0

Coal, Biomass and Alternative Fuels

1998;():V003T05A001. doi:10.1115/98-GT-021.
FREE TO VIEW

Through its New Energy and Industrial Technology Development Organization (NEDO) the Japanese government is sponsoring the World Energy Network (WE-NET) Program. WE-NET is a 28-year global effort to define and implement technologies needed for hydrogen-based energy systems. A critical part of this effort is the development of a hydrogen-fueled combustion turbine system to efficiently convert the chemical energy stored in hydrogen to electricity when hydrogen is combusted with pure oxygen.

A Rankine cycle, with reheat and recuperation, was selected by Westinghouse as the general Reference System. Variations of this cycle have been examined to identify a Reference System having maximum development feasibility, while meeting the requirement of a minimum of 70.9% low heating value (LHV) efficiency. The strategy applied by Westinghouse was to assess both a near-term and long-term Reference Plant. The near-term plant requires moderate development based on extrapolation of current steam and combustion turbine technology. In contrast, the long-term plant requires more extensive development for an additional high-pressure reheat turbine, and is more complex than the near-term plant with closed-loop steam cooling and extractive feedwater heating. Trade-offs between efficiency benefits and development challenges of the near-term and long-term reference plant are identified. Results of this study can be applied to guide the future development activities of hydrogen-fueled combustion turbine systems.

Commentary by Dr. Valentin Fuster
1998;():V003T05A002. doi:10.1115/98-GT-025.
FREE TO VIEW

Because of the serious consequences of turbomachinery erosion on their performance and life, it is important to have reliable methods for predicting their erosion when solid particles are ingested with the incoming flow. This is a very challenging problem since turbomachinery erosion is affected by many factors such as blade passage geometry, blade row location, rotational speed flow conditions, blade material and particles’ characteristics. Several studies which are essential to predicting blade surface erosion intensity and pattern, have been conducted at the University of Cincinnati’s Propulsion Laboratory over the past twenty-five years. This paper describes only some of the work done on erosion testing at high temperatures and velocities for different materials and coatings. The testing has been performed with a special high temperature erosion wind tunnel which simulates the aerodynamic conditions on the blades. The coated substrates reveal one order of magnitude less wear compared to some commercial non-CVD coatings on the same alloys. This study demonstrates that some coatings provide an excellent erosion resistance to INCO 718, LNCO 738, MAR-246, X-40 and Tungsten Carbide.

Commentary by Dr. Valentin Fuster
1998;():V003T05A003. doi:10.1115/98-GT-062.
FREE TO VIEW

Sawmills in the Canadian province of British Columbia (BC) will soon be confronted with a collective wood waste disposal problem (bark, sawdust and shavings) of about 3 Million bone dry tonne/y at an average wet basis moisture content of about 45%. About 40 existing sawmill beehive burners presently incinerate this waste. Emissions from these beehive burners exceed current provincial particulate limits. Markets for the waste — board plants, sawdust digesters, charcoal plants, etc. — are limited. The economics of 25 to 50 MW wood-fired, steam power plants is poor. 6¢/kW.h is needed to finance a plant; the major utility offers 2¢/kW.h.

This paper describes a 3 MW to 5 MW, Nuovo Pignone PGT-5 co-generation plant fuelled by the 2000°F (1093°C) exhaust from a Heuristic EnvirOcycler, a two-stage, wood waste incinerator. The exhaust meets BC’s particulate limit of 0.052 grains/dscf (120 mg/Nm3). 37 to 65 Million Btu/h (39 to 68 GJ/h) of waste heat can be recovered from the system exhausts. In this application the PGT-5’s external combustion chamber is replaced by a “recuperator”, i.e., a high temperature, gas-to-air, heat exchanger.

Two variations of the basic system are examined. One features a larger than necessary EnvirOcycler to generate additional steam in the waste heat boiler. The other variation discusses heating 1,550°F (843°C) air from the recuperator up to 1,796°F (980°C) with natural gas. The extra power generated can cost as little as 1.9¢/kW.h.

Commentary by Dr. Valentin Fuster
1998;():V003T05A004. doi:10.1115/98-GT-063.
FREE TO VIEW

Alternative fuel projects often require substantial amounts of oxygen. World scale gas-to-liquids (GTL) processes based on the partial oxidation of natural gas, followed by Fischer-Tropsch chemistry and product upgrading, may require in excess of 10,000 tons per day of pressurized oxygen. The remote location of many of these proposed projects and the availability of low-cost natural gas and byproduct steam from the GTL process disadvantages the use of traditional, motor-driven air separation units in favor of steam or gas turbine drive facilities. Another process of current interest is the partial oxidation of waste materials in industrial areas to generate synthesis gas. Synthesis gas may be processed into fuels and chemicals, or combusted in gas turbines to produce electricity. A key to the economic viability of such oxygen-based processes is cost effective air separation units, and the manner in which they are integrated with the rest of the facility. Because the trade-off between capital and energy is different for the remote gas and the industrial locations, the optimum integration schemes can also differ significantly. This paper examines various methods of integrating unit operations to improve the economics of alternative fuel facilities. Integration concepts include heat recovery, as well as several uses of byproduct nitrogen to enhance gas turbine operation or power production. Start-up, control and operational aspects are presented to complete the review of integrated designs.

Commentary by Dr. Valentin Fuster
1998;():V003T05A005. doi:10.1115/98-GT-102.
FREE TO VIEW

Reducing emissions is an important issue facing gas turbine manufacturers. Almost all of the previous and current research and development for reducing emissions has focused, however, on flow, heat transfer, and combustion behavior in the combustors or on the uniformity of fuel injection without placing strong emphasis on the flow uniformity entering the combustors. In response to the incomplete understanding of the combustor’s inlet air flow field, experiments were conducted in a 48% scale, 360° model of the diffuser-combustor section of an industrial gas turbine. In addition, the effect of air extraction for cooling or gasification on the flow distributions at the combustors’ inlets was also investigated. Three different air extraction rates were studied: 0% (baseline), 5% (airfoil cooling), and 20% (for coal gasification). The flow uniformity was investigated for two aspects: (a) global uniformity, which compared the mass flow rates of combustors at different locations relative to the extraction port, and (b) local uniformity, which examined the circumferential flow distribution into each combustor. The results indicate that even for the baseline case with no air extraction there was an inherent local flow aonuniformity of 10 ∼ 20% at the inlet of each combustor due to the complex flow field in the dump diffuser and the blockage effect of the cross-flame tube. More flow was seen in the portion further away from the gas turbine center axis. The effect of 5% air extraction was small. Twenty-percent air extraction introduced approximately 35% global flow asymmetry diametrically across the dump diffuser. The effect of air extraction on the combustor’s local flow uniformity varied with the distances between the extraction port and each individual combustor. Longer top hats were installed with the initial intention of increasing flow mixing prior to entering the combustor. However, the results indicated that longer top hats do not improve the flow uniformity; sometimes, adverse effects can be seen. Although a specific geometry was selected for this study, the results provide sufficient generality to benefit other industrial gas turbines.

Commentary by Dr. Valentin Fuster
1998;():V003T05A006. doi:10.1115/98-GT-103.
FREE TO VIEW

Biomass and sewage sludge are attracting increasing interest in power plant technology as a source of carbon dioxide-neutral fuels. A new way to reduce the consumption of fossil fuels could be the co-combustion or co-gasification of coal and biomass or coal and sewage sludge. In both cases, pyrolysis is the first step in the technical process. In order to obtain detailed information about the pyrolysis of coal/biomass and coal/sewage sludge mixtures as well as unblended fuels, the ‘Institut für Verfahrenstechnik und Dampfkesselwesen (IVD)’ at the University of Stuttgart has carried out investigations using an electrically heated entrained flow reactor.

One application of substitution of fossil fuels could be the utilization of pyrolysis gas or gas generated in a gasification process as a reburn fuel in conventional boilers fired with fossil fuels. Investigation showed that generated gas from coal, biomass and sewage sludge pyrolysis and gasification have high NOx reduction efficiencies compared to methane or low calorific gases using it as a reburn fuel in coal fired boilers. In order to take advantage of this pretreatment process the release of organic as well as of mineral compounds during the pyrolysis or gasification has to be investigated. For coal pyrolysis and gasification the reactions are known since there was a lot of research all over the world. Biomass or sewage sludge have other structures compared to fossil fuels and contain alkali, chlorine and other problematic compounds, like heavy metals. The release of those elements and of the organic matter has to be investigated to characterize the gas and the residual char. The optimum process parameters regarding the composition of the generated gas and the residual char have to be found out.

The IVD has studied the co-pyrolysis of biomass and sewage sludge together with a high volatile hard coal. The main parameters to be investigated were the temperature of the pyrolysis reactor (400°C–1200°C) and the coal/biomass and coal/sewage sludge blends. Besides co-pyrolysis experiments test runs with unmixed main fuels were carried out with the hard coal, straw as biomass, and a sewage sludge. It was expected that the high reactivity of biomass and sewage sludge would have an effect on the product composition during co-pyrolysis.

The test runs provided information about fuel conversion efficiency, pyrolysis gas and tar yield, and composition of pyrolysis gas and tar. Besides gas and tar analysis investigations regarding the path of trace elements, like heavy metals, alkali, chlorine and nitrogen components, during the pyrolysis process varying different parameters have been carried out. The fuel nitrogen distribution between pyrolysis gas, tar and char has been analyzed as well as the ash composition and thus the release of mineral components during pyrolysis.

Increasing reaction temperatures result in a higher devolatilization for all fuels. Biomass shows a devolatilization of up to 80% at high temperatures. Hard coal shows a weight toss of approx. 50% at same temperatures. Sewage sludge devolatilizes also up to 50%, which is nearly a total release of organic matter, because of the high ash content of about 50% in sewage sludge.

Gaseous hydrocarbons have a production maximum at about 800°C reaction temperature for all feedstocks. Carbon monoxide and hydrogen are increasingly formed at high pyrolysis temperatures due to gasification reactions.

Mineral elements are released during straw pyrolysis, but within the hot gas filtration unit further recombination reactions and condensation of elements on panicles take place. There is no release of mineral elements during sewage sludge pyrolysis and only a slight release of heavy metals at high pyrolysis temperatures.

The effect of co-pyrolysis depends on the feedstocks used in association with the panicle size. The co-pyrolysis test runs showed that a synergetic effect exists when using sewage sludge and hard coal. There is a higher char production related to the unmixed fuels; gas and tar formation are lowered. Co-pyrolysis test runs with biomass and coal did not show this effect on the pyrolysis products. Reasons for this behaviour could be a difference in particle size and material structure which influences the devolatilization velocity of the fuels used or the relatively short residence time in the entrained flow reactor. It seems possible that coal pyrolysis is influenced by the reaction atmosphere, generated in co-pyrolysis. In the co-pyrolysis of coal and sewage sludge, the sludge degases much faster than coal because of the structure of sewage sludge and its small panicle. The coal pyrolysis taking place afterwards in the reaction tube occurs in a different atmosphere, compared to the mono-pyrolysis experiments. The devolatilization of coal in the co-pyrolysis experiments together with straw was not disturbed by the gaseous products of straw pyrolysis, because the large straw particles showed a delayed degasing compared to the coal particles.

Topics: Sewage , Biomass , Coal , Pyrolysis
Commentary by Dr. Valentin Fuster
1998;():V003T05A007. doi:10.1115/98-GT-122.
FREE TO VIEW

Tightly regulated and state-controlled utilities are rapidly changing into a competitive, market-driven industry, as private power development is being actively pursued worldwide. Accelerated economic growth in developing countries has fueled a massive growth in the power sector. Gas turbine based power plants have become an attractive option; however, many of these developing countries have limited supplies of conventional gas turbine fuels, namely natural gas or distillate oil. Therefore, power developers are seeking alternative fuels. This paper discusses the balance-of-plant (BOP) considerations and economics of using alternative fuels such as liquefied natural gas (LNG), liquefied petroleum gas (LPG), naphtha, and crude/heavy oils.

Commentary by Dr. Valentin Fuster
1998;():V003T05A008. doi:10.1115/98-GT-131.
FREE TO VIEW

The Department of Energy’s Federal Energy Technology Center (FETC) is sponsoring the Combustion 2000 Program aimed at introducing clean and more efficient advanced technology coal-based power systems in the early 21st century. As part of this program, the United Technologies Research Center has assembled a seven member team to identify and develop the technology for a High Performance Power Systems (HIPPS) that will provide in the near term, 47% efficiency (HHV), and meet emission goals only one-tenth of current New Source Performance Standards for coal-fired power plants. In addition, the team is identifying advanced technologies that could result in HIPPS with efficiencies approaching 55% (HHV).

The HIPPS is a combined cycle that uses a coal-fired High Temperature Advanced Furnace (HITAF) to preheat compressor discharge air in both convective and radiant heaters. The heated air is then sent to the gas turbine where additional fuel, either natural gas or distillate, is burned to raise the temperature to the levels of modern gas turbines. Steam is raised in the HITAF and in a Heat Recovery Steam Generator for the steam bottoming cycle. With state-of-the-art frame type gas turbines, the efficiency goal of 47% is met in a system with more than two-thirds of the heat input furnished by coal. By using advanced aeroderivative engine technology, HIPPS in combined-cycle and Humid Air Turbine (HAT) cycle configurations could result in efficiencies of over 50% and could approach 55%.

The following paper contains descriptions of the HIPPS concept including the HITAF and heat exchangers, and of the various gas turbine configurations. Projections of HIPPS performance, emissions including significant reduction in greenhouse gases are given. Application of HIPPS to repowering is discussed.

Commentary by Dr. Valentin Fuster
1998;():V003T05A009. doi:10.1115/98-GT-159.
FREE TO VIEW

In the framework of a multi-national European Joule project, experimental research and modeling concerning co-gasification of biomass and coal in a bubbling pressurized fluidized bed reactor is performed. The impact of fuel characteristics (biomass type, mixing ratio) and process conditions (pressure, temperature, gas residence time, air-fuel ratio and air-steam ratio) on the performance of the gasifier (carbon conversion, fuel gas composition, non-steady state behaviour) was studied experimentally and theoretically.

Pelletized straw and miscanthus were used as biomass fuels. The process development unit has a maximum thermal capacity of 1.5 MW and was operated at pressures up to 10 bar and bed temperatures in the range of 650 °C–900 °C. The bed zone of the reactor is 2 m high with a diameter of 0.4 m and is followed by an adiabatic freeboard, approximately 4 m high with a diameter of 0.5 m.

Time-averaged as well as time-dependent characteristics of the fuel gas were determined experimentally. The results will be compared with the gas turbine requirements provided by a gas turbine manufacturer, one of the partners in the project. The evaluation of the results will ultimately be used to implement and test an adequate control strategy for the pressurized fluidized bed gasifier integrated with a gas turbine combustion chamber.

Commentary by Dr. Valentin Fuster
1998;():V003T05A010. doi:10.1115/98-GT-160.
FREE TO VIEW

Low calorific value fuel gas, obtained by pressurized fluidized bed gasification of coal/biomass mixtures, is combusted at 0.8 MPa with air or oxygen in a vertical cylindrical chamber (D = 0.28 m, L = 2.0 m). The fuel (T = 1060 K) and oxydizer (air at 350 K, oxygen at 460 K) are injected coaxially, resulting in an essentially axissymmetric flow pattern. Particles have been removed from the fuel gas stream by a cyclone, mounted between the gasifier and the combustor. A two-dimensional model, implemented in the CFD code FLUENT was developed for the calculation of temperatures, flow patterns and species concentrations throughout the combustor. The calculated results are compared with experimental data for two low calorific value fuel gas compositions and two oxidizer compositions at two axial combustor locations (X/L = 0.175 and X/L = 1). The results appear to justify further investigation of the applicability of the model to low calorific value fuel gas fired gas turbine combustors.

Commentary by Dr. Valentin Fuster
1998;():V003T05A011. doi:10.1115/98-GT-161.
FREE TO VIEW

The catalytic indirectly heated gasification of bagasse was investigated in this study. The quality of the gaseous fuel was assessed using the total energy analysis of the gas, in which both heat content and total yields of the gas produced from the gasification of bagasse are analyzed at temperatures ranging from 700 to 1000°C. Untreated bagasse gasification was used as a baseline for the investigation of the effect of catalysts on the gasification process. The total energy analysis showed a significant improvement of gas quality due to increase of temperature and due to the use of alumina-zinc based catalysts at temperatures below 900°C. The presence of these catalysts in the gasification process affected the quality of the gases formed, mainly by increasing the hydrogen production, reduction of the gas dilution by carbon dioxide and a slightly higher production of carbon monoxide. Above 900°C, temperature dominates the gasification reaction mechanisms causing the catalysts to have little or no significant effect. Thermal cracking of tar is of major importance on the gasification process, as the tar yields reduce from 42.1 to 24.7% of the bagasse original weight with the increase of the gasifier temperature from 700 to 1000°C. However, the solid residue reduced only from 16 to 13.3%. Hence, the increase in the gaseous yields at high temperature appeared to be due to the gasification of tar with some contribution from secondary reactions involving char. The result was the production of a medium heat content gaseous fuel.

Commentary by Dr. Valentin Fuster
1998;():V003T05A012. doi:10.1115/98-GT-162.
FREE TO VIEW

The co-processing of coal-biomass mixtures under inert and reducing atmospheres has been studied in a bench scale fixed-bed (‘hot-rod’) reactor. The aim was to look for evidence of synergistic effects during the co-gasification of coal and biomass. Total volatile release, tar and char yields from mixtures of Daw Mill coal and Silver Birch wood (alone and in mixtures of 25, 50 and 75 % by weight), have been determined as a function of temperature (850 and 1000 °C) and pressures (up to 25 bar) under He-pyrolysis and CO2-gasification conditions. The total volatile yields of mixtures have been found to match those calculated theoretically from pure coal and biomass values under all conditions attempted, thus suggesting a lack of synergy in the amount of char produced. However, char reactivity measurements in an atmospheric thermogravimetric analyser (isothermal combustion in air at 500 °C) indicate that chars of coal-biomass mixtures have higher combustion reactivities than would be expected from the reactivities of the raw fuels alone. Similarly, the tar yields from mixtures are also somewhat higher than those predicted from the individual contributions of coal and biomass.

