0

Industrial and Cogeneration

2005;():1-8. doi:10.1115/GT2005-68014.

Methods of compressor performance maintenance for large utility combustion turbines continue to evolve. On-line water wash systems used to recover performance loss due to fouling are evolving that use less water. This paper derives a water wash model based on a thin film of water covering the airfoil surfaces. The economic potential for recovering “unrecoverable” losses due to increased roughness and erosion is evaluated. As an outage is needed to remove the compressor cover and perform the maintenance, the approach is to identify the most beneficial maintenance actions and an optimal maintenance interval.

Commentary by Dr. Valentin Fuster
2005;():9-18. doi:10.1115/GT2005-68026.

In the paper, Neural Network (NN) models for gas turbine diagnostics are studied and developed. The analyses carried out are aimed at the selection of the most appropriate NN structure for gas turbine diagnostics, in terms of computational time of the NN training phase, accuracy and robustness with respect to measurement uncertainty. In particular, feed-forward NNs with a single hidden layer trained by using a back-propagation learning algorithm are considered and tested. Moreover, Multi-Input/Multi-Output NN architectures (i.e. NNs calculating all the system outputs) are compared to Multi-Input/Single-Output NNs, each of them calculating a single output of the system. The results obtained show that NNs are robust with respect to measurement uncertainty, if a sufficient number of training patterns are used. Moreover, Multi-Input/Multi-Output NNs trained with data corrupted with measurement errors seem to be the best compromise between the computational time required for NN training phase and the NN accuracy in performing gas turbine diagnostics.

Commentary by Dr. Valentin Fuster
2005;():19-29. doi:10.1115/GT2005-68027.

In the paper, Neuro-Fuzzy Systems (NFSs) for gas turbine diagnostics are studied and developed. The same procedure used previously for the set up of Neural Network (NN) models was used. In particular, the same database of patterns was used for both training and testing the NFSs. This database was obtained by running a Cycle Program, calibrated on a 255 MW single shaft gas turbine working in the ENEL combined cycle power plant of La Spezia (Italy). The database contains the variations of the Health Indices (which are the characteristic parameters that are indices of gas turbine health state, such as efficiencies and characteristic flow passage areas of compressor and turbine) and the corresponding variations of the measured quantities with respect to the values in new and clean conditions. The analyses carried out are aimed at the selection of the most appropriate NFS structure for gas turbine diagnostics, in terms of computational time of the NFS training phase, accuracy and robustness towards measurement uncertainty during simulations. In particular, Adaptive Neuro-Fuzzy Inference System (ANFIS) architectures were considered and tested, and their performance was compared to that obtainable by using the NN models. An analysis was also performed in order to identify the most significant ANFIS inputs. The results obtained show that ANFISs are robust with respect to measurement uncertainty, and, in all the cases analyzed, the performance (in terms of accuracy during simulations and time spent for the training phase) proved to be better than that obtainable by MIMO and MISO Neural Networks trained and tested on the same data.

Commentary by Dr. Valentin Fuster
2005;():31-38. doi:10.1115/GT2005-68120.

This paper presents a methodology of diagnostic investigations for gas turbines. The key feature is that the analysis is carried out in two modes: off-line and on-line. The first mode is performed periodically. It involves detailed measurements. Values obtained from measurements create the input data for further analysis. Health state of a gas turbine is then evaluated. The evaluation bases on calculation of several health state parameters. The on-line diagnostic mode uses these parameters as a reference state. The usual lack of measurements available in the on-line investigations creates the need for additional input data for the analysis. Therefore diagnostic investigations are supported by the results from the off-line mode. One of the main problems to be solved in diagnostic analysis is the appropriate modeling of gas turbine operation. An approach presented here regards the operation in various conditions, meaning also off-design operation.

Commentary by Dr. Valentin Fuster
2005;():39-46. doi:10.1115/GT2005-68184.

Operations have a key role to play in today’s highly competitive environment, as they must provide the utmost effectiveness in asset management practices to boost the Return on Capital Employed (ROCE) back to the asset Owners. Asset Management Effectiveness (AME) can be rooted in a certain number of Operator-controllable leverages, that this paper is aimed to show thru a simple model. The model ties the identified operational leverages to the ROCE, providing guidance for the Operators to communicate properly with the asset Owners for demonstrating the financial benefits of their operational practices. A certain number of advanced services available to the Operators are shown and described, that can support both Operations and asset Owners in their search for the highest financial returns from existing assets.

Topics: Turbomachinery
Commentary by Dr. Valentin Fuster
2005;():47-53. doi:10.1115/GT2005-68247.

The major challenges before the design engineers of a gas turbine plant and its variants are the enhancement of power output, substantial reduction in NOx emission and improvement in plant thermal efficiency. There are various possibilities to achieve these objectives and humid air gas turbine cycle power plant is one of them. The present study deals with the thermodynamic study of humid air gas turbine cycle power plants based on first law. Using the modeling and governing equations, the parametric study has been carried out. The results obtained will be helpful in designing the humid air gas turbines, which are used as peaking units. The comparison of performance of humid air gas turbine cycle shows that it is superior to basic gas turbine cycle but inferior and more complex to steam injected cycle.

Commentary by Dr. Valentin Fuster
2005;():55-62. doi:10.1115/GT2005-68250.

In this work, the effects of inlet air cooling by vapor compression refrigeration cycle and evaporative water-cooling system, the cooling of blade coolant air by fuel before entering the combustor and recuperation on the combined cycle power plant performance have been studied. The present results show substantial improvements in the value of specific work and plant efficiency in the presence of cooling of inlet air and blade coolant air compared to a system without such cooling effects. However, implementation of recuperation alone reduces plant efficiency and specific work but recuperation combined with cooling effects increases plant efficiency. Design engineers might find presented results useful in optimizing a combined cycle system.

Commentary by Dr. Valentin Fuster
2005;():63-70. doi:10.1115/GT2005-68266.

It is well established that sub-micron ambient aerosol contamination of the intake air can produce fouling of the gas turbine compressor and result in a reduction of power output. Application of electrospun nanofibers of 0.25 micron diameter to a conventional filter media substrate has been demonstrated to improve the efficiency of gas turbine intake filters to remove sub-micron contaminate. The benefits of nanofiber filtration have been proven through use in gas turbine intake air filtration and other industrial and defense filtration applications for over twenty years. Recent advancements in electrospun nanofiber media technology have increased the filtering efficiency of gas turbine intake filters, with minimal differences in filter element pressure loss. These advances have also improved the durability of nanofibers in high temperature and high humidity applications. This paper discusses the laboratory testing that demonstrates these performance and durability improvements. A comparative field test program demonstrates the capability of nanofiber filtration to significantly reduce the fouling of gas turbine compressors.

Commentary by Dr. Valentin Fuster
2005;():71-82. doi:10.1115/GT2005-68336.

Ambient temperature strongly influences gas turbine power output causing a reduction of between 0.50% to 0.90% for every 1°C of temperature rise. There is also a significant increase in the gas turbine heat rate as the ambient temperature rises, resulting in an increased operating cost. As the increase in power demand is usually coincident with high ambient temperature, power augmentation during the hot part of the day become important for independent power producers, cogenerators and electric utilities. Evaporative and overspray fogging are simple, proven and cost effective approaches for recovering lost gas turbine performance. A comprehensive review of the current understanding of the analytical and experimental and practical aspects of high-pressure inlet fogging technology is provided. A discussion of analytical and experimental results relating to droplet dynamics, factors affecting droplet size, and inlet configuration effects on inlet evaporative fogging are covered in this paper. Commonly used fogging nozzles are also described and experimental findings presented.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
2005;():83-90. doi:10.1115/GT2005-68337.

The strong influence of ambient temperature on the output and heat rate on a gas turbine has popularized the application of inlet fogging and overspray for power augmentation. One of the main advantages of overspray fogging is that it enhances power output as a result of decrease in compression work associated with the continuous evaporation of water within the compressor due to fog intercooling. A comprehensive review on the current understanding of the analytical and experimental aspects of overspray fogging technology as applied to gas turbines is presented in this paper.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
2005;():91-97. doi:10.1115/GT2005-68346.

Application of inlet air fogging to gas turbine engines for power augmentation, has become common practice, with more than a 1000 installations worldwide including a large number of advanced gas turbines. In this paper an experimental investigation and empirical analysis of key operating parameters on the performance of impaction pin nozzles will be investigated. To this date no such correlation is available for impaction pin nozzles, which are currently used in about 75% of this type of applications. The correlations are developed from a series of experiments conducted in a wind tunnel equipped with the Malvern Spraytec droplet size measurement system. The conducted analysis covered a wide range of the relevant parameters. Mainly the water flow rate from the nozzle orifice was (0.00126 1.s−1 to 0.00063 1.s−1 : 0.02 gpm to 0.1 gpm), the operating pressure was (34.5 bars to 204.1 bars: 500 psi to 3000 psi), the airflow velocity was (1.5 m.s−1 to 15.2 m.s−1 : 295 fpm to 3000 fpm), the distance between the nozzle orifice and the location of measurement was (0.0127m to 0.508 m: 0.5” to 20”). Other parameters such as the plume spray one angle and the surrounding ambient psychrometric conditions, which may affect the droplet size for impaction pin nozzles is also discussed.

Topics: Gas turbines , Nozzles
Commentary by Dr. Valentin Fuster
2005;():99-107. doi:10.1115/GT2005-68368.

High fogging (wet compression, spray inter-cooling) is a technology used for gas turbine (GT) power augmentation. By evaporative spray inter-cooling of the air during compression, which is the main physical effect associated with the HF, a 5–7% power boost of the GT (for each percent of injected water per mass of air) is achieved. HF of a gas turbine can be accomplished using different spray technologies. In this study three different, commercially available spray technologies — pressure-swirl, hot water injection and air-assisted atomization — are compared regarding both technical and economical benefits and risks. The comparison is based on droplet sizing results, system complexity, the feasibility of system integration into the GT’s control and plant operation concept, GT performance and operational and additional O&M costs. It is also known that high fogging carries certain risks to the safe operation of a GT, such as compressor blades erosion, reduction in compressor surge margin and cooling airflows. To minimize the negative impact of high fogging, it is therefore important to select the most appropriate high fogging system as well as to provide for its full engine integration.

Commentary by Dr. Valentin Fuster
2005;():109-118. doi:10.1115/GT2005-68378.

This Paper contains technical and procedural information related to the overhauling of axial compressor being integral part of GE-built Frame No. 7, model EA CGT. Four CGT-driven electric generators are in operation in Watson Cogeneration Company (WCC), Carson, California, USA. CGT units were commissioned during fall of 1987 and winter of 1988 and have been operating under the base load since. Original overhauls of CGT rotors were performed by GE. Later on a decision was made to include third Parties capable to overhaul axial compressor and GT rotors. Overhaul Specifications (1) were generated in 1998 and were included into the Bid Packages. These specifications were intended to be used as a guide line only. The particular axial compressor overhaul efforts described in this Paper were the third since the commissioning. Step-by-step overhaul efforts of axial compressor assembly are described in order to understand the complexities and risks associated with the repair procedures.

Commentary by Dr. Valentin Fuster
2005;():119-125. doi:10.1115/GT2005-68380.

This Paper describes aerodynamic silencer that is presently installed in one GT Unit in Watson Cogeneration Company, Carson, California, USA. The silencer, made from three sections, was fit into existing Intake Air Duct System (IADS) during March, 2004 Major Inspection (MI). The purpose for original silencer replacement was to reduce IA pressure drops and to minimize continuous dirt carry-over from the old silencer panels into the compressor suction. GT is very sensitive to IA pressure drops and this sensitivity is manifested in GT power output reduction. IA pressure drop reduction alone justified expenditures in replacing the old silencer by new, aerodynamic silencer. Simplified EA presented in this Paper demonstrates good profit in replacing the old silencer with the new, redesigned silencer. In addition to improvements of IA aerodynamics, silencer was designed to capture IA cooling water that did not evaporate and to drain the collected water to designated locations. This Paper also incorporates field test data results, completed by others.

Commentary by Dr. Valentin Fuster
2005;():127-133. doi:10.1115/GT2005-68392.

