments. The testing demonstrated that the use of trip-strips for backside cooling provides an effective means of reducing CO emissions. The lower CO levels can be exploited by lowering flame temperatures to achieve lower NOx emissions. Reaction quenching associated with film cooling is indicated as the cause of the higher CO emissions in more conventional liners. Cyclic rig testing showed the TBC to have good short-term durability. Long-term field testing is getting underway.

Commentary by Dr. Valentin Fuster
1999;():V002T02A038. doi:10.1115/99-GT-240.
FREE TO VIEW

AlliedSignal’s F124 combustor is analyzed using CFD as part of an effort to improve the design of the combustor. A reduction of soot emissions, without negative impact on other performance features such as liner life and lean stability, was the primary objective. The existing F124 combustor (TFE1042) was modeled using the commercial CFD-ACE+ software package to validate the CFD results and provide a basis for comparison for the modified design. Two design of experiment (DOE) matrices of the redesigned combustor were analyzed using CFD modeling. The results of the CFD solutions led to the selection of two configurations for combustor rig experimental testing. The test configurations were selected based on CFD predicted trends for smoke, ignition, lean stability and pattern factor. Engine tests demonstrated a smoke number reduction from more than 40 to less than 10. Lean stability was degraded as a result of a leaner primary zone, but adequate lean stability margin was maintained.

Commentary by Dr. Valentin Fuster
1999;():V002T02A039. doi:10.1115/99-GT-241.
FREE TO VIEW

This paper describes the FT8-2 Dry Low NOx (DLN) combustor development process and reviews the development history and initial field experience at a natural gas pipeline station in Germany. The development process is primarily focused on defining a fuel nozzle or injector, investigating emissions, fuel-air mixing, flame stability, acoustics, flashback resistance, and flame disgorgement. Empirical development tools including single nozzle and sector combustion rigs, as well as flame imaging techniques, are discussed. A summary of in-house engine development testing is provided. The control methodology used to meet emissions, while maintaining combustor pressure pulsations at an acceptable level, is provided. The natural gas compressor station design and operational experience with a GHH BORSIG compressor driven by the FT8 engine in Werne, Germany is summarized. Also presented are details of the very short conversion period from Standard to DLN combustor with the first successful ignition of the engine 26 days after work had begun.

Commentary by Dr. Valentin Fuster
1999;():V002T02A040. doi:10.1115/99-GT-269.
FREE TO VIEW

To extend the stable operating range of a lean premixed combustion system, variable geometry can be used to adjust the combustor air flow distribution as gas turbine operating conditions vary. This paper describes the design and preliminary testing of a lean premixed fuel injector that provides the variable geometry function. Test results from both rig and engine evaluations using natural gas are presented. The variable geometry injector has proven successful in the short-term testing conducted to date. Longer term field tests are planned to demonstrate durability.

Commentary by Dr. Valentin Fuster
1999;():V002T02A041. doi:10.1115/99-GT-270.
FREE TO VIEW

The response of NOX to fuel type is determined for lean-prevaporized-premixed combustion in on atmospheric pressure jet-stirred reactor (JSR). Pure, straight chain, alkane fuels from C1 (methane) to C16 (hexadecane) and methanol are tested. Testing is conducted under three conditions, each of which is obtained using a particular inlet-jet nozzle for the JSR. For a selected condition, the fuels are burned under nearly identical macroscopic thermal and fluid mechanical fields in the JSR. A constant mixture inlet total temperature is used that provides for complete vaporization of all of the fuels tested without causing pre-flame reactions in the premixer. Two of the nozzles used have single, centered jets and provide a nominal combustion temperature of 1790 K. The third nozzle has eight diverging jets. In this case, the fuel-air equivalence ratio is increased, providing a nominal combustion temperature of 1850 K.

Lowest levels of NOX are measured when methanol is burned. Upon switching to methane, the NOX concentration increases by 62±10%. For the alkane fuels, methane combustion yields the least amount of NOX. Upon switching from methane to ethane, the NOX increases by 22±2%. Further increases in the size of the alkane fuel molecule (from C2 to C16) show small increases (8±5%) in the NOX concentration. Although the JSR-condition selected affects the absolute NOX concentration, the trends in NOX concentration with respect to fuel-type are the same for the three conditions used.

The JSR is modeled as a single perfectly stirred reactor (PSR) operating at the experimental residence time and combustion temperature. This modeling shows that the increase in the NOX measured for the alkane fuels may be explained in terms of the increase in O-atom concentration with increasing C/H ratio. In order to explore the effect of Fenimore prompt NOX, a two-PSRs-in-series model is used. The first PSR is assigned a residence time equal to 5% the total residence time of the reactor. The second PSR accounts for the remaining 95% of the reactor residence time. Because of its short lifetime, the CH radical is effectively restricted to the first PSR, and prompt NOX mainly forms in this zone. Mechanisms directly dependent on the O-atom form NOX throughout the reactor. The modeling suggests that prompt NOX is responsible for the greater concentration of NOX found in the methane combustion compared to the methanol combustion.

Commentary by Dr. Valentin Fuster
1999;():V002T02A042. doi:10.1115/99-GT-271.
FREE TO VIEW

In the numerical simulation of turbulent reacting flows, the high computational cost of integrating the reaction equations precludes the inclusion of detailed chemistry schemes, therefore reduced reaction mechanisms have been the more popular route for describing combustion chemistry, albeit at the loss of generality. The in situ adaptive tabulation scheme (ISAT) has significantly alleviated this problem by facilitating the efficient integration of the reaction equations via a unique combination of direct integration and dynamic creation of a look-up table, thus allowing for the implementation of detailed chemistry schemes in turbulent reacting flow calculations. In the present paper, the probability density function (PDF) method for turbulent combustion modeling is combined with the ISAT in a combustor design system, and calculations of a piloted jet diffusion flame and a low-emissions premixed gas turbine combustor are performed. It is demonstrated that the results are in good agreement with experimental data and computations of practical turbulent reacting flows with detailed chemistry schemes are affordable.

Commentary by Dr. Valentin Fuster
1999;():V002T02A043. doi:10.1115/99-GT-272.
FREE TO VIEW

Based on the characteristics of turbulent combustion in lean-premix combustion chambers, this paper presents a combustion model which solves transport equations for six chemical species. The source terms are calculated by probability weighted integration of 35 elementary reaction rates. The model presented here does not include any adjustable parameters. Therefore, it is universal in its character for conditions of highly turbulent premixed lean to stoichiometric combustion.

The model is applicable to fuel compositions including methane, carbon monoxide, and hydrogen. The application is shown for a test case burning methane in lean-premixed mode.

Commentary by Dr. Valentin Fuster
1999;():V002T02A044. doi:10.1115/99-GT-273.
FREE TO VIEW

An advanced design system has been developed for combustor flow analysis. The system is based on the finite-volume methodology and is of second-order numerical accuracy. Use of co-located grids and Cartesian velocities offers significant advantages over previous staggered-grids, covariant-velocities based schemes. The physicochemical effects are simulated by the standard k-ε model for turbulence, the eddy-breakup model with a two-step general hydrocarbon chemistry for combustion, and a stochastic Lagrangian transport and evaporation model for spray. The developed design system has been applied to analyze a production gas turbine combustor configuration and several design changes. The calculated exit-plane temperature profiles compare well against full-scale rig data. The trends of the exit temperature profiles, showing the effect of design changes to the geometry and flow-splits of various combustor features, are well predicted. The study demonstrates the developed design system to be a robust and viable tool for analyzing and guiding combustor design.