Commentary by Dr. Valentin Fuster
1998;():V003T05A013. doi:10.1115/98-GT-163.
FREE TO VIEW

Gasification combined cycle has the potential to provide a clean, high efficiency, low environmental impact power generation system. A prime fuel for such systems is coal but there is scope in part to utilise renewable energy sources including biomass and waste materials such as sewage sludge or even oil residues. There is considerable scope to improve the performance of the first generation systems of gasification combined cycle plant, both through design changes and through the continued development towards second generation plant. Such improvements offer the prospect of even better efficiency, coal/biomass/waste utilisation flexibility, lower emissions especially of CO2, and lower economic cost of power generation. There have been several major R&D initiatives, supported in part by the European Commission, which have been designed to meet these aims. The approach adopted has been to form multi-partner project teams comprising industry, industrial research organisations and selected universities. The main technical issues that have been considered include co-gasification, e.g. co-feeding, fuel conversion, gas quality, contaminants, component developments, and the integration of hot fuel gas cleaning systems for removal of solid particles, control of sulphur emissions, control of fuel bound nitrogenous species, removal of halides and control of alkali species. The technical R&D activities have been underpinned by several major techno-economic assessment studies. This paper provides an overview of these various activities which either form part of the European Commission JOULE Coal R&D Programme or were supported under an APAS special initiative.

Commentary by Dr. Valentin Fuster
1998;():V003T05A014. doi:10.1115/98-GT-164.
FREE TO VIEW

The use of biomass fuels in gas turbine engines requires an examination of the effect of the fuel on the engine materials. While the fuel may be more environmentally friendly than conventional fuels, it has the potential to produce serious life limiting corrosion within gas turbine engines. The effects of the high alkali and low sulphur content of the fuel on its corrosive properties needs to be examined. To determine the extent and type of corrosion typical material systems were exposed to bio-fuel combustion products in a flame tunnel and furnace under conditions designed to promote high temperature corrosion. The preliminary results indicate that type I hot corrosion is certainly occurring with some signs of type II hot corrosion in certain material systems.

Commentary by Dr. Valentin Fuster
1998;():V003T05A015. doi:10.1115/98-GT-226.
FREE TO VIEW

Many biomass power plants operating today are small plants characterized by low efficiencies. The average biomass power plant is 20 MW with a biomass-to-electricity efficiency of about 20 percent. Small biomass power plants are also costly to build. Co-firing biomass with coal in existing large, low cost, base load pulverized coal (PC) power plants has been suggested as a cost-effective, near term opportunity for biomass power. However, co-firing of biomass in PC boilers requires addition of a separate biomass feed system. The proposed concept avoids a separate feed system by converting biomass to charcoal for co-firing with coal. Fuel supply reliability would be improved by producing and stockpiling charcoal at dedicated facilities located off the power plant site. With an energy density similar to coal, charcoal could be transported more economically than biomass. Overall costs for co-firing charcoal and coal would be lower than systems co-firing biomass. Investment in Clean Coal Technologies could also be leveraged for biomass energy use by co-firing charcoal with coal in Integrated Gasification Combined Cycle (IGCC) and Pressurized Fluid Bed Combustion (PFBC) power systems.

Topics: Biomass , Coal , Firing
Commentary by Dr. Valentin Fuster
1998;():V003T05A016. doi:10.1115/98-GT-231.
FREE TO VIEW

While natural gas is achieving unrivalled penetration in the power generation sector, especially in gas-turbine combined cycles (CCGT), an increasing number of alternative fuels are in a position to take up the ground left vacant by this major primary energy. In particular, within the thriving family of liquid fuels, the class of volatile products opens interesting prospects for clean and efficient power generation in CCGT plants. Therefore, it has become a necessity for the gas turbine industry to extensively evaluate such new fuel candidates, among which: naphtha’s; kerosines; gas condensates; Natural Gas Liquids (NGL) and alcohols are the most prominent representatives. From a technical standpoint, the success of such projects requires both a careful approach to several specific issues (eg: fuel handling & storage, operation safety) and a clear identification of technological limits. For instance, while the purity of gas condensates meets the requirements of heavy-duty technologies, it generally appears unsuitable for aeroderivative machines.

This paper offers a succinct but comprehensive technical approach and overviews some experience acquired in this area with heavy duty gas turbines. Its aim is to inform gas turbine users/engineers and project developers who envisage volatile fuels as alternative primary energies in gas turbine plants.

Topics: Fuels , Gas turbines
Commentary by Dr. Valentin Fuster
1998;():V003T05A017. doi:10.1115/98-GT-233.
FREE TO VIEW

Performance and design data are presented for energy conversion of biomass in modified HAT cycles in a combined heat and power plant. The computations are performed by the heat balance software GateCycle and based on a gas turbine LM 2500 from General Electric. The processing of biomass includes drying of wet biomass fuel, gasification at atmospheric pressure, gas cooling and wet gas cleaning of the product gas and gas compression to pressure requirement of the gas turbine combustion system.

A further problem in biomass fuelled power systems and humidified gas turbine (HAT cycle) is to accomplish stable combustion in a gas turbine combustor with humidified air along with product gas having low heating value. To avoid the problem, the air humidification is performed only on a fraction of the compressed air stream, which makes it possible to use dry air (unhumidified air) for the combustion in the primary zone and humidified air for diluting and cooling. It is shown that part-airflow humidification gives similar or slightly improved performance (specific power and cycle efficiency) and also reduces the need of area for heat exchanging and recuperating.

Commentary by Dr. Valentin Fuster
1998;():V003T05A018. doi:10.1115/98-GT-294.
FREE TO VIEW

Biomass derived fuels are an essential alternative for heat and energy production, in order to minimise environmental impact, since they make no net contribution to the increase of CO2 emissions into the atmosphere. In certain countries, biofuels are also interesting since they are available as waste products from the agricultural or forestry industry. Unfortunately, combustion of biofuels often results in high emissions levels of pollutants such as NOx, CO and unburned hydrocarbons.

In gas turbines, catalytic combustion of biofuels has the potential to reduce emissions of these undesired species. The ULECAT project (Ultra Low Emissions CATalytic combustor) described in this paper is the first step of a program aiming at the development of an ultra-low emission gas turbine in the range of 1 to 5 MWe, able to run with both biomass-derived gases and liquid fuels. The objective of the project is to assess the feasibility of a dual fuel catalytic combustor.

Combustor design issues are investigated at full and part load conditions. For the comparison of combustor configuration, modelling provides a useful help for catalytic section design, in particular for the estimation of catalytic activity and wall temperature which strongly influence catalyst life time.

Catalyst development is one of the main topics of this project. It is mainly focused on high temperature catalyst durability and the reduction of NOx formation. This last point is of primary importance in biofuels combustion and certain catalysts have shown an important potential in reducing ammonia conversion into NOx in some operating conditions. Catalyst performances are evaluated at lab scale and also pilot scale in representative gas turbine combustor conditions with both Diesel fuel and biomass derived fuels.

Commentary by Dr. Valentin Fuster
1998;():V003T05A019. doi:10.1115/98-GT-295.
FREE TO VIEW

An increasing demand for efficient and environmentally clean use of biomass and wood waste byproducts as fuel requires major developments in gas turbines. Gas turbines are designed primarily to handle either natural gas or in some cases diesel fuels. Introducing low BTU, contaminants containing biofuels into a gas turbine would require proper understanding of fuel characteristics, combustor capability to burn these fuels, compressor surge margins, and ability of the turbine section to withstand deposition, erosion and corrosion.

Allison Engine Company (Allison), in cooperation with the U.S. Department of Energy and other partners, has initiated a bioturbine development project which would lead to commercialization of a bioturbine to operate on major categories of biofuels. The project will address six key issues:

• Quantify chemical, physical and combustion characteristics of biofuels, gasifiers, and the mass volume

• Conduct emission modeling of existing combustor with low BTU fuels

• Conduct rig tests

• Modify current design of the combustor to handle low BTU fuels

• Evaluate compressor surge margins to handle increased mass flows

• Conduct full scale engine field test

The total cost of this two and a half years project is approximately $8 million. The DOE will contribute over $3 million. Allison and partners will contribute the remaining $5 million.

There is an additional vital task which must be performed, but is not a part of the current project. The capability of the turbine to withstand deposition, erosion and corrosion must also be evaluated in order to protect the turbine, and provide long term, uninterrupted operation of the gas turbine on biofuels.

An important first step is to obtain quantitative data on gasified biofuels, including the contaminants. This information will be used in combustor modeling and to develop rig tests. The combustor will then be modified and made capable of handling these fuels. Allison will use the 601-K engine combustor (similar to the RB211 DLE combustor), and modify its hardware and software as required. The combustor modification will involve modeling, rig test, hardware and software modifications, and final engine test The entire project is expected to be complete during the second half of 1999.

Concurrent with these tasks, Allison will evaluate the options available to increase the capability of engine mass flow due to low BTU fuels. A parallel task of “ruggedizing” the turbine section is also planned. The resulting turbine is expected to be comparable to natural gas fired commercial gas turbines in performance, durability, reliability and major overhaul cycles.

Topics: Fuels , Turbines
Commentary by Dr. Valentin Fuster
1998;():V003T05A020. doi:10.1115/98-GT-315.
FREE TO VIEW

A new generation of small scale (less than 20 MWe) biomass fueled, power plants are being developed based on a gas turbine (Brayton cycle) prime mover. These power plants are expected to increase the efficiency and lower the cost of generating power from fuels such as wood. The new power plants are also expected to economically utilize annual plant growth materials (such as rice hulls, cotton gin trash, nut shells, and various straws, grasses, and animal manures) that are not normally considered as fuel for power plants. This paper summarizes the new power generation concept with emphasis on the engineering challenges presented by the gas turbine component.

Topics: Biomass , Gas turbines
Commentary by Dr. Valentin Fuster
1998;():V003T05A021. doi:10.1115/98-GT-327.
FREE TO VIEW

Utilising biomass in the CHP production is not without difficulties: the chemical and physical characteristics of the biofuels; corrosion, slagging and fouling; and working environment. An in-situ high-temperature corrosion monitoring test system was successfully developed. Furthermore, activities have been launched to develop a straw pre-processing method separating the aggressive substances from straw.

As a result of the gasification projects (straw, coal-straw, wood chips) it was concluded that it is possible to gasify straw — probably for 100% straw and definitely for 50/50 blends, although with some difficulties — and for wood chips deposit formation was a major obstacle. Further R&D is definitely needed, but with the limited international interest the gasification technology seems to have reached a dead end in Denmark.

Another focal point is the working environment, where care must be taken to limit any potential health hazards resulting from the handling of long-term stored biofuels.

Topics: Biofuel
Commentary by Dr. Valentin Fuster
1998;():V003T05A022. doi:10.1115/98-GT-331.
FREE TO VIEW

The development of integrated coal gasification combined cycle (IGCC) systems ensures cost-effective and environmentally sound options for supplying future coal utilizing power generation needs. The Japanese government and the electric power industries in Japan promoted research and development of an IGCC system using an air-blown entrained-flow coal gasifier. We worked on developing a low-Btu fueled gas turbine combustor to improve the thermal efficiency of the IGCC by raising the inlet-gas temperature of gas turbine.

On the other hand, Europe and the United States are now developing the oxygen-blown IGCC demonstration plants. Coal gasified fuel produced in an oxygen-blown entrained-flow coal gasifier, has a calorific value of 8.6MJ/m 3 which is one fifth that of natural gas. However, the adiabatic flame temperature of oxygen-blown medium-Btu coal gaseous fuel is higher than that of natural gas and so NOx production from nitrogen fixation is expected to increase significantly. In the oxygen-blown IGCC system, a surplus nitrogen in quantity is produced in the oxygen-production unit. When nitrogen premixed with coal gasified fuel is injected into the combustor, the power to compress nitrogen increases. A low NOx combustion technology which is capable of decreasing the power to compress nitrogen is a significant advance in gas turbine development with an oxygen-blown IGCC system. We have started to develop a low NOx combustion technology using medium-Btu coal gasified fuel produced in the oxygen-blown IGCC process.

In this paper, the effect of nitrogen injected directly into the combustor on the thermal efficiency of the plant is discussed. A 1300 °C-class gas turbine combustor with a swirling nitrogen injection function designed with a stable and low NOx combustion technology was constructed and the performance of this combustor was evaluated under atmospheric pressure conditions. Analyses confirmed that the thermal efficiency of the plant improved by 0.2 percent (absolute), compared with a case where nitrogen is premixed with coal gasified fuel before injection into the combustor. Moreover, this new technique which injects nitrogen directly into the high temperature region in the combustor results in a significant reduction in NOx production from nitrogen fixation. We estimate that CO emission concentration decreases to a significant level under high pressure conditions, while CO emission concentration in contrast to NOx emission rises sharply with increases in quantity of nitrogen injected into the combustor.

Commentary by Dr. Valentin Fuster
1998;():V003T05A023. doi:10.1115/98-GT-335.
FREE TO VIEW

The Minnesota Valley Alfalfa Producers (MnVAP), a farmer owned cooperative, is developing a 75 MW combined cycle power plant integrated with alfalfa processiag facilities in southwestern Minnesota. The Minnesota Agri-Power (MAP) project is supported by the U. S. Department of Energy and a project development team that includes Stone & Webster, the University of Minnesota, United Power Association, Carbona Corporation/Kvaerner Pulping Inc. and Westinghouse. Alfalfa processing facilities separate the fibrous stem material from the protein-rich leaf fraction. The resulting alfalfa leaf meal (ALM) is further processed into a variety of valuable livestock feed products. Alfalfa stem material is gasified using air-blown fluidized bed technology to produce a hot, clean, fuel gas. The fuel gas is fired in a combustion turbine and the exhaust heat is used to produce steam to power a steam turbine. At base load, the electric power plant will consume 1000 tons per day of biomass fuel. This paper briefly describes the project development activities of the alfalfa feed trials and the combined cycle power plant. This commercial scale demonstration represents an important milestone on a continuing pathway towards environmentally and economically sustainable energy systems.

Commentary by Dr. Valentin Fuster
1998;():V003T05A024. doi:10.1115/98-GT-337.
FREE TO VIEW

The Brazilian Wood Biomass Demonstration Project (WBP) Phase II was contracted with the United Nations Development Programme - PNUD, Setor Comerical Norte, Quadra 2 - BLOCO A, EDF.Corporation-7° Andar, Brasilia - DF Brasil 70712-900 and General Electric Marine & Industrial Engines to develop the gas turbine equipment necessary to utilize fuel produced by the gasification of wood products. The program included performance studies, control specification requirements, bleed and fuel valve specifications, a modified dual gas fuel nozzle for fuel delivery to the combustor and results of two (2) combustor component tests utilizing biomass simulated fuel. This paper will deal primarily with the fuel nozzle design elements, the setup and evaluation of the component combustor tests and resulting emissions produced by the simulated Biomass fuel. Details of the combustor test arrangement, facilities and special test equipment needed to complete the evaluation will be presented. In addition, background on the two types of combustor testing will be discussed.

Commentary by Dr. Valentin Fuster
1998;():V003T05A025. doi:10.1115/98-GT-338.
FREE TO VIEW

There is an increasing interest in catalytic combustors fuelled by low-heating value (LHV) gases, with a LHV of 5–7 MJ/Nm3. This is because catalytic combustion could be advantageous compared to flame combustion with respect to stable combustion of LHV-gases and low conversions of fuel-N (mainly NH3) to NOx. In the present project, funded by the EU Joule Programme, catalytic combustion of gasified wood for gas turbine applications is studied. A synthetic gas mixture of H2, CO, CO2, H2O, CH4, N2 and NH3, that resembles the output from a fluidized bed gasifier using biomass as raw material, is used. The gas mixture is mixed with air at atmospheric pressure and combusted over washcoated cordierite monoliths in a bench-scale laboratory quartz-reactor.

The objectives of the work described here are twofold. To begin with, improvement of the thermal stability of hexaaluminate washcoats by substitutions of rare earth or transition metal compounds is being studied. Secondly, catalytic combustion of gasified biomass over these washcoats has been studied in a bench-scale unit.

In this on-going project, obtained result show that it is possible to improve the surface area of hexaaluminate compounds up to 17 m2/g after careful synthesis and calcination up to 1400 °C for four hours. The selectivity of NH3-conversion to N2 is at present at 60 percent, but varies strongly with temperature. Fuel components such as H2, CO, C2H4 and NH3 ignite at temperatures close to compressor outlet temperatures. This means that a pilot-flame may not be needed for ignition of the fuel. A comparison between a Pd-impregnated lanthanum hexaaluminate and a Mn-substituted lanthanum hexaaluminate showed that the ignition temperature and the NOx-formation varied strongly over the two different catalysts.

Commentary by Dr. Valentin Fuster
1998;():V003T05A026. doi:10.1115/98-GT-339.
FREE TO VIEW

Kraft pulp and paper mills generate large quantities of black liquor and byproduct biomass suitable for gasification. These fuels are used today for onsite cogeneration of heat and power in boiler/steam turbine systems. Gasification technologies under development would enable these fuels to be used in gas turbines. This paper reports results of detailed full-load performance modeling of pulp-mill cogeneration systems based on gasifier/gas turbine technologies and, for comparison, on conventional steam-turbine cogeneration technologies. Pressurized, oxygen-blown black liquor gasification, the most advanced of proposed commercial black liquor gasifier designs, is considered, together with three alternative biomass gasifier designs under commercial development (high-pressure air-blown, low-pressure air-blown, and low-pressure indirectly-heated). Heavy-duty industrial gas turbines of the 70-MWe and 25-MWe class are included in the analysis. Results indicate that gasification-based cogeneration with biomass-derived fuels would transform a typical pulp mill into a significant power exporter and would also offer possibilities for net reductions in emissions of carbon dioxide relative to present practice.

Commentary by Dr. Valentin Fuster
1998;():V003T05A027. doi:10.1115/98-GT-340.
FREE TO VIEW

Several advanced, coal- and biomass-based combustion turbine power generation technologies using solid fuels (IGCC, PFBC, Topping-PFBC, HIPPS) are currently under development and demonstration. A key developing technology in these power generation systems is the hot gas filter. These power generation technologies must utilize highly reliable and efficient hot gas filter systems if their full thermal efficiency and cost potential is to be realized. This paper reviews the recent test and design progress made by Westinghouse in the development and demonstration of hot gas ceramic barrier filters toward the goal of reliability. The objective of this work is to develop and qualify, through analysis and testing, practical hot gas ceramic barrier filter systems that meet the performance and operational requirements for these applications.