It is important to design and operate energy conversion systems such as gas turbine cogeneration ones optimally from the thermoeconomic viewpoint. However, an energy conversion system has a complex network structure, and it takes much time to create its model for the thermoeconomic analysis and optimization. In this paper, a systems approach is presented for the performance analysis and optimization of mechanical systems with network structures, and it is applied to the thermoeconomic analysis and optimization of a gas turbine cogeneration unit. The system modeling for the performance analysis is conducted by a building block approach. Static and dynamic problems for the performance analysis are formulated as sets of nonlinear algebraic and differential algebraic equations, and are solved by the Newton-Raphson method and a hierarchical combination of the Runge-Kutta and Newton-Raphson methods, respectively. The performance optimization is conducted to determine design and operation conditions which optimize performance criteria. This problem is formulated as a nonlinear programming one and is solved by a global optimization method. In the application, the cycle analysis is conducted to determine mass flow rates, pressures, and temperatures, which is followed by the exergy and cost analyses to determine exergy flow rates and efficiencies, and capital costs, respectively. In addition, design and operation conditions are determined to maximize the exergy efficiency or minimize the annual total cost based on the results of the cycle, exergy, and cost analyses. Through a numerical study, it turns out that the proposed systems approach enables one to conduct the thermoeconomic analysis and optimization efficiently.

Commentary by Dr. Valentin Fuster
2005;():135-141. doi:10.1115/GT2005-68425.

The overall system efficiency of a microturbine generator system, and hence the economic viability, are greatly enhanced through co-generation. The utilization of the exhaust energy for heating needs is well understood and has been extensively implemented. Potentially more desirable but having found far less implementation is the integration of microturbines with absorption chillers to generate cold water. Long recognized as a tremendous opportunity in HVAC systems, the integration of a microturbine with an absorption chiller is complicated by the relatively low quality and quantity of heat available from the traditionally recuperated microturbine system. This paper address the design process and the issues encountered in developing a 160 ton absorption chiller system integrated with an array of eight, 60-kW Capstone microturbines. The designed system is not a pre-packaged turbine/chiller system but rather a system that integrates a mix of existing and new microturbines with a commercial gas-to-liquid heat exchanger and a commercial single effect Li-Br absorption chiller. The goal of the effort is demonstrate a cost effective retrofit installation of an absorption chiller with existing microturbines to effectively provide base load chilling for the building’s HVAC needs.

Commentary by Dr. Valentin Fuster
2005;():143-150. doi:10.1115/GT2005-68451.

The use of fuel cell systems for distributed generation represents an interesting option due to the intrinsic high efficiency and the potential to reduce the environmental impact of power supply in comparison with thermoelectric plants. In this paper the study of a cogenerative energy system based on a Proton Exchange Membrane fuel cell stack, that should satisfy a small electric utility, is reported; the capability of this cogenerative system to supply electrical and thermal power demand of a civil user has been investigated. In this research the electric efficiency has been calculated as net electric power on chemical power given to the system and the thermal efficiency as thermal power given to user on chemical power in input. Moreover, an energy saving index has been introduced to assess the cogenerative performance of this energy system. The investigation has been developed by experimenting an existing stack of fuel cell and studying its behaviour with a variable power demand. In particular, all the input and output mass flows have been evaluated to have parameters through which the operation of the whole cogenerative system, made by fuel cell stack and all the auxiliaries like compressor and pumps, could be simulated.

Commentary by Dr. Valentin Fuster
2005;():151-159. doi:10.1115/GT2005-68550.

Micro gas turbine may represent a successful energy system for distributed combined heat and power generation, if its energetic and economic performances become competitive, especially in cogenerative application. Furthermore, in the last years, the Inverted Brayton Cycle gas turbine has been reconsidered as a potential solution for the simple cycle gas turbine performance increase. In the present paper, the employment of an Inverted Brayton Cycle at the micro gas turbines discharge is investigated. The results obtained show that the electric and thermal performances of the energy system increase; moreover, the reduction of the maximum hot gas temperature in the recuperator (usually micro gas turbines are recuperated), due to the below ambient pressure expansion, may reduce the recuperator thermal stress problems. The economic impact of the repowering of simple cycle micro gas turbines with the Inverted Brayton Cycle is also investigated in the present study; the carried out analysis shows that the benefit of the Inverted Brayton Cycle depends on the fuel price and the electric energy tariff and the turbine discharge pressure can be optimized to maximize the economic performance.

Commentary by Dr. Valentin Fuster
2005;():161-168. doi:10.1115/GT2005-68649.

Wet compression is an effective way of gas turbine power augmentation, but at the same time it causes some new thermodynamic problems. How to evaluate the thermodynamic parameters and how to improve the wet compression process and wet compression gas turbine cycle are key to the actual applications of the technique in industries. From the entropy and exergy views, wet compression processes are analyzed in this paper. By calculating the entropy and exergy variations with the established thermodynamic models of wet compression process, we presented some interesting results that could used to reduce the increase of entropy and the loss of exergy in the wet compression process. On the basis of thermodynamic analysis of wet compression, the entropy and exergy effects at different evaporative rates, different pressure ratios etc. are further investigated in this paper.

Commentary by Dr. Valentin Fuster
2005;():169-176. doi:10.1115/GT2005-68726.

In general, the compressor takes approximately 2/3 of the expansion work delivered by the turbine. Any reduction of compressor work would improve the power output of the gas turbine. One way of doing this is to inject an amount of water in the inlet air. The amount of water that can be absorbed by the air is limited due to the relative humidity. The SwirlFlash™ system of Alpha Power Systems is an over-spray injection system, which delivers an over-saturated mixture to the compressor. Adiabatic compression will heat up the mixture and the water will evaporate inside the compressor. The heat required for evaporation will cool down the air, which results in a lower compressor discharge temperature. This lower discharge temperature leads to a reduction of compressor work. The lower compressor work and the increased fuel flow will raise the gas turbine power output. This paper is the result of the evaluation of the SwirlFlash™ system installed at the Herdersbrug power plant of Electrabel Belgium. The wet compression system was commissioned in 2003. In this paper, the principle of the wet compression system and the Herdersbrug installation are described as well as the influence of this system on several parameters such as the compressor discharge temperature, the gas flow, the steam production, the power gain and the heat rate. The influence on the emissions, humming and the interference with anti-icing are also discussed. Finally, the material related aspects and the vibration behavior are treated.

Commentary by Dr. Valentin Fuster
2005;():177-185. doi:10.1115/GT2005-68771.

Inlet fogging of gas turbine engines has attained considerable popularity due to the ease of installation and the relatively low first cost compared to other inlet cooling methods. With increasing demand for power and with shortages envisioned especially during the peak load times during the summers, there is a need to boost gas turbine power. In Taiwan, most gas turbines operate with combined cycle for base load. Only a small portion operates with simple-cycle for peak load. To recover lost power output due to increased ambient temperature in hot days, the power augmentation strategies for combined-cycles need to be studied in advance. Therefore, the objective is to study the effects caused by adding inlet fogging to an existing gas turbine-based combined-cycle power plant. Simulation runs were made for adding inlet fogging to a combined cycle with two Alstom gas turbines, two heatrecovery steam generators, and one steam turbine. Results show that the power output will be increased by 1% to 5% in typical hot summer days. Since there are seventeen combined-cycle power plants located in different areas in Taiwan, total extra power output gained by inlet fogging can make up the power loss in hot summer days. This paper also includes a parametric study of performance to provide guidelines for combined-cycle power augmentation by inlet fogging.

Topics: Cycles
Commentary by Dr. Valentin Fuster
2005;():187-193. doi:10.1115/GT2005-68818.

This Paper describes the efforts implemented in performance enhancements of CGT, Frame No7, Model EA. The CGT is operating in Watson Cogeneration Company (WCC), Carson, California, USA. The better-than-expected improvements were achieved after MI of CGT Unit 4, in October 2004. The significant increase in CGT power and simultaneous reduction in HR was obtained by introduction of OEM-improved design details, by increasing FT from 2,020 degrees F to 2,055 degrees F and also by reduction of parasitic, internal to CGT, losses.

Commentary by Dr. Valentin Fuster
2005;():195-200. doi:10.1115/GT2005-68937.

Accurate on-line plant and equipment performance evaluation is becoming critical in the power generation industry as operators seek to optimize their plants, particularly in competitive power markets. The analysis accuracy of an on-line performance monitoring system is directly dependent on the quality of the input data and usually suffers because installed plant sensors are not high-precision instruments. The inherent measurement uncertainties can be overcome by using a readily available heat balance program in combination with a least square solver. This data reconciliation system will provide the performance evaluation system with data that better reflects the plant’s current operating point, thus improving the performance evaluation system’s output and allowing for better plant optimization. Additionally, the reconciliation system can identify broken, biased or highly noisy sensors. These improvements can be obtained without installing additional precision sensors or putting unreasonable efforts into sensor calibration.

Commentary by Dr. Valentin Fuster
2005;():201-208. doi:10.1115/GT2005-68964.

Empirical correlations to predict (1) drain mass flow rate, (2) temperature depression of intake air and (3) power increase to a given ambient and operating condition are derived. The cooling efficiency and drain ratio are expressed using separation of variables, i.e. relative spray amount, ambient wet bulb depression and fogger configuration (droplet spectrum, geometry of fog rack etc.). The correlation for power increase obtained as function of net specific spray water flow can be applied to overspray mode of operation. It is confirmed that the power increase predicted matches field data within error less than 10% in the range of net specific water amount up to 0.4% overspray and ambient wet bulb depression from 3.35 to 13.8deg.

Commentary by Dr. Valentin Fuster
2005;():209-213. doi:10.1115/GT2005-69134.

The Mercury™ 50 gas turbine operates with a recuperated cycle to produce 4600 kW with a high thermal efficiency and very low emissions. Commercialized in 2003, the Mercury 50 gas turbine has completed an extensive design, development, and field evaluation program that insured a reliable and durable product that is easy to operate and maintain. This results in maximizing operation reliability and availability and lowering maintenance cost, critical factors in the distributed generation and cogeneration power generation markets. 6 Sigma, robust engineering and Kaizen methodologies were used during the design and pre-production phases to collect valuable input from users and prove out the durability, operability and maintainability of the product. This paper will review the Mercury 50 design for durability and ease of operation and maintenance.

Commentary by Dr. Valentin Fuster
2005;():215-226. doi:10.1115/GT2005-69144.

The strong influence of ambient temperature on the output and heat rate of a gas turbine has popularized the application of inlet fogging and overspray for power augmentation. This paper focuses on practical considerations, for implementation of fogging technology, such as water quality requirements, foreign object damage, gas turbine inlet icing, intake duct design, changes in compressor performance characteristics, and blade coating distress problems. It also provides a checklist for users and project developers to facilitate the design and implementation of fogging systems. In addition, this paper covers operational experience and reviews the work pursued by gas turbine OEMs in the field of fogging technology. A list of unresolved issues and ongoing research related to the fogging technology is also provided.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster

Oil and Gas Applications

2005;():227-235. doi:10.1115/GT2005-68003.

Offshore oil and gas production requires both electric and mechanical power for various applications. Traditionally, gas turbines have been the driver of choice for both the power generation, compression and larger pump applications. Today, an alternate approach of electric motor drivers is sometimes considered for pump and compressor drivers. In this case, the power for the electric motor and utility power is supplied by larger gas turbines used as a central power generation plant. This is sometimes referred to as the “All Electric” solution. There are several important factors to be evaluated when considering options and selecting the optimum solution for this type of application. Based on general assumptions on the parameters and characteristics of possible solutions, decision criteria are derived and the sensitivity of the results relative to varying assumptions is determined.

Commentary by Dr. Valentin Fuster
2005;():237-245. doi:10.1115/GT2005-68007.

Gas turbine power enhancement technologies such as inlet fogging, interstage water injection, saturation cooling, inlet chillers, and combustor injection are being employed by end-users without evaluating the potentially negative effects these devices may have on the operational integrity of the gas turbine. Particularly, the effect of these add-on devices, off-design operating conditions, non-standard fuels, and compressor degradation/fouling on the gas turbine’s axial compressor surge margin and aerodynamic stability is often overlooked. Nonetheless, compressor aerodynamic instabilities caused by these factors can be directly linked to blade high-cycle fatigue and subsequent catastrophic gas turbine failure; i.e., a careful analysis should always proceed the application of power enhancement devices, especially if the gas turbine is operated at extreme conditions, uses older internal parts that are degraded and weakened, or uses non-standard fuels. This paper discusses a simplified method to evaluate the principal factors that affect the aerodynamic stability of a single shaft gas turbine’s axial compressor. As an example, the method is applied to a frame type gas turbine and results are presented. These results show that inlet cooling alone will not cause gas turbine aerodynamic instabilities but that it can be a contributing factor if for other reasons the machine’s surge margin is already slim. The approach described herein can be employed to identify high-risk applications and bound the gas turbine operating regions to limit the risk of blade life reducing aerodynamic instability and potential catastrophic failure.

Commentary by Dr. Valentin Fuster
2005;():247-253. doi:10.1115/GT2005-68015.