Commentary by Dr. Valentin Fuster
1999;():V002T02A045. doi:10.1115/99-GT-274.
FREE TO VIEW

To analyze combustion oscillation in the premixed combustor, a large-eddy simulation program for premixed combustion flow was developed. The subgrid scale (SGS) model of eddy viscosity type for compressible turbulence (Speziale et al., 1988) was adopted to treat the SGS fluxes. The fractal flamelet model, which utilizes the fractal properties of the turbulent premixed flame to obtain the reaction rate, was developed.

Premixed combustion without oscillation was analyzed to verify the present method. The computational results showed good accordance with experimental data (Rydén et al., 1993).

The combustion oscillation of an “established buzz” type in the premixed combustor (Langhorne, 1988) was also analyzed. The present method succeeded in capturing the oscillation accurately. The detailed mechanism was investigated. The appearance of the non-heat release region, which is generated because the supply of the unburnt gas into the combustion zone stagnates, and its disappearance play an important role.

Commentary by Dr. Valentin Fuster
1999;():V002T02A046. doi:10.1115/99-GT-275.
FREE TO VIEW

A method is presented by which the product composition and temperature of constant pressure combustion reactions can be calculated for non equilibrium conditions, by constraining the products free energy and entropy. The calculations for a hydrogen/ oxygen system are compared with chemical kinetic predictions. From the calculated compositions the relationship between free energy and extent of reaction are derived and thence how the individual product mole fractions vary with extent of reaction. The application of these techniques to modelling combustion chemistry is discussed.

Commentary by Dr. Valentin Fuster
1999;():V002T02A047. doi:10.1115/99-GT-276.
FREE TO VIEW

This paper involves a theoretical analysis of the propagation of sound within chemically reacting hydrocarbon combustion products. A new procedure for determining the sound speed and absorption coefficient was developed. Using a similar approach can do analyses for other chemically reacting gas mixture. One-dimensional sound propagation was assumed. Molecular diffusion effects, such as viscous stress, heat conduction, and mass diffusion were appropriately neglected along with possible effects due to vibrational non-equilibrium. In this way only the effects of chemical reactions on sound propagation was considered. The expressions for sound speed and absorption coefficient were derived as a function of frequency.

Commentary by Dr. Valentin Fuster
1999;():V002T02A048. doi:10.1115/99-GT-277.
FREE TO VIEW

The extent of air and fuel mixing prior to combustion in lean premixed combustion has been shown to drastically affect combustor performance, both in terms of emissions and stability. Standard extraction probes are often used for measuring the spatial (average over time) distribution of fuel concentration. However, the temporal fluctuations in fuel mole fraction, which are averaged by conventional extraction probes, have been also shown to drastically affect combustor performance. Several methods have been developed to measure the fluctuations in fuel mole fraction both temporally and spatially. These techniques are often difficult or impossible to apply to an operating combustor at high pressures and temperatures.

In this paper, we describe a Fast Response Extraction Probe (FREP) which is capable of measuring temporal mole fraction fluctuations up to frequencies of 1 kHz. A short, capillary tube is inserted into the premixing passage (which is at high pressure) for sampling the flow. The capillary tube is connected to a small gas cell at atmospheric pressure. Light from a 3.39 μm He-Ne laser is passed through the gas cell. By measuring the absorption of laser light, the concentration of hydrocarbons can be determined. The temporal response of the system is dependent on the geometry of the gas cell and sampling tube and the pressure drop through the sampling tube. A model for predicting the time response of the FREP is presented and compared to a laboratory scale experiment. The FREP was also tested in a high-pressure combustion rig at United Technologies Research Center. It is shown that the FREP is capable of measuring fuel-air fluctuations at gas turbine conditions. It is also shown that, at the conditions studied, there is a relationship between pressure oscillations and fuel mole fraction oscillations.

Commentary by Dr. Valentin Fuster
1999;():V002T02A049. doi:10.1115/99-GT-278.
FREE TO VIEW

Premixed flames have shown to provide sufficient low NOx emission levels in gas turbine combustor. To achieve this goal several issues are to be addressed. Highly uniform fuel air ratio along with proper velocity profiles at the flame front has to be achieved. Vortex generators are commonly employed inside lean premixed combustor to generate a swirling flow that can assure the two aspects above. An experimental parametric study of the isothermal swirl flow of a non-confined prototype premixer is presented. Several different geometrical details have been considered for two different fuel air ratios. Velocity fields and mixing distribution between fuel and air at the premixer exit are described. Velocity and turbulence data can be related to formation of flashback and temporal homogeneity of the mixture respectively while the degree of spatial mixing, in terms of segregation factor, is connected with NOx rate of formation. In both cases the vigorous and high responsive character of the vortex generated by the swirler plays a major role. The final objective is twofold: to define an optimum geometrical arrangement to be tested in subsequent combustion tests and to demonstrate the feasibility of a rapid assessment method of degree of mixing between fuel and air.

Commentary by Dr. Valentin Fuster
1999;():V002T02A050. doi:10.1115/99-GT-295.
FREE TO VIEW

Combustor hardware employing catalytic combustion technology has been developed for a 1.5 MW gas turbine. This system, combined with state of the art catalyst technology, was used to demonstrate ultra-low emissions on the engine. The demonstrator combustor utilizes a two stage lean premix preburner system to obtain the required catalyst inlet temperatures and low NOx over the operating load range. The performance of the preburner system was characterized during engine tests by measuring temperature rise and emissions just downstream of the preburner. A fuel schedule for the primary and secondary stages was selected to give NOx emissions below 2 ppmv at the engine exhaust. Overall engine performance was measured over the full load range. Emissions of NOx < 3 ppmv and CO and UHC < 5 ppmv were obtained at 72% to 100% load. Combustor dynamics were shown to be less than 0.3 psi(rms). This combustor operated for 1000 hours on a dynamometer test facility and showed low emissions performance over this period.

Commentary by Dr. Valentin Fuster
1999;():V002T02A051. doi:10.1115/99-GT-296.
FREE TO VIEW

Calculations of NOx emissions were made for the original high pressure combustor and for the original and a modified design of the low pressure combustor used in a Compressed Air Energy Storage (CAES) plant. All were typical diffusion flame combustors. Since a CAES plant has an independent air supply, the relationship between combustor inlet temperature and pressure is not typical for gas turbines, and the pressure level for the HP combustor is unusually high (up to 4.5 MPa). Vitiated air from HP combustor exhaust is used as combustion air in the LP combustor. The NOx emissions prediction method, which was used, for calculations is based on a flamelet model which takes detailed kinetic schemes for fuel oxidation, NOx generation and turbulence/chemistry interaction into account. Site measurements over the entire load curve confirmed the numerical predictions for both the original combustors and the newly developed LP combustor design.

Topics: Nitrogen oxides
Commentary by Dr. Valentin Fuster
1999;():V002T02A052. doi:10.1115/99-GT-297.
FREE TO VIEW

A high-pressure facility capable for testing Lean Premixed Pre-vaporized (LPP) combustor geometries is reported. The proposed design enables the use of both single-point and whole-field non-intrusive measuring techniques when the combustor operates under conditions representative of gas turbines or jet engines operation, thereby enabling the complete characterization of the turbulent flow field that develops inside LPP combustors. The individual control of the air and fuel supplies, provided, respectively, by a number of industrial compressors and a specific LPG supply facility, allows the investigation of a wide range of operating conditions. In this paper, the facility specifications are explained and adequately justified. Preliminary results are presented, showing the suitability of the design.