Commentary by Dr. Valentin Fuster
1998;():V003T05A028. doi:10.1115/98-GT-341.
FREE TO VIEW

Fuel gas cleanup processing significantly influences overall performance and cost of IGCC power generation. The raw fuel gas properties (hearing value, sulfur content, alkali content, ammonia content, “tar” content, particulate content) and the fuel gas cleanup requirements (environmental and turbine protection) are key process parameters. Several IGCC power plant configurations and fuel gas cleanup technologies are being demonstrated or are under development. In this evaluation, air-blown, fluidized-bed gasification combined-cycle power plant thermal performance is estimated as a function of fuel type (coal and biomass fuels), extent of sulfur removal required, and the sulfur removal technique. Desulfurization in the fluid bed gasifier is combined with external hot fuel gas desulfurization, or, alternatively with conventional cold fuel gas desulfurization. The power plant simulations are built around the Westinghouse 501F combustion turbine in this evaluation.

Commentary by Dr. Valentin Fuster
1998;():V003T05A029. doi:10.1115/98-GT-346.
FREE TO VIEW

Black liquor, the lignin-rich byproduct of kraft pulp production, is burned in boiler/steam turbine cogeneration systems at pulp mills today to provide heat and power for onsite use. Black liquor gasification technologies under development would enable this fuel to be used in gas turbines. This paper reports preliminary economics of 100-MWe scale integrated black-liquor gasifier/combined cycles using alternative commercially-proposed gasifier designs. The economics are based on detailed full-load performance modeling and on capital and operating and maintenance costs developed in collaboration with engineers at Bechtel Corporation and Stone and Webster Engineering. Comparisons with conventional boiler/steam turbine systems are included.

Commentary by Dr. Valentin Fuster
1998;():V003T05A030. doi:10.1115/98-GT-349.
FREE TO VIEW

Biomass is a fuel of increasing interest in power generation since it is clean and renewable. Besides conventional power generating systems biomass fuel will be utilized in Integrated Gasification Combined Cycle (IGCC) power plants in the near future.

Carbona Inc. (the successor to Enviropower Inc.) is commercializing a biomass fueled IGCC system. This system is based on a simplified IGCC process which applies the gasification technology originally developed by the Institute of Gas Technology (IGT) and further developed by Enviropower before licensing the technology to Carbona and an advanced hot gas clean-up system.

An extensive pilot test program has been carried out by Enviropower/Carbona covering all aspects of a biomass based gasification process. More than 5000 tons of different biomass feedstocks have been gasified at the pilot plant in Tampere, Finland. The pilot plant converts 15 MW (51 MMBtu/h) thermal input of fuel to product gas. Several biomass qualities/mixtures have been used during the test runs including hard wood, soft wood with and without branches, needles and bark. Short rotation biomass like willow and alfalfa have also been tested.

This paper concentrates on the results and differences in gasification of different biomass materials with special emphasis on the suitability of product gas for gas turbines, the fate of ammonia, vapor phase alkali metals and air toxics. The development of demonstration projects is also discussed in this paper.

Commentary by Dr. Valentin Fuster
1998;():V003T05A031. doi:10.1115/98-GT-350.
FREE TO VIEW

One of the main limiting factors when a green field pulp mill intends to increase its production is the capacity of the Tomlinson boiler by which the cooking chemicals are recovered. Since a boiler of this kind has a service life of some thirty years and represents a substantial investment, it is not an easy decision to replace it for a minor production increase. This study takes a novel approach to the problem. The excess black liquor is assumed to be recovered in a black liquor gasifier, an emerging technology for alternative black liquor recovery. Black liquor gasification in itself has been studied considerably during the last decade. In this study it is combined with the gasification of biomass, i.e. bark and wood, which is already available at a pulp mill. The resulting hybrid energy system, with a gas turbine and a gas expander, offers interesting possibilities: good off-design performance and high efficiency with respect to the fuel quality.

Commentary by Dr. Valentin Fuster
1998;():V003T05A032. doi:10.1115/98-GT-351.
FREE TO VIEW

A new, efficient process for reducing the ash content, drying and fractionating raw lignocellulosic materials into chemicals and a dry solid end product, eminently suitable as a fuel for conventional boilers or for milling to a fine powder for gas turbine firing, shows strong potential for renewable power generation. The dry, low ash solids, termed “Cellulig™”, will also be suitable for gasification and to drive gas turbines. Sustainable liquid and gaseous fuels will become increasingly necessary in the 21st century to reduce dependence on imported fuels, to replace dwindling supplies of oil and natural gas and to avoid environmental damage from green house gases. Convertech Group Ltd. has built a demonstration biomass processing plant at Burnham, Canterbury, New Zealand, with investment from the energy industry and the Australian Energy Research and Development Council. The essential chemical and process engineering elements are described and the current and future development opportunities outlined.

Topics: Fuels
Commentary by Dr. Valentin Fuster
1998;():V003T05A033. doi:10.1115/98-GT-359.
FREE TO VIEW

Elsta B.V. Co., C.V., requested that the General Electric Company and Turbotecnica/Nuovo Pignone propose expanding fuels flexibility for the three MS9001E, DLN-1 (TM) units (being) installed at Terneuzen, The Netherlands. The major challenge was the use of a process gas rich in hydrogen for mixing with normal plant natural gas fuel supply so that molecular hydrogen in the mixture reached 10 percent by volume. Evaluation of the gas turbine combustor operation, based on the criteria of emissions, stability and turndown, required extensive testing, first at the General Electric Corporate Research and Development Center, then at the Schenectady (Building 262) Combustion Laboratory, a full pressure combustor test facility, and finally at Terneuzen with the gas turbines and plant owner’s fuel mixing and forwarding systems. Results of this field testing, built upon the earlier work, are detailed below.

Commentary by Dr. Valentin Fuster
1998;():V003T05A034. doi:10.1115/98-GT-472.
FREE TO VIEW

Biomass integrated gasification-gas turbine (BIG-GT) technology offers the opportunity for efficient and environmentally sound power generation from biomass fuels. Since biomass is ‘carbon-neutral’ it can be used in power generation equipment without contributing to the ‘greenhouse effect’ if it is grown sustainably.

The Brazilian BIG-GT initiative is one of a number of initiatives world-wide aimed at demonstrating, and thereby establishing, biomass as an energy resource for power production. The goal of the Brazilian BIG-GT project is to confirm the commercial viability of producing electricity from wood through the use of biomass-fuelled integrated gasification combined-cycle (BIG-GT) technology. To fulfil this goal a 32 MWe eucalyptus-fuelled demonstration power plant will be built in Brazil on the basis of a design made by TPS Termiska Processer AB (TPS).

The first two phases of the project, which included experimental and engineering studies and the basic engineering of the plant, were completed in 1997. The next phase of the project, the construction and commissioning of the plant, is the recipient of a U.S. $35 million grant from the Global Environmental Facility (GEF) of the United Nations Development Programme (UNDP), in addition to financing from the World Bank (WB).

The plant will be built in Bahia, north-eastern Brazil. The customer of the plant is a consortium, SER - Sistemas de Energia Renovável, comprising of CHESF (Companhia Hidro Elétrica do São Francisco), a federally-owned electricity generation and distribution company, Eletrobras (Centrais Elétricas Brasileiras), a holding company comprising of the main Brazilian companies from the electric generation and distribution sector, and Shell Brasil. Start-up of the plant is scheduled for the year 2000.

The plant will be based on a TPS designed atmospheric-pressure gasification/gas cleaning process. The product gas will be fired in a modified GE LM 2500 gas turbine. The gasification and gas cleaning process is based on the use of a circulating fluidised bed gasifier, secondary stage catalytic tar cracker and conventional cold filter and wet scrubbing technology. The feedstock to the plant will be mainly eucalyptus wood from a dedicated plantation which is harvested on a three-year cycle.

This paper describes the background of the project leading up to the technology selection, the technology that will be employed in the plant and the outline of the economics of this ‘first-of-a-kind’ plant. The progress made in establishing the organisation and the formal framework (e.g. securing the electricity and fuel contracts) are also reported. Future projections of likely technological improvements and cost reductions, and their effect on the overall economics of an ‘Nth’ plant, are presented.

Topics: Wood products
Commentary by Dr. Valentin Fuster
1998;():V003T05A035. doi:10.1115/98-GT-574.
FREE TO VIEW

To support the further development of indirectly heated gasifiers, intended to provide fuels for advanced gas turbines, several laboratory indirectly heated gasifiers were constructed. During many comparative tests advantages and problems with each system were observed. The most useful systems make use of laboratory tube furnaces in conjunction with temperature, time and pressure or volume yield measuring systems and a gas chromatograph with a thermal conductivity detector. In this paper high temperature pyrolysis results obtained with the latest system are presented. Contrasting feedstocks suitable for commercial systems separately or in blends are used. Yield vs. time measurements are used to determine relevant rate constants and outputs. Since the rate constants are mainly reflective of heat transfer effects, cylindrical dowel sticks of varying radii were volatilized. The data set leads to an analytic heat transfer model that considers the hemicellulose, cellulose, and lignin components of the dowels. Also developed from the dowel experiments is an approximate procedure for estimating the proportionate releases of CO, CO2, CH4 and H2 for any type of biomass whose component proportions are known.

Topics: Feedstock
Commentary by Dr. Valentin Fuster

Combustion and Fuels

1998;():V003T06A001. doi:10.1115/98-GT-073.
FREE TO VIEW

To help understand how staged combustion aids in reducing emissions of oxide of nitrogen from gas turbines, measurements and computations are made of structures of two-stage counterflow methane-air flames at nomal atmospheric pressure and a feed-stream temperature of about 300 K. The fuel stream is partially premixed, with equivalence ratios from 1.5 to 3.0. To the air stream is added up to 10% by mass of water spray, carbon dioxide, or nitrogen. Flame structures, including formation of species containing two carbon atoms, are measured by gas chromatography of samples withdrawn by fine quartz probes and are calculated by numerical integrations of the conservation equations employing an updated elementary chemical-kinetic data base. The same sampling system is employed with a low-flow-rate NOx analyzer to obtain profiles of nitric oxide and nitrogen dioxide, which are also calculated in the numerical integrations. The two-stage flame exhibits a green fuel-rich premixed flame and a blue diffusion flame with the maximum NOx concentrations found near the blue flame. At an air-side strain rate of 50 s−1, for fuel-side equivalence ratios of 1.5, 2.0, 2.5 and 3.0, respectively, measured peak NOx concentrations were about 70, 90, 100 and 90 ppm, reduced to 60, 70, 50, and 40 ppm, respectively, when 5% water by mass was added to the air stream. Results of the numerical integrations were in reasonable agreement with these experimental results when suitable selections were made of certain critical elementary reaction-rate constants. These new NOx measurements and computations help to increase understanding of influences of staging and diluent addition, identify important reactions for pollutant formation and suggest means to reduce emissions.

Commentary by Dr. Valentin Fuster
1998;():V003T06A002. doi:10.1115/98-GT-074.
FREE TO VIEW

In recent years, the possibility of climate change has begun to be considered seriously. Options available today can help reduce emissions at relatively little overall cost but may be able to achieve only moderate reductions. If it becomes necessary to reduce emissions further, it is likely there will be opportunities for new technologies as well as making greater use of existing ones. Bearing in mind the time required to develop and deploy new energy supply technologies on a large-scale, it is only sensible to adopt a precautionary stance. This requires better understanding of the potential of technologies not yet in widespread use and stimulation of the development and deployment of promising ones. The EEA Greenhouse Gas R&D Programme is working to improve understanding of technologies for reducing greenhouse gas emissions from fossil fuels. This is an example of effective co-operative action between different countries and industries. Membership is worldwide; through this work, members are able to learn about new technologies and share experiences. This paper reviews the work of the IEA Greenhouse Gas R&D Programme.

The established options for reducing emissions include improving energy efficiency, substitution of lower-carbon fuels for high-carbon fuels, and introduction of alternative energy sources. If deep reductions in emissions are required, discussion tends to focus on alternatives to fossil fuels even though the latter provide a very large proportion of the energy used today. To avoid disruptive changes, the world will need to be able to continue using fossil fuels but in a climate-friendly way.

Capture and storage of carbon dioxide could deliver deep reductions in emissions from fossil fuels but the technology is still in its infancy — this is the subject of on-going work by the IEA Greenhouse Gas R&D Programme. Enhancement of natural sinks, such as forests, could also help by sequestering atmospheric carbon dioxide. Use of biomass for power generation has also been examined to see how it compares as a large-scale mitigation option compared with capture and storage. Methane is another important greenhouse gas, produced by many human activities. Technology can help reduce emissions of methane; examples of some of these technologies will be described. The mechanism of Activities Implemented Jointly is potentially important for application of all of these options and the Greenhouse Gas Programme is working to improving understanding about viable options and methods of delivering successful projects.

Commentary by Dr. Valentin Fuster
1998;():V003T06A003. doi:10.1115/98-GT-075.
FREE TO VIEW

The advent of dry, low-emissions combustion systems for gas turbine applications and the trend towards requiring emissions monitoring and lower NOx limits by regulatory agencies, have created renewed interests in the uncertainty of NOx emissions measurements. This paper addresses the uncertainty of measuring NOx emissions from gas turbines in the field, including gas turbines equipped with conventional combustion systems, with or without water injection, with dry, low-emissions combustion systems and with exhaust clean-up systems.

The sources of errors, using current state-of-the-art instruments, in field emissions testing or continuous emission monitoring of gas turbines to meet specific emission (ppmvd @ 15% O2) as well as mass emission rate (kg/hr) limits are identified. The uncertainty of measuring NOx emissions from gas turbines is estimated and compared with Geld data. The effect of NOx emission levels on measurement uncertainty is also addressed. The minimus NOx measurement uncertainty is determined and is in good agreement with what is currently allowed by regulatory agencies.

Commentary by Dr. Valentin Fuster
1998;():V003T06A004. doi:10.1115/98-GT-178.
FREE TO VIEW

The paper describes results of a parametric study obtained while using an analytical model described earlier (Bunama and Karim, 1997b) investigating the combined effects of mass, energy and momentum transfer with variable transport and thermodynamic properties on the formation of fuel vapour-air mixtures above a stagnant liquid fuel surface within the confines of a vertical cylindrical vessel. This was done mainly to examine the establishment of the formation of flammable mixtures and their changes in size and location with time within liquid fuel tanks that are partially empty. The effects of changes in the ambient and wall temperatures, presence of liquid on the walls and vessel geometry were considered. Moreover, the results of a corresponding experimental investigation are presented. Much of the data relates to the high volatility fuel n-pentane that represents the tighter fuel fractions in commercial fuels which through their early evaporation contribute much to the fire hazards in fuel tanks.

Topics: Fuels , Vessels
Commentary by Dr. Valentin Fuster
1998;():V003T06A005. doi:10.1115/98-GT-179.
FREE TO VIEW

The flammability limits of methane, ethylene, propane and hydrogen were determined experimentally at elevated initial mixture temperatures up to 350°C at atmospheric pressure for upward flame propagation in a steel test tube apparatus. The existence of preignition reactions at these levels of temperatures that may influence the value of the flammability limits was also investigated. The fuel-air mixtures were exposed to elevated temperatures over different periods of time before spark ignition (up to 2 hours). It was shown that the flammability limits for methane widened approximately linearly with an increase in the initial mixture temperature over the entire range of temperatures tested and were not affected by the length of the exposure time to these temperatures before spark ignition. However, different behaviour was observed for the flammability limits of the other tested fuels — ethylene, propane and hydrogen. At higher temperatures the flammability limits narrowed and were very significantly affected by the exposure time. The longer was the exposure time of fuel-air mixtures to the elevated temperatures, the narrower were their flammability limits.

Commentary by Dr. Valentin Fuster
1998;():V003T06A006. doi:10.1115/98-GT-180.
FREE TO VIEW

Advances in gas turbine technology have led to levels of turbine inlet temperature that preclude the use of thermocouple and simple gas analysis techniques for gas temperature determination. Simple gas analysis schemes rely on the measurement of a very limited range of species in the gas sample, typically CO2, CO and hydrocarbons (UHC). A method of estimating the other important species is required. Simple gas analysis schemes that rely only on elemental mass balance equations to determine the concentration of species are inadequate where high temperature results in significant dissociation.

A method has been developed to enable temperature determination at levels that render simple schemes inaccurate. The procedure is based on the measurement of CO2, CO, UHC and oxides of nitrogen in the exhaust gas. Other species concentrations are calculated using an assumption of partial thermodynamic equilibrium. This allows the calculation of many important combustion parameters. The method has been implemented as a computer code, with an object orientated design approach using the C++ language.

The paper details the theory behind the approach and its implementation. The expected errors for practical applications are discussed and quantified. The method is illustrated by an exhaust temperature pattern factor investigation of an annular combustor. Temperatures determined by thermocouples are compared with those calculated from gas samples.

Commentary by Dr. Valentin Fuster
1998;():V003T06A007. doi:10.1115/98-GT-184.
FREE TO VIEW

Gas turbine combustor CFD modeling has become an important combustor design tool in the past few years, but CFD models are generally limited to the flow field inside the combustor liner or the diffuser/combustor annulus region. Although strongly coupled in reality, the two regions have rarely been coupled in CFD modeling. A CFD calculation for a full model combustor from compressor diffuser exit to turbine inlet is described. The coupled model accomplishes two main objectives: 1) implicit description of flow splits and flow conditions for openings into the combustor liner, and 2) prediction of liner wall temperatures. Conjugate heat transfer with nonluminous gas radiation (appropriate for lean, low emission combustors) is utilized to predict wall temperatures compared to the conventional approach of predicting only near wall gas temperatures. Remaining difficult issues such as generating the grid, modeling swirler vane passages, and modeling effusion cooling are also discussed.

