During the prototype shop tests, the Model V84.3A ring combustor gas turbine unexpectedly exhibited a noticeable “humming” caused by self-excited flame vibrations in the combustion chamber for certain operating conditions. The amplitudes of the pressure fluctuations in the combustor were unusually high when compared to the previous experience with silo combustor machines. As part of the optimization program, the humming was investigated and analyzed. To date, combustion instabilities in real, complex combustors cannot be predicted analytically during the design phase. Therefore, and as a preventive measure against future surprises by “humming,” a feedback system was developed which counteracts combustion instabilities by modulation of the fuel flow rate with rapid valves (active instability control, AIC). The AIC achieved a reduction of combustion-induced pressure amplitudes by 86 percent. The Combustion instability in the Model V84.3A gas turbine was eliminated by changes of the combustor design. Therefore, the AIC is not required for the operation of customer gas turbines.

During the prototype shop tests, the Model V84.3A ring combustor gas turbine unexpectedly exhibited a noticeable “humming” caused by self-excited flame vibrations in the combustion chamber for certain operating conditions. The amplitudes of the pressure fluctuations in the combustor were unusually high when compared to the previous experience with silo combustor machines. As part of the optimization program, the humming was investigated and analyzed. To date, combustion instabilities in real, complex combustors cannot be predicted analytically during the design phase. Therefore, and as a preventive measure against future surprises by “humming”, a feedback system was developed which counteracts combustion instabilities by modulation of the fuel flow rate with rapid valves (Active Instability Control, AIC). The AIC achieved a reduction of combustion-induced pressure amplitudes by 86%. The combustion instability in the Model V84.3A gas turbine was eliminated by changes of the combustor design. Therefore, the AIC is not required for the operation of customer gas-turbines.

No abstract

Catalytic combustion has been the subject of thorough research work for over two decades, mainly in the U.S. and Japan. However, severe material problems in the ceramic or metallic monolith prevented regular operation in most cases. Still, during these two decades, turbine inlet temperatures were raised remarkably, and lean premix combustors have become standard in stationary gas turbines. In view of these facts, a simple “monolith-in-tube” concept of a catalytic combustor was adapted for the use in high-temperature gas turbines. Its essential feature is the fact that a considerable portion of the homogeneous gas phase reaction is shifted to the thermal reactor, thus lowering the catalyst temperature. This is achieved by the employment of very short catalyst segments. The viability of this concept has been demonstrated for a variety of pure hydrocarbons, alcohols as well as common liquid fuels. Extensive experimental investigations of the atmospheric combustor led to the assessment of parameters such as reference velocity, fuel-to-air ratio, and fuel properties. The maximum combustor exit temperature was 1673 K with a corresponding catalyst temperature of less than 1300 K for diesel fuel. Boundary conditions were in all cases combustion efficiency (over 99.9 percent) and pressure loss (less than 6 percent). Additionally, a model has been developed to predict the characteristic values of the catalytic combustor such as necessary catalyst length, combustor volume, and emission characteristics. The homogeneous reaction in the thermal reactor can be calculated by a one-dimensional reacting flow model.

Catalytic combustion has been the subject of thorough research work for over two decades, mainly in the U.S. and Japan. However, severe material problems in the ceramic or metallic monolith prevented regular operation in most cases. Still, during these two decades, turbine inlet temperatures were raised remarkably, and lean premix combustors have become standard in stationary gas turbines. In view of these facts, a simple “monolith-in-tube” concept of a catalytic combustor was adapted for the use in high-temperature gas turbines. Its essential feature is the fact that a considerable portion of the homogeneous gas phase reaction is shifted to the thermal reactor, thus lowering the catalyst temperature. This is achieved by the employment of very short catalyst segments. The viability of this concept has been demonstrated for a variety of pure hydrocarbons, alcohols as well as common liquid fuels. Extensive experimental investigations of the atmospheric combustor lead to the assessment of parameters such as reference velocity, fuel-to-air ratio and fuel properties. The maximum combustor exit temperature was 1,673 K with a corresponding catalyst temperature of less than 1,300 K for Diesel fuel. Boundary conditions were in all cases combustion efficiency (over 99.9%) and pressure loss (less than 6%). Additionally, a model has been developped to predict the characteristic values of the catalytic combustor such as necessary catalyst length, combustor volume and emission characteristics. The homogeneous reaction in the thermal reactor can be calculated by a one-dimensional reacting flow model.