Commentary by Dr. Valentin Fuster
1999;():V002T02A053. doi:10.1115/99-GT-298.
FREE TO VIEW

The autoxidation of Jet A, dodecane, and a dodecane-15%-cumene blend doped with sulfur compounds were studied at 433 K. Oxygen, hydro peroxide and soluble gum were monitored during the autoxidation. Dodecane, cumene, and the dodecane-15%-cumene blend autoxidized rapidly, while Jet A had an induction period followed by a relatively slow post autoxidation. The results suggest that an inhibitor formed early in the post autoxidation of Jet A. Gum formed in the autoxidation of Jet A, whereas none was detected in dodecane, cumene, or dodecane-15% cumene. However, gum was detected in dodecane and dodecane-15% cumene doped with thiols and disulfides. Alkyl thiols and disulfides reduced the rate of autoxidation of dodecane, and there was an induction period in the formation of gum. Traces of sulfur (≈4 ppm) inhibited the autoxidation of dodecane-15% cumene in a way that resembled the post autoxidation of Jet A. Adding an organic base increased the rate of post autoxidation in Jet A and prevented formation of the oxidation inhibitor. An inhibition mechanism is proposed in which phenois are formed via acid-catalyzed decomposition of benzylic hydro peroxides.

Commentary by Dr. Valentin Fuster
1999;():V002T02A054. doi:10.1115/99-GT-299.
FREE TO VIEW

It is well known that fuel preparation, its method of injection into a combustor and its atomization characteristics have a significant impact on emissions. A simple dilute spray model which assumes that droplet heating and vaporization occur in sequence has been implemented in the past within computational fluid dynamics (CFD) codes at GE and has been used extensively for combustion applications. This spray model coupled with an appropriate combustion model makes reasonable predictions of the combustor pattern factor and emissions. In order to improve upon this predictive ability, a more advanced quasi-steady droplet vaporization model has been considered. This paper describes the evaluation of this advanced model. In this new approach, droplet heating and vaporization take place simultaneously (which is more realistic). In addition, the transport properties of both the liquid and vapor phases are allowed to vary as a function of pressure, gas phase temperature and droplet temperature. These transport properties which are most up to date have been compiled from various sources and appropriately curve-fit in the form of polynomials. Validation of this new approach for a single droplet was initially performed. Subsequently, calculations of the flow and temperature field were conducted and emissions (NOx, CO and UHC) were predicted for a modern single annular turbofan engine combustor using both the standard spray model and the advanced spray model. The effect of the number of droplet size ranges as well as the effect of stochastic treatment of the droplets were both investigated.

Commentary by Dr. Valentin Fuster
1999;():V002T02A055. doi:10.1115/99-GT-300.
FREE TO VIEW

In this paper CFD analysis of the steady two-phase turbulent combusting flow in a single annular low-NOx combustor is presented. For this purpose the commercial code CFD-ACE (1998) was used, where Eulerian equations are solved for the gas phase and the liquid spray fuel droplets are treated in a Lagrangian frame of reference allowing for evaporation of droplets and providing source terms for the gas phase. The standard k-ε model was used for turbulence and an assumed shape probability density function was used for the instantaneous chemistry in the conserved scalar combustion model. Thermal NOx is assumed to be the only source of NOx production and is decoupled from the gas phase reacting flow and calculated in a postprocessing step. The calculation is done on a block structured multi-domain computational grid. Particular attention has be paid to the detailed modeling of the fuel injector having multiple air swirler passages starting from the trailing edge of the air swirler vanes and utilizing up to 400000 computational grid cells for the entire model.

The model represents the single annular low-NOx combustor for the BR700 aircraft engine family, which is based on a Rich Burn - Quick Quench - Lean Burn (RQL) concept. CFD analysis is done for high power reduced take off conditions and is compared with full annular rig test results for the temperature traverse and the integral EINOx. The results imply satisfactory prediction capability for the EINOx and the average radial temperature distribution. The prediction of the details of the temperature traverse is not satisfactory and will remain a challenge for the future.

Commentary by Dr. Valentin Fuster
1999;():V002T02A056. doi:10.1115/99-GT-301.
FREE TO VIEW

This paper reports a numerical investigation of the transcritical and supercritical droplet vaporization phenomena. The simulation is based on the time-dependent conservation equations for liquid and gas phases, pressure-dependent variable thermophysical properties, and a detailed treatment of liquid-vapor phase equilibrium at the droplet surface. The numerical solution of the two-phase equations employs an arbitrary Eulerian-Lagrangian, explicit-implicit method with a dynamically adaptive mesh. Three different equations of state (EOS), namely the Redlich-Kwong (RK), the Peng-Robinson (PR) and Soave-Redlich-Kwong (SRK) EOS, are employed to represent phase equilibrium at the droplet surface. In addition, two different methods are used to determine the liquid density. Results indicate that the predictions of RK-EOS are significantly different from those obtained by using the RK-EOS and SRK-EOS. For the phase-equilibrium of n-heptane-nitrogen system, the RK-EOS predicts higher liquid-phase solubility of nitrogen, higher fuel vapor concentration, lower critical-mixing-state temperature, and lower enthalpy of vaporization. As a consequence, it significantly overpredicts droplet vaporization rates, and underpredicts droplet lifetimes compared to those predicted by PR- and SRK-EOS. In contrast, predictions using the PR-EOS and SRK-EOS show excellent agreement with each other and with experimental data over a wide range of conditions. A detailed investigation of the transcritical droplet vaporization phenomena indicates that at low to moderate ambient temperatures, the droplet lifetime first increases and then decreases as the ambient pressure is increased. At high ambient temperatures, however, the droplet lifetime decreases monotonically with pressure. This behavior is in accord with the reported experimental data.

Topics: Fuels , Drops , Evaporation
Commentary by Dr. Valentin Fuster
1999;():V002T02A057. doi:10.1115/99-GT-302.
FREE TO VIEW

Most types of combustion-driven devices experience combustion instabilities. For aero-engine combustors, the frequency of this oscillation is typically in the range 60–120Hz and is commonly called ‘rumble’. The rumble oscillations involve coupling between the air and fuel supplies and unsteady flow in the combustor. Essentially pressure fluctuations alter the inlet fuel and air, thereby changing the rate of combustion, which at certain frequencies further enhances the pressure perturbation and so leads to self-excited oscillations. The large residence time of the liquid fuel droplets, at idle and sub-idle conditions, means that liquid and gaseous phases must both be considered. In the present work, we use a numerical model to investigate forced unsteady combustion due to specified time-dependent variations in the fuel and air supplies. Harmonic variations in inlet air and fuel flows have been considered and the resulting unsteady combustion calculated. The influence of droplet size distribution has also been investigated. The calculations provide insight into understanding the interaction between atomization, unsteady combustion and flow oscillations.