Commentary by Dr. Valentin Fuster
1998;():V003T06A008. doi:10.1115/98-GT-185.
FREE TO VIEW

It is known that many of the previously published global methane oxidation mechanisms used in conjunction with computational fluid dynamics (CFD) codes do not accurately predict CH4 and CO concentrations under typical lean-premixed combustion turbine operating conditions. In an effort to improve the accuracy of the global oxidation mechanism under these conditions, an optimization method for selectively adjusting the reaction rate parameters of the global mechanisms (e.g., pre-exponential factor, activation temperature and species concentration exponents) using chemical reactor modeling is developed herein. Traditional global mechanisms involve only hydrocarbon oxidation; that is, they do not allow for the prediction of NO directly from the kinetic mechanism. In this work, a two-step global mechanism for NO formation is proposed to be used in combination with a three-step oxidation mechanism. The resulting five-step global mechanism can be used with CFD codes to predict CO, CO2, and NO emission directly. Results of the global mechanism optimization method are shown for a pressure of 1 atmosphere and for pressures of interest for gas turbine engines. CFD results showing predicted CO and NO emissions using the five-step global mechanism developed for elevated pressures are presented and compared to measured data.

Commentary by Dr. Valentin Fuster
1998;():V003T06A009. doi:10.1115/98-GT-216.
FREE TO VIEW

An application of a simplified combustion model to pollutant emissions prediction in gas turbines is presented here. A critical analysis of the Rizk and Mongia semi-analytical model is conducted, and some corrections are accomplished to obtain a better agreement with experimental data. Special attention is devoted to the temperature equation, which is carefully modified, and to the schematization of several kinds of combustors. The simulation of some conventional pollutant reduction techniques, such as inert injection or exhaust gases re-circulation, is conducted with this corrected model. The results show a good agreement with experimental data both in conventional, and in innovative combustors, like Lean-Premixed or Rich-Lean concepts. The model needs a very short computation time and is likely to allow a simultaneous solution of chemical and fluid-dynamic aspect of the combustor.

Commentary by Dr. Valentin Fuster
1998;():V003T06A010. doi:10.1115/98-GT-217.
FREE TO VIEW

Computational combustion dynamics simulations have been used widely for the design and analysis of conventional rich dome combustors using a fast chemistry assumed shape PDF approach (Shyy et al. 1986) and/or an eddy-breakup model (Valachovic, 1993, Danis et al., 1996). The application of these tools to ultra-low emissions combustors such as the GE LM6000 DLE has been hampered by the inadequacies of the eddy break-up combustion model. In the present work, a partially-premixed laminar flamelet combustion model, based initially on the model of Müller et al. (1994), is applied to an LM6000 single cup combustor. The basic fluid mechanical code is ACC, using the k-ε turbulence model (Prakash, et al., 1998). Assumed-shape PDF models are used for mixture fraction Z(x), and the scalar field G(x), whose level surfaces G = G0 represent the flame location. The model includes the effects of local strain rate on flame propagation rate and extinction through modification of the turbulent flame speed correlation, which determines the rate of propagation of the scalar field variable G. The effects of variable inlet fuel/air ratio variance (unmixedness) on predicted NOx emissions are included through the moments of a calculated NO source term on the PDF’s of Z, and include the contributions of flame-front production of NO in premixed flames. Comparisons to measured velocity and emissions data are shown.

Commentary by Dr. Valentin Fuster
1998;():V003T06A011. doi:10.1115/98-GT-218.
FREE TO VIEW

During the preliminary design and analysis phase of a gas turbine combustor, trade studies of the effects of design variables on emissions and operability are necessary to ensure a successful design. Due to the considerable resources required for full computational combustion dynamics simulations, simplified design tools are required for rapid analysis of a large number of design variable combinations. In a previous paper, a semi-analytical model of a gas turbine combustor was described (Tonouchi, et al., 1997). The model employs a gas particle Monte Carlo technique to simulate the effects of finite-rate micro- and macro-mixing, including full detailed chemical kinetics (Bowman, et al., 1997). Initial model validation work focused on emissions calculations for conventional rich dome combustors. This work presents NOx and CO emissions calculations for a single-cup natural gas-fired dry low emissions (DUE) combustor, and comparison to experimental data. The effect of parametric variation of the micro- and macro-mixing model constants, assigned volumes of the primary and secondary zones, and inlet unmixedness on the results are also presented.

Commentary by Dr. Valentin Fuster
1998;():V003T06A012. doi:10.1115/98-GT-225.
FREE TO VIEW

This paper presents the effect of flame dome depth on the total pressure performance and flow behavior in a sudden expansion region of the combustor diffuser without flow entering the dome head. The mean velocity and turbulent Reynolds stress profiles in the sudden expansion region were measured by a Laser Doppler Velocitmetry (LDV) system. The experiments show that total pressure loss is increased, when flame dome depth is increased. Installation of an inclined combuster wall in the sudden expansion region is suggested from the viewpoint of a control of the reattaching flow. The inclined combustor wall is found to be effective in improvement of the diffuser performance. Better characteristics of the flow rate distribution into the branched channels are obtained in the inclined wall configuration, even if the distorted velocity profile is provided at the diffuser inlet.

Commentary by Dr. Valentin Fuster
1998;():V003T06A013. doi:10.1115/98-GT-227.
FREE TO VIEW

A subgrid scalar mixing and combustion model originally developed for gas phase combustion has been extended to include the liquid phase. This approach includes a more fundamental treatment of the effects of the final stages of droplet vaporization, molecular diffusion, chemical reactions and small scale turbulent mixing than other LES closure techniques. As a result, Reynolds, Schmidt and Damkohler number effects are explicitly included. This model has been implemented within an Eulerian-Lagrangian two phase large-eddy simulation (LES) formulation. In this approach, the liquid droplets are tracked using the Lagrangian approach up to a pre-specified cut-off size. The evaporation of the Lagrangian droplets and the evaporation and mixing of the droplets smaller than the cutoff size is modeled within the subgrid using an Eulerian two-phase model. The issues related to the implementation of this subgrid model within the LES are discussed in this paper along with some preliminary results that demonstrate its capabilities.

Commentary by Dr. Valentin Fuster
1998;():V003T06A014. doi:10.1115/98-GT-228.
FREE TO VIEW

Radiation heat transfer in flames depends strongly on local quantities such as pressure, temperature and concentration of participating species. In the present study, 3D numerical calculations of radiative heat transfer together with the reacting flow field are compared to detailed measurements of the velocity, temperature and spectral radiation field of a model combustor.

The geometry of the combustion chamber (dch = 0.5m), the flame configuration (type-II swirling, diffusion flame) and the highly turbulent flow conditions resemble the characteristics of industrial combustors.

The concentrations of CO2, H2O, CO, CH4, NO, NOx, O2 and H2 as well as local mean temperatures and their fluctuations were recorded at 300 locations at 14 axial planes. The radiation intensity incident on the wall was measured spectrally and time resolved at 11 axial planes within the spectral range of 1.4 to 5.4 μm.

For numerically solving the reacting flow field, spectral methods for calculating the radiative heat transfer were coupled to fluid mechanical methods for calculating the reacting flow.

The agreement between numerical prediction and measurements for the reacting flow field as well as for the radiative heat transfer is reasonably good. The numerical computations show that radiative transfer is of major importance. The temperature in the hot reaction zone was found to be lowered by approximately 400 K by radiative losses.

Commentary by Dr. Valentin Fuster
1998;():V003T06A015. doi:10.1115/98-GT-229.
FREE TO VIEW

The Computational Fluid Dynamics (CFD) code that has been used extensively for combustion applications within GE is CONCERT. CONCERT is a fully elliptic body-fitted CFD code based on pressure correction techniques that solves the three-dimensional (3-D) flow in an aircraft engine combustor. The geometry representation is given by single block structured grids. This paper gives an overview and presents some early applications of the Advanced Combustion Code (ACC) which is currently being developed to further advance the use of Computational Combustion Dynamics (CCD) at GE. ACC will have complex geometry representation capability via unstructured multiblock grids, and will comprise of a number of advanced combustion/spray/turbulence models that are being developed in a modularized fashion. The ACC solver is based on a fully implicit pressure correction algorithm with colocated grids. The development of ACC is taking place by comparing, as far as possible, the results against corresponding CONCERT calculations. Results of a systematic evaluation of various combustion models within the framework of a non-swirling jet diffusion flame using ACC will be presented. Full 3-D combustion calculations have been performed in the dome of single and multiple annular combustors: the corresponding flow and temperature fields and emissions predictions will be presented.

Commentary by Dr. Valentin Fuster
1998;():V003T06A016. doi:10.1115/98-GT-230.
FREE TO VIEW

The chemically reacting flow in the mixing and secondary zone of the improved RQL combustor sector of the DLR has been numerically investigated. The combustion model is adapted to meet experimental conditions and proves to underpredict temperature fluctuations. The pressure dependence of the NOx emission index for the quench and lean zone of the combustor can be described with an exponential factor of 0.66. Although the level of agreement between measured and calculated results is not consistently good, the code can be used to study the effect of various operating conditions as well as the effect of geometrical changes on the flow and the emissions.

Commentary by Dr. Valentin Fuster
1998;():V003T06A017. doi:10.1115/98-GT-232.
FREE TO VIEW

One of the major aims of research in gas turbine combustor systems is the minimisation of non-desirable emissions. The primary method of reducing pollutants such as soot and NOx has been to run the combustion primary zone lean. Unfortunately, this causes problems when the combustor is run under idle and relight conditions as the primary zone air fuel ratio (AFR) can exceed the flammability limit.

Altering this AFR directly affects the primary zone aerodynamics through changes in the spray profile. One method of determining the influence of changes in AFR upon the primary zone is to use Computational Fluid Dynamic (CFD) models. However, to model the flow through an air-blast fuel injector and accurately predict the resulting primary zone aerodynamics requires hundreds of thousands, if not millions, of cells. Therefore, with current computer capabilities simplifications need to be made.

One simplification is to model the primary zone as a 2-D case. This reduces the number of cells to a computationally solvable level. However, by reducing the problem to 2-D the ability to accurately model air-blast fuel injectors is lost as they are intrinsically 3-D devices. Therefore, it is necessary to define boundary conditions for the fuel injector.

Currently, due to difficulties in obtaining experimental measurements inside a air-blast fuel injector, these boundary conditions are often derived using semi-empirical methods. This paper presents and compares two such models; the model proposed by Crocker et al. in 1996 and one developed at DERA specifically for modelling air-blast fuel injectors.

The work also highlights the importance of the often neglected radial component upon the primary zone aerodynamics.

Commentary by Dr. Valentin Fuster
1998;():V003T06A018. doi:10.1115/98-GT-235.
FREE TO VIEW

A method has been found to reduce the high measurement uncertainty of NOx emissions of gas turbines.

Calculation of the measurement uncertainty of NOx emission measurements shows that the high uncertainty is not only due to the uncertainty of the NOx emission measurement itself but is also due to the correction to 15 % O2 in the flue gases.

A second, independent method to determine the percentage oxygen in the flue gases is introduced in addition to the direct sampling method. This second method provides additional information resulting in a significant reduction of the measurement uncertainty.

The solution found has the advantage that the additionally required measurement equipment is kept to a minimum.

Experience with the new method in field tests are excellent.

Commentary by Dr. Valentin Fuster
1998;():V003T06A019. doi:10.1115/98-GT-258.
FREE TO VIEW

Described are sub-scale tests that successfully demonstrate active feedback control as a means of suppressing damaging combustion oscillations in natural-gas-fueled, lean-premix combustors. The control approach is to damp the oscillations by suitably modulating an auxiliary flow of fuel injected near the flame. The control system incorporates state observer software that can ascertain the frequency, amplitude, and phase of the dominant modes of combustion oscillation, and a sub-scale fuel flow modulator that responds to frequencies well above 1 kHz.

The demonstration was conducted on a test combustor that could sustain acoustically coupled combustion instabilities at preheat and pressurization conditions approaching those of gas-turbine engine operation. With the control system inactive, two separate instabilities occurred with combined amplitudes of pressure oscillations exceeding 70 kPa (10 psi). The active control system produced four-fold overall reduction in these amplitudes. With the exception of an explainable control response limitation at one frequency, this reduction represented a major milestone in the implementation of active control.

Commentary by Dr. Valentin Fuster
1998;():V003T06A020. doi:10.1115/98-GT-267.
FREE TO VIEW

A practical active control system for the mitigation of combustion instability has been designed and demonstrated in a lean, premixed, single -nozzle combustor at realistic engine operating conditions. A full -scale engine fuel nozzle was modified to incorporate a simple fuel flow actuator. Results indicate that the system was capable of reducing pressure fluctuations by 82% (15 dB or 5.6X) while maintaining or reducing NOx and CO emissions levels.

Commentary by Dr. Valentin Fuster
1998;():V003T06A021. doi:10.1115/98-GT-268.
FREE TO VIEW

This paper describes an analytical and experimental investigation to enhance combustion system operability using side branch resonators. First, a simplified model of the combustion system dynamics is developed in which the large amplitude pressure oscillations encountered at the operability limit are viewed as limit cycle oscillations of an initially linear instability. Under this assumption, increasing the damping of the small amplitude combustion system dynamics will increase combustor operability. The model is then modified to include side branch resonators. The parameters describing the side branch resonators and their coupling to the combustion system are identified, and their influence on system stability is examined. The parameters of the side branch resonator are optimized to maximize damping augmentation and frequency robustness.

Secondly, the model parameters for the combustor and side branch resonator dynamics are identified from experimental data. The analytical model predicts the observed trends in combustor operability as a function of the resonator parameters and is shown to be a useful guide in developing resonators to improve the operability of combustion systems.

Commentary by Dr. Valentin Fuster
1998;():V003T06A022. doi:10.1115/98-GT-269.
FREE TO VIEW

Lean premixed combustors, such as those used in industrial gas turbines to achieve low emissions, are often susceptible to thermoacoustic combustion instabilities, which manifest themselves as pressure and heat release oscillations in the combustor. These oscillations can result in increased noise and decreased durability due to vibration and flame motion. A physically based nonlinear parametric model has been developed that captures this instability. It describes the coupling of combustor acoustics with the rate of heat release. The model represents this coupling by accounting for the effect of acoustic pressure fluctuations on the varying fuel/air ratio being delivered to the flame, causing a fluctuating heat release due to both fuel air ratio variations and flame front oscillations. If the phasing of the fluctuating heat release and pressure are proper, an instability results that grows into a limit cycle. The nonlinear nature of the model predicts the onset of the instability and additionally captures the resulting limit cycle.

Tests of a lean premixed nozzle run at engine scale and engine operating conditions in the UTRC Single Nozzle Rig, conducted under DARPA contract, exhibited instabilities. Parameters from the model were adjusted so that analytical results were consistent with relevant experimental data from this test. The parametric model captures the limit cycle behavior over a range of mean fuel air ratios, showing the instability amplitude (pressure and heat release) to increase and limit cycle frequency to decrease as mean fuel air ratio is reduced.

Commentary by Dr. Valentin Fuster
1998;():V003T06A023. doi:10.1115/98-GT-272.
FREE TO VIEW

Research and development project of ceramic gas turbines (CGT) was started in 1988 promoted by the Ministry of International Trade and Industry (MITI) in Japan. The target of the CGT project is development of a 300kW-class ceramic gas turbine with a 42 % thermal efficiency and a turbine inlet temperature (TIT) of 1350°C. Three types of CGT engines are developed in this project. One of the CGT engines, which is called CGT302, is a recuperated two-shaft gas turbine for co-generation use. In this paper, we describe the research and development of a combustor for the CGT302.

The project requires a combustor to exhaust lower pollutant emissions than the Japanese regulation level. In order to reduce NOx emissions and achieve high combustion efficiency, lean premixed combustion technology is adopted.

Combustion rig tests were carried out using this combustor. In these tests we measured the combustor performance such as pollutant emissions, combustion efficiency, combustor inlet/outlet temperature, combustor inlet pressure and pressure loss through combustor. Of course air flow rate and fuel flow rate are controlled and measured, respectively.

The targets for the combustor such as NOx emissions and combustion efficiency were accomplished with sufficient margin in these combustion rig tests. In addition, we report the results of the tests which were carried out to examine effects of inlet air pressure on NOx emissions here.

Commentary by Dr. Valentin Fuster
1998;():V003T06A024. doi:10.1115/98-GT-292.
FREE TO VIEW

A wet low-NOx combustion system being developed for the AlliedSignal ASE40 industrial gas turbine is assessed using advanced 3-D CFD analysis. A PDF combustion-turbulence interaction model was modified to allow analysis of simultaneous injection of water with gaseous or liquid fuel. To the authors’ knowledge, such a CFD analysis is unique in the open literature. Analyses of the wet low-NOx combustion system were performed with and without water injection at full power engine conditions. Good qualitative agreement between engine emission data and predictions was seen. NOx reductions of 58% and 77% were measured for water-to-natural gas mass ratios of 0.5 and 1.0, respectively, compared to 75% and 93% for CFD calculations. Corresponding CO levels were measured to increase by factors of 3 and 9, compared to CFD predictions of 4 and 7. Similar trends were predicted for water injection with DF-2 diesel fuel. Predicted overall flow patterns were not significantly changed with water injection. NOx reductions were caused by a reduction in maximum flame temperatures in the primary and intermediate zones when water was injected. CO increases were caused by a reduction of CO oxidation downstream of the dilution zone (in the turn-around duct) due to lower gas temperatures with water injection.

Commentary by Dr. Valentin Fuster
1998;():V003T06A025. doi:10.1115/98-GT-303.
FREE TO VIEW

Successful NOx measurements in the end plane of the primary zone of a small tubular gaseous fuelled combustor of conventional gas turbine design, employing acoustic driving of the combustor via the air inlet pipe, have been made at scaled 1/8 load operating conditions. The acoustic drive caused partial blockage of the combustor primary zone air flow which increased the equivalence ratio in the plane of the NOx measurements. The mixing was acoustically augmented which together with the blockage richening significantly changed the combustor mean NOx–mean equivalence ratio characteristic in the end plane of the combustor primary zone. Under lean-conditions at 1/8 load and “with-drive” at 246 Hz mean NOx was reduced indicating that a value of 10ppm (50% reduction) is possible, confirming previous results. Under rich-conditions NOx “with-drive” at 246 Hz might be reduced by 23%, relative to previous results, and even might reach the “no-drive” value of about 25ppm, which was possibly due to acoustic augmentation of the primary zone aerodynamics. The NOx at the combustor exhaust “with-drive” should be even lower than that measured “with-drive” because of normal dilution. Therefore, the technique has the potential to create a low NOx conventional combustor without requiring the complicated design-changes of current industry efforts.