This paper encapsulates generalized considerations of power turbine matching with aeroderivative gas generator at high power settings. A computation route is set up to estimate the magnitude of the desired parameters from design point knowledge of a gas generator. Then, a method is delineated to verify matching of power turbine inlet nozzle area with exhaust of gas generator by measuring tangible tested parameters. Data manipulation revealed that there exists a favorable correlation between pressure ratio of high pressure turbine and gas generator speed that may directly reflect the influence of physical area change of power turbine inlet nozzle area. A practical example is presented to demonstrate the procedure. From engine design to retirement, the generalized considerations may be applied on several occasions where question of matching may become important and require explanation for performance and financial justifications. Some generalized rules of matching are condensed and their applications are suggested.

Topics: Turbines , Generators
Commentary by Dr. Valentin Fuster
2005;():255-266. doi:10.1115/GT2005-68030.

In the paper, self-adapting models capable of reproducing time-dependent data with high computational speed are investigated. The considered models are recurrent feed-forward neural networks (RNNs) with one feedback loop in a recursive computational structure, trained by using a back-propagation learning algorithm. The data used for both training and testing the RNNs have been generated by means of a non-linear physics-based model for compressor dynamic simulation, which was calibrated on a multi-stage axial-centrifugal small size compressor. The first step of the analysis is the selection of the compressor maneuver to be used for optimizing RNN training. The subsequent step consists in evaluating the most appropriate RNN structure (optimal number of neurons in the hidden layer and number of outputs) and RNN proper delay time. Then, the robustness of the model response towards measurement uncertainty is ascertained, by comparing the performance of RNNs trained on data uncorrupted or corrupted with measurement errors with respect to the simulation of data both uncorrupted and corrupted with measurement errors. Finally, the best RNN model is tested on field data taken on the axial-centrifugal compressor on which the physics-based model was calibrated, by comparing physics-based model and RNN predictions against measured data. The comparison between RNN predictions and measured data shows that the agreement can be considered acceptable for inlet pressure, outlet pressure and outlet temperature, while errors are significant for inlet mass flow rate.

Commentary by Dr. Valentin Fuster
2005;():267-274. doi:10.1115/GT2005-68031.

In this paper, a methodology for the optimization of a single off-shore gas compression station is developed. The station is composed of three gas turbines, each one driving a centrifugal compressor. The study concerns the feasibility of the most suitable arrangement to face the depletion of wells and the consequent reduction of the head top pressure. Once the arrangement is chosen, an optimization procedure is developed and carried out. The procedure, which is aimed at obtaining either high production rates or good station efficiency, is based on knowledge of the centrifugal compressor characteristics and on the availability of gas turbine thermodynamic cycle program, the latter allowing the definition of the machine actual operating state.

Commentary by Dr. Valentin Fuster
2005;():275-283. doi:10.1115/GT2005-68053.

GE Energy’s new gas turbine, the MS5002E, is a 30 MW-class industrial gas turbine for mechanical drive and power generation applications. The MS5002E (fig.1) is the latest in the Frame5 two-shaft family and, while it retains some features from previous versions, the machine has been specifically designed for low environmental impact and high reliability, in direct response to customer demand for high efficiency and availability [1] & [2]. Main features for the MS5002E are: • 32 MW base load power at ISO inlet conditions (no losses); • 36% thermal efficiency; • 11-stage axial compressor and 17:1 pressure ratio; • reverse flow, six cans, Dry Low NOx (DLN2 technology) combustion system; • two-stages reaction type HP turbine; • two-stages PT leveraged from the LM2500+ HSPT (High Speed Power Turbine); • HP speed operating range 90% (6709rpm) / 101% (7529rpm); • PT speed operating range 50% (2857rpm) / 105% (6000rpm); • exhaust gas temperature (EGT): ∼510°C; • two-baseplates configuration (gas turbine flange-to-flange unit and auxiliary system); • integrated enclosure and baseplate, providing maximum accessibility for maintenance. The design of the MS5002E has been validated through an extensive test program which has included some key-test rigs such as the Rotordynamic Test, the CTV Test (full-scale axial compressor test) and numerous component and full-scale combustion tests in laboratory, conducted in advance of the First Engine to Test (FETT). The MS5002E First Engine to Test was initially started in January 2003 and the validation program has been completed with a full gas turbine teardown, dirty layout (visual and dimensional inspections for each major gas turbine component in as-is conditions) and NDT inspection in June 2004. During engine teardown, disassembly/assembly procedures and tools have been tested and validated. Additional endurance and operability testing is ongoing and will be completed by the end of 2005. The First Engine to Test is a complete equivalent-to-production package including gas turbine, auxiliaries and control system. For the test, a dedicated plateau has been built in Massa, Italy [3]. The gas turbine has been equipped with over 1400 direct measurement points (for a total of more than 2400 direct and indirect measurements) covering the flange-to flange, the package and auxiliaries. All critical-to-quality parameters, such as turbine gas path components temperatures and stresses, combustor temperatures and dynamics, performances and emissions, have been carefully verified by means of redundant instrumentation. This paper presents how the test program has been built on the GE Energy NPI (New Product Introduction) Development Process and how results from tests are fed back to the gas turbine design process. The paper discusses test rig and facilities layout, gas turbine operation experience and lessons learned. Results from the tests and measurements are also discussed.

Topics: Gas turbines , Testing
Commentary by Dr. Valentin Fuster
2005;():285-295. doi:10.1115/GT2005-68093.

Modern high-pressure water mist systems are an advanced choice for rotating machinery fire protection. High-pressure water mist systems can provide: • Proven extinguishing efficiency, • Proven capability to protect equipment from thermal stresses, • Tolerance to poor enclosure integrity, • A safe and reliable alternative to gaseous systems, and • An environmentally friendly alternative to dry chemicals, halons and halon alternatives. Generally, the systems have total flooding design, which is the most appropriate for protecting rotating equipment in their purpose-built enclosures. Fine water mist with a specific application rate, droplet size distribution and high discharge momentum is used to fill the enclosure quickly and completely. For all fire protection systems, third party testing and appraisal is important. FM and VdS have approved gas turbine fire protection systems for enclosures up to 500m3 , while systems for enclosures up to 3300 m3 are (2004) within approval process. This paper explains the water mist system basic terminology and fundamentals. The paper then discusses system design requirements and features. In the end, health and safety, as well as environmental aspects are reviewed.

Commentary by Dr. Valentin Fuster
2005;():297-301. doi:10.1115/GT2005-68129.

Atlantic LNG has developed a proven method for LNG turbine maintenance, which will revolutionize the industry. Over the life of Atlantic’s 4-Train facility, this method will reduce planned downtime by over a year. This paper details the innovative approach that the authors took to reduce Atlantic LNGs first major gas turbine overhaul from 29 days to 23 days, and the second major overhaul from 23 days to 18 days, with the ultimate goal of a 12 day turnaround (TAR). Atlantic LNG will soon begin a cycle in which a major overhaul is required approximately every 6 months as we cycle through our fleet of 26 Frame 5s. Using Gantry System technology, the company will be able to achieve a 50% reduction in downtime, which equates to approximately 1 year of additional production over the life of the plant.

Commentary by Dr. Valentin Fuster
2005;():303-308. doi:10.1115/GT2005-68170.

Due to enormous material losses in the case of emergency, it is vital to ensure the operation reliability of the natural gas pipeline compressor stations (CS). The risk of breakdown is rather high for gas turbines (GT) with total operation time approaching the design-estimated life and particularly for those in which the actual period of operation exceeds this value. Over 25% of turbine drives working on natural gas transportation net in Russia have exceeded their design life [1]. For instance, around 600 gas turbines of the GTC-10-4 type (10MW power) are still in service despite their 120,000–160,000 hours of operation (more than 1,000 gas turbines GTC-10 type have been made and installed at natural gas pumping stations in the seventies in Russia). These gas turbines contain several critical components. Most of them are related to the high temperature parts, including inner high-temperature turbine casing (ITC). This ITC is a kind of a collector (duct) connecting a combustion chamber outlet and the turbine’s entry. Combined with an insulation layer, it serves as a protective shield for outer (main) turbine casing against the effect of hot gases. Notwithstanding the fact that the GTC-10-4 turbine has a modest inlet gas temperature (TIT∼800°C), there are various problems with the ITC shape and state during the turbine’s operation. The ITC operates under conditions of dramatic temperature changes, pressure drops, extended periods of high temperature. All these factors can cause the ITC shell deformations, which results in poor turbine performances. Regular maintenance inspections including opening a turbine do not permit to establish reasons for dramatic changes in the ITC shape. A detailed numerical analysis has been performed to better understand the ITC dynamics over its service period of operation. Moreover, it should be observed that ITC forms a flow prior to entering a turbine. Then, gas flow is directed to the first stage nozzles of the turbine. Advanced numerical flow investigation methods were applied to improve hot gas distribution in front of the turbine. A considerable decrease in velocity nonuniformity was achieved both radially and circumferentially through the ITC shape optimization. Great need in this component stimulated introduction of a new manufacturing technology aimed at production of new ITCs and replacement of numerous defective ones still used at natural gas pumping stations across Russia. Results of thermo-deformation analysis and numerical flow investigation for various ITC configurations are presented in the paper. It also contains proposals for improving the state of the ITC and outer turbine casing (OTC) in the result of the fixing unit development and applying a new insulation material.

Commentary by Dr. Valentin Fuster
2005;():309-316. doi:10.1115/GT2005-68349.

Ramgen Power Systems, Inc. (RPS) is developing a family of high performance supersonic compressors that combine many of the aspects of shock compression systems commonly used in supersonic flight inlets with turbo-machinery design practices employed in conventional axial and centrifugal compressor design. The result is a high efficiency compressor that is capable of single stage pressure ratios in excess of those available in existing axial or centrifugal compressors. A variety of design configurations for land-based compressors utilizing this system have been explored. A proof-of-concept system has been designed to demonstrate the basic operational characteristics of this family of compressors when operating on air. The test unit was designed to process ∼1.43 kg/s and produce a pressure ratio across the supersonic rotor of 2.41:1. Based on the results from that effort a compressor specifically designed for the high pressure ratios required to support CO2 liquification has been proposed. The basic theory of operation of this new family of compressors will be reviewed along with the performance characteristics and conceptual design features of the proposed CO2 compressor systems.

Commentary by Dr. Valentin Fuster
2005;():317-325. doi:10.1115/GT2005-68436.

Knowledge of compressor system behaviour during trip is essential to obtain reliable operation. Experience at Troll Kollsnes gas treatment plant has shown that the compressor can suffer from surge and rotating stall in situations with voltage drops in the electricity grid. The electric motor driven compressor and the protection valves and piping has been modelled and dynamic simulations has been done to reveal the transient response. An elaborate plant model has been created with the dynamic simulation tool OTISS by AspenTech and tuned to represent the plant. The paper focuses on the impact of different compressor discharge volumes, dimensions of the recycle piping as well as the sizing, characteristics and trip signal delay of the recycle protection valve(s). Sensitivity analyses for the actual plant reveal the response and protection during compressor driver trip.

Topics: Compressors
Commentary by Dr. Valentin Fuster
2005;():327-337. doi:10.1115/GT2005-68701.

Gas turbine performance deterioration can be a major economic factor. An example is within offshore installations where a degradation of gas turbine performance can mean a reduction of oil and gas production. This paper describes the test results from a series of accelerated deterioration tests on a GE J85-13 jet engine. The axial compressor was deteriorated by spraying atomized droplets of saltwater into the engine intake. The paper also presents the overall engine performance deterioration as well as deteriorated stage characteristics. The results of laboratory analysis of the salt deposits are presented, providing insight into the increased surface roughness and the deposit thickness and distribution. The test data show good agreement with published stage characteristics and give valuable information regarding stage-by-stage performance deterioration.

Topics: Compressors
Commentary by Dr. Valentin Fuster
2005;():339-347. doi:10.1115/GT2005-68702.

This paper reports the results of a series of online water wash tests of a GE J85-13 jet engine at the test facilities of the Royal Norwegian Air Force. The engine performance was deteriorated by injecting atomized saltwater at the engine inlet. Then the engine was online washed with water injected at three different droplet sizes (25, 75 and 200 μm) and at water-to-air ratios ranging from 0.4% to 3% by mass. Engine performance was measured using standard on-engine instrumentation. Extra temperature and pressure sensors in the compressor section provided additional information of the propagation of deposits in the aft stages. The measurements were supported by visual observations. The overall engine performance improved rapidly with online wash. The build-up of deposits in the aft stages was influenced both by the droplet size and the water-to-air ratio. The water-to-air ratio was the most important parameter to achieve effective online washing.

Commentary by Dr. Valentin Fuster
2005;():349-359. doi:10.1115/GT2005-68758.