Commentary by Dr. Valentin Fuster
1999;():V002T02A058. doi:10.1115/99-GT-356.
FREE TO VIEW

Reviewed are power system concepts employing the solid oxide fuel cell (SOFC) at atmospheric pressure in simple cycle; in an atmospheric pressure hybrid cycle with a gas turbine (SOFC/GT); and in a pressurized SOFC/GT hybrid (PSOFC/GT). Estimates of power system performance are presented and discussed. Simple atmospheric pressure SOFC systems designed for combined heat and power (CHP) application can approach 50% electric generation efficiency (net AC/LHV) and 80% fuel effectiveness [(net AC + useful heat)/LHV]. Pressurized SOFC systems with intercooled, recuperated, and SOFC-reheated GT cycles can approach 70% electric generation efficiency, while the atmospheric pressure SOFC/GT hybrid cycle and a simple pressurized SOFC/GT cycle can approach 55% and 60% generation efficiency, respectively. These high levels of efficiency are extraordinary in that they are achievable at the MW capacity level.

Commentary by Dr. Valentin Fuster
1999;():V002T02A059. doi:10.1115/99-GT-357.
FREE TO VIEW

A methodology is presented in this paper on the modeling of NOx formation in diffusion flame combustors where both droplet burning and partially premixed reaction proceed simultaneously. The model simulates various combustion zones with an arrangement of reactors that are coupled with a detailed chemical reaction scheme. In this model, the primary zone of the combustor comprises a reactor representing contribution from droplet burning under stoichiometric conditions and a mixing reactor that provides additional air or fuel to the primary zone. The additional flow allows forming a fuel vapor/air mixture distribution that reflects the unmixedness nature of the fuel injection process. Expressions to estimate the extent of deviation in fuel/air ratios from the mean value, and the duration of droplet burning under stoichiometric conditions were derived. The derivation of the expressions utilized a data base obtained in a parametric study performed using a conventional gas turbine combustor where the primary zone equivalence ratio varied over a wide range of operation. The application of the developed model to a production combustor indicated that most of the NOx produced under the engine takeoff mode occurred in the primary as well as the intermediate regions. The delay in NOx formation is attributed to the operation of the primary zone under fuel rich conditions resulting in a less favorable condition for NOx formation. The residence time for droplet burning increased with a decrease in engine power. The lower primary zone gas temperature that limits the spray evaporation process coupled with the leaner primary zone mixtures under idle and low power modes increases the NOx contribution from liquid droplet combustion in diffusion flames. Good agreement was achieved between the measured and calculated NOx emissions for the production combustor. This indicates that the simulation of the diffusion flame by a combined droplet burning and fuel vapor/air mixture distribution offers a promising approach for estimating NOx emissions in combustors, in particular for those with significant deviation from traditional stoichiometry in the primary combustion zone.

Commentary by Dr. Valentin Fuster
1999;():V002T02A060. doi:10.1115/99-GT-358.
FREE TO VIEW

Aircraft fitted with afterburner systems for increased thrust have been observed to have NOx emissions with a higher proportion of nitrogen dioxide (NO2) than non-augmented aircraft. These emissions are generally characterised by a brown plume and has implications for aircraft visibility and stealth as well as environmental considerations. This paper describes the CFD modelling of NOx emissions from a modern afterburner system with particular emphasis on the formation of nitric oxide (NO) and the subsequent conversion of NO to NO2.

A commercial CFD code, was used to solve a three dimensional model of a “burn then mix” afterburner system under investigation. A post processor package has been developed and was used to calculate both NO and NO2 concentrations. Four reheat settings were investigated; minimum, 25%, 50% and maximum reheat. For all conditions investigated the bulk of NOx emission was found in the core, stemming from the vitiated combustor air flow. NOx was also formed in the bypass stream, the production zone was found to be close to the fuel sprayers and flame stabiliser at minimum reheat, but moved downstream towards the exit nozzle as reheat power was increased. The model showed that for all the conditions under investigation, over 90% of the NOx produced in the reheat system was formed via the thermal-NO route.

The model has been compared with centre-line traverse data measured at the exit nozzle of the engine on a sea level static test bed. The predicted NOx emissions agreed quantitatively with the experimental measurements to within ± 5%.

Commentary by Dr. Valentin Fuster
1999;():V002T02A061. doi:10.1115/99-GT-359.
FREE TO VIEW

In view of increasingly stringent NOx emissions regulations on stationary gas turbines, lean combustion offers an attractive option to reduce reaction temperatures and thereby decrease NOx production. Under lean operation, however, the reaction is vulnerable to blowout. It is herein postulated that pilot hydrogen dopant injection, discretely located, can enhance the lean blowout performance without sacrificing overall performance. The present study addresses this hypothesis in a research combustor assembly, operated at atmospheric pressure, and fired on natural gas using rapid mixing injection, typical of commercial units. Five hydrogen injector scenarios are investigated. The results show that (1) pilot hydrogen dopant injection, discretely located, leads to improved lean blowout performance and (2) the location of discrete injection has a significant impact on the effectiveness of the doping strategy.

Commentary by Dr. Valentin Fuster
1999;():V002T02A062. doi:10.1115/99-GT-360.
FREE TO VIEW

This paper describes safety issues important to the operation of combined fuel cell and gas turbine (hybrid) systems, and provides motivation for building dynamic modeling tools to support their development. It also describes two models — a steam reformer and a fuel cell — that will be used to investigate the dynamic performance of a hybrid system. The present goals are to develop dynamic models for these two components, ensure their reliability, and obtain a basic understanding of their performance prior to integration into a complete hybrid system model. Because of the large physical domain to be analyzed in the integrated hybrid system, both reformer and fuel cell models are simplified to a one-dimensional system of equations. Model results are presented for a tubular, counterflow steam reformer showing methane conversion and temperature behavior during initial startup, and following several step change perturbations. For the fuel cell model, a generic planar type is analyzed showing voltage and current behavior following step changes in load resistance and fuel input. The results provide confidence in each model’s reliability, enabling them to be integrated for hybrid system simulation. Results from the integrated simulations will provide guidance on future hybrid technology development needs.

Commentary by Dr. Valentin Fuster
1999;():V002T02A063. doi:10.1115/99-GT-361.
FREE TO VIEW

In the deregulated electric utility environment, distributed power is expected to play a major role in meeting national power generation requirements of the 21st century. Compliance with environmental regulations and ultra-high efficiency will be attributes in penetrating distributed power markets. Fuel cell/turbomachinery hybrids are poised to become the first choice for these applications, yielding emissions of nitrogen oxides below 2 ppmv and low CO2 emissions due to high efficiency (>70% in 20 MW power packages). The challenges faced by hybrid systems are: system integration, power electronics/power conditioning, system controls, operational and transient dynamics, and commercial system cost. Opportunities to use existing gas turbines as the selected turbomachinery are being investigated, prompted by size flexibility and the savings in time and investment that this offers. This paper is an attempt to address the opportunities and challenges posed by a variety of fuel cell/turbomachinery hybrids.

Commentary by Dr. Valentin Fuster
1999;():V002T02A064. doi:10.1115/99-GT-399.
FREE TO VIEW

The ability of a lean-premixed combustion system to minimize emissions while maintaining combustion stability over the operating curve relies upon how well the fuel nozzle premixes the fuel and air. As the level of premixing increases, NOx emissions at a given flame temperature decrease until a perfectly premixed condition is achieved. The objective of this paper is to quantify the level of premixing achieved by a premixing nozzle using an acetone fluorescence technique and determine its relationship to NOx emissions and combustion stability.