Commentary by Dr. Valentin Fuster
1998;():V003T06A026. doi:10.1115/98-GT-304.
FREE TO VIEW

Lean premixed combustion has emerged as a method of achieving low pollutant emissions from gas turbines. A common problem of lean premixed combustion is combustion instability. As conditions inside lean premixed combustors approach the lean flammability limit, large pressure variations are encountered. As a consequence, certain desirable gas turbine operating regimes are not approachable. In minimizing these regimes, combustor designers must rely upon trial and error because combustion instabilities are not well understood (and thus difficult to model). When they occur, pressure oscillations in the combustor can induce fluctuations in fuel mole fraction that can augment the pressure oscillations (undesirable) or dampen the pressure oscillations (desirable). In this paper, we demonstrate a method for measuring the fuel mole fraction oscillations which occur in the premixing section during combustion instabilities produced in the combustor that is downstream of the premixer.

The fuel mole fraction in the premixer is measured with kHz resolution by the absorption of light from a 3.39 μm He-Ne laser. A sudden expansion combustor is constructed to demonstrate this fuel mole fraction measurement technique. Under several operating conditions, we measure significant fuel mole fraction fluctuations that are caused by pressure oscillations in the combustion chamber. Since the fuel mole fraction is sampled continuously, a power spectrum is easily generated. The fuel mole fraction power spectrum clearly indicates fuel mole fraction fluctuation frequencies are the same as the pressure fluctuation frequencies under some operating conditions.

Commentary by Dr. Valentin Fuster
1998;():V003T06A027. doi:10.1115/98-GT-310.
FREE TO VIEW

Lower Emissions have become key characteristics of most new gas turbine engines over the last several years. The ‘lean premixed’ approach has been used in the development of the Dry Low Emissions (DLE) technology. The LM6000 and the LM2500 combustors employ a triple dome design with staging of fuel and air flows to achieve lean-premixed operation from light-off to full power. This technology permits the operator to run with reduced emissions of NOx as well as CO and UHC over a wide load setting. Emissions goals of 25 ppm have been successfully met at site rating conditions for the entire family of LM DLE products. The DLE combustor operates on the mid dome at light-off, the inner and the outer domes are brought on progressively, as the engine is loaded. The combustor utilizes a small quantity of air for dome and liner cooling as most of the combustor air is mixed with fuel in the premixers. Backside cooling enhancements permit the reduction of film cooling, which can cause quenching of CO oxidation reactions. Combustion acoustics are controlled by the use of passive devices on the exterior of the engine as well as by fuel staging within premixers and by the use of a control system which senses and alters the combustor operation to limit acoustics. The DLE technology meets the emissions and reliability needs of the industry with limited package modifications. This paper describes the DLE technology, developed to meet the needs of the industry. Critical design features including the Double Annular Counter-Rotating Swirler (DACRS) premixer, the triple annular dome design, the heat shield design and the staging sequence are discussed, in addition to the field experience gained on the LM2500 and LM6000 DLE models.

Commentary by Dr. Valentin Fuster
1998;():V003T06A028. doi:10.1115/98-GT-323.
FREE TO VIEW

Environmental compatibility requires low emission burners for gas turbine power plants as well as for jet engines. In the past significant progress has been made developing low NOx and CO burners by introducing lean premixed techniques. Unfortunately these burners often have a more pronounced tendency than conventional burner designs to produce combustion driven oscillations. The oscillations may be excited to such an extent that strong pulsation may possibly occur; this is associated with a risk of engine failure and higher NOx emissions.

In order to describe the acoustical behaviour of the complete burner system the determination of the transfer function of the flame itself is crucial. Using a new method which was presented by Bohn, Deutsch and Krüger (1996) and Bohn, Li, Krüger and Matousckek (1997), the dynamic flame behaviour can be predicted by means of a full Navier-Stokes-simulation of the complex combustion process for the steady-state as well as for the transient situation.

This method has been successfully used by the authors to obtain the frequency response of turbulent diffusion flames and laminar premixed flames. For the application in modern gas turbines the influence of turbulence on the dynamic behaviour of premixed flames is of big interest.

Therefore, this paper presents numerical studies of a turbulent premixed flame configuration for which experimental data is available in the literature. Two different combustion models have been used for the steady-state as well as for the transient calculations. With the improved model, which takes into account the chemical kinetics and the interaction between turbulence and kinetics, good agreement has been found for the steady-state results and for the frequency response of the flame.

Topics: Turbulence , Flames
Commentary by Dr. Valentin Fuster
1998;():V003T06A029. doi:10.1115/98-GT-357.
FREE TO VIEW

The development of a low NOx combustor is underway at the Advanced Material Gas-Generator Research Institute (AMG). To achieve low NOx emissions, a lean-premixed-prevaporized (LPP) swirling flow combustor was selected. There are two objectives for this combustor: a uniform fuel/air mixture coupled with well-atomized fuel for low NOx emissions, and a combustion stabilization of this new concept combustor. The spray characteristics test and the combustion test have carried out. This paper describes the evaluation of these results.

Commentary by Dr. Valentin Fuster
1998;():V003T06A030. doi:10.1115/98-GT-360.
FREE TO VIEW

In gas turbine combustors, optimum arrangement between a fuel nozzle and a swirler/prefilmer module must be sought to achieve satisfactory ignition and stability characteristics in addition to reduced level of emissions. However, due to thermal expansion of the combustor or misalignment of the fuel nozzle, the location of a fuel nozzle may vary. Displacement of a fuel nozzle may change the amount of fuel injected to the pre-filming device (usually the inner swirler wall) and the location of attachment, which in turn affects the thickness of pre-filming liquid sheet on the wall. As a result, the spray structure formed by pre-filming airblast atomization may be significantly changed.

An experimental investigation is carried out to study the effects of fuel nozzle displacement on the structure of a spray formed by a dual orifice pressure atomizer and a counter-rotating dual swirler. The inner wall of the swirler is designed to be used as a pre-filming device. The behavior of droplets, the flow characteristics of the swirling air flow, and the interaction between droplets and the flow are studied. Optical diagnostic methods including a flow visualization and an Adaptive Phase/Doppler technique are used. Distributions of droplet size, number density, and liquid phase volume flux are presented for various fuel nozzle displacements, in addition to gas phase velocity.

Topics: Fuels , Nozzles , Displacement
Commentary by Dr. Valentin Fuster
1998;():V003T06A031. doi:10.1115/98-GT-381.
FREE TO VIEW

A catalytically assisted ceramic combustor for a gas turbine was designed to achieve low NOx emission under 5ppm at a combustor outlet temperature over 1300°C. This combustor is composed of a burner system and a ceramic liner behind the burner system. The burner system consists of 6 catalytic combustor segments and 6 premixing nozzles, which are arranged in parallel and alternately. The ceramic liner is made up of the layer of outer metal wall, ceramic fiber and inner ceramic tiles.

Fuel flow rates for the catalysts and the premixing nozzles are controlled independently. Catalytic combustion temperature is controlled under 1000°C, premixed gas is injected from the premixing nozzles to the catalytic combustion gas and lean premixed combustion over 1300°C is carried out in the ceramic liner. This system was designed to avoid catalytic deactivation at high temperature and thermal and mechanical shock fracture of the honeycomb monolith of the catalyst.

A combustor for a 10MW class, multi-can type gas turbine was tested under high pressure conditions using LNG fuel. Measurements of emission, temperature, etc. were made to evaluate combustor performance under various combustion temperatures and pressures.

This paper presents the design features and the test results of this combustor.

Commentary by Dr. Valentin Fuster
1998;():V003T06A032. doi:10.1115/98-GT-382.
FREE TO VIEW

An experimental and numerical investigation into the attenuation of combustion induced pressure oscillations in a single nozzle rig was undertaken at the United Technologies Research Center. Results from these investigations indicated a high combustor exit Mach number, similar to that used in a gas turbine engine, was required to correctly simulate the combustor dynamics and evaluate acoustic characteristics of lean premixed fuel injectors. Comparisons made between aerodynamically stabilized and bluff-body stabilized nozzles and the use of premixed and diffusion pilots showed that small levels of diffusion piloting behind a bluff-body yielded the best acoustic/emission performance. Their success is due to increased flame stabilization (superior anchoring ability) which reduced flame motion and thermal/acoustic coupling. For cases where diffusion piloting was not present, both designs exhibited similar dynamical behavior. Increases in the combustor exit Mach number and reductions in the inlet air temperature were shown to degrade acoustic performance of both nozzle designs. The bluff-body configuration with small levels of diffusion piloting, however, was found to be less sensitive to these changes when compared to its aerodynamic counterpart.

Commentary by Dr. Valentin Fuster
1998;():V003T06A033. doi:10.1115/98-GT-388.
FREE TO VIEW

The benefits of thermal barrier coatings for protection of combustor walls are well known. However, the trend to higher combustor inlet temperatures and the reduced availability of cooling air leads to a demand for better insulation performance from the thermal barrier coating (TBC). This is of particular benefit for low emission combustors where wall quenching effects need to be minimised and often hot side cooling is not permissible.

A combustor can, for advanced stationary gas turbines, with 1.8 mm thick thermal barrier was designed and tested. The can was compared to a combustor can with a thermal barrier coating sprayed with current state-of-the-art methods, but to the same thickness.

Steps to optimise performance were taken in all development stages. The design allowed easy spray geometries, improved edges and no film cooling. Spraying was optimised in order to achieve a segmented microstructure for reduction of stresses (by decrease of the Young’s Modulus in the coating) and increase compliance of the coating. Testing in component test rigs showed excellent results. The lifetime of the optimised combustor can was beyond test capabilities, whereas the reference combustor failed immediately.

Metallographic and X-ray characterisation before and after component rig testing was performed and revealed features that explain the superiority of the segmented thermal barrier coating.

This work has been funded by the CEC under the contract BRE-CT94-0936.

Commentary by Dr. Valentin Fuster
1998;():V003T06A034. doi:10.1115/98-GT-389.
FREE TO VIEW

The three-dimensional unsteady flow in a gas turbine combustor was studied using CFD means. The flow structure around a fuel spoke is of interest not only because of pollutant issues, but also because of combustor operating issues such as combustion acoustics and potential flame-holding in the premixer. The CFD model was tested extensively in terms of grid density and lime-marching step size before the final calculation was made. It was shown that when a swirling flow crosses over a cylindrical fuel spoke, wake vortices are formed and a strong secondary flow is generated along the spanwise direction. A secondary vortex existed near the tip of the spoke. This complex flow structure affects the quality of fuel and air mixing and can be addressed by CFD-based design methods.

Commentary by Dr. Valentin Fuster
1998;():V003T06A035. doi:10.1115/98-GT-390.
FREE TO VIEW

A catalytic combustor was designed and tested for a small recuperated gas turbine engine for use in a hybrid electric vehicle (HEV). Combustor rig and engine tests were performed with DF-2 diesel fuel, kerosene, and automotive gasoline. Rig test steady-state emissions were measured over the full engine operating range. Nitrogen oxides (NOx) emissions were insensitive to operating condition, and were almost strictly a function of fuel nitrogen content. For low-nitrogen kerosene, NOx emissions less than 1 ppm(vol) (at 15 percent O2) were demonstrated. Startup emissions were measured for conditions modeling engine spoolup from ignition to full speed.

Laboratory engine tests on DF-2 over a range of speeds and loads demonstrated tailpipe emissions less than 10 ppm(vol) NOx and less than 1 ppm(vol) unburned hydrocarbons (HC). NOx, HC, and carbon monoxide (CO) emissions were less than the California State Ultra-Low Emissions Vehicle (ULEV) standards for steady-state operation, corrected for assumed vehicle load and fuel consumption rate.

Commentary by Dr. Valentin Fuster
1998;():V003T06A036. doi:10.1115/98-GT-433.
FREE TO VIEW

The effect of the residence time variation on NOx formation in high-intensity, lean-premixed (LP) methane combustion is explored through experiments conducted in a high-pressure jet-stirred reactor (HP-JSR) operated at 6.5 atm pressure. The residence time is varied between 0.5 ms and 4 ms, holding the measured reactor recirculation zone temperature constant at 1803 K. Air preheat is not used. The results indicate a minimum NOx level of 3.5 ppmvd (15% O2) for reactor mean residence times between 2 and 2.5 ms. As the residence time is reduced from 2.0 ms to 0.5 ms, the NOx increases, consistent with a spreading of super-equilibrium concentrations of free-radicals throughout the reactor. For the shortest residence times examined, PSR modeling agrees with the NOx measurements. At long residence times, (i.e., above 2.5 ms), the measured CO behavior indicates the super-equilibrium free radicals, and thus the rapid NOx production, are confined mainly to the jet zone of the reactor. For the long residence time range, the measured NOx increases with increasing residence time, and is significantly less than the PSR predictions. A simple two-zone model of the HP-JSR is used to interpret and evaluate the NOx formation.

Experiments exploring the effect of inlet temperature on NOx are conducted in an atmospheric pressure, methane-fired, jet-stirred reactor (A-JSR). The reactor temperature is held constant at 1788 K, and the inlet mixture temperature is varied between the no-preheat case and 623 K. These experiments show that increasing the inlet air temperature over the full range tested decreases the NOx by about 30%. Several explanations are offered for the behavior. For both reactors, i.e., the HP-JSR and A-JSR, single inlet jet nozzles are used. The results lead to a practical conclusion that very low NOx levels can be achieved for combustion in strongly back-mixed reaction cavities adjusted to optimal residence time and inlet temperature.

Commentary by Dr. Valentin Fuster
1998;():V003T06A037. doi:10.1115/98-GT-440.
FREE TO VIEW

To address the complex effect of inlet parameters on combustor performance, a statistically based technique is applied to a model, premixed natural gas fired combustor. In this way, multiple parameters are simultaneously investigated for their contribution to combustion performance. Atmospheric tests are performed at conditions otherwise representative of industrial combustors: 670 K. inlet preheat and an equivalence ratio of 0.47. Experimental results, in combination with CFD modeling, reveal that (1) the statistical approach is an effective tool by which parameters that dominate performance can be identified, (2) the principal statistically significant parameter linked to NOx production is the inlet fuel distribution, (3) the principal statistically significant parameter linked to CO production is swirl solidity, and (4) an inlet fuel distribution that features a concentration peak in line with the shear layer of the recirculation zone yields NOx levels comparable to a well premixed case.

Commentary by Dr. Valentin Fuster
1998;():V003T06A038. doi:10.1115/98-GT-441.
FREE TO VIEW

The performance of liquid fuel atomizer has direct effects on combustion efficiency, pollutant emission and stability. Pressure swirl atomizer, or simplex atomizer, is widely used in liquid fuel combustion devices in aircraft and power generation industry. A computational, experimental, and theoretical study is conducted to predict its performance. The Arbitrary-Lagrangian-Eulerian method with finite volume scheme is employed in the CFD model. Internal flow characteristics of the simplex atomizer as well as its performance parameters such as discharge coefficient, spray angle and film thickness are predicted. A temporal linear stability analysis is performed for cylindrical liquid sheets under 3-D disturbance. The model incorporates swirling velocity component, finite film thickness and radius which are essential features of conical liquid sheets emanating from simplex atomizers. It is observed that the relative velocity between liquid and gas phase, density ratio and surface curvature enhance the interfacial aerodynamic instability. As Weber number and density ratio increase, both the wave growth rate and the unstable wave number range increase. Combination of axial and swirling velocity components is more effective than single axial component for disintegration of liquid sheet. A breakup model for conical liquid sheet is proposed. Combining the breakup model with linear stability analysis, mean drop sizes are predicted. The theoretical results are compared with measurement data and agreement is very good.

Commentary by Dr. Valentin Fuster
1998;():V003T06A039. doi:10.1115/98-GT-442.
FREE TO VIEW

The objective of the present investigation is to provide better understanding of the hybrid atomization process in an effort to support the development of fuel injectors for future high performance / low emissions gas turbine combustors. A specially designed atomizer that incorporated two swirling air streams, and a prefilming device located upstream of the atomizer exit section was tested under a combined hybrid airblast and high liquid pressure mode. The experiments focused on evaluating the effects of several operating parameters, in particular the air / liquid ratio, on the atomization quality. The results demonstrated that, to accurately determine the role of the air / liquid ratio in the atomization process, the effects of liquid injection velocity and the relative air–liquid velocity need to be separated from that of the air / liquid ratio. Two approaches were used in the present investigation to deduce the actual effect of the air / liquid ratio: first, by reducing the air swirler flow areas, and second, by increasing the number of liquid injection holes. Both approaches enabled changing the air / liquid ratio without changing the air or liquid velocities. The atomization results indicate that changes in swirler flow area produce a stronger effect of the air / liquid ratio than that when liquid hole number was changed. For fixed air / liquid ratio, better atomization quality was achieved when both levels of air flow and liquid flow were high compared to when both flow rates were low. Also, the atomizer demonstrated a continuous improvement in atomization quality under very high air pressure drop, indicating a better utilization of the air kinetic energy over conventional airblast atomizers. The other important observation was that the dependency of the atomization process on air velocity was not constant, but rather changed with liquid pressure, air flow rate, and air pressure drop.

Topics: Pressure
Commentary by Dr. Valentin Fuster
1998;():V003T06A040. doi:10.1115/98-GT-444.
FREE TO VIEW

Computational Combustion Dynamics has been used extensively at General Electric Company for combustion applications. This paper demonstrates an application of Advanced Combustion Code to GE’s lean premixed dry low NOx emissions LM2500 and LM6000 gas turbine combustors. A methodology for anchoring the Double Annular Counter-Rotating Swirler (DACRS) exit conditions to Laser Doppler Velocity data from a reacting single cup experiment is described. The DACRS exit velocity profiles and turbulence parameters are inlet boundary conditions for the annular combustor simulation. Since over 80 per cent of the total air enters the combustor via the premixers, inaccuracies in these boundary conditions have a significant impact on the predicted flame shape, liner temperatures and emissions.