An optimised inlet air system design is an important factor in the gas turbine (GT) industry. Optimising the design of the air intake system is an increasingly challenging process as both the layout complexity and range of features that can be included in the intake system expands. These may include a combination of insect or trash screens, weather protection and filtration systems, silencers, anti-icing systems, ventilation system off takes and inlet heating or cooling systems for power augmentation. Poor designs can result in inefficient use of these components as well as losses in engine performance due to excessive pressure losses or distortion in the flow entering the gas turbine. High flow distortion, velocity, pressure or temperature, can induce compressor surge and high acromechanical stresses in compressor blades and vanes. In extreme cases this may result in blade or vane failures. Computational Fluid Dynamics (CFD) analysis is a powerful tool for visualisation of the predicted flow through a hypothetical air inlet system prior to manufacture. The CFD output plots include flow streamlines and contours, of pressure, velocity or temperature, at any plane in the model. These enable pressure losses, flow distortion issues, potential recirculation areas and high local velocities within the system to be reviewed. This allows optimisation of the installation design to minimise system pressure loss and flow distortion, both through the components and at the engine interface. This paper, with reference to case studies of gas turbine applications, highlights the impact that CFD analysis can have on the design of intake systems to ensure that the best overall performance is obtained. The process of developing the CFD geometry and how significant features of an installation are modeled is outlined. Environmental and operational conditions, such as cross winds can impact the flow through an intake system; therefore, incorporation of such factors into the model boundary conditions are covered. Typical output metrics from the CFD analysis are shown from selected case studies; total pressure drop and flow distortion at the interface plane between the intake system and gas turbine. The importance of experienced interpretation of the CFD output to define potential intake design modifications to improve system performance is highlighted. In specific cases model testing has been carried out to validate CFD results. Case study examples are used to show the improvements made in air intake performance that contribute to increased operational efficiency of the gas turbine application.

Commentary by Dr. Valentin Fuster

Structures and Dynamics

2005;():361-371. doi:10.1115/GT2005-68048.

Design optimization has become increasingly important in today’s world. The ability to develop products that offer the best possible solution, distinguish industry leaders from those that lag behind. To reach this goal, optimization techniques are required which provide solutions in a timely and cost effective manner. This paper addresses a specific optimization process for designing isolation mount systems for gas turbine engine accessory components. This process enables the designer to quickly select an isolation system that will reduce the loads on components without the use of a time consuming Finite Element Analysis (FEA). Commercially available tools such as MATLAB [7] and MSC-WORKING MODEL 2D [6] are used to study a range of mount systems and help the designer focus his attention on the best choice of design variables. Gas Turbine engine accessory mount systems are generally sized by emergency conditions such as Fan Blade Out (FBO). These emergency conditions are rarely seen in service, but since they can drive the cost and weight of the mount system, an optimization process is needed to select the best configurations. References [8] through [10] discuss this in detail. Design Cycle time is just as important as cost and weight. The ability to size and package components quickly and accurately is vital to the design process. Poor utilization of space can drive cost and weight as much as poor component design. Knowing the correct size of the mount system in a rapid fashion offers further opportunities for surrounding components & systems to be optimized.

Commentary by Dr. Valentin Fuster
2005;():373-380. doi:10.1115/GT2005-68127.

FMM is a reduced order model for efficiently calculating the forced response of a mistuned bladed disk. FMM ID is a companion program that determines the mistuning in a particular rotor. Together, these methods provide a way to acquire data on the mistuning in a population of bladed disks and then simulate the forced response of the fleet. This process is tested experimentally, and the simulated results are compared with laboratory measurements of a “fleet” of test rotors. The method is shown to work quite well. It is found that accuracy of the results depends on two factors: the quality of the statistical model used to characterize mistuning, and the sensitivity of the system to errors in the statistical modeling.

Commentary by Dr. Valentin Fuster
2005;():381-390. doi:10.1115/GT2005-68128.

A method for predicting the vibratory response of bladed disks under high engine acceleration rates is developed. The method is based on the Fundamental Mistuning Model, an existing reduced order model for predicting the steady-state vibratory response. In addition, a criterion is developed for a critical engine acceleration rate, above which transient effects play a large role in the response. It is shown that military engines operate at acceleration rates above this critical value and therefore transient effects are important in practice.

Commentary by Dr. Valentin Fuster
2005;():391-396. doi:10.1115/GT2005-68217.

An analysis has been developed to simulate the time-transient vibratory motion of a general configuration engine turbomachinery free-standing blade when subjected to in-service blade-on-casing tip-rub events. The analysis imports the at-speed stress-stiffened blade stiffness matrix and lumped mass matrix from a finite element model of the actual blade. Formulation and computational approaches are presented. Correct characterization of the blade tip-surface rub mechanics tribology models necessitates using empirical information that is currently being acquired from single-blade spin-pit tests now in progress in a parallel companion phase of this research. Output results for validation cases are presented. The analysis efficiently simulates complete transients involving multiple successive incursions (blade on casing hits), tracking the blade tip contact force distribution and blade motion throughout the simulated time frame including blade motion during, between and after successive casing hits.

Commentary by Dr. Valentin Fuster
2005;():397-404. doi:10.1115/GT2005-68423.

An energy-based fatigue life prediction framework has been developed for prediction of axial and bending fatigue life at various stress ratios. The framework for the prediction of fatigue life via energy analysis was developed in accordance with the approach in our previous study which states: the total strain energy dissipated during a monotonic fracture process is a material property that can be determined by measuring the area underneath the monotonic true stress-strain curve. The framework consists of the following two elements: (1) Development of a bending fatigue criterion by observing the total strain energy of the effective volume, which is achieved by computing the total plastic strain energy with consideration of the stress gradient influence through the thickness of a specimen, in the fatigue area, during cyclic loading. A comparison between the prediction and the experimental results from 6061-T6 aluminum specimens was conducted and shows that the new energy-based fatigue criterion is capable of predicting accurate fully reversed bending fatigue life. (2) Development of a new life prediction criterion for axial fatigue at various stress ratios. The criterion was constructed by accounting for both the residual energy dissipated, monotonically, due to the mean stress, and the incorporation of the mean stress effect into the total strain energy density dissipated per cycle. The performance of the criterion was demonstrated by experimental results from 6061-T6 aluminum dog-bone specimens subjected to axial stress at various stress ratios. The comparison shows very good agreement, thus validating the capability of producing accurate fatigue life predictions.

Topics: Fatigue life
Commentary by Dr. Valentin Fuster
2005;():405-412. doi:10.1115/GT2005-68434.

First, this paper established the seal structural 2D axisymmetric model of a certain Solid Rocket Booster (SRB) and calculated the deformation and stresses at ignition through a large displacement, incompressible, contact finite element analysis. The results show that the maximum contact stress appears at the contact area and the maximum shear stress at groove notch. Then, some typical parameters of the seal structure which might have the impact on the sealing performance, such as the gap breadth, initial compressibility, fillets of the groove notch and bottom, groove width, were analyzed. We can find that the gap breadth and initial compressibility do great contributions to the maximum contact normal stress, and the groove notch and bottom fillets act upon the maximum shear stress obviously. In order to verify the validity of the 2D axisymmetric model, 3D structural finite element analysis of the structure was conducted, and the results indicate that in service, the upper flange is inclined relative to the nether flange, which seems to mean that the gap breadth can not be considered as a constant during the 2D axisymmetric analysis. However further calculations say that if using the minimum gap breadth gotten in 3D analysis as its constant gap value, the above 2D axisymmetric model can rationally take the place of 3D model to analyze the sealing performance. Finally, the failure modes & criteria of the O-ring seals based on the maximum contact normal stress and shear stress were determined to ensure the reliability of this structure.

Topics: Seals , Design , Failure
Commentary by Dr. Valentin Fuster
2005;():413-427. doi:10.1115/GT2005-68516.

This paper investigates the feasibility of using contact and non-contact sensors to develop an ultra high frequency (UHF) vibration monitoring system for prognostics/diagnostics of turbine engine bearings. The authors have developed ImpactEnergy™, a feature extraction and analysis driven system that integrates high frequency vibration/acoustic emission data, collected using accelerometers and a laser interferometer to assess the health of bearings and gearboxes in turbine engines. ImpactEnergy™ combines advanced diagnostic features derived from waveform analysis, high-frequency enveloping, and more traditional time domain processing like root mean square (RMS) and kurtosis with classification techniques to provide bearing health information. The intelligent UHF concepts based system (System) presented in this paper is tested and validated in a laboratory environment by monitoring multiple bearings on test rigs that replicate the operational loads of a turbomachinery environment. The UHF system is also applied to data collected on test rigs at original equipment manufacturer (OEM) locations.

Commentary by Dr. Valentin Fuster
2005;():429-440. doi:10.1115/GT2005-68560.

One of the most common failure modes for turbomachinery wheels is associated to high-cycle fatigue of blades. A classical way to extend the working life of those structures is obtained through the introduction of specific devices in order to reduce vibrational amplitudes during resonance. Different kinds of components are used such as shrouds and wires within power industry and under platform limiters for aeronautics. Dry friction between the devices and blades induces non linear behaviors and flattens the associated frequency response functions (FRF). Even if this phenomena is now well known, different interpretations are presented in bibliography to explain the origin of this flattening. The most common one is based on the dissipated energy while more recent studies propose a different approach and explain peak flattening by changes in boundary conditions induced by the stick/slip phenomenon. The objective of the proposed study is to progress towards a better understanding of the flattening phenomena during vibration of bladed assemblies in presence of dry friction. A simple case is analyzed in order to show the contribution of respectively energy dissipation and changes of contact state on peak levels.

Commentary by Dr. Valentin Fuster
2005;():441-449. doi:10.1115/GT2005-68585.

The impending application of on-board sensors for detecting and sizing material defects and evaluating their consequences will lead to improved forecasting of readiness, as well as improved safety, retirement-for-cause, and management of assets. This research looks at the consequences of multiple, i.e., continual, on-board inspections on the cumulative probability of detection (CPOD) of the system; that is the probability of detecting a defect considering all previous inspections. In particular, modeling and simulation of the CPOD is examined as a function of the degree of correlation between subsequent inspections. A surface crack in a turbine disk is used as a test case with loading from a typical stress spectrum from a fighter engine. The analysis indicates that a significant difference in detectability is achieved through multiple inspections depending upon the degree of correlation between inspections, with statistically independent inspections exhibiting a “dramatically” improved CPOD over dependent inspections. In particular, if each inspection is statistically independent 1) it is the left tail of the parent POD that defines the CPOD, 2) for the same median value, a higher coefficient of variation of the parent POD generates a significantly more effective CPOD, and 3) if enough inspections are performed, the CPOD curve becomes a step function at the first non-zero value in the parent POD curve, thereby giving orders-of-magnitude improvement in detectibility over the parent POD. The critical issue of statistical independence of multiple inspections is investigated by examining the CPOD as a function of correlation between inspections. The results indicate that the effectiveness of continual inspections on the CPOD varies from a correlation coefficient of zero (independent), which gives a dramatic improvement compared to the parent POD, to a correlation coefficient of one (dependent), which reverts to the parent POD. In summary, the correlation between inspections is a critical component that determines the effectiveness of continual inspections.

Commentary by Dr. Valentin Fuster
2005;():451-455. doi:10.1115/GT2005-68667.

Anti-symmetrically laminated composites have coupling effects between tensile stress and twisting deformation, and are very attractive as fan blade materials of aircraft engines. Blades fabricated by anti-symmetrically laminated composites can automatically adjust the stagger angle to better aerodynamic conditions with change of axial force or rotational speed owing to the coupling effects. Thus, the anti-symmetrically laminated composite blades are expected to improve aerodynamic efficiency and the stability of aircraft engines. In this paper, the mechanical behavior of anti-symmetrically laminated composite blades is evaluated by spin tests and finite element analyses. Three kinds of blades fabricated by carbon/epoxy laminated composites in different anti-symmetrical stacking sequences were tested. A non-contact measurement technique using a multi-channel optical fiber sensor was used for measurements of blade deformations at high-speed rotating conditions, up to 10,000 rpm. The twisted angle change at the blade tip could be successfully measured. The twisted angle change increased in proportion to the second power of rotational speed, and the maximum angle change was about 4 degree at 10,000 rpm. The finite element analysis results agreed well with the spin test results. Furthermore, the three-dimensional deformation of the test blades was evaluated based on finite element analyses.

Commentary by Dr. Valentin Fuster
2005;():457-462. doi:10.1115/GT2005-68760.

DARWIN™ (Design Assessment of Reliability With INspection) is a simulation-based computer program for probabilistic fatigue life prediction of rotors and disks in commercial aircraft jet engines. This program is being developed by Southwest Research Institute® (SwRI®) and a team of major aircraft gas turbine engine manufacturers (General Electric, Pratt & Whitney, Honeywell, and Rolls Royce Indianapolis) as a major research and development initiative. This paper is a presentation of the experience of Honeywell in the use of DARWIN to assess probability of fracture (POF) due to surface damage in a highly stressed bolthole in a nickel component.