A catalytic combustor concept with short catalyst segments and a thermal reactor is investigated with regard to NO, production of this concept under high-temperature conditions. The maximum combustor exit temperature was more than 1800 K with catalyst temperatures below 1300 K. For combustion of iso-octane, NO emissions of 4 ppm (day. 15% 02) at a flame temperature of 1800 K were measured. No significant influence of catalyst length, reference velocity and overall residence time on NO" emissions was observed.Additionally, the test combustor was fuelled with commercial diesel and kerosene (Jet-A). In this case, NO emissions were noticeable higher due to fuel-bound nitrogen. The emissions measured were for diesel, 12 ppm, and for kerosene, 7 ppm, (each dry, 15% 02), again at a flame temperature of 1800 K. To evaluate the conversion ratio of fuel-bound nitrogen to NO,, isooctane was doped with various amounts of ammonia and metyhlamine. The conversion rates were 70 to 90%, with a slight tendency to lower values (50%) for nitrogen mass fractions above 0.1%.Considering the NO emission level of actual premix burners, the lower emission value of the presented catalytic combustor results from a perfect premixed plug-flow combustion system incorporating a catalyst herein and not from a specific advantage of the principle of catalytic combustion itself. Again similar to a premix-combustor are the NO emission characteristics in the case of lean combustion of nitrogen bound fuels, which yield very high conversion rates.

Experimental investigations on shock-induced flutter in a linear transonic turbine cascade are presented. To examine the relation between trailing edge shock oscillations on adjacent blades in transonic flow and observed turbine blade vibrations, an elastic suspension system has been developed so that only aerodynamic coupling occurs in the system. The experimental investigations have been performed on a linear test rig with superheated steam as working fluid. The test facility enables Mach and Reynolds numbers to be varied independently. The investigated cascade consists of seven blades which are taken from the tip section of a transonic low pressure steam turbine blade. Each blade is attached by an elastic spring system which allows the respective blade to vibrate in a mode equal to the real blade’s first bending mode. By varying the individual spring stiffness it is possible to either get a tuned or mistuned cascade. The examinations mainly deal with the oscillatory behavior of the blades with respect to a variation in the isentropic outlet Mach number. In addition, the complex shock-boundary-layer interaction on the blades’ suction sides is described. An important result is that the maximum blade oscillation amplitude can be related to a specific outlet Mach number. At this Mach number all seven blades are vibrating with exactly the same frequency. This phenomenon is observed at both the tuned and the mistuned cascades. Spectrum analysis shows that one of the major shock oscillation frequencies corresponds to the flutter frequency. In addition to this frequency the spectrum analysis of the blade oscillation shows the dominant frequencies of the shock oscillation which are not natural blade frequencies. The experimental results show that oscillating shocks in a linear cascade give high potential for aeroelastic excitation of transonic blades under certain flow conditions. Blade oscillations and shock characteristics are discussed in detail.

The gap flow across the tips of cooled rotor blades of combustion turbines is of crucial importance for the thermal load of the tip section and the stage efficiency. Various blade tip section designs with different cooling concepts were tested in a two dimensional steam cascade test rig with superheated steam as working fluid. The blade tip surface temperature distributions were measured with a thermography system and the gap flow was determined by 105 static wall pressure measurements at the opposite wall. The experimental program included systematic variations of decisive flow parameters like Mach and Reynolds numbers of the main and leakage flow as well as variations of the gap width in a range of 1.6% to 4.8% of the chord length. The performance of a simple flat blade tip served as a baseline for the comparison with advanced grooved tip designs developed in earlier studies. In addition attention was given to a systematic investigation of the influence of the coolant mass flow ejected out of the blade into the grooved tip section. The paper presents an overview of the experimental program and results as well as a discussion of the influence of the main flow and geometry parameters on the leakage flow characteristics and the accompanying thermal load of the tip section.