The paper shows comparisons between measured and predicted velocity in a rectangular duct equipped with a single DACRS. The k-ε turbulence model and the two-step eddy break up/eddy dissipation combustion models are used to predict the reacting flow field of the natural gas/air flame. The inlet velocity profiles are developed first to match the LV data and the observed flame impingement location at nominal settings of the inlet turbulence parameters. The sum square error between measured and predicted velocity is used as the optimization function. Next, a design of experiment computational study is conducted to determine the inlet turbulence length scale and kinetic energy in order to further improve the data match. The eddy break up model is shown to be more robust than the eddy dissipation model. The eddy dissipation model resulted in slow combustion rates, and high fuel and carbon monoxide emissions.

Topics: Modeling , Emissions
Commentary by Dr. Valentin Fuster
1998;():V003T06A041. doi:10.1115/98-GT-445.
FREE TO VIEW

Airblast atomisation drop size is a function of the liquid and gas flow conditions. It is also subject to the atomisation geometry, or more specifically the jet breakup mechanism. Plain jet atomisation featuring coaxial air and fuel flows has been investigated to assess the injector geometry effect on the spray characteristics. Results from various flow conditions and atomiser configurations suggest that a prompt atomisation correlation that was evaluated for prefilming injectors can be applied to plain jet airblast atomisation, in a slightly modified form. Changes in the velocity term are necessary to fit the measured data. A scaling factor has been established to compensate for the velocity term change. This factor may also imply the underlying difference between flat sheet and round jet atomisation. The liquid atomisation mode is dependent not only on the manner of geometrical air-liquid contact but also on flow conditions. In this study, the combined air-fuel velocity ratio VR and Weber number (WeVR) is found to be a criteria that determines the air flow pattern influence on atomisation. Data from this experiment show that a small change in the axial distance between the liquid jet and air orifice entrance results in marked difference in spray drop mean size under low air momentum flow conditions.

Topics: Ejectors , Geometry
Commentary by Dr. Valentin Fuster
1998;():V003T06A042. doi:10.1115/98-GT-454.
FREE TO VIEW

The knowledge of the hot gas side heat load is a necessary prerequisite for the design of the advanced cooling scheme of a dry low-NOx combustor and the mechanical integrity (MIT) calculations of the combustor walls. The magnitude and the spatial distribution of the heat fluxes has to be known in the very early phase of the design, where there is no hardware available.

The evaluation of a combustor wall design has to be based on known process data, thermodynamic and combustion parameters and has to rely on computational methods and experience.

A stepwise computational approach is presented to reach this target utilising

• 1D flow and Nusselt-Number correlations

• 2-D boundary layer code

• computational fluid dynamics (CFD)

For the validation of the method atmospheric and pressurised single burner combustor tests were performed.

The relative merits and pitfalls of the different methods are discussed in detail. Recommendations for their utilisation within the design process and for their further development are given.

Commentary by Dr. Valentin Fuster
1998;():V003T06A043. doi:10.1115/98-GT-487.
FREE TO VIEW

A novel fuel-air mixing technique on the basis of vortex generators has been developed and successfully implemented in the worlds first lean-premix reheat combustor of ABB’s GT24/GT26 series industrial gas turbines. This technique uses a special arrangement of delta-wing type vortex generators to achieve rapid mixing through longitudinal vortices, which produce low pressure drop and no recirculation zones along the mixing section. In this paper, after a short introduction to the topic, the motivation for utilizing vortex generators and the main considerations in their design are explained. A detailed analysis of the flow field, pressure drop and the strength of the vortices generated by a single vortex generator are presented as one of the three main geometrical parameters is varied. The results obtained through water model tests indicate that an optimum vortex generator geometry exists, which produces the maximum circulation at a relatively low pressure drop price. Moreover, the axial velocity distribution along the mixing section stays uniform enough to assure flash-back free operation despite the elevated inlet temperatures encountered in a reheat combustor. After selecting this optimized geometry, the process of the arrangement of multiple vortex generators in an annular combustor segment is described. The optimum arrangement presented here is suitable both for gaseous and liquid fuel injection, since it requires only one injection location per combustor segment.

Commentary by Dr. Valentin Fuster
1998;():V003T06A044. doi:10.1115/98-GT-492.
FREE TO VIEW

Combustion dynamics (or combustion oscillations) have emerged as a significant consideration in the development of low-emission gas turbines. To date, the effect of premix fuel nozzle geometry on combustion dynamics has not been well-documented. This paper presents experimental stability data from several different fuel nozzle geometries (i.e., changing the axial position of fuel injection in the premixer, and considering simultaneous injection from two axial positions). Tests are conducted in a can-style combustor designed specifically to study combustion dynamics. The operating pressure is fixed at 7.5 atmospheres and the inlet air temperature is fixed at 588K (600F). Tests are conducted with a nominal heat input of 1MWth (3MBTUH). Equivalence ratio and nozzle reference velocity are varied over the ranges typical of premix combustor design. The fuel is natural gas. Results show that observed dynamics can be understood from a time-lag model for oscillations, but the presence of multiple acoustic modes in this combustor makes it difficult to achieve stable combustion by simply re-locating the point of fuel injection. In contrast, reduced oscillating pressure amplitude was observed at most test conditions using simultaneous fuel injection from two axial positions.

Commentary by Dr. Valentin Fuster
1998;():V003T06A045. doi:10.1115/98-GT-493.
FREE TO VIEW

Combustion flows with swirl are investigated in the context of a lean premixed pre-vaporized combustor. The turbulent flame speed closure model implemented into the Kiva program appears to be efficient in utilizing computing time and memory. It also predicts a larger flame spread than a distributed reactor model. The effect of different turbulence and combustion models on flow recirculation patterns and heat release is reported.

Commentary by Dr. Valentin Fuster
1998;():V003T06A046. doi:10.1115/98-GT-500.
FREE TO VIEW

Advanced prefilming airblast atomizers are widely used for low emission combustors since they deliver a fine spray almost independently of the fuel flow rate. The droplet spectrum produced by this type of atomizer results from the aerodynamic forces at the atomizer edge and from the fuel properties prior to the film disintegration. Therefore, the wall film temperature is an important parameter affecting the fuel properties and in turn the atomization quality. Even though this atomizer type became well investigated (Lefebvre 1989, Rizk et al. 1987, Sattelmayer et al. 1989), still no general quantitative relationship between atomizer design and spray quality could be established since the fuel state at the atomizer edge cannot be precisely predicted yet.

In extending earlier experimental and theoretical work on airblast atomizers (Sattelmayer et al. 1989, Himmelsbach et al. 1994, Willmann et al. 1997) and recent advances in the numerical modeling of wall film flows (Rosskamp et al. 1997a), this paper presents a numerical approach to judge the effect of fuel mass flow, air flow and the film length (i. e. length of atomizer lip) on the temperature of the liquid at the atomizer edge. The computer code developed provides detailed information on the wall film flow and the nozzle wall temperature. For the prediction of heat transfer to the film a new model has been developed which is based on measurements of the internal film flow (Elsäßer et al 1997).

This new numerical approach can serve as a design tool to evaluate the effects of design modifications during atomizer development with view to their effect on atomization performance. The paper includes the theory for two-phase flow modeling and a generic parameter study that points out that the liquid loading and the length of the atomizer lip are important parameters in atomizer design. The calculations presented in the paper emphasize the necessity of coupled two-phase flow calculations because the film strongly interacts with the gas phase and the predicted atomizer performance is sensitive to changes in the air flow.

Commentary by Dr. Valentin Fuster
1998;():V003T06A047. doi:10.1115/98-GT-502.
FREE TO VIEW

With the advent of today’s dry low emission (DLE) combustors for industrial gas turbines (GT), an additional type of load case other than thermal loading on the combustor structure, has put itself heavily into focus, namely pulsation loading. Although recognised for decades in rocket engines and ram-jets, it was not until the incorporation of diluted flames and completely closed combustors, used to reduce NOx emissions, that thermo-acoustically excited pressure pulsations became an issue in the design of industrial GT combustors.

This paper presents the computational methods that are available for the analysis of this phenomenon, and their application in the development of ABB’s latest annular DLE combustor. Among these are calculations of burner excitation frequency, pressure wave propagation within the combustor, acoustic and structural eigen frequencies, and coupled acoustic structural analysis.

Guidelines for the combustor design to prevent instabilities and coupling of air and structure pulsations are suggested.

Commentary by Dr. Valentin Fuster
1998;():V003T06A048. doi:10.1115/98-GT-517.
FREE TO VIEW

Experimental and computational (CFD) investigations were conducted on an Airblast Simplex (ABS) nozzle. A Phase Doppler Particle Analyzer (PDPA) was used to study the main characteristics of the nozzle. Water was used as the test liquid with pressure drops of 68.9, 172.3 & 689 kPa, while the air pressure drops were 2.49 & 34.5 kPa. PDPA data were obtained at three different axial distances from the nozzle exit (x = 2.5 (or 7), 17 & 40 mm). In the numerical simulations, the three-dimensional flow field for both air and liquid was computed. The combination of detailed experimental measurements and CFD provided excellent insight into the physics of the atomizer. The three-dimensional CFD simulation showed distinct jets originating the air swirler and secondary flows that cause multiple peaks in number density, volume flux and velocity distributions. The jets suggest why asymmetric distributions are common in sprays. The jetting effect from the swirler vanes emphasizes the importance of swirler and airflow design on the performance of a spray nozzle. The discrete jetting effect is not visually observable, nor is it easily measured using PDPA.

The ABS nozzle examined has several distinct advantages over a pressure swirl atomizer, including, (1) the distinct sheet from the hollow cone spray is dispersed within 17 mm in all cases studied herein, and (2) the spray volume does not change significantly with increasing air or liquid pressure drop. Therefore, careful design of ABS nozzle results in a spray with nearly constant spray volume and SMD over a wide range of Air/Liquid Ratios.

Commentary by Dr. Valentin Fuster
1998;():V003T06A049. doi:10.1115/98-GT-519.
FREE TO VIEW

A dual fuel burner has been developed to meet stringent NOx goals without the use of water or steam injection. This combustion system is based on the proven ABB EV burner dry low NOx technology and uses the same type of aerodynamic vortex breakdown flame stabilization. A more advanced aerodynamic design improves the quality of the fuel air mixture for both gaseous and liquid fuels. The design of the liquid fuel injection and the fuel-air-mixture preparation is described in this paper. Fuel air mixture homogeneity was improved with the help of experimental and numerical tools. This way an optimization in fuel atomizer design was possible. Distinct differences in fuel distribution were observed for different designs of pressure atomizers. Combustion tests of the burner were performed at pressures up to 20 bars. The NOx levels measured under gas turbine full load conditions are <25 vppm using oil no. 2 and <10 vppm using natural gas. These results highlight the potential for achieving similar combustion low emission performance for gaseous and liquid fuels near perfect lean premix conditions.

Operating parameters and test results at part load conditions are discussed as well in this paper. The wide operating range of the burner in the full premix mode restricts the need for pilot application or burner staging to low load (<50 %) conditions. This allows for low emissions on NOx, CO and UHC in the entire load range.

Commentary by Dr. Valentin Fuster
1998;():V003T06A050. doi:10.1115/98-GT-537.
FREE TO VIEW

A parametric experimental study has been conducted to measure the discharge coefficient, the flow number, the film thickness, the spray angle, the velocity coefficient and droplet size of a large-scale simplex nozzle using ultrasonic and optical techniques. Seventeen nozzle geometries have been studied for three mass flow rates. The large-scale nozzle provides adequate resolution for measurements of film thickness, spray angle, and droplet size. The experimental data collected have been used to derive new and improved correlations for nozzle flow and breakup parameters. It is found that the atomizer constant (ratio of total inlet area to product of the swirl chamber and orifice diameter) is the primary parameter affecting the atomizer performance. As the atomizer constant increases, the discharge and velocity coefficients increase and the spray angle decreases.

Commentary by Dr. Valentin Fuster
1998;():V003T06A051. doi:10.1115/98-GT-552.
FREE TO VIEW

Lean premixed (LP) combustion is a proven concept for NOx-reduction in gas turbine systems. Lean premixed liquid fuel systems, also known as LPP (lean premixed prevaporized) systems, however, are more complex, as the fuel has to be atomized, evaporated and premixed prior to combustion and there is a higher risk of autoignition and coking.

The Siemens LPP Hybrid burner, which was introduced into the Halmstadvaerket gas turbine of Sydkraft AB in Sweden in 1993, has since proven its reliability. As the primary design criteria have been maximum reliability and security against flashback, it still has a considerable potential for NOx-Emissions reduction.

The objectives of this paper are to summarize the activities connected with the further development of the LPP Hybrid burner based on the Halmstadvaerket operation experience. These include the introduction of the LPP burner into a new machine generation with ring combustors as well as investigations into the two phase flow phenomena inside the burner.

When the Model VX4.3A gas turbine family was introduced, the LPP Hybrid burner had to be scaled down to fit the different machine sizes. The paper reports on the development steps and first operation experiences with the LPP Hybrid burner system in HBR-combustor machines.

In a concluding chapter, the paper outlines the next steps in the LPP development program. These include further experimental and numerical studies which are partly conducted in a co-operation between Sydkraft AB and Siemens AG.

Commentary by Dr. Valentin Fuster
1998;():V003T06A052. doi:10.1115/98-GT-553.
FREE TO VIEW

AlliedSignal Engines is addressing critical concerns slowing commercialization of structural ceramics in gas turbines. The AlliedSignal 331-200[CT] APU test bed features ceramic first-stage nozzles and blades. Fabrication of ceramic components provides manufacturing process demonstration scale-up to minimum levels for commercial viability. Endurance tests and field testing in commercial aircraft will demonstrate component reliability.

Manufacturing scale-up activities showed significant progress in 1997. Subcontractors AlliedSignal Ceramic Components (CC, Torrance, CA) and Kyocera Industrial Ceramics Corporation (KICC, Vancouver, WA), transitioned process refinements to demonstration. CC initiated trial production of 100 nozzles/month. These suppliers are also developing fixed processes to fabricate ceramic integrally-bladed turbine rotors (“blisks”).

Ceramic design technology advanced with carbon particle impact testing supporting impact model verification, and 300 hours successful engine testing of longer-life inserted blade attachment compliant layers. Ceramic turbine nozzles were readied for planned field demonstrations with 473 hours of engine testing.

This work was funded as part of the Turbine Engine Technologies Program by the DoE Office of Transportation Technologies under Contract No. DE-AC02-96EE50454.

Commentary by Dr. Valentin Fuster
1998;():V003T06A053. doi:10.1115/98-GT-560.
FREE TO VIEW

The prevaporized, partially premixed, swirl-stabilized n-heptane flame in an atmospheric pressure combustor is investigated using a tuneable KrF excimer laser. Flashing the flame with a laser sheet tuned to the P2(8) line of OH (hydroxyl radical), single-shot images of the laser-induced signals are taken simultaneously with two ICCD-cameras aligned to the same measuring volume. One camera detects mainly the laser-induced fluorescence (LIF) from the 3→2 band of the OH plus signals from UHC (unburned hydrocarbons). Only broad-band emission from UHC is imaged onto the other camera. Comparing the two images, signals stemming from OH and UHC, respectively, can be distinguished. Pictures, taken in various planes along the main flow direction, reveal highly turbulent structures in the flame. High fluorescence signals from OH can obviously be found both in filament-like flame fronts lying between fresh combustible mixtures and hot combustion products as well as in broadened reaction regions.

Topics: Flames , Imaging , Heptane
Commentary by Dr. Valentin Fuster
1998;():V003T06A054. doi:10.1115/98-GT-568.
FREE TO VIEW

Flame emittance images phase-locked to acoustic fluctuations have been acquired to study flame structure variations in lean premixed gas-fueled swirl-stabilized combustor systems used in industrial gas turbines. Images have been acquired through a specially-designed optical probe that makes use of the limited optical access typical of these combustors. A number of observations appear to be consistent under widely varying operating conditions. While flame stabilization was anticipated in the central vortex and outer recirculation zone of the nozzle, no optical emissions have been observed in the outer region. Oscillations in the flame emittance appear to be primarily in intensity rather than structure suggesting that the major contributor to acoustic coupling may be equivalence ratio variations rather than the motions of large scale fluid structures.

Commentary by Dr. Valentin Fuster
1998;():V003T06A055. doi:10.1115/98-GT-576.
FREE TO VIEW

Combustion-zone stoichiometry and fuel-air premixing were actively controlled to optimize the combustor performance over a range of operating conditions. The objective was to maximize the combustion temperature, while maintaining NOx within a specified limit. The combustion system consisted of a premixer located coaxially near the inlet of a water-cooled shroud. The equivalence ratio was controlled by a variable-speed suction fan located downstream. The split between the premixing air and diffusion air was governed by the distance between the premixer and shroud. The combustor performance was characterized by a cost function evaluated from time-averaged measurements of NOx and oxygen concentrations in products. The cost function was minimized by the downhill simplex algorithm employing closed-loop feedback. Experiments were conducted at different fuel flow rates to demonstrate that the controller optimized the performance without prior knowledge of the combustor behavior.

Topics: Combustion
Commentary by Dr. Valentin Fuster
1998;():V003T06A056. doi:10.1115/98-GT-581.
FREE TO VIEW

A new analytical time lag flame model based on Bloxidge’s flame model was introduced for calculating combustion oscillation of premixed flame to take into account the distribution of heat release rate and flame speed which was calculated by analytical formulas dependent on pressure, temperature, fuel-to-air ratio and velocity. The transfer matrix technique using the new flame model was applied to the calculation of acoustic resonance. To verify the model, combustion oscillation experiments were performed for methane-air premixed flames stabilized by a swirl burner at elevated pressures between 0.6–0.9MPa. Fluctuating pressure had the maximum peak at the specific value of f. Here f is the frequency of resonance and τf is the passing time of premixed gas through flame length. The analysis could simulate the dependency of fuel-to-air ratio and static pressure for dynamic pressure local peak.