Commentary by Dr. Valentin Fuster
2005;():463-466. doi:10.1115/GT2005-68770.

Application of advanced Nickel based alloys in gas turbine rotors has risen significantly in the last two decades. It was shown by other authors that the deterministic lifing approach may be inadequate for a range of advanced turbine alloys under certain service conditions, and that a probabilistic approach results in a more relevant physics-based predictive model for such cases. Additionally, recent changes in the engine certification and design guidelines published by the FAA, Air Force, Army and other agencies call for increased application of probabilistic analysis. Thus, probabilistic lifing models become one of the key enablers for successful design and application of advanced turbine engine materials. This paper will discuss the main elements of probabilistic lifing system development for a new alloy, including material characterization requirements, selection of appropriate modeling techniques and validation plans.

Topics: Alloys , Rotors , Turbines
Commentary by Dr. Valentin Fuster
2005;():467-475. doi:10.1115/GT2005-68872.

For probabilistic designs or assessments to be acceptable, they must have the statistically robust confidence intervals provided by sampling methods. However, sample-based analyses require the number of function evaluations to be so great as to be impractical for many complex engineering applications. Efficient sampling methods allow probabilistic analysis on more applications than basic methods, although they still require a significant computational budget. This paper reviews a series of tools that aim to reduce variance in individual failure rate estimates which would reduce the confidence interval for the same number of evaluations. Several methods share a common goal, lowering the sample discrepancy within the sample space, that will create near optimal low-discrepancy sample sets. The optimization approaches include evolutionary algorithms, piecewise optimization, and centroidal Voronoi tessellation. The results of the optimization procedures show a much lower discrepancy than previous methods.

Commentary by Dr. Valentin Fuster
2005;():477-482. doi:10.1115/GT2005-68899.

A multidimensional finite element model of a viscoelastic sliding bearing is presented. The model resulted into a simplified finite element model composed by an elastic matrix and a damping matrix. These matrices are independent from each other since the viscoelastic material properties are assumed to be of Kelvin type material. Kelvin type materials are approximated as a linear combination of an elastic modulus and a viscous coefficient. This simple model describes accurately most rubbers used in machine components. The model combines the linearity of the Kelvin type material plus the finite element interpolation scheme. Thus, the advantages of the finite element discretization can be applied to any geometry. In order to obtain Kelvin’s coefficients a test rig was built. Material properties were experimentally determined and the model was validated. Afterwards, a discretized model was developed for a radial support bearing with and embedded sandwich-like rubber band. From this model, it was possible to analyze bearing stiffness and damping properties. Then, damping and stiffness coefficients were input to a rotordynamic model of a single-mass rotor with a slender shaft to assess imbalance response characteristics obtained with viscoelastic sliding bearings. This procedure allows the designer to evaluate alternative damping mechanisms that can be added to sliding bearings.

Topics: Bearings
Commentary by Dr. Valentin Fuster
2005;():483-494. doi:10.1115/GT2005-68935.

An effective method is proposed to calculate sensitivity of nonlinear forced response levels for bladed discs with friction contact interfaces to variation of parameters of these interfaces, including clearances and interferences. First and second order sensitivity coefficients together with ranges of high fidelity for forced response and the sensitivity coefficients prediction are determined. Numerical investigations of the sensitivity of the multiharmonic steady-state forced response of bladed discs with friction contacts and gaps have been performed showing the capabilities and efficiency of the method proposed.

Commentary by Dr. Valentin Fuster
2005;():495-504. doi:10.1115/GT2005-68936.

In this paper, an approach is developed to analyse the multiharmonic forced response of large-scale finite element models of bladed discs taking account of the nonlinear forces acting at the contact interfaces of blade roots. Area contact interaction is modelled by area friction contact elements which allow for friction stresses under variable normal load, unilateral contacts, clearances and interferences. Examples of application of the new approach to analysis of root damping and forced response levels are given and numerical investigations of effects of contact conditions at root joints and excitation levels are explored for practical bladed discs.

Topics: Damping , Disks , Blades
Commentary by Dr. Valentin Fuster
2005;():505-512. doi:10.1115/GT2005-68982.

Rare anomalies may be introduced during the metallurgical or manufacturing processes that may lead to uncontained failures of aircraft gas turbine engines. The risk of fracture associated with these anomalies can be quantified using a probabilistic fracture mechanics approach. In this paper, a general probabilistic framework is presented for risk assessment of gas turbine engine components subjected to either inherent or induced material anomalies. A summary of efficient computational methods that are applicable to this problem is also provided.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster
2005;():513-521. doi:10.1115/GT2005-68983.

This paper summarizes the development of a probabilistic micromechanical code for treating fatigue life variability resulting from material variations. Dubbed MicroFaVa (Micro mechanical Fa tigue Va riability), the code is based on a set of physics-based fatigue models that predict fatigue crack initiation life, fatigue crack growth life, fatigue limit, fatigue crack growth threshold, crack size at initiation, and fracture toughness. Using microstructure information as material input, the code is capable of predicting the average behavior and the confidence limits of the crack initiation and crack growth lives of structural alloys under LCF or HCF loading. This paper presents a summary of the development of the code and highlights applications of the model to predicting the effects of microstructure on the fatigue crack growth response and life variability of the α + β Ti-alloy Ti-6Al-4V.

Commentary by Dr. Valentin Fuster
2005;():523-531. doi:10.1115/GT2005-69004.

This paper deals with the development of a procedure to model geometric variations of blades. Specifically, vibratory parameters of blades are extracted from CMM data on an integrally bladed rotor (IBR). The method is based on proper orthogonal decomposition (POD) of CMM data, solid modeling and finite element techniques. In addition to obtaining natural frequencies and mode shapes of each blade on an IBR, statistics of these modal parameters are also computed and characterized. Numerical results are validated by comparison with experimental results.

Commentary by Dr. Valentin Fuster
2005;():533-541. doi:10.1115/GT2005-69022.

This paper defines a probabilistic High Cycle Fatigue (HCF) assessment process for a fan or compressor Integrally Bladed Rotor (IBR). It identifies key design variables, how they are statistically modeled, the probabilistic integration technique, and the physics-based modeling process. It defines how previous eigensensitivity based reduced order models cannot be used for IBR assessment and validates an alternate approach. An autoregressive model accounts for correlation between IBR blade-to-blade variabilities. An approach is also defined to combine sector tuned stress variation and mistuning amplifications. Predicted stress variations integrate with a probabilistic Goodman Diagram to allow an IBR risk assessment. The paper concludes by summarizing several remaining areas that are necessary for a practical assessment process. These areas are probabilistic fluid dynamic prediction, probabilistic mission analysis, propagating model error, and the need for an effective validation strategy.

Topics: Fatigue , Rotors , Cycles
Commentary by Dr. Valentin Fuster
2005;():543-551. doi:10.1115/GT2005-69050.

The paper summarises the results of a study investigating the correlation between the aerodynamic blade-to-blade coupling in mistuned bladed disc assemblies and the level of forced response. The focus was placed on the torsional mode of vibration, where blades’ coupling through the disc is weak, exposing the effects of aerodynamic coupling. The forced response of a large number of mistuned discs was computed, using whole-annulus finite element (FE) models. The unsteady aerodynamic forces that act upon the disc were calculated using a three-dimensional Reynolds-averaged Navier-Stokes CFD code. The results show the difference in response with and without aerodynamic coupling, exposing higher blade-to-blade interaction in the latter case. The statistical results show the change of the forced response distribution curves with the introduction of aero-coupling and their deviation from Gaussian distribution.

Topics: Disks
Commentary by Dr. Valentin Fuster
2005;():553-561. doi:10.1115/GT2005-69062.

The major objective of this paper is to evaluate a stand-point for integral shroud coupling, regarding the complex problem of nonlinear resonance vibrations of a shrouded blade with friction and impact effects. Following the load sequence in the start-up and further uploading to base load, a nonlinear cyclic FE static computation with friction forces at the shroud interface delivers contact stress results essential for assessment of a reliable shroud coupling. The FE refinement study at the shroud interface proves the reliability of the computed eigenfrequencies with respect to the harmonic engine excitation. Using nonlinear dynamic simulations of the shroud connection with friction forces, contact stiffness, surface roughness and impacts, the decoupling between the static and dynamic motions at the shroud interface is demonstrated. Based on the one-dimensional description of vibration characteristics for the shrouded blade, the resulting normal and tangential contact stiffness are evaluated from the computed 3D FE nodal diameter diagrams. The excitation forces acting on the blade are determined with the stimulus concept, in which an empirical factor is estimated from pulsation measured in the combustor chamber over the frequency range of the blade vibrations. The entire process is illustrated for the redesigned Z-lock interface on the shroud of a gas turbine stage whose contact surfaces had shown fretting problems. The numerical results confirm possible contact failures for the old shroud configuration. The blade calculated with the modified shroud connection shows numerically, stable dynamic behavior and will therefore prevent further fretting contact problems.

Commentary by Dr. Valentin Fuster
2005;():563-570. doi:10.1115/GT2005-69070.

This paper examines the nature of the statistical distribution of the peak maximum amplitude of the forced response of a mistuned bladed disk. A range of values of the structural coupling between blades and the standard deviation of mistuning have been used, and it is determined if it is correct to describe the distribution of the peak maximum amplitude as Weibull. Also, using a neural network, a functional relationship between the peak maximum amplitude distribution and input parameters (structural coupling between blades and standard deviation of mistuning) is sought via coefficients of Hermite polynomials and 99 percentile peak maximum amplitude.

Topics: Disks
Commentary by Dr. Valentin Fuster
2005;():571-577. doi:10.1115/GT2005-69124.

A fracture mechanics-based methodology is presented for determining equivalent initial flaw size (EIFS) distributions from field inspection data accounting for different service histories of fracture critical components. The EIFS distribution based on inspection data from a subset of the fleet allows for an assessment of initial material/manufacturing quality and enables a probabilistic fracture mechanics-based fatigue life prediction for the fleet as a whole. The methodology is demonstrated on an industrial gas turbine blade experiencing in-service cracking.

Commentary by Dr. Valentin Fuster
2005;():579-588. doi:10.1115/GT2005-68157.

An unsteady, three dimensional Navier-Stokes solution in rotating frame formulation for turbomachinery applications has been described. Casting the governing equations in a rotating frame enables the freezing of grid motion and results in substantial savings in computer time. Heat transfer to a gas turbine blade was computationally simulated by finite element methods and probabilistically evaluated in view of the several uncertainties in the performance parameters. The interconnection between the CFD code and finite element structural analysis code was necessary to couple the thermal profiles with the structural design. The stresses and their variations were evaluated at critical points on the turbine blade. Cumulative distribution functions and sensitivity factors were computed for stresses due to the aerodynamic, geometric, material and thermal random variables. These results can be used to quickly identify the most critical design variables in order to optimize the design and make it cost effective. The analysis leads to the selection of the appropriate materials to be used and to the identification of both the most critical measurements and parameters.

Commentary by Dr. Valentin Fuster
2005;():589-600. doi:10.1115/GT2005-68221.

Due to the trend in the design of modern aeroengines to reduce weight and to realize high pressure ratios, fan and first stage compressor blades are highly susceptible to flutter. At operating points with transonic flow velocities and high incidences stall flutter might occur involving strong shock-boundary layer interactions, flow separation and oscillating shocks. In this paper, results of unsteady Navier-Stokes flow calculations around an oscillating blade in a linear transonic compressor cascade at different operating points including near stall conditions are presented. The nonlinear unsteady Reynolds-averaged Navier-Stokes equations are solved time-accurately using implicit time-integration. Different Low-Reynolds-Number turbulence models are used for closure. Furthermore, empirical algebraic transition models are applied to enhance the accuracy of prediction. Computations are performed two-dimensionally as well as three-dimensionally. It is shown that, for the steady calculations, the prediction of the boundary layer development and the blade loading can be substantially improved compared with fully turbulent computations when algebraic transition models are applied. Furthermore, it is shown that the prediction of the aerodynamic damping in the case of oscillating blades at near stall conditions can be dependent on the applied transition models.

Commentary by Dr. Valentin Fuster
2005;():601-612. doi:10.1115/GT2005-68439.