The technique of using acetone fluorescence has been used as a fast and quantitative diagnostic to map the fuel-air distribution. This technique has been applied to the development of a lean premixing nozzle to measure the fuel air distribution at the fuel nozzle exit plane. In this study, the fuel air distribution is presented as 2-D images. The average fuel/air ratio and the standard deviation are calculated at various annular regions to determine the distribution as a function of radius. A single unmixedness parameter (σ/μ) over the entire annulus is also calculated to allow relative ranking of the various fuel nozzle configurations.

The fluorescence data is acquired for various nozzle hardware configurations in an atmospheric test facility. Fuel and air flow conditions are determined by scaling engine conditions to cold flow conditions and matching the fuel to air momentum ratio at the fuel injection site. Measured fuel/air distributions, 6 mm downstream of the nozzle exit plane, from the acetone fluorescence technique are correlated to emissions and acoustic measurements made at full pressure and temperature conditions in a single-nozzle test rig.

The paper includes a description of the acetone fluorescence technique, the method for optimizing the fuel/air distribution through changes to the main gas fuel injection array, and correlations made between the fuel/air distribution, nozzle geometry, power setting, emissions and combustor acoustics.

Topics: Fluorescence , Fuels , Nozzles
Commentary by Dr. Valentin Fuster
1999;():V002T02A065. doi:10.1115/99-GT-400.
FREE TO VIEW

Fuel cells are emerging as a major new power generation technology that is particularly suitable for distributed power generation, high-efficiency, and low pollutant emission. An interesting combined cycle, the “HYBRID,” has recently been scoped “on paper” that portends the potential of ultra-high efficiency (approaching 80%) in which a gas turbine is synergistically combined with a fuel cell into a unique combined cycle. This paper introduces hybrid technology to the gas turbine community as a whole, and summarizes the current and projected activities associated with this emerging concept.

Commentary by Dr. Valentin Fuster
1999;():V002T02A066. doi:10.1115/99-GT-419.
FREE TO VIEW

The concept of hybrids combining fuel cell and gas turbine systems is without question neoteric, and probably is less than eight years old. However, this concept is in a sense a logical development derived from the many early systems that embodied the key features of rotating machinery to compress air. It was the introduction of high temperature fuel cells such as the solid oxide fuel cell (SOFC) that allowed the concept of hybrid gas turbine fuel cell systems to take root. The SOFC with an operating temperature circa 1000° C matched well with small industrial gas turbines that had firing temperatures on the same order.

The recognition that the SOFC could be substituted for the gas turbine combustor was the first step into the realm of fuel cell topping systems. Fuel cells in general were recognized as having higher efficiencies at elevated pressures. Thus the hybrid topping system where the gas turbine pressurized the fuel cell and the fuel cell supplied the hot gases for expansion over the turbine promised to provide a high level of synergy between the two systems.

Bottoming systems using the exhaust of a gas turbine as the working fluid of a fuel cell such as the molten carbonate fuel cell (MCFC) have been identified and are potential future power generation hybrid systems. The MCFC is especially well suited to the bottoming role because of the need to have carbon dioxide present in the inlet air stream. The carbon dioxide in the gas turbine exhaust allows the high temperature blower, normally used to recirculate and inject exhaust products into the inlet air, to be eliminated.

Hybrid systems have the potential of achieving fossil fuel to electricity conversion efficiencies on the order of 70% and higher. The costs of hybrid systems in dollars per kilowatt are generally higher than say an advanced gas turbine that is available today but not by much. The net energy output over the life of a hybrid topping system is similar to that of a recuperated gas turbine but possibly lower than a high-efficiency simple-cycle machine, depending on the efficiency of the hybrid.

Methodologies to aid in the selection of the hybrid system for future development have to be developed and used consistently. Life cycle analyses (LFA) provide a framework for such selection processes. In particular the concept of net energy output provides a mechanism to assign relative worth to competing concepts.

Commentary by Dr. Valentin Fuster
1999;():V002T02A067. doi:10.1115/99-GT-430.
FREE TO VIEW

The integration of a fuel cell and a gas turbine is a natural evolution in the quest for improved generation efficiency with clean emissions. Integration is achieved by using the gas turbine compressor as the air mover for the fuel cell, and using the high temperature exhaust of the fuel cell to supplant the gas turbine combustor.

Edison Technology Solutions (ETS), the California Energy Commission, the U. S. Department of Energy (DOE), and Siemens Westinghouse Power Corporation are jointly sponsoring a project to design and fabricate a hybrid cycle power system that couples a Pressurized Solid Oxide Fuel Cell (SOFC) generator module with a Micro Turbine Generator (MTG), to yield a system with a nominal capacity of 250 kW and a power generation efficiency approaching 60%. The SOFC will supply approximately 80% of the output power and the MTG 20%. The MTG functions primarily as a turbo charger for the SOFC with some additional shaft power available to turn an integral generator. The demonstration will be conducted at the University of California at Irvine. Startup is scheduled for the summer of 1999.

This project is expected to be the first demonstration of a hybrid cycle employing a pressurized SOFC to supplant the combustor of a gas turbine generator. The parameters of evaluation are the power output, electric generation efficiency, degradation characteristics, operability, and operating power range.

The primary project objective is to demonstrate successful startup and operation over the design power range at efficiencies approaching 60%. Secondary objectives are to evaluate operability, component reliability, and to conduct a design evaluation to develop improvements for subsequent designs.

Commentary by Dr. Valentin Fuster
1999;():V002T02A068. doi:10.1115/99-GT-431.
FREE TO VIEW

The objective of this work is to investigate the finite rate kinetics effect on NO formation in turbulent non-premixed flames by using the presumed pdf method with two scalars where one scalar accounts for finite-rate chemistry. To overcome the increase of computational time demanded by the above method, an adaptive non-structured table look-up procedure has been implemented to interpolate thermochemical properties. These quantities are obtained from the look-up table during the integration of the turbulent reacting flow thus replacing a very computationally onerous integration with a less expensive interpolation. To study the non-equilibrium effects in non-premixed combustion and to assess the capabilities of the adaptive table, the numerical simulation of a turbulent syngas-air flame (experimentally studied by Drake et al.[4]) has been performed under equilibrium and non-equilibrium conditions.

Commentary by Dr. Valentin Fuster

Oil and Gas Applications

1999;():V002T03A001. doi:10.1115/99-GT-012.
FREE TO VIEW

In the design and testing of gas compressors, the correct determination of the thermodynamic properties of the gas. such as enthalpy, entropy and density from pressure, temperature and composition, plays an important role. Due to the wide range of conditions encountered, pressure, specific volume and temperature (p-v-T) equations of state (EOS) and ideal gas heat capacities, along with measured data, are used to determine the isentropic efficiency of a compressor configuration and to model the actual behavior of real gases and compressors.

There are many possible model choices. The final selection should depend on the applicability of the EOS to the gas and the temperature dependence of the heat capacities, as well as the particular process of interest along with the range of pressures and temperatures encountered. This paper compares the thermodynamic properties from five commonly used equations in the gas compressor industry: the Redlich-Kwong (RK), Redlich-Kwong-Soave (RKS), Peog-Robinson (PR), Benedict-Webb-Rubin-Starling (BWRS), and Lee-Kesler-Plocker (LKP) models. It also compares them with a high accuracy EOS for methane from Wagner and Setzmann in the common range for gas compressors. The validity of a linear temperature dependence for ideal gas heat capacities is also evaluated. The objective was to determine if the models give significant differences in their predicted efficiencies.