Commentary by Dr. Valentin Fuster
1998;():V003T06A057. doi:10.1115/98-GT-582.
FREE TO VIEW

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 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 are coupled via a four element transfer matrix. This approach is similar to the “black box” theory in communication engineering. To determine ail four coefficients, two independent test states have to be created. This is achieved by using acoustic sources upstream and downstream of the burner, respectively. 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 and to investigate the influence of various parameters such as acoustic amplitude and equivalence ratio. The treatment of burners as acoustic two-port with feedback used to model the thermoacoustic behavior of combustion chambers and the experimental determination of the burner transfer matrix is novel.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
1998;():V003T06A058. doi:10.1115/98-GT-596.
FREE TO VIEW

Lean premixed, prevaporized (LPP) high temperature combustor designs as explored for Advanced Subsonic Transport (AST) and High Speed Civil Transport (HSCT) combustors can achieve low NOx emission levels. An enabling device is needed to arrest flashback and inhibit autoignition at high power conditions and during transients (surge and rapid spool down). A novel flashback arrestor design has demonstrated the ability to arrest flashback and inhibit downstream flame holding in a 4.6 cm diameter tubular reactor at full power inlet temperatures (725 °C) using Jet-A fuel at 0.4 ≤ ϕ ≤ 3.5. Several low pressure loss (0.2% to 0.4% at 30 m/s) flashback arrestor designs were developed which arrested flashback at all of the test conditions. Flame holding was also inhibited off the flash arrestor face or within the downstream tube at low velocities (≤ 3 to 6 m/s), thus protecting the flashback arrestor and combustor premixer components. Upstream flow conditions influence the specific configuration based on using either a 45% or 76% upstream geometric blockage. Stationary, lean premixed dry low NOx gas turbine combustors would also benefit from this low pressure drop flashback arrestor design which can be easily integrated into new and existing designs.

Commentary by Dr. Valentin Fuster

Oil and Gas Applications

1998;():V003T07A001. doi:10.1115/98-GT-001.
FREE TO VIEW

Field testing of gas turbine compressor packages requires the accurate determination of efficiency, capacity, head, power and fuel flow in sometimes less than ideal working environments. Nonetheless, field test results have significant implication for the compressor and gas turbine manufacturers and their customers. Economic considerations demand that the performance and efficiency of an installation are verified to assure a project’s return on investment. Thus, for the compressor and gas turbine manufacturers, as well as for the end-user, an accurate determination of the field performance is of vital interest. This paper describes an analytic method to predict the measurement uncertainty and, thus, the accuracy, of field test results for gas turbine driven compressors. Namely, a method is presented which can be employed to verify the validity of field test performance results.

The equations governing the compressor and gas turbine performance uncertainties are rigorously derived and results are numerically compared to actual field test data. Typical field test measurement uncertainties are presented for different sets of instrumentation. Test parameters that correlate to the most significant influence on the performance uncertainties are identified and suggestions are provided on how to minimize their measurement errors. The effect of different equations of state on the calculated performance is also discussed.

Results show that compressor efficiency uncertainties can be unacceptably high when some basic rules for accurate testing are violated. However, by following some simple measurement rules and maintaining commonality of the gas equations of state, the overall compressor package performance measurement uncertainty can be limited and meaningful results can be achieved.

Commentary by Dr. Valentin Fuster
1998;():V003T07A002. doi:10.1115/98-GT-053.
FREE TO VIEW

This paper investigates the effects of multistage centrifugal compressor fouling on the performance of machine and plant operation. A thermodynamic model for the compressor has been developed for studying in-situ compressor performance deterioration due to internal component fouling at high power settings. The four casings, six sections compressor train is one of the critical process loop components in meeting ethylene production targets in a large scale petrochemical complex. The compressor handles process gas which is a complex mixtures of cracked gases (17 constituents) containing measurable concentrations of high molecular weight hydrocarbons. The model requires minimum inputs (pressure, temperature, flow, speed) from readily available instrument sensors and derives all the necessary information that predicts overall performance. Various parameters characterizing multistage compressor health condition were investigated. The study revealed that fouling of early stages adversely affect the polytropic efficiency of later stages. The model is cost effective in operation and is fully capable of detecting small changes occurring periodically in the compressor. A simple predictive monitoring scheme was developed for assessing the degree of compressor fouling that will strongly help engineers responsible for decision making.

Topics: Compressors
Commentary by Dr. Valentin Fuster
1998;():V003T07A003. doi:10.1115/98-GT-068.
FREE TO VIEW

This paper reports an experimental investigation on centrifugal compressor surge. The compression system consists of a four-stage blower with vaned diffusers and a large plenum discharging into the atmosphere through a throttle valve. Measurements of unsteady pressure and flow rate in the plant, and of instantaneous velocity in the diffusers of the first and fourth compressor stage are performed during deep surge, at several valve settings and three different rotation speeds. Additional tests have been carried out on a different system configuration, i.e., without plenum, in order to obtain the steady-state compressor characteristics and to collect reference data on stall in surge-free conditions. In this configuration, a fully developed rotating stall was detected in the compressor diffusers, while during surge it affects only a limited part of the surge cycle.

The goal of the present experimental work was to get a deeper insight into unstable operating conditions of multi-stage centrifugal compressors and to validate a theoretical model of the system instability to be used for the design of dynamic control systems.

Topics: Compressors , Surges
Commentary by Dr. Valentin Fuster
1998;():V003T07A004. doi:10.1115/98-GT-069.
FREE TO VIEW

This paper describes, from a theoretical point of view, the behaviour of compression systems during surge and the effect of passive and active control devices on the instability limit of the system.

A lumped parameter model is used to simulate the compression system described in Part I of this work (Arnulfi et al., 1998), based on an industrial multistage centrifugal compressor. A comparison with experimental results shows that the model is accurate enough to describe quantitatively all the features of the phenomenon.

A moveable wall control system is studied in order to suppress surge in the compressor. Passive and active control schemes are analysed; they both address directly the dynamic behaviour of the compression system to displace the surge line to lower flow rates. The influence of system geometry and compressor speed is investigated; the optimum values of the control parameters and the corresponding increase in the extent of the stable operating range are presented in the paper.

Commentary by Dr. Valentin Fuster
1998;():V003T07A005. doi:10.1115/98-GT-083.
FREE TO VIEW

The Norwegian Offshore Market has shown itself to be a market leader in introducing new requirements and technological innovations in the gas turbine packages used on new installations.

This paper will review the changes that Altair Norge A/S has been a party to in the areas of “acoustics”, “reduction in pressure losses”, “reductions in weight”,” and an overall effort to reduce the “costs”.

The systems effected by the above changes are Inlets, Ventilation, Enclosures, and Exhausts. The time span of this review is from 1981 to the present; with the emphasis on the last 5 years.

It will be shown how the requirements generated by the authorities to protect the employees offshore generate benefits technologically.

Commentary by Dr. Valentin Fuster
1998;():V003T07A006. doi:10.1115/98-GT-270.
FREE TO VIEW

A one dimensional model based on fundamental principles of gas turbine thermodynamics and combustion processes was constructed to quantify the principle of exhaust gas recirculation (EGR) for NOx reduction. The model utilizes the commercial process simulation software ASPEN PLUS®. Employing a set of 8 reactions including the Zeldovich mechanism, the model predicted thermal NOx formation as function of amount of recirculation and the degree of recirculate cooling. Results show that addition of sufficient quantities of uncooled recirculate to the inlet air (i.e. EGR>∼4%) could significantly decrease NOx emissions but at a cost of lower thermal efficiency and specific work. Cooling the recirculate also reduced NOx at lower quantities of recirculation. This has also the benefit of decreasing losses in the thermal efficiency and in the specific work output. Comparison of a ‘rubber’ and ‘non-rubber’ gas turbine confirmed that residence time is one important factor in NOx formation.

Commentary by Dr. Valentin Fuster
1998;():V003T07A007. doi:10.1115/98-GT-271.
FREE TO VIEW

To be suitable for gas turbine use, renewable energy resources are typically processed to produce a cleaned gaseous fuel which tends to be of low energy density relative to natural gas. Depending on the process and source, the derived fuel gas may also contain a large fraction of Hydrogen. Such gases may require enrichment to permit stable combustion and also carry some hazard of flashback due to the Hydrogen content. Previous experience also suggests that even small traces of ash which could be carried through the gas cleaning have the potential to cause airfoil fouling.

Allison Engine Company produces engines for electric power generation in the range up to 7 MW and expects to offer derivatives up to 12 MW by the year 2000. Two series of engines are involved, offering different virtues and challenges in renewable fuels use. It is planned to extend the range of renewable fuels utilized by all these engines by carrying out a rig and engine demonstration program. A major goal is to operate a gas turbine using the product gas from a high quality biomass gasifier.

Topics: Fuels , Gas turbines
Commentary by Dr. Valentin Fuster
1998;():V003T07A008. doi:10.1115/98-GT-287.
FREE TO VIEW

Since the introduction of emission standards for gas turbines in the late ‘70s and early ‘80s, the gas turbine industry has responded with a variety of combustion and cleanup alternatives that have improved emissions. While the emissions were being reduced, the cost of control, and the negative environmental impacts were often significant.

Thanks to a technological breakthrough, catalytic combustion has now been achieved, and can fulfill the promise of low cost NOx elimination without the high cost of SCR or the operational problems associated with Lean Pre-Mix.

Commentary by Dr. Valentin Fuster
1998;():V003T07A009. doi:10.1115/98-GT-302.
FREE TO VIEW

Conventional engine shut-down systems for vibration protection of aero-derivative gas generators and power turbines are usually not appropriate for protecting the machines against component malfunction primary damage.

Consequences often seen are substantial secondary damages to the engine as a result of e.g. ballor roller bearing failure. Effects typically vary from various blade or vane damage, root fretting or blade rubbing, etc.

Conventional engine shut-down systems do not activate until a certain unbalance of energy is released. The nature of the signals related to bearing and bladepass are too small in energy, and lies in too high a frequency range to be monitored by conventional engine protection systems.

This paper highlights the importance of the early detection and trending of vibrations, and taking necessary actions to prevent secondary damage to the engine, and describes the experience from a dedicated vibration and acoustic analysis system designed for early warning of common faults.

Commentary by Dr. Valentin Fuster
1998;():V003T07A010. doi:10.1115/98-GT-314.
FREE TO VIEW

Natural gas transmission systems have many sources of fugitive methane emissions that have been difficult to eliminate. This paper discusses an option for dealing with one such source for operations using turbo-compressor units fitted with dry gas seals. Dry seals rely on a small leakage of process gas to maintain the differential pressure of the process against the atmosphere. The seal leakage ultimately results in waste gas that is emitted to the atmosphere through the primary vent. A simple, cost effective, emission disposal mechanism for this application is to vent the seal gas into the gas turbine’s air intake. Explosion hazards are not created by the resultant ultra-lean fuel/air mixture, and once this mixture reaches the combustion chamber, where sufficient fuel is added to create a flammable mixture, significant oxidation of the seal vent gas is realized. Background of the relevant processes is discussed as well as a review of field test data. Similar applications have been reported [1] for the more generalized purpose of Volatile Organic Compound (VOC) destruction using specialized gas turbine combustor designs. As described herein, existing production gas turbine combustors are quite effective at fugitive methane destruction without specialized combustor designs.

Commentary by Dr. Valentin Fuster
1998;():V003T07A011. doi:10.1115/98-GT-535.
FREE TO VIEW

This paper discusses the design and development program that is taking place to enable the availability in mid 1999 of a unit designated the Coberra 6761. This features the aircraft derivative Rolls-Royce RB211-24G upgrade gas generator and a new close coupled Cooper-Bessemer RT61 three stage power turbine. The paper describes the upgrade of the gas turbine from 28.4MW (38 000 SHP) to 31.8MW (42 600 SHP) ISO output power at over 40% thermal efficiency. Measures taken to maximize reliability and maintainability while reducing cost of ownership are described. The improvements in the gas generator compressors and turbines are detailed. The new design features of the power turbine are reviewed including a new support structure, modular service features and 3D orthogonal airfoil designs. The forthcoming validation program for the entire gas turbine unit is also discussed.

Commentary by Dr. Valentin Fuster
1998;():V003T07A012. doi:10.1115/98-GT-590.
FREE TO VIEW

Solar Turbines Incorporated has extended the size-range of its’ turbomachinery products with development of the Titan™ 130 industrial gas turbine. The 13 megawatt-class, simple-cycle machine is designed to produce 13.3 MW (17,800 hp) output power with a thermal efficiency of 34.5% at ISO inlet conditions with no losses. The larger gas turbine is intended to meet increasing market demands in gas compression and mechanical pump-drive industrial applications.

The overall engine design is based on aerodynamic-scale of the 7 MW-class Taurus™ 70 gas turbine with similar operating cycle parameters. The engine configuration consists of hardware that has been scaled from the Taurus 70 and components that are common with the Mars® gas turbine. As with the Taurus 70 and Mars products, the Titan 130 gas turbine features a low emissions combustion system based on Solar’s proven dry, lean-premix, pollution-prevention technology. The enhance system is capable of reducing pollutant emissions over an extended operating range from part-load to full-load conditions.

This paper discusses the evolutionary design of the Titan 130 from the Taurus 70 and Mars gas turbines products. Descriptions of the basic configuration, component scaling techniques and modular design construction are presented.

Commentary by Dr. Valentin Fuster

Cycle Innovations

1998;():V003T08A001. doi:10.1115/98-GT-033.
FREE TO VIEW

This paper addresses the use of the partial oxidation process within a gas turbine cycle. In a partial oxidation reaction fuel is oxidized in a sub-stoichiometric atmosphere at a pressure and a temperature similar to those in the combustion chamber of a conventional turbine. Following the sub-stoichiometric stage, oxidation of the fuel is completed in the final stage. The absence of excess air in the first stages makes it possible to reduce the work required by the compressor and to decrease NOx formation. To assess the partial oxidation concept, several gas turbine systems were simulated and optimized. The results of the comparison study and exergy analysis of the cycles considered are presented.

Commentary by Dr. Valentin Fuster
1998;():V003T08A002. doi:10.1115/98-GT-035.
FREE TO VIEW

Significant research effort is currently centered on developing advanced gas turbine systems for electric power generation applications. A number of innovative gas turbine cycles have been proposed lately, including the Humid Air Turbine (HAT), and the Chemically Recuperated Gas Turbine (CRGT). The potential of the CRGT cycle lies in the ability to generate power with a high efficiency while achieving ultra-low NO emissions without the need for selective catalytic reduction of the exhaust gases. However, much of the work that has been published on such cycles is restricted to a discussion of the thermodynamic potential of the cycle, and little work has focussed on discussion of some of the specific design issues associated with such a cycle.

More specifically, design of the chemical recuperation heat recovery device involves a complex design trade-off in order to achieve a design with acceptable hot and cold-side pressure drops and acceptable overall dimensions.

The design of such a heat recovery device is more complex than that of a traditional heat recovery steam generator (HRSG), since the methane steam reformer must not only allow sufficient heat transfer to occur, but also allow a sufficient cold side residence time, so that the methane steam reforming reactions can come close to equilibrium, ensuring maximal methane conversion. In this work, the authors present a code capable of performing the design of a methane steam reformer heat recovery device based on a heat exchanger geometry similar to that of a traditional HRSG. The purpose of the paper is to discuss the key parameters relevant to the design of a CRGT MSR reactor, and how these parameters interact with the rest of the cycle. Various design options are discussed, and the results of a parametric analysis are presented, leading to the identification of several suitable geometries.

Commentary by Dr. Valentin Fuster
1998;():V003T08A003. doi:10.1115/98-GT-036.
FREE TO VIEW

Significant research effort is currently centered on developing advanced gas turbine systems for electric power generation applications. A number of innovative gas turbine cycles have been proposed lately, including the Humid Air Turbine (HAT), and the Chemically Recuperated Gas Turbine (CRGT). The potential of the CRGT cycle lies in the ability to generate power with a high efficiency while achieving ultra-low NO emissions without the need for selective catalytic reduction of the exhaust gases. Much of the research work published on the CRGT cycle is restricted to an analysis of the thermodynamic potential of the cycle. However, a detailed performance analysis of such cycles requires the development of a suitable cycle simulation code, capable of simulating cycle operation at the design point and in part load conditions. In this paper, the authors present a modular code for complex gas turbine cycle simulations. The code includes a module for design and off-design simulation of the methane-steam reformer chemical heat recovery device of a CRGT cycle. The code is then used to perform a detailed design and off-design performance analysis of a CRGT cycle based on the LM2500-STIG cycle adapted for chemical recuperation.

Topics: Design , Gas turbines , Cycles
Commentary by Dr. Valentin Fuster
1998;():V003T08A004. doi:10.1115/98-GT-037.
FREE TO VIEW

A number of different innovative gas turbine cycles are currently being developed for future high performance power generation applications. Of particular interest are new cycles such as the Humid Air Turbine (HAT) cycle and the Chemically Recuperated Gas Turbine (CRGT) cycle. The HAT cycle has reached the stage of commercial demonstration, whereas the CRGT cycle is still in the research stage.

Performance predictions for CRGT cycles show cycle performance levels higher than those of the basic Steam Injected (STIG) cycle, but lower than those of state-of-the-art combined cycles. Thus, CRGT cycles are most likely to be of interest for intermediate load operation or cogeneration. For such applications, part load performance is a key performance criterion. Part load operation is generally detrimental to power plant performance, and it is thus important to assess the part load performance of such a plant under various operating conditions. The emphasis of this paper is on the comparison of part load performance of various advanced cycles (STIG, CRGT, Combined Cycle). The results show that the part load performance levels of a CRGT cycle deteriorate less fast than those of a combined cycle, whereas part load behavior is similar to that of a STIG cycle. However, this study also indicates that CRGT part load performance is less attractive than published performance curves of Westinghouse’s innovative CHAT cycle, thus confirming that the key advantage of the CRGT cycle is the projected ultra-low emissions characteristics of the cycle.

Topics: Stress , Gas turbines , Cycles
Commentary by Dr. Valentin Fuster
1998;():V003T08A005. doi:10.1115/98-GT-057.
FREE TO VIEW

The power-speed requirements of warships and the poor part load efficiency of simple cycle gas turbines has given rise to the design of many ship installations where two types of gas turbines are used. A large type for high speed, at full power, and a small one for cruise. It is common to mount two units of each type. This design results in a large amount of bulky and heavy ducting, much more voluminous and heavy than the gas turbines themselves.

The present paper outlines an investigation into a novel intercooled split-cycle with some deck mounted components. This reduces the requirement for internal ducts in the ships hull, essentially, to those needed by the cruise engine.