One of the challenges in the design of rotating machinery is the issue of vibrations. The structure, no matter if talking about compressors or turbines, is subject to various sources of excitation that lead to vibrations under resonance conditions. This paper deals with the question of turbine blade vibrations. It describes practical examples of the implementation of unsteady computational fluid dynamics (CFD) and forced response calculation by means of finite element methods (FEM) applied to the design and development procedure of axial turbocharger turbines. The four examples deal with various questions which rise at different stages in the development process of turbines. One example concerns to the expected excitation of the rotor due to the stator. It demonstrates the advantages of using CFD in the prediction of this kind of excitation. Another one deals with an engine application, for which the influence of the inlet housing on the blade excitation had to be assessed. Both examples rely on the comparison of calculated excitation to the corresponding experimental strain gauge measurement for a reference case. This reference case can be used for calibration. A further case study concerns to blade vibrations in pulse charging systems. It was the intention not only to determine a spatial resolution of the excitation, but also to calculate true stresses by means of forced response calculations with FEM. In this example first bending mode shapes of the turbine blade of a rather simple type were investigated. Higher, more complex mode shapes were also investigated to prove the method. In this example, dynamic stresses were also estimated, using calculated excitation as input for forced response calculations. The results show that the use of modern numerical methods reduces cost and required time in the design of axial turbocharger turbines. They help to substantially reduce the experimental effort, while even more complete information concerning excitation and response of the structure is made available for the designer.

Commentary by Dr. Valentin Fuster
2005;():613-623. doi:10.1115/GT2005-68453.

An assessment and validation of a numerical prediction tool for flutter are made using new experimental data from experiments on turbine blades in a sector cascade. The 3D geometry is that of a low-pressure (LP) turbine blade with twist and a profile that changes along span in an annular sector cascade. The numerical model is a linear harmonic Euler equation solver. Rig results are obtained for the blade by oscillating 1 blade out of 7 in the annular sector cascade. The blade is oscillated in the rig using a mechanical type of actuator to control the mode. The mode shapes in the rig consist of torsion and bending modes around a pivot mechanism fixed inside the hub end wall. The frequencies obtained in the rig are in the range up to 219 Hz, or reduced frequency based on full chord k = 0.5, which covers the range of useful reduced frequencies typically found in turbine designs. Under reference running conditions the unsteady pressure responses are found qualitatively in line with the experiment. The test case is shown to be challenging to the numerical tool in terms of effects of tip clearance as well as off-design effects. In order to improve results tip clearance modeling and inclusion of viscous terms are identified as key factors.

Commentary by Dr. Valentin Fuster
2005;():625-633. doi:10.1115/GT2005-68665.

Unsteady aerodynamic characteristics of an oscillating cascade composed of DCA (Double Circular Arc airfoil) blades were studied both experimentally and numerically. The test cascade was operated in high subsonic flow fields with incidence angles up to 5 degrees. Above 3 degrees of the incidence, a separation bubble was produced at the leading edge. The principal concern of the present study was placed on the influence of the separated region on the vibration instability of the cascade blades. The experiment was conducted in a linear cascade wind tunnel in which seven DCA blades were equipped. The central one could be oscillated in a pitching mode. The influence coefficient method was adopted for the measurement, where the unsteady aerodynamic moments were measured on the central blade and neighboring ones. For the numerical analysis, a quasi 3-D N-S code with k–ε turbulence model was developed. The experimental and numerical results complemented each other to obtain detailed understanding of the unsteady aerodynamic behavior of the cascade. It was found that the separation bubble at the leading edge governed the vibration characteristics of blades through the oscillation of the separation bubble itself on the blade surfaces. From the results of parametric studies, the phase shift of the oscillation of the separation bubble was found to be a key factor for determining the unsteady aerodynamic characteristics of the oscillating blades.

Commentary by Dr. Valentin Fuster
2005;():635-649. doi:10.1115/GT2005-68813.

Modern computer simulations can predict some aspects of the unsteady aerodynamic phenomena associated with turbomachinery blade rows. This allows analysts to investigate aeroelastic phenomena, such as flutter, and blade-row interactions, such as forced response and unsteady effects on performance. This paper describes tools and design processes used to numerically investigate unsteady aerodynamic phenomena in heavy-duty gas turbines. A linearized Navier-Stokes method from the DLR has been used to predict the aerodynamic damping of both compressor and turbine airfoils under a variety of operating conditions. Some of these predictions were validated with engine experience. Other CFD codes, including TRACE from the DLR and ITSM3D from the University of Stuttgart, have been used to predict blade-row interaction. This includes the prediction of forced response due to rotor-vane interaction and unsteady effects on performance. The effects of airfoil clocking, including the effects of cooling flow injection, have also been investigated.

Commentary by Dr. Valentin Fuster
2005;():651-660. doi:10.1115/GT2005-68075.

Self bearing motors have been identified as a potential technology for oil free turbomachinery where integral starter generator (ISG) technology may be used, among others. One of the key advantages of self bearing motors is that they reduce the shaft length of machines, hence increasing the rotordynamic performance. Until now, research has yielded self bearing motors that produce motoring torque and active force control of either the radial forces or the thrust force. In this paper, a new self bearing motor winding is analyzed for its potential to simultaneously produce actively controlled torque, radial forces and thrust force. The benefit of such a design is that the shaft length can be reduced even further as both the radial bearing and thrust bearing functions are performed by the motor. Experimental results are presented that verify the force and torque production capability of the motor to within 14% of the theoretical predictions. Another benefit of the motor is its ability to be used for large axial (thrust) movements, such that it could be used in precision axial positioning of high speed rotating loads over a large range (up to 25 mm or so for a 25 mm diameter rotor).

Commentary by Dr. Valentin Fuster
2005;():661-670. doi:10.1115/GT2005-68078.

In our companion paper, the trapezoidal winding is proposed for use as a self bearing motor that is capable of actively controlling motoring torque, radial force and axial thrust. Of course, this assumes that sensor signals are available for control of these axes. Conventionally, this is done using encoders and proximity sensors. In this paper the authors propose a self sensing approach that uses the back-EMF waveforms of the trapezoidal winding to estimate the angular, radial and thrust positions of the rotor, as well as rotor speed. Both simulated and experimental results are presented that demonstrate the potential of the approach and identify its limitations when applied to a closed loop feedback system. The experimental results indicate that the x, y, z and θ positions can be completely decoupled and estimated using the phase difference information in the back-EMF waveforms. A test actuator demonstrates that large axial displacements in excess of 5–10mm are achievable with the new actuator layout.

Commentary by Dr. Valentin Fuster
2005;():671-678. doi:10.1115/GT2005-68177.

Passenger vehicle turbochargers (TCs) offer increased engine power and efficiency in an ever-competitive marketplace. Turbochargers operate at high rotational speeds and use engine oil to lubricate fluid film bearing supports (radial and axial). However, TCs are prone to large amplitudes of sub-synchronous shaft motion over wide ranges of their operating speed. Linear rotordynamic tools cannot predict the amplitudes and multiple frequency shaft motions. A comprehensive nonlinear rotordynamics model coupled to a complete fluid-film-bearing model solves in real time the dynamics of automotive turbochargers. The computational design tool predicts the limit cycle response for several inner and outer film clearances and operating conditions including rotor speed and lubricant feed pressure. Substantial savings in product development and prototype testing are the benefits of the present development. The paper presents predictions of the linear and nonlinear shaft motion of an automotive turbocharger supported on a semi-floating ring bearing. The shaft motion predictions are compared to measurements of shaft motion at the compressor nose for speeds up to 240 krpm, and for lubricant inlet pressure of 4 bar at 150°C. Linear and nonlinear rotordynamic models reproduce very well the test data for synchronous response to imbalance. The nonlinear results show two sub-synchronous whirl frequencies whose large magnitudes agree well with the measurements. A large side load predicted for this turbocharger must be considered for accurate prediction of the rotordynamic response.

Commentary by Dr. Valentin Fuster
2005;():679-688. doi:10.1115/GT2005-68199.

The high-speed micro hydrostatic gas journal bearings used in the high-power density MIT micro-engines are of very low aspect ratio with an L/D of less than 0.1 and are running at surface speeds of order 500 m/s. These ultra-short high-speed bearings exhibit whirl instability limits and a dynamic behavior much different from conventional hydrostatic gas bearings. The design space for stable high-speed operation is confined to a narrow region and involves singular behavior (Spakovszky and Liu (2003)). This together with the limits on achievable fabrication tolerance that can be achieved in the silicon chip manufacturing technology severely affects bearing operability and limits the maximum achievable speeds of the micro turbomachinery. This paper introduces a novel variation of the axial-flow hydrostatic micro-gas journal bearing concept which yields anisotropy in bearing stiffness. By departing from axial symmetry and introducing biaxial symmetry in hydrostatic stiffness, the bearing’s top speed is increased and fabrication tolerance requirements are substantially relieved making more feasible extended stable high-speed bearing operation. The objectives of this work are: (1) to characterize the underlying physical mechanisms and the dynamic behavior of this novel bearing concept, and (2) to report on the design, implementation and test of this new micro-bearing technology. The technical approach involves the combination of numerical simulations, experiment, and simple, first principles based modeling of the gas bearing flow field and the rotordynamics. A simple description of the whirl instability threshold with stiffness anisotropy is derived explaining the instability mechanisms and linking the governing parameters to the whirl ratio and stability limit. An existing analytical hydrostatic gas bearing model is extended and modified to guide the bearing design with stiffness anisotropy. Numerical simulations of the full non-linear governing equations are conducted to validate the theory and the novel bearing concept. Experimental results obtained from a micro-bearing test device are presented and show good agreement between the theory and the measurements. The theoretical increase in achievable bearing top speed and the relief in fabrication tolerance requirements due to stiffness anisotropy are quantified and important design implications and guidelines for micro gas journal bearings are discussed.

Commentary by Dr. Valentin Fuster
2005;():689-700. doi:10.1115/GT2005-68222.

One of the major challenges for the successful operation of high-power-density micro-devices lies in the stable operation of the bearings supporting the high-speed rotating turbomachinery. Previous modeling efforts by Piekos [1], Liu et al. [2] and Spakovszky and Liu [3] have mainly focused on the operation and stability of journal bearings. However, since thrust bearings play the vital role of providing axial support and stiffness, there is a need to gain a fuller understanding of their behavior. In this work, a rigorous theory is presented to analyze the effects of compressibility in micro-flows (characterized by low Reynolds numbers and high Mach numbers) through hydrostatic thrust bearings for application to microturbomachines. The analytical model, which combines a 1-D compressible flow model with Finite-Element Analysis, serves as a useful tool for establishing operating protocols and assessing the stability characteristics of hydrostatic thrust bearings. The model is capable of predicting key steady-state performance indicators, such as bearing mass flow, axial stiffness and natural frequency as a function of the hydrostatic supply pressure and thrust bearing geometry. The model has been applied to investigate the static stability of hydrostatic thrust bearings in micro-turbine-generators, where the electrostatic attraction between the stator and rotor gives rise to a negative axial stiffness contribution and may lead to device failure. Thrust bearing operating protocols have been established for a micro-turbopump, where the bearings also serve as an annular seal preventing the leakage of pressurized liquid from the pump to the gaseous flow in the turbine. The dual role of the annular pad poses challenges in the operation of both the device and the thrust bearing. The operating protocols provide essential information for the required thrust bearing supply pressures and axial gaps required to prevent the leakage of water into the thrust bearings for various pump outlet pressures. Good agreement is observed between the model predictions and experimental results. In addition, a dynamic stability analysis is also performed, which indicates the occurrence of unstable axial oscillations due to flow choking effects in both forward and aft thrust bearings. These a-priori dynamic stability predictions were subsequently verified experimentally on a micro-turbocharger. The frequencies of unstable axial oscillations predicted using the model compare favorably to those determined experimentally, thus vindicating the validity of the model. A simple and useful dynamic stability criterion is established, where the occurrence of flow choking in both thrust bearings give rise to dynamic instability.

Commentary by Dr. Valentin Fuster
2005;():701-714. doi:10.1115/GT2005-68223.

The MIT microengine rotors are supported by hydrostatic gas journal and hydrostatic gas thrust bearings. Due to the low length-to-diameter ratio of the devices, the thrust bearings play an important role in providing sufficient tilting stiffness to resist any tilting motion about the spinning axis of the rotor. The performance of the thrust bearings can be influenced by geometric nonuniformities such as thrust bearing clearances and orifice diameters and profiles which arise in the process of microfabrication. To enable stable high speed operation of the micro-devices, it is important to quantify these effects. Furthermore, a thrust bearing analysis tool needs to be developed that is able to explore different thrust bearing arrangements and configurations. In this work, an analytical model is established for analyzing the effects of rotor tilt and geometric non-uniformities in micro-hydrostatic gas thrust bearings for application to microturbomachinery. A previously developed model (Teo and Spakovszky [1]) is generalized and extended for application to thrust bearings with orifices arranged in non-axisymmetric configurations. As a consequence of rotor tilt or geometric non-uniformities, the flow through individual orifices of the thrust bearing becomes non-uniform. The orifice flows are in turn coupled to the hydrostatic pressure field in the thrust bearing pad, and a Green’s function approach is adopted to solve the coupled system. The hydrodynamic thrust bearing forces induced by the pumping action of the rotor rotation are determined by solving the Reynolds equation. The model is able to predict thrust bearing tilting stiffness and variations in the thrust bearing mass flow rates as a function of rotor tilting angle for a variety of orifice arrangements. The model can be applied to analyze the effects of non-uniformities in orifice diameter and the presence of clogged orifices on tilting and the concomitant reduction in tilting stiffness. In addition, the effects of orifice taper are analyzed using an influence-coefficient technique for 1-D compressible flows. Results obtained for various taper ratios are presented and discussed. The model serves as a useful tool for specifying design tolerances during the fabrication of micro-hydrostatic gas thrust bearings and is used in the experiments to estimate the tilting angle of the rotor during operation.