It was found that different EOS gave somewhat different enthalpy changes for methane, ethane and nitrogen for real compressions. This appeared to be connected to the different densities given by the models. Interestingly, the isentropic enthalpy changes are quite similar, suggesting that the effect is canceled out when two properties are involved. However, since the efficiency is the ratio of isentropic enthalpy change to actual enthalpy change, the EOS yield different efficiencies. These differences are on the same order as the typical tolerances allowed for prediction and testing of industrial gas compressors (3 to 5%) and comparisons with the highly accurate equation of state for pure methane from Wagner and Setzmann (1991) showed similar differences.

Commonly, the ideal gas heat capacity is assumed linear in temperature from 10 to 150°C (50 to 300°F). Comparison of this form with a quadratic expression from the literature and the highly accurate equation of Wagner and Setzmann for methane, showed insignificant differences among the methods for temperatures up to 600°K (1080°R).

Commentary by Dr. Valentin Fuster
1999;():V002T03A002. doi:10.1115/99-GT-051.
FREE TO VIEW

This paper evaluates and demonstrates how the public domain data provided by individual interstate pipeline companies to FERC, when combined with individual company equipment lists, can be used to regress industry information on cost of operations and maintenance, fuel gas used, and cost of fuel and power. The paper describes the methods of analysts and identifies their limitations. The paper presents results of such regression analysis as average and variance of cost and fuel usage for industrial gas turbines and aeroderivative gas turbines. It provides further comparisons between gas turbine prime movers, reciprocating engine prime movers, and electric motor drives, and presents annual costs per installed horsepower as a function of turbine size. The paper is based on work performed for PRC International and the Gas Research Institute.

Commentary by Dr. Valentin Fuster
1999;():V002T03A003. doi:10.1115/99-GT-091.
FREE TO VIEW

The Rolls-Royce Avon aeroderivative gas generator has amassed over 45 million hours of operation in industrial applications since its introduction in 1964. However, its reputation for reliability and long life has kept it in the forefront in many markets. With the increasing world-wide requirements for lower emissions, the decision was taken to incorporate a Dry Low Emissions (DLE) combustion system to provide operators with increased options. It was also decided that the retrofit market would be fully addressed, thus driving the need to minimise changes to the basic engine. How this project has been approached, with a low risk solution both to the mechanical and operational integrity, is the basis for this paper.

Topics: Emissions
Commentary by Dr. Valentin Fuster
1999;():V002T03A004. doi:10.1115/99-GT-173.
FREE TO VIEW

A thermodynamic, environmental and economic assessment of an exhaust gas recirculation (EGR) system for NOx reduction has been carried out on an RB211 gas turbine based compressor station. The configured system was evaluated using a commercial process simulation software ASPEN PLUS® for the EGR process, along with a one dimensional model for the prediction of NOx. The assessment was focused on a realistic system of 20% gas recirculation cooled 300 °C with an aerial cooler. Detailed economic analysis based on present value cost per unit mechanical energy (kWh), showed that there is no economic advantage in implementing an EGR system in an existing gas turbine based station. Although the environmental cost was lower with the EGR system, it was offset by the cost of the EGR system itself combined with the additional incremental cost of fuel due to the decrease in the thermal efficiency.

Commentary by Dr. Valentin Fuster
1999;():V002T03A005. doi:10.1115/99-GT-185.
FREE TO VIEW

This paper presents a method for the analysis of gas turbine operating state, which uses the Cycle-Deck developed by gas turbine manufacturers and the measurements taken by means of the standard machine instrumentation. The gas turbine operating condition analysis is performed “adapting” the characteristic geometric and performance parameters (i.e., characteristic flow passage areas and efficiencies of the compressor and turbine, combustor efficiency, etc …), used as inputs by the Cycle-Deck, until the computed estimates of the measurable parameters agree with the values measured on the gas turbine. This is done by minimizing an objective function built as the sum of the squared residuals between the computed and measured values of the same parameters. The analysis of the variations between computed and expected values of the characteristic parameters allows the localization of inefficient operations due to deterioration and faults.

Topics: Gas turbines , Cycles
Commentary by Dr. Valentin Fuster

Cycle Innovations

1999;():V002T04A001. doi:10.1115/99-GT-004.
FREE TO VIEW

Conceptual engineering design of a chemically recuperated gas turbine (CRGT) system was carried out for an existing GE-LM2500 based natural gas compressor station. The configured system was evaluated using the commercial process simulation software ASPEN PLUS®. The results of the thermodynamic assessment showed an increase in thermal efficiency from 34% to 38% and an increase of the site actual output power by 15% at full load and for the same mass flow of air through the gas turbine compressors. The mass flow rate through the combustor and turbines increases only by 5%, which can be accommodated with the existing driver hardware. The present paper provides equipment identification, costs and economic evaluation for this specific system. The water needed for the CRGT system was drawn from an aquifer source through two wells capable of a continuous combined throughput of 2.8 liters/s (i.e. ∼22 GPM each). Environmental assessment of the system showed that the equivalent CO2 emission at full load decreases from ∼ 600 kg/MW.hr to 520 kg/MW.hr (i.e., 13%). The overall system economics shows that a return on capital investment is attainable in 5 years.

Commentary by Dr. Valentin Fuster
1999;():V002T04A002. doi:10.1115/99-GT-005.
FREE TO VIEW

In the paper two environomic procedures for the analysis and the optimization of energy systems, where environmental considerations are taken into account together with thermodynamic and economic ones, are presented. The aim is the assignment to energy plants of costs linked to their pollutant activities. The problem is faced with two different environomic approaches: (a) a method assigning a cost to the pollutant emissions (environmental cost); (b) a method assigning a cost to the exergy destroyed inside the system and rejected in the biosphere with the plant wastes (efficiency penalty).

As an example the environomic analysis is developed considering the pollutant emissions (CO, NOx and SOx) of an existing 700 MW combined power plant. Finally, a procedure to determine an efficiency penalty to the emitted CO2 is presented, and a comparison is developed between the results obtained by the two environomic approaches.

Commentary by Dr. Valentin Fuster
1999;():V002T04A003. doi:10.1115/99-GT-065.
FREE TO VIEW

During the last years, two new subjects among the others have risen interest in the field of small scale electric power generation: advanced microturbines and Solid Oxide Fuel Cells.

This paper investigates the thermodynamic potential of the integration of the Solid Oxide Fuel Cell technology with microturbine systems, in order to obtain ultra-high efficiency small capacity plants, generating electric power in the range of 250 kW with 65% LHV net electrical efficiency and with the possibility of cogenerating heat. A detailed description of the calculation model is presented, capable of full and part-load performances analysis of the microturbine and of the integrated SOFC+microturbine system.

Commentary by Dr. Valentin Fuster
1999;():V002T04A004. doi:10.1115/99-GT-066.
FREE TO VIEW

The modem combined cycle power plants achieved thermal efficiency of 50–55% by applying bottoming multistage Rankine steam cycle. At the same time, the Brayton cycle is an attractive option for a bottoming cycle engine. In the author’s US Patent No. 5,442,904 is described a combined cycle system with a simple cycle gas turbine, the bottoming air turbine Brayton cycle, and the reverse Brayton cycle.