The engine performance has been predicted and a comparison is carried out between a traditional installation and the one investigated. An estimate has been carried out of the flow conditions of the duct to assess the change in losses for operation in the cruise and the full power condition. The new scheme appears to be promising.

Commentary by Dr. Valentin Fuster
1998;():V003T08A006. doi:10.1115/98-GT-058.
FREE TO VIEW

It is well known that a cycle performance can be improved considerably by adopting humid air to a simple gas turbine. Further improvement can be achieved by utilizing LNG (liquified natural gas) cold energy which is obtained during vaporization process of natural gas from liquid to gas state. Qualitatively well known fact of high specific power and improvement of efficiency are analyzed quantitatively for various cases. These include comparisons of power, efficiency and other important operating parameters for the cases of a simple cycle and HAT cycle with and without utilization of LNG cold energy. Compared with simple cycle, HAT cycle got 48% increase in total work, 16% increase in efficiency and HAT-LNG cycle got each 54%, 17% increases at 10 pressure ratio.

An analysis shows that a reasonable matching exists between the amount of LNG as fuel and the energy required to control inlet air temperature. It should be also admitted that use of a high cost liquified natural gas is inevitable for transportation of fuel from production site to consumer.

Commentary by Dr. Valentin Fuster
1998;():V003T08A007. doi:10.1115/98-GT-059.
FREE TO VIEW

The next civil supersonic aircraft project will pose a number of challenges. The propulsion system for this aircraft will have to achieve economic operation for both supersonic and subsonic regime while meeting the intended noise and pollutant emissions regulation. Whilst there are a number of proposed engines for the next generation civil supersonic aircraft, they all exhibit difficulties inherent in the engine duty. The present paper offers a simple solution based on retractable fans. The key for the success of the concept is the single stage double pass tip turbine that drives the fan. Characteristics of this unique turbine such as extra-high power output at reasonable efficiencies and low metal temperatures along with some performance aspects of the power plant are discussed. Although further investigation is still required, the performance of the system merits consideration. The work was undertaken by one of the authors as part of his MSc research project.

Commentary by Dr. Valentin Fuster
1998;():V003T08A008. doi:10.1115/98-GT-060.
FREE TO VIEW

Since the humid air turbine (HAT) cycle was first presented by Rao and Joiner (1990), several modifications were proposed to the original configuration to further improve its efficiency. In the last years, the attention was focused in the water recovery from flue gas and in determining the most suitable systems to separate water from gas and solving the problem of low temperature at the stack.

In all the above studies it was shown that condensing water from flue gas requires a significant flow rate of a cooling medium (generally water) which is needed to remove condensation heat which must then be disposed in the environment. This worsens power plant performance because large cooling towers are needed. On the other hand, the reduced cost of water treatment may compensate the additional costs of the condensation equipment.

In this paper, the introduction of an Organic Rankine Cycle (ORC), which transforms in mechanical power a fraction of the heat recovered from the HAT cycle, both in the water recovery system and in other heat exchangers, is presented. Results were obtained by using three different fluids and maximizing the ORC input exergy. The substances which were used are the conventional R502 refrigerant fluid, ammonia and the new HF134a, which is replacing phased-out CFCs in refrigeration systems.

Commentary by Dr. Valentin Fuster
1998;():V003T08A009. doi:10.1115/98-GT-142.
FREE TO VIEW

A variable cycle engine (VCE) is a very complex concept. Extensive variable geometry features are required to ensure safe handling characteristics in the two or more operating modes that are envisaged. These may include variable compressor staters, variable nozzle areas and variable bypass valves.

In this paper a method for assessing the optimum setting of control variables is described. The starting point of the analysis is one of the operating modes of the engine, and an optimisation is carried out to decide the new values of the control variables within that operating mode or when changing to another.

The sensitivity of the cycle, in the initial mode, is assessed and an influence coefficient vector is created. This is then employed to decide what setting changes should be implemented to the control variables to achieve the desired result, respecting operating constraints such as surge margins, gas temperatures and rotor speeds.

The engine selected for analysis is the selective bleed engine, although the method could be used for other engine geometries after suitable adaptation. There are two very useful features of this method. The first is that it can give a view of the variable cycle capabilities of an engine with variable geometry features. The second is that it identifies the constraints, enabling the designer to focus the development effort where it is most needed.

Commentary by Dr. Valentin Fuster
1998;():V003T08A010. doi:10.1115/98-GT-143.
FREE TO VIEW

This paper describes the current status and anticipated improvements in the Cascaded Humidified Advanced Turbine (CHAT) technology. With these improvements, CHAT will provide an alternative approach to achieving the goals of the Department of Energy’s Advanced Turbine Systems (ATS) project of a 60% thermal cycle efficiency for natural gas fired combined cycle power plants.

The traditional approach to increasing the efficiency of simple cycle and combined cycle power plants was to increase the firing temperature and pressure of the basic Brayton cycle. However, every increase in the CT firing temperature required progressively higher development cost, associated with new materials and cooling techniques, and increased NOx control challenges.

CHAT is a gas turbine based cycle with intercooling, reheat, recuperation and humidification. It is based upon the integration of an existing design heavy duty combustion turbine with an additional high pressure shaft comprised of industrial compressors and expander. The current CHAT plant design includes an HP expander inlet temperature of 871 C (1600 F), which represents the level of the combustion turbine technology of the late 1960’s – early 1970’s. The expander on the power generation shaft (LP) is based upon current combustion turbine technology with turbine inlet temperature of 1400 C (2550 F). One of the most effective ways to increase the CHAT plant efficiency is to increase the HP expander inlet temperature. By increasing this temperature to a relatively low 1150 C (2100 F), and maintaining the current inlet temperatures on the LP expander, the CHAT plant could achieve the ATS program target efficiency.

The paper presents the current CHAT plant’s performance and cost characteristics, and the initial findings of a project co-sponsored by the Electric Power Research Institute (EPRI) and Energy Storage and Power Consultants, (ESPC) for the development of an HP expander with increased inlet temperatures.

Topics: Combustion , Turbines
Commentary by Dr. Valentin Fuster
1998;():V003T08A011. doi:10.1115/98-GT-144.
FREE TO VIEW

Two different power plant configurations based on a Semi-Closed Gas Turbine (SCGT) are analyzed and compared in terms of First and Second Law analysis. SCGT plant configurations allow the application of CO2 separation techniques to gas-turbine based plants and several further potential advantages with respect to present, open-cycle solutions. The first configuration is a second-generation SCGT/CC (Combined Cycle) plant, which includes inter-cooling (IC) between the two compression stages, achieved using spray injection of water condensed in a separation process removing vapor from the flue gases. The second configuration (SCGT/RE) combines compressor inter-cooling with the suppression of the heat recovery steam generator and of the whole bottoming cycle; the heat at gas turbine exhaust is directly used for gas turbine regeneration.

The SCGT/CC-IC solution provides good efficiency (about 55%) and specific power output figures, on account of the spray inter-cooling; however, with this configuration the cycle is not able to self-sustain the CO2 removal reactions and amine regeneration process, and needs a substantial external heat input for this purpose.

The SCGT/RE solution is mainly attractive from the environmental point of view: in fact, it combines the performance of an advanced gas turbine regenerative cycle (efficiency of about 49%) with the possibility of a self-sustained CO2 removal process. Moreover, the cycle configuration is simplified because the HRSG and the whole bottoming cycle are suppressed, and a potential is left for cogeneration of heat and power.

Commentary by Dr. Valentin Fuster
1998;():V003T08A012. doi:10.1115/98-GT-165.
FREE TO VIEW

The application of recuperators in advanced thermodynamic cycles is growing due to stronger demands of low emissions of pollutants and the necessity of improving the cycle efficiency of power plants to reduce the fuel consumption.

This paper covers applications and types of heat exchangers used in gas turbine units. The trends of research and development are brought up and the future need for research and development is discussed. Material aspects are covered to some extent.

Attempts to achieve compact heat exchangers for these applications are also discussed. With the increasing pressure ratio in the gas turbine cycle, large pressure differences between the hot and cold sides exist. This has to be accounted for.

The applicability of CFD (Computational Fluid Dynamics) is discussed and a CFD–approach is presented for a specific recuperator. This recuperator has narrow wavy ducts with complex cross-sections and the hydraulic diameter is so small that laminar flow prevails. The thermal-hydraulic performance is of major concern.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
1998;():V003T08A013. doi:10.1115/98-GT-166.
FREE TO VIEW

The CHAT (Cascade Humid Air Turbine) cycle introduction has recently been proposed for a more simple and profitable application of the humid air turbine. The very interesting performance announced for this plant has been evaluated in this paper, particular attention is devoted to the multi-stage evaporation process and its thermodynamic limits.

A detailed thermodynamic analysis of the most important cycle parameters, like various pressure levels, fire temperatures and blade coolant bleeding can permit the evaluation of better plant performances. The results show a substantial agreement with other published data and they confirm the good efficiency and high specific power of the CHAT cycle.

Considering the proposed compressor and turbine for the CHAT plant an off design simulation of the plant is also realized to estimate the real behaviour of turbomachinery components.

Moreover this study is based on ESMS code already developed by the authors and the new components model (thermodynamic, design and off-design simulation) introduced for this work are presented.

Topics: Design , Cycles
Commentary by Dr. Valentin Fuster
1998;():V003T08A014. doi:10.1115/98-GT-167.
FREE TO VIEW

The paper presents the first results of a research, in progress at the Istituto di Macchine e Sistemi Energetici of Genova University, about a gas-steam combined cycle with indirect fired gas-turbine.

The study has dealt with the gas turbine plant which is the most innovative part of the process. To select the design point of every component, a thermodynamic parametric analysis over a large range of pressure ratios and a large range of mass flow ratio through the two characteristic by-passes has been developed.

The off-design performances of the gas plant have been investigated and finally a proposal of control system has been made, tested using a simple mathematical model, able to analyse in the time domain the dynamic behaviour of the controlled gas plant.

Commentary by Dr. Valentin Fuster
1998;():V003T08A015. doi:10.1115/98-GT-290.
FREE TO VIEW

The present paper describes the application of the exergy method to several power cycles of current interest in the Brazilian power scene. The ability of the exergy method to highlight component irreversibilities is of particular interest in this investigation.

In this paper the exergy analysis for a simple gas turbine cycle, a combined gas/steam cycle, a combined gas/steam/freon cycle and a chemically recuperated gas turbine have been performed.

As a yardstick for comparison a standard gas turbine engine with and without a steam bottoming cycle has been employed. The fuel considered is natural gas. The analysis of this system has been carried out using the exergy method.

For the simple gas turbine cycle a biomass fuel has been employed as an alternative. The attraction of this fuel is its low impact on the environment and its plentiful supply in many regions in Brazil.

Commentary by Dr. Valentin Fuster
1998;():V003T08A016. doi:10.1115/98-GT-345.
FREE TO VIEW

The effects of three types of cooling systems on the calculated operating performances of a hydrogen-fueled thermal power plant with a 1,700°C-class gas turbine were studied with the goal of attaining a thermal efficiency of greater than 60%. The combination of a closed-circuit water cooling system for the nozzle blades and a steam cooling system for the rotor blades was found to be the most efficient, since it eliminated the penalties of a conventional open-circuit cooling system which ejects coolant into the main hot gas stream.

Based on the results, the water cooled first-stage nozzle blade and the steam cooled first-stage rotor blade were designed. The former features array of circular cooling holes close to the surface and uses a copper alloy taking advantage of recent coating technologies such as thermal barrier coatings (TBCs) and metal coatings to decrease the temperature and protect the blade core material. The later has cooling by serpentine cooling passages with V-shaped staggered turbulence promoter ribs which intensify the internal cooling.

Commentary by Dr. Valentin Fuster
1998;():V003T08A017. doi:10.1115/98-GT-352.
FREE TO VIEW

The effect of heat exchanger effectiveness on cycle efficiency is well known. But the relationship between exergy loss in heat exchangers and the effectiveness is less well documented. In this paper the relationship is explored; it is shown how the exergy loss in the heat exchanger is changed as effectiveness is altered. It is also shown how the exergy losses in other components of the recuperative gas turbine cycle are changed, together with the overall cycle performance, as the effectiveness is varied.

Commentary by Dr. Valentin Fuster
1998;():V003T08A018. doi:10.1115/98-GT-353.
FREE TO VIEW

Static process calculations done by various institutions have shown competitive thermal efficiencies for the evaporative cycle, also known as the HAT-cycle, when compared to the combined gas and steam cycle. The cycle seems also attractive from other standpoints such as lower investment costs and excellent part load performance. With this background several companies and organisations in Sweden have jointly started an evaporative gas turbine project.

The participating organisations agreed to design and erect a pilot plant at the Lund Institute of Technology. In co-operation with the Royal Institute of Technology in Stockholm various tests will be performed with the overall goal to verify efficiency improvements and functional performance. The test schedule started with the gas turbine in dry stand alone operation. From this initial phase the process will be developed, first step is the introduction of a recuperator. The final process includes a water circuit and condensing of water vapour in exhaust gases for recycling to the humidifier.

This paper presents a dynamic model of the gas turbine unit with and without the recuperator. Algebraic models are used for the gas turbine components combined with conservation equations in differential forms for volumes. The approximation used for volumes is the lumped parameter model meaning that properties within the volume are uniform. Heat flow between gas and solid structure is included in the recuperator model but is elsewhere neglected. A comparison between measurements and calculations is presented for the gas turbine without recuperator.

Commentary by Dr. Valentin Fuster
1998;():V003T08A019. doi:10.1115/98-GT-383.
FREE TO VIEW

In this paper, a novel technology based on the zero CO2 emission MATIANT (contraction of the names of the 2 designers: MATHIEU and IANTOVSKI) cycle is presented. This latter is basically a gas cycle and consists of a supercritical CO2 Rankine-like cycle on top of regenerative CO2 Brayton cycle. CO2 is the working fluid and O2 is the fuel oxidizer in the combustion chambers. The cycle uses the highest temperatures and pressures compatible with the most advanced materials in the steam and gas turbines. In addition, a reheat and a staged compression with intercooling are used. Therefore the optimized cycle efficiency rises up to around 45% when operating on natural gas. A big asset of the system is its ability to remove the CO2 produced in the combustion process in liquid state and at high pressure, making it ready for transportation, for reuse or for final storage. The assets of the cycle are mentioned. The technical issues for the predesign of a prototype plant are reviewed.

Topics: Cycles , Emissions
Commentary by Dr. Valentin Fuster
1998;():V003T08A020. doi:10.1115/98-GT-384.
FREE TO VIEW

This paper analyzes the fundamentals of IGCC power plants where carbon dioxide produced by syngas combustion can be removed, liquefied and eventually disposed, to limit the environmental problems due to the “greenhouse effect”. To achieve this goal, a semiclosed-loop gas turbine cycle using an highly-enriched CO2 mixture as working fluid was adopted. As the oxidizer, syngas combustion utilizes oxygen produced by an air separation unit. Combustion gases mainly consist of CO2 and H2O: after expansion, heat recovery and water condensation, a part of the exhausts, highly concentrated in CO2, can be easily extracted, compressed and liquefied for storage or disposal.

A detailed discussion about the configuration and the thermodynamic performance of these plants is the aim of the paper. Proper attention was paid to: (i) the modelization of the gasification section and of its integration with the power cycle, (ii) the optimization of the pressure ratio due the change of the cycle working fluid, (iii) the calculation of the power consumption of the “auxiliary” equipment, including the compression train of the separated CO2 and the air separation unit. The resulting overall efficiency is in the 38–39% range, with status-of-the-art gas turbine technology, but resorting to a substantially higher pressure ratio. The extent of modifications to the gas turbine engine, with respect to commercial units, was therefore discussed. Relevant modifications are needed, but not involving changes in the technology.

A second plant scheme will be considered in the second part of the paper, using air for syngas combustion and a physical absorption process to separate CO2 from nitrogen-rich exhausts. A comparison between the two options will be addressed there.

Commentary by Dr. Valentin Fuster
1998;():V003T08A021. doi:10.1115/98-GT-385.
FREE TO VIEW

This paper analyzes the fundamentals of IGCC power plants with carbon dioxide removal systems, by a cycle configuration alternative to the one discussed in Part A (with Oxygen-Blown Combustion). The idea behind this proposal is to overcome the major drawbacks of the previous solution (large oxygen consumption and re-design of the gas turbine unit), by means of a semiclosed cycle using air as the oxidizer. Consequently, combustion gases are largely diluted by nitrogen and cannot be simply compressed to produce liquefied CO2 for storage or disposal. However, CO2 concentration remains high enough to make separation possible by a physical absorption process. It requires a re-pressurization of the flow subtracted from the cycle, with relevant consequences on the plant energy balance.

The configuration and the thermodynamic performance of this plant concept are extensively addressed in the paper. As in the first part, the influence of the pressure ratio is discussed, but values similar to the ones adopted in commercial heavy-duty machines provide here acceptable performance. Proper attention was paid to the impact of the absorption process on the energy consumption. The resulting net overall efficiency is again in the 38–39% range, with assumptions fully comparable to the ones of Part A. Finally, we demonstrated that the present scheme enables the use of unmodified machines, but large additional equipment is required for exhausts treatment and CO2 separation. A final comparison between the two semiclosed cycle concepts was therefore addressed.

Commentary by Dr. Valentin Fuster
1998;():V003T08A022. doi:10.1115/98-GT-386.
FREE TO VIEW

The design-point performance of various gas turbine cycles such as simple, regenerative, and intercooled-regenerative, is well understood. It is also understood that more-elaborate shaft arrangements such as one, two or three concentric or non-concentric shafts, and a separate power turbine shaft, provide better starting and operational flexibility, and wider plateaus of high off-design performance. However, the types of different off-design performance one can obtain with these different shaft arrangements has not been previously reported. In this paper we use a computer program to investigate the design-point and off-design-point performance of engines with: one single shaft joining the compressor, turbine and load; one shaft joining compressor and turbine, and one shaft for the power turbine; two shafts for compressor and turbine, and one shaft for the power turbine; and three shafts joining the compressor and turbine, and one shaft for the power turbine. This is done by specifying typical compressor and turbine maps, and computing various aspects of off-design performance. The advantages and disadvantages of the various arrangements ore investigated and discussed.

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