Commentary by Dr. Valentin Fuster
2005;():715-724. doi:10.1115/GT2005-68296.

Reliable gas bearings will enable the rapid deployment of high speed oil-free micro-turbomachinery. This paper presents analysis and experiments of the dynamic performance of a small rotor supported on Rayleigh step gas bearings. Comprehensive tests demonstrate that Rayleigh step hybrid gas bearings exhibit adequate stiffness and damping capability in a narrow range of shaft speeds, up to ∼ 20 krpm. Rotor coast down responses were performed with two test bearing sets with nominal radial clearance of 25 μm and 38 μm. A near-frictionless carbon (NFC) coating was applied on the rotor to reduce friction at liftoff and touchdown. However, the rotor could not lift easily and severe rubbing occurred at shaft speeds below ∼ 4,000 rpm. The tests show that the supply pressure raises the rotor critical speed and decreases the system damping ratio, while only affecting slightly the rotor-bearing system onset speed of instability. Whirl frequencies are nearly fixed at the system natural frequency (∼ 120 Hz) with subsynchronous amplitude motions of very large magnitude that prevented rotor operation above ∼ 20 krpm. The geometry of the Rayleigh steps distributed on the rotor surface generates a time varying pressure field, resulting in a sizable 4X super synchronous component of bearing transmitted load. Predictions show the synchronous stiffness and damping coefficients decrease with shaft speed. Predicted threshold speeds of instability are much lower than measured values due to the analytical model limitations assuming a grooved stator. The predicted synchronous responses to imbalance correlate well with the measurements. The Rayleigh step gas bearings are the most unreliable rigid bearing configuration tested to date.

Topics: Rotors , Gas bearings
Commentary by Dr. Valentin Fuster
2005;():725-736. doi:10.1115/GT2005-68343.

Experimental dynamic force coefficients are presented for a flexure-pivot-tilting-pad (FPTP), bearing in load-between-pad (LBP) configuration for a range of rotor speeds and bearing unit loadings. The bearing has the following design parameters: 4 pads with pad arc angle 72° and 50% pivot offset, pad axial length 0.0762 m (3 in), pad radial clearance 0.254 mm (0.010 in), bearing radial clearance 0.1905 mm (0.0075 in), preload 0.25 and shaft nominal diameter of 116.84 mm (4.600 in). Measured dynamic coefficients have been compared with theoretical predictions using an isothermal analysis for a bulk-flow Navier-Stokes model. Predictions from two models — the Reynolds equation and a bulk-flow Navier-Stokes (NS) equation model are compared with experimental, complex dynamic stiffness coefficients (direct and cross-coupled) and show the following results: (i) The real part of the direct dynamic-stiffness coefficients is strongly frequency dependent because of pad inertia, support flexibility, and the effect of fluid inertia. This frequency dependency can be accurately modeled for by adding a direct added mass term to the conventional stiffness/damping matrix model. (ii) Both models underpredict the identified added-mass coefficient (∼32 kg), but the bulk-flow NS equations predictions are modestly closer. (iii) The imaginary part of the direct dynamic-stiffness coefficient (leading to direct damping) is a largely linear function of excitation frequency, leading to a constant (frequency independent) direct damping model. (iv) The real part of the cross-coupled dynamic-stiffness coefficients shows larger destabilizing forces than predicted by either model. The direct stiffness and damping coefficients increase with load, while increasing and decreasing with rotor speed, respectively. As expected, a small whirl frequency ratio (WFR) was found of about 0.15, and it decreases with increasing load and increases with increasing speed. The two model predictions for WFR are comparable and both underpredict the measured WFR values. Rotors supported by either conventional tilting PAD bearings or FPTP bearings are customarily modeled by frequency-dependent stiffness and damping matrices, necessitating an iterative calculation for rotordynamic stability. The present results show that adding a constant mass matrix to the FPTP bearing model produces an accurate frequency-independent model that eliminates the need for iterative rotordynamic stability calculations.

Commentary by Dr. Valentin Fuster
2005;():737-746. doi:10.1115/GT2005-68384.

High performance oil-free turbomachinery implements gas foil bearings (FBs) to improve mechanical efficiency in compact units. FB design, however, is still largely empirical due to their mechanical complexity. The paper provides test results for the structural parameters in a bump-type foil bearing. The stiffness and damping (Coulomb or viscous type) coefficients characterize the bearing compliant structure. The test bearing, 38.1 mm in diameter and length, consists of a thin top foil supported on bump-foil strips. A prior investigation identified the stiffness due to static loads. Presently, the test FB is mounted on a non-rotating stiff shaft and a shaker exerts single frequency loads on the bearing. The dynamic tests are conducted at shaft surface temperatures from 25 °C to 75°C. Time and frequency domain methods are implemented to determine the FB parameters from the recorded periodic load and bearing motions. Both methods deliver identical parameters. The dry friction coefficient ranges from 0.05 to 0.20, increasing as the amplitude of load increases. The recorded motions evidence a resonance at the system natural frequency, i.e. null damping. The test derived equivalent viscous damping is inversely proportional to the motion amplitude and excitation frequency. The characteristic stick-slip of dry friction is dominant at small amplitude dynamic loads leading to a hardening effect (stiffening) of the FB structure. The operating temperature produces shaft growth generating a bearing preload. However, the temperature does not affect significantly the identified FB parameters, albeit the experimental range was too small considering the bearings intended use in industry.

Commentary by Dr. Valentin Fuster
2005;():747-755. doi:10.1115/GT2005-68401.

Hydro-inertia gas bearing is a type of static air bearing, which supports the rotor by suction force generated by supersonic flow in large bearing clearance [1]. A tool to analyze the flow inside the clearance of hydroinertia gas bearings have been developed, and validated by experiment. A tool to estimate the load capacity and the bearing stiffness of the hydroinertia gas bearing based on experimental data has also been developed. A micro spinner test rig has been fabricated to test an hydroinertia gas bearings designed by the developed tools, and stable operation of 4mm diameter shaft at 1,200,000 rpm has successfully been achieved. A micro-high-speed bearing test rig to test a rotor for micromachine gas turbine has been designed and fabricated. Current micromachine gas turbine’s configuration requires a rotor with 10mm diameter compressor and turbine impellers on each end of 4mm diameter shaft to operate stably at 870,000rpm. Based on the achievement of stable operation at the high-speed of 1,200,000 rpm, hydro-inertia gas bearing has been selected as a candidate for both the bearings for micromachine gas turbine. Currently, the rotor speed as high as 770,000rpm has been achieved in this test rig.

Commentary by Dr. Valentin Fuster
2005;():757-762. doi:10.1115/GT2005-68484.

Historical attempts to measure forces in magnetic bearings (MBs) have experienced limited success as a result of relatively high uncertainties. Recent advances in strain-gauge technology have provided a new method for measuring MB forces. Fiberoptic strain gauges (FOSGs) are roughly 100 times more sensitive than conventional strain gauges and are not affected by electromagnetic interference. At the Texas A&M University (TAMU) Turbomachinery Laboratory, installing FOSGs in MBs has produced force measurements with low uncertainties. Dynamic flexibility transfer functions (DFTFs) exhibiting noticeable gyroscopic coupling have been identified and compared with finite element predictions. Comparison has verified the effectiveness of using MBs as calibrated exciters in rotordynamic testing. Many applications including opportunities for testing unexplained rotordynamic phenomena are now feasible.

Commentary by Dr. Valentin Fuster
2005;():763-771. doi:10.1115/GT2005-68486.

Widespread usage of gas foil bearings (FBs) into micro turbomachinery to midsize gas turbine engines requires accurate performance predictions anchored to reliable test data. The paper presents a simple yet accurate model predicting the static and dynamic force characteristics of gas FBs. The analysis couples the Reynolds equation for a thin gas film to a simple elastic foundation model for the top foil and bump strip layer. An exact flow advection model is adopted to solve the partial differential equations for the zeroth- and first- order pressure fields that render the FB load capacity and frequency dependent force coefficients. As the static load imposed on the foil bearing increases, predictions show the journal center displaces to eccentricities exceeding the bearing nominal clearance. A nearly constant FB static stiffness, independent of journal speed, is estimated for operation with large loads; and approaching closely the structural stiffness derived from contact operation at null rotor speed. Predicted minimum film thickness and journal attitude angle demonstrate good agreement with archival test data for a first-generation gas FB. The bump-foil strip structural loss factor, exemplifying a dry-friction dissipation mechanism, aids to largely enhance the bearing direct damping force coefficients. At high loads, the bump-foil structure influences most the stiffness and damping coefficients. The FB whirl frequency ratio (WFR) is examined to ensure its dynamically stable operation. The predictions demonstrate that FBs have greatly different static and dynamic force characteristics when operating at journal eccentricities in excess of the bearing clearance from those obtained for operation at low loads, i.e. small journal eccentricities.

Topics: Bearings
Commentary by Dr. Valentin Fuster
2005;():773-781. doi:10.1115/GT2005-68522.

This paper considers optimization of rotor system design using stability and vibration response criteria. The initial premise of the study is that the effect of certain design changes can be parameterized in a system dynamic model through their influence on the system matrices obtained by finite element modeling. A suitable vibration response measure is derived by considering an unknown axial distribution of unbalance components having bounded magnitude. It is shown that the worst-case unbalance response is given by an absolute row-sum norm of the system frequency response matrix. The minimization of this norm is treated through the formulation of a set of linear matrix inequalities (LMIs) that can also incorporate design parameter constraints and stability criteria. The formulation can also be extended to cover uncertain or time-varying system dynamics arising, for example, due to speed-dependent bearing coefficients or gyroscopic effects. Numerical solution of the matrix inequalities is tackled using an iterative method that involves standard convex optimization routines. The method is applied in a case study that considers the optimal selection of bearing support stiffness and damping levels to minimize the worst-case vibration of a flexible rotor over a finite speed range. The main restriction in the application of the method is found to be the slow convergence of the numerical routines that occurs with high-order models and/or high problem complexity.

Commentary by Dr. Valentin Fuster
2005;():783-790. doi:10.1115/GT2005-68577.

In a rotor-bearing system, there are usually some under- or un-modelled dynamic components which are considered frequency dependent, such as foundations, bearings, and seals. This paper presents a method to identify the dynamic behavior of these components using an otherwise accurate engineering model of the system in combination with available measurements of system frequency response functions. The approach permits treatment of flexible rotors and allows that the system test excitations and measurement sensors are not collocated. Because all engineering models contain some residual error and all measurements incorporate an element of noise or uncertainty, the quality of the identified parameters must be estimated. This paper introduces application of μ-analysis to solve this problem, resulting in acceptable solution time and hard guarantees of solution reliability. Two illustrative examples are provided, showing that the presented approach is an efficient method to identify and bound these parameters.

Topics: Rotors
Commentary by Dr. Valentin Fuster
2005;():791-799. doi:10.1115/GT2005-68583.

The case for installing auxiliary bearings in parallel with magnetic bearings is often made with regard to touchdown, when a complete system failure occurs. The work reported in this paper focuses on the case when rotor/auxiliary bearing contact occurs, but the magnetic bearings retain their functionality. One may envisage future transportation applications in which this situation would occur, for example, during high acceleration levels induced by turbulence. An understanding of the rotor dynamic response during contact conditions could enable auxiliary bearing life expectancy to be extended using appropriate control action from the still functional magnetic bearings. To achieve this, a system model is required for control strategy design purposes. This paper considers the development of a non-linear system model for predicting the contact dynamics in a flexible rotor/magnetic/auxiliary bearing system. Previous experimental work produced similar contact dynamic response characteristics; whether due to unbalance or circular forcing through a magnetic bearing. Initial model-based predictions of these tests did not provide sufficiently accurate reproduction of the measured orbits, particularly in the presence of auxiliary bearing misalignment and multi-plane rotor contact. Parameter variations are thus undertaken to investigate the reasons for these differences. Contrary to expectations, uncertainty in the magnetic bearing characteristics during contact conditions appears to offer an explanation.

Topics: Bearings , Rotors
Commentary by Dr. Valentin Fuster
2005;():801-806. doi:10.1115/GT2005-68593.