In this system, air turbine Brayton cycle produces mechanic power using exergy of gas turbine exhaust gases, while the reverse Brayton cycle refrigerates gas turbine inlet air. Using this system, supercharging of gas turbine compressor becomes possible.

In the paper, thermodynamic optimization of the system is done, and the system techno-economic characteristics are evaluated.

Commentary by Dr. Valentin Fuster
1999;():V002T04A005. doi:10.1115/99-GT-067.
FREE TO VIEW

In the emerging deployment of microturbines (25–75Kw), a recuperator is mandatory to achieve thermal efficiencies of 30 percent and higher, this being important if they are to successfully penentrate the market currently dominated by Diesel generator sets. This will be the first application of gas turbines for electrical power generation, where recuperators will be used in significant quantities. The experience gained with these machines will give users’ confidence that recuperated engines will meet performance and reliability goals. The latter point is particularly important, since recuperated gas turbines have not been widely deployed for power generation, and early variants were a disappointment. Recuperator technology transfer to larger engines will see the introduction of advanced heat exchanged industrial gas turbines for power generation in the 3–15 Mw range.

After many decades of development, existing recuperators of both primary surface and plate-fin types, have demonstrated acceptable thermal performance and integrity in the cyclic gas turbine environment, but their capital costs are high.

A near-term challenge to recuperator design and manufacturing engineers is to establish lower cost metallic heat exchangers that can be manufactured using high volume production methods. A longer term goal will be the development and utilization of a ceramic recuperator, since this is the key component to realize the full performance potential of very small and medium size gas turbines.

Commentary by Dr. Valentin Fuster
1999;():V002T04A006. doi:10.1115/99-GT-119.
FREE TO VIEW

This paper describes a design point study of different power generation systems using the gasification of sugar cane bagasse, which produces a low calorific value fuel.

The biomass gasification is viewed as a process of drying the solid fuel and heating it up before the gasification reactions take place. The process is represented by equilibrium conditions. Varying pressure, temperature and feed composition in the gasifier controls the fuel gas composition.

Five alternative arrangements of an integrated gasification combined cycle (IGCC) power plant have been analysed using the exergy method. These alternatives comprise a combined gas / steam cycle using a simple gas turbine and also a reheat gas turbine, a combined gas / air cycle using a simple gas turbine and also a reheat gas turbine, and a combined gas / air / freon cycle using a reheat gas turbine.

The combined gas / steam cycles are more efficient rather than the combined gas / air cycles. Using a reheat gas turbine at the topping cycle of the combined cycles analysed contribute to increase the overall plant exergetic efficiency.

The exergy analysis has been carried out along with the performance assessment of power plants and takes into account the irreversibility of their components and the exhaust losses. The gasifier is the component of the power plant that destroys maximum exergy.

Three computer codes have been developed. One of them has the ability of simulating the thermodynamic properties of the low calorific value fuel in the gasification process and the other two are related to the performance analysis of the different power cycles using the exergy method.

The production of energy from the gasification of sugar cane bagasse has been of increasing importance to the Brazilian energetic matrix. This paper has been focused on the GEOPHILES - αlfa Programme, a project to train Latin America engineers on biomass gasification for power generation. The Commission of Energy of the European Community has supported this project.

Commentary by Dr. Valentin Fuster
1999;():V002T04A007. doi:10.1115/99-GT-279.
FREE TO VIEW

A new type of internal combustion engine and its thermodynamic cycle are introduced. The core of the engine is a nutating non-rotating disk, with the center of its hub mounted in the middle of a Z-shaped shaft. The two ends of the shaft rotate, while the disk nutates. The motion of the disk circumference prescribes a portion of a sphere. A portion of the area of the disk is used for intake and compression, a portion is used to seal against a center casing, and the remaining portion is used for expansion and exhaust. The compressed air is admitted to an external accumulator, and then into an external combustion chamber before it is admitted to the power side of the disk. The accumulator and combustion chamber are kept at constant pressures. The engine has a few analogies with piston-engine operation, but like a gas turbine it has dedicated spaces and devices for compression, burning and expansion. The thermal efficiency is similar to that of comparably-sized simple-cycle gas turbines and piston engines. For the same engine volume and weight, this engine produces less specific power than a simple-cycle gas turbine, but approximately twice the power of a two-stroke engine and four times the power of a four-stroke engine. The engine has advantages in the 10 kW to 200 kW power range. This paper introduces the geometry and thermodynamic model for the engine, presents typical performance curves, and discusses the relative advantages of this engine over its competitors.

Topics: Engines , Disks
Commentary by Dr. Valentin Fuster
1999;():V002T04A008. doi:10.1115/99-GT-312.
FREE TO VIEW

The thermoeconomic analysis of gas turbine based cycles is presented and discussed in this paper. The thermoeconomic analysis has been performed using the ThermoEconomic Modular Program (TEMP V.5.0) developed by the Authors (Agazzani and Massardo, 1997). The modular structure of the code allows the thermoeconomic analysis for different scenarios (turbine inlet temperature, pressure ratio, fuel cost, installation costs, operating hours per year, etc.) of a large number of advanced gas turbine cycles to be obtained in a fast and reliable way. The simple cycle configuration results have been used to assess the cost functions and coefficient values.

The results obtained for advanced gas turbine based cycles (intercooled, re-heated, regenerated and their combinations) are presented using new and useful representations: cost vs. efficiency, cost vs. specific work, and cost vs. pressure ratio. The results, including productive diagram configurations, are discussed in detail and compared to one another.

Topics: Gas turbines , Cycles
Commentary by Dr. Valentin Fuster
1999;():V002T04A009. doi:10.1115/99-GT-368.
FREE TO VIEW

A general 2-step approach for advanced gas turbine cycle design and finding optimal layouts is used to study chemically recuperated and evaporative gas turbine cycles. The analysis assumes compression with intercooling, the possibility of reheat in the turbine and no bottoming cycle. In a first step, different cycle layouts are first optimized for pressure ratio by considering a general black box heat recovery system while its exergy destruction is monitored. In a second step, after cycle optimization, corresponding heat recovery systems are designed using insights provided by a composite curve analysis. All cycle simulations are performed with ASPEN+, enhanced with in-house exergy subroutines, composite curve tools and a general gas turbine cooling model. Comparative results from this analysis are presented and the important role of pressure ratio is discussed.

Commentary by Dr. Valentin Fuster
1999;():V002T04A010. doi:10.1115/99-GT-369.
FREE TO VIEW

A new recuperator has been developed at Rolls-Royce, specifically designed to address the problems recuperators traditionally experience in withstanding thermal cycling and high cost associated with production complexity. Unlike existing recuperators it is made from two continuous sheets of metal wound into a spiral form. It is believed to be the only recuperator in the world manufactured as a continuous process and therefore has inherently lower cost. A significant increase in thermal fatigue resistance relative to state of the art recuperators is achieved due to the compliant structure and careful design of the heat exchanger matrix.

A technology acquisition programme has been carried out which has involved the design, manufacture and rig testing of recuperator cores. The success of the work is now leading to the continuation of the programme to address the application of this technology to specific engine developments from microturbines of 40kW to large industrial and marine turbines of 20MW.

Commentary by Dr. Valentin Fuster
1999;():V002T04A011. doi:10.1115/99-GT-370.
FREE TO VIEW

This paper assesses performances and economic viability of CO2 removal by chemical absorption from the flue gases of natural gas-fired Combined Cycles, more specifically for two configurations: one where CO2 is removed ahead of the stack without modifying the power cycle; the other where part of the flue gases is recirculated to the gas turbine, thereby reducing the flow to be treated by chemical absorption. In both cases sequestered CO2 is made available at conditions suitable to storage into deep oceanic waters.