Presented here are results from an experimental study investigating the reduction of subsynchronous vibrations in rotating machinery by adding a single active magnetic bearing actuator to a flexible rotor-bearing system. In this scenario, the Active Magnetic Bearing (AMB) actuator is used as an Active Magnetic Damper (AMD) and is not utilized for rotor support. The AMD can be used to increase stability margins by adding more damping in strategic locations on a rotor allowing for increased tolerance to instability mechanisms and enabling increased performance and efficiency in turbomachinery. Results from an experimental 3-mass test rig supported in fluid-film bushings are presented here. The study shows that subsynchronous vibrations are reducible with an AMD located near the mid-span of the rotor and up to a 98% reduction in the amplitude of subsynchronous vibrations is demonstrated. The overall results from this work demonstrate that reduction in subsynchronous response is feasible and that full rotor dynamic analysis and design is critical for the successful application of this approach as critical speed locations can be altered.

Commentary by Dr. Valentin Fuster
2005;():807-814. doi:10.1115/GT2005-68641.

In a previous ASME paper the second author reported experiments on wire mesh bearing dampers (WMD) incorporated in a power turbine rotor-bearing system in order to enable a direct comparison between WMD and squeeze film dampers (SFD). The results showed that both WMD and SFD perform equally well for reducing the rotordynamic amplitudes of vibration. Moreover the WMD were found to have significant advantages over SFD. The damping provided by the wire mesh is independent of temperature changes and presence of turbine oil. Experiments by another investigator showed that WMD are capable of sustaining more than twice the unbalance as compared to SFD, which promises possible application to withstand blade loss loads. This paper presents empirically developed non-dimensional design equations for WMD, capable of predicting stiffness and damping for a wire mesh ‘donut’ subject to changes in various design, installation, and operational parameters.

Commentary by Dr. Valentin Fuster
2005;():815-823. doi:10.1115/GT2005-68692.

The purpose of this study is to investigate the nonlinear axial response of a thrust bearing-rotor system, which is subjected to an axial harmonic force. For the axial vibration of the rotor, the system forces include the external axial harmonic force and the reacting oil film forces, which are obtained by solving a time-dependent Reynolds Equation within the thrust pads of the thrust bearing. The time-dependent Reynolds Equation is solved by a finite difference method, and the system equation of motion is solved by the fourth-order Runge-Kutta method. A linear analysis is attempted in to evaluate its suitability for the situation under consideration. And the bearing stiffness and damping coefficients are investigated with parameters including the dimensionless wedge thickness, the initial oil film thickness and the rotor spin speed. The results show that the average steady state response will decrease as the harmonic axial force intensifies its fluctuating magnitude. The results also indicate that it will induce ultra-super harmonics when the axial harmonic force intensifies its fluctuating magnitude.

Commentary by Dr. Valentin Fuster
2005;():825-831. doi:10.1115/GT2005-68732.

The main goal of this paper is to improve identification methods for rotordynamic coefficients of labseals for turbines. This aim was achieved in joint effort of the Technische Universität München, working on experimental identification methods for rotordynamic coefficients, the University of Technology, Darmstadt, working on prediction methods, and Siemens AG, realizing the results. The paper focuses on a short comb-grooved labyrinth seal. Short labseals, amongst others the above mentioned comb-grooved labyrinth, were examined. by means of a very accurately measuring test rig. The rotor was brought into statically eccentric positions relative to the stator, in order to measure the circumferential pressure distribution as a function of pressure, rotating speed and entrance swirl. The data collected were used to validate results obtained with a numerical method. The theoretical approach is based on a commercial CFD tool, which solves the Navier Stokes equations using numerical methods. As a result, a detailed model of the flow within the test rig is produced. The efforts of computation here are greater than when compared with the likewise wide-spread Bulk flow models, however improved accuracy and flexibility is expected. As the validation of the model is successful, it could then be used to gain further insight in the flow within the seal, and to understand the results better. This showed that rotordynamic coefficients of labseals gained from different test rigs are not necessarily comparable.

Commentary by Dr. Valentin Fuster
2005;():833-839. doi:10.1115/GT2005-68804.

This paper describes the design and manufacturing of an experimental facility for measurement of equivalent stiffness and damping of air bearings. For these preliminary tests, the shaft moves only in two perpendicular directions, laying in the rotation plane, thus producing 2×2 characteristic matrices. However, the rig can be easily modified to measure rotordynamic characteristics related to angular motion of the journal and measuring 4×4 matrices. The testing facility uses an experimental magnetic bearing suspension system that allows imposing any given orbit to the shaft, during the testing experiments. All individual parts, as well as the assembly, were dynamically studied to determine their modal response and optimize it according to the test rig’s operating frequency range. The principle of operation is to produce a shaft orbit using the magnetic suspension system and measuring the forces generated on the test bearing housing. Then, the stiffness and damping coefficients are calculated using an iterative parameter identification algorithm (a modification of the IVF method). The force measurement is performed via three load cells placed in a triangle configuration around the test bearing housing. All data is gathered and processed using PC based data acquisition boards and software. The present design allows testing air bearings up to 44 mm in external diameter and a bandwidth of 0 Hz to 1.000 Hz. Preliminary testing was performed on this research that demonstrates the capability of the apparatus to measure the dynamic properties with ease and accuracy.

Topics: Bearings
Commentary by Dr. Valentin Fuster
2005;():841-850. doi:10.1115/GT2005-68839.

The coupling between lateral and torsional vibrations has been investigated for a rotor dynamic system with breathing crack model. The stiffness matrix has been developed for the shaft element which accounts for the effect of the crack and all six degrees of freedom per node. Since the off-diagonal terms of the stiffness matrix represent the coupling of the respective modes, the special attention has been paid on accurate determination of their values. Based on the concepts of fracture mechanics, the variation of the stiffness matrix over the full shaft revolution is represented by the truncated cosine series where the fitting coefficient matrices are extracted from the stiffness matrices of the cracked shaft for a number of its different angular positions. The variation of the system eigenfrequencies and dynamic response of the rotor with two cracks have been studied for various shaft geometries, crack axial locations, and relative phase of cracks.

Commentary by Dr. Valentin Fuster
2005;():851-860. doi:10.1115/GT2005-68885.

In this paper, the application of Neural Networks and Fuzzy Logic to the diagnosis of Faults in Rotating Machinery is investigated. The Learning-Vector-Quantization (LVQ) Neural Network is applied in series and in parallel to a Fuzzy inference engine, to diagnose 1x faults. The faults investigated are unbalance, misalignment, and structural looseness. The method is applied to a test rig [1], and the effectiveness of the integrated Neural Network and Fuzzy Logic method is illustrated.

Commentary by Dr. Valentin Fuster
2005;():861-870. doi:10.1115/GT2005-68887.

In this paper, a method for calibrating rotordynamic models of speed dependent systems supported on anisotropic supports is presented. The method is based on the comparison between the calculated eigenvalues and the ones extracted from the synchronous frequency response functions. An eigensensitivity analysis is conducted to calculate the sensitivity of the computed eigenvalues to the selected elements to be updated. This method is suitable for field application since it requires simple coast down tests. The method is illustrated on a test rig with fluid film bearings, and is shown to be effective in the calibration of rotordynamic models at the speeds of the modes excited within the operating speed range.

Topics: Calibration
Commentary by Dr. Valentin Fuster
2005;():871-882. doi:10.1115/GT2005-68981.

Deterioration of a rotordynamic system changes its modal properties. This paper initiates a study of the degree to which such changes can be detected by monitoring modal metrics obtained by a modern technique for experimental modal analysis (EMA). The eigenvalues and residues associated with a complex modal description of the frequency response are identified by processing response data derived from an analytical model. This model, which features an elastic shaft with attached rigid rotor and supported by hydrodynamic bearings, was previously used by Wagner and Ginsberg [Proc. of the 23rd International Modal Analysis Conf. (IMAC), forthcoming, 2005] to explore the merits of using standard or directional frequency response functions to perform EMA. The techniques used there, specfically the original version of the Algorithm of Mode Isolation (AMI) for FRFs and Two-Sided AMI for dFRFs, are used to extract the modal properties from the model’s frequency domain response. The modal eigenvalues and residues are identified for a range of bearing clearances within the limit of acceptable wear. One set of metrics that are considered describes the behavior of the system’s eigenvalues as clearance increases. Another set of metrics describes the modal residue factors, which depend on the drive and response locations. A defect is considered to be detectable if the change in the value of a metric due to deterioration exceeds the uncertainty in that metric’s value associated with the inexact nature of EMA. Although the eigenvalues are identified with great accuracy, they are found to be relatively insensitive to bearing clearance, so that metrics derived from them do not meet the detectability criterion. In contrast, the residue values are identified less accurately, but they are highly sensitivity to the clearance. It is concluded that metrics describing the behavior of the modal residue factors can unambiguously indicate bearing wear that is large, but still acceptable for continued operation. It also is found that it is preferable to monitor the residues obtained by processing standard FRFs using the original AMI version, rather than using Two-Sided AMI to process dFRFs.

Commentary by Dr. Valentin Fuster
2005;():883-889. doi:10.1115/GT2005-68988.

The backstepping method is applied to design a controller which is capable of actively suppressing the oscillations induced by compressor surge using the sensitivity of the centrifugal compressor characteristic to the tip clearance of the unshrouded impeller. This is achieved using a magnetic thrust bearing to modulate the tip clearance of the impeller. The controller is designed with the objective that system trajectories remain on the compressor characteristic curve in the presence of disturbances downstream of the compressor. This ensures zero steady state offset of the impeller, which maintains the efficiency of the compressor. Results from simulation of the nonlinear model for a single stage high speed centrifugal compressor show that using backstepping control, mass flow and pressure oscillations associated with compressor surge are quickly suppressed with acceptable control requirements of typical machines. Further, the stable region of the compressor operation is significantly increased and the compression system will not enter surge cycles for a wide variety of disturbances downstream of the compressor.

Commentary by Dr. Valentin Fuster
2005;():891-898. doi:10.1115/GT2005-69013.

The present work studies the behavior of a magnetic bearing supported rotor when the flow of electric current to the magnetic actuator is suppressed In this condition the rotor is supported by the auxiliary bearing, which has looseness with the rotor, generating a series of impacts between these components. For the study of this state, a model of a flexible rotor is proposed, and the impacts are simulated using kinematical restitution coefficient theory. The results obtained from the theoretical model are compared with experimental data taken on a test rig using tools for non linear systems analysis such bifurcation diagrams. The comparison shows that, besides the simplification of the contact, the model predicts ranges chaotic, quasi-periodic, and periodic motions in the test rig.

Commentary by Dr. Valentin Fuster
2005;():899-906. doi:10.1115/GT2005-69048.

This study utilizes genetic algorithm to minimize the condition number of Hermitian matrix of influence coefficient (HMIC) to reduce the computation errors in balancing procedure. Then, the optimal locations of balancing planes and sensors would be obtained as fulfilling optimization. The finite element method is used to determine the steady-state response of flexible rotor-bearing systems. The optimization improves the balancing accuracy, which can be validated by the experiments of balancing a rotor kit.

Commentary by Dr. Valentin Fuster
2005;():907-915. doi:10.1115/GT2005-69104.

Damper seals (such as honeycomb and hole pattern seals) have been widely used in the turbomachinery industry to counterbalance destabilizing aerodynamic forces acting on the rotor system and to provide the necessary amount of damping for stable operation of the machine. Recent experience and research developments have focused the interest of the turbomachinery community on the dynamic characteristics of the divergent-taper damper seal. In addition to offering a review of the current literature on the subject of interest (in order to provide a comprehensive vision of the overall phenomenon), this paper extends the discussion for a better understanding of the mechanism and impact of damper seal clearance divergence on the rotordynamics of a rotor-bearing-seal system (as applied to centrifugal compressors). The analysis of a damper seal alone is insufficient to assess the influence of this component on the rotordynamic stability of a turbocompressor. This paper will show the variation of a complete rotor-bearing-seal system’s logarithmic decrement as a function of the hole pattern seal clearance divergence for a sample centrifugal compressor application, as analyzed with the proprietary, state-of-the-art rotordynamic software suite from the author’s company. This divergence-based transition of system logarithmic decrement from positive to zero (thereby implying the onset of instability) leads to the definition of a “damper seal divergence stability threshold.” Divergence in a damper seal can originate from numerous sources, including taper produced during manufacture, and pressure or thermal-driven distortion of the seal under operation. While small damper seal divergence may first produce an increase of the system logarithmic decrement, additional seal divergence has a dramatic effect on the first forward whirling mode natural frequency, as well as on the logarithmic decrement of that mode. This paper describes the analytical methods used to derive the stability of the rotor-bearing-seal system. It also presents a practical experience that stresses the necessity for a sound approach to properly evaluate the impact of a divergent damper seal on the stability of centrifugal compressors.

Commentary by Dr. Valentin Fuster

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

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