Performances and cost of electricity are evaluated for systems based on large, heavy-duty turbines representative of state of the art “FA” technology. Carbon sequestration reduces net plant efficiency and power output by about 10% and increases the cost of electricity from 36 to about 50 mills/kWh. Flue gas recirculation warrants slightly higher efficiencies and lower costs.

CO2 removal is eventually compared with other strategies for the reduction of CO2 emissions, like switching existing coal-fired steam plants to natural gas or replacing existing steam plants with conventional CCs. At current fuel prices the latter appears the option of choice, with a cost of about 25 $ per tonn of avoided CO2 emission.

Commentary by Dr. Valentin Fuster
1999;():V002T04A012. doi:10.1115/99-GT-371.
FREE TO VIEW

The thermodynamic performances and relative sizing of small single shaft and two spool semi-closed cycle turbine engines (SCTE), for propulsion needs were studied relative to the conventional intercooled recuperative (ICR) open cycle gas turbine. The study indicated that the basic attributes of the SCTE are higher specific power, hp/pps inlet flow, a flat specific fuel consumption (SFC) characteristic, and potentially reduced exhaust emissions, at the expense of slightly higher SFC, lower power to weight ratio, and increased cost $/hp relative to the ICR.

The combustor is paramount to the successful operation of the SCTE in that the feedback (recirculated) flow, must be clean and devoid of particulates that could deposit on the heat transfer surfaces and turbo machinery. Additionally if high specific power is the primary design constraint high reticulation flow ratio combustor technology must be researched.

A conservative cycle analysis was conducted with current small gas turbine state-of-art component efficiencies and temperature limits for metallic hot-end components. The analysis was simplified by using existing gas properties for hydrocarbon air mixtures without oxygen depletion, or exhaust products in compression. Although the assumption that the intercooler and recuperator effectiveness and pressure drops remained constant provided a common datum, the part load performance analysis could benefit from the supplementation of improved heat exchanger algorithms.

Topics: Design , Gas turbines , Cycles
Commentary by Dr. Valentin Fuster
1999;():V002T04A013. doi:10.1115/99-GT-372.
FREE TO VIEW

A detailed study is presented in this paper concerning the changes required to transform a GE/Nuovo Pignone PGT2 gas turbine into a humid air turbine (HAT). The PGT2 is a 2 MW single-shaft gas turbine with a two stage centrifugal compressor and single can combustor. These features of the compressor and the combustion chamber are very important, because they reduce to a minimum the modifications necessary to add all the components required for operation in HAT cycle mode.

This paper shows the performance of the PGT2 gas turbine in three different modified configurations: 1) regenerated cycle, 2) intercooled-regenerated cycle and 3) HAT cycle with flue gas condensation. A detailed economic study is presented of the regenerated and the HAT cycles. The intercooled-regenerated cycle was not considered in the economic analysis, because it does not reach the performance of the HAT cycle but it has not the advantage of being a dry cycle.

All the components being added to the PGT2 are sized to determine with a sufficient accuracy their cost. The economic analysis presents a detailed cost analysis of the optimal configurations for maximum efficiency, and allows the calculation of the capital cost of the modified gas turbine. Return of investment is also presented in different economic situations with varying electricity and fuel prices.

Commentary by Dr. Valentin Fuster
1999;():V002T04A014. doi:10.1115/99-GT-373.
FREE TO VIEW

An investigation was conducted to examine the effects of a variable flow low pressure turbine on a variable cycle engine’s performance.

One of the greatest challenges, during the design of a variable cycle engine is how to optimise the various cycles and then to match then to the capabilities of the engine components, the use of extensive variable geometry is required to achieve this.

A method of matching variable cycle engines that was developed Cranfield University was adapted to cater for the use of a variable flow low pressure turbine. It was discovered that the implementation of variable geometry within the low pressure turbine could significantly reduce the requirements for variable geometry within the compressor system, at the cost of replacing well proven compressor variable geometry with high risk technology within the LP turbine.

Utilising the variable flow turbine to expand the bypass ratio range of the engine was studied. Increasing the LPM bypass ratio to 1.1 and 1.2 yielded SFC reductions of 3% and 5% respectively, reducing the bypass ratio of the HPM to 0.1 gave a 20% increase in specific thrust. It was found that the performance benefits gained from expanding the bypass ratio are large enough to warrant further investigation into this concept.

Topics: Engines , Turbines , Cycles
Commentary by Dr. Valentin Fuster
1999;():V002T04A015. doi:10.1115/99-GT-374.
FREE TO VIEW

This paper outlines the handling of a semi closed cycle gas turbine, its working fluid is carbon dioxide and the fuel is low heating value gas from coal. At startup however, air and natural gas are used. The objective of the machine is to produce clean electricity with the smallest efficiency penalty.

Many aspects of the operation of the engine are examined here; these include starting requirements, stator vane and bleed valve scheduling and the working fluid transition from air to carbon dioxide. Other features highlighted are the compressor operating lines and surge margins. The present paper describes the salient features of the three main stages into which the engine operation has been divided. These stages are: startup to synchronous idle, change of working fluid (from air to Carbon Dioxide-Argon) and fuel (from natural gas to coal synthetic gas) at synchronous idle and part load operation.

Preliminary findings show that engine handling can be carried out effectively with variable stators. This is possible because of the two shaft gas generator. Another point of interest is the large increase of corrected speed relative to rotational speed experienced when the working fluid changes from air to carbon dioxide. In general the control of the engine does not seem to present any insurmountable problems despite the complexities arising from the need to change working fluid and fuel.

Commentary by Dr. Valentin Fuster
1999;():V002T04A016. doi:10.1115/99-GT-375.
FREE TO VIEW

Steady-state and transient performance analysis programs for 200kw-class small turboshaft engine with free power turbine were developed. An existing turbojet engine was used for the gas generator of the developed turboshaft engine, and it was modified to satisfy performance requirements of this turboshaft engine. To verify the availability of steady-state performance program for this engine: the program was applied to the same type gas turbine test unit, and the analysis results were compared to experimental results.

The developed transient performance analysis program using the CMF (Constant Mass flow) method was utilized to analysis in the cases of fuel step increase and the ramp increase.

Commentary by Dr. Valentin Fuster
1999;():V002T04A017. doi:10.1115/99-GT-396.
FREE TO VIEW

An ASME performance test code. PTC 47, for integrated gasification combined cycle, IGCC. plants is currently being written. This code will include definitions of the significant overall plant and plant component performance results — input, output, and effectiveness. It will indicate the measurements and measurement techniques required to calculate these results, and estimate systematic uncertainties associated with each measurement, as well as random uncertainties due to the measuring device or variations in plant conditions during the test. The code writing committee is making use of mathematical models of the performance of various IGCC power plant configurations to aid in the work of code preparation and testing.

The objective of this paper is to describe how computerized system performance models are now being used in the preparation of PTC 47, and to indicate how such models might be used in the conduct of performance tests in the future. The general form of a system performance model is discussed to clarify how it relates to correction factors for system performance parameters — input, output, and effectiveness: but the paper does not include defining or solving the actual equations of a system performance model.

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