Performance Testing of Axial Compressors

Performance testing is a key part of the design and development process of advanced axial compressors.  These are widely used in the modern world and can be found in nearly every industry, and include the core compressor for aeropropulsion turbofan engines, as well as aeroderivative gas turbine engines for power generation.  An example of this are the turbine engines shown in Figure 1 and 2, which feature an industrial gas turbine and a high bypass ratio turbofan engine with a multistage high-pressure core compressor. The development time of these machines can involve numerous expensive design-build-test iterations before they can become an efficient and competitive product. This places a great importance on the accuracy of the data taken during the performance tests during the development of the compressor since the test data taken is often used to anchor the loss models within the design tools. Modern axial compressors typically have high aerodynamic loadings per stage for improved system efficiency and requires precise aerodynamic matching of the stages to achieve the required pressure ratio with high efficiency. Variable geometry inlet guide vanes and stators in the first few stages are typically required to provide acceptable operability while maintaining high efficiency and adequate stall margin.

Industrial gas turbine for power generation.
Figure 1. Industrial gas turbine for power generation. Source
Figure 2. Turbofan engine for aeropropulsion.
Figure 2. Turbofan engine for aeropropulsion. Source

Performance Testing of Axial Compressors

Axial compressors all undergo a thorough design and development phase in which performance testing is vital to their ultimate success as a product. Performance testing during the development phase of these high-power density machines can ensure that the design meets the specified requirements or can identify a component within the turbomachine which falls short of its expected performance, and may require further development, and possible redesign. Performance testing can also ensure that the unit can meet all the conditions specified and not merely the guaranteed condition. Aerodynamic performance testing multistage axial compressors during the early part of development is often done in phases. The development test program is planned and executed with a design of experiments approach and includes varying the air flow and shaft rotational speed as well as the variable geometry schedule in order to fully characterize the compressor. In the first phase, the front block of the compressor is built and tested at corrected (referenced) air flow rate, inlet pressure, temperature and shaft rotational speed. Instrumentation includes utilizing traditional rakes and surveys at the exit, to obtain spanwise distributions of pressure, temperature, and flow angles. Testing in phases is typically done for two reasons. Firstly, core compressors of these types of units are high power density machines and testing the entire multistage axial compressor in full geometric scale is usually prohibitive since they require enormous horsepower from the drive unit. It is often customary to test compressors at a reduced geometric scale in order to stay within the power limitations of the test facility drive motor or turbine. However, it is more difficult to obtain accurate test data of the scaled down compressor due to small blade passages. Additionally, a reduced scale model may not enable maintaining the running clearances, and the smaller blade leading and trailing edge thicknesses can also be an issue from a manufacturing perspective. The second reason for testing the front and rear compressor blocks separately during the early phases of development is the possibility that in a multistage axial compressor, the actual performance of the front block may fall short of the design goal, and the overall performance would be adversely effected due to aerodynamic mismatch between stages. In this way the designers can take a mid-course correction and tweak the design of either the front or rear block blades, prior to committing to the expense of fabricating the final blades of the rear block.

Figure 3. CAD representation of a multistage axial compressor core compressor from a turbine engine.

Testing of these compressor components is done with precise, calibrated instrumentation, and high response data acquisition systems. Figure 4 illustrates a computer flow model of the core compressor, as well as the front and the rear blocks, which are typically performance tested separately.

Component Flow Model
Figure 4. Computer flow model of a core compressor, showing the front and rear blocks, which are performance tested separately in high speed rotating test rigs.

Front Block Compressor Exit Profile of Key Parameters.

Figure 5 illustrates a highly instrumented test rig of the front block of a high-speed rotating multistage axial compressor. Test instrumentation includes traditional steady measurements of inlet and exit pressures and temperatures, and exit flow angle, as well as the radial distributions of these parameters.

Figure 5. Axial compressor test rig.
Figure 5. Axial compressor test rig. Source

Figure 6 shows an example of the radial profile data as measured by rakes and survey probes, at the exit plane of the front block compressor. The measured values from the test are typically compared to the values obtained from the computer flow model of the compressor. Knowing the measured spanwise gradients of key aerodynamic parameters at the exit of the front block compressor is critical before finalizing the blading design of the rear block compressor. Obtaining accurate test data is also important for the calibration of loss models within the compressor design and flow analysis code.

Figure 6. Test data of spanwise total pressure, temperature and absolute flow angle, at the exit of the axial compressor front block.
Figure 6. Test data of spanwise total pressure, temperature and absolute flow angle, at the exit of the axial compressor front block.

The spanwise distributions of pressure, temperature and flow angle results obtained from testing the front block are utilized to validate or improve the predictive accuracy of the design and analysis codes, which can, if necessary, be used to redesign blades in the front block in order to improve its performance. Accurate measurements of the spanwise gradients of pressure, temperature and flow angles  at the design operating condition are necessary in order for the design of the rear block compressor blades.

Another key objective of performance testing during the development phase of axial compressors is to experimentally determine the optimum variable geometry schedule that results in the highest level of overall efficiency for a range of flow rates and shaft rotational speeds along the engine operating line. The reset angles of inlet guide vane and variable stators are varied and the resulting aerodynamic performance is measured for a range of constant speed lines. This data is particularly important to obtain for the validation and calibration of the loss and deviation angle models within the design and flow analysis codes at off design operating conditions. Unsteady pressure data measured with high response transducers is also taken to determine limits of stall, rotating stall and compressor surge. This is also done for health monitoring of the test rig to avoid operating in full compressor surge, since operating there for prolonged times can be harmful to the compressor test rig and the compressor blades.

Compressor Overall Characteristic Performance Maps

The overall performance of compressors are usually represented by a characteristic map, for example, such as the compressor pressure ratio and efficiency maps illustrated in Figures 7. Characteristic maps typically show the comparison of the measured test data to the predicted simulation results obtained from the flow model of the machine. The pressure ratio and efficiency maps are typically plotted in terms of corrected flow rate and corrected shaft rotative speed, which are referenced to standard pressure and temperature conditions. The speed lines in Figure 7 are in terms of the ratio of corrected speed to the design corrected speed. Performance testing of the rear compressor block is done in a similar manner to the front block. The inlet conditions of pressure and temperature are at ambient values, but the corrected air flow rate and corrected shaft rotative speed of the first stage of the rear block need to be the same value as in the full engine system.

Figure 7. Pressure ratio and efficiency test data comparison to computer model predictions.
Figure 7. Pressure ratio and efficiency test data comparison to computer model predictions. Source

The performance of an axial compressor test must take into consideration the composition of the air, that is, the percentage of relative humidity in the air. The fluid properties of air changes with the local temperature as well as with the water vapor content. Thus, the fluid properties of the wet air must be calculated with great accuracy and the test data reduction should be performed using the actual fluid properties of the air with humidity. This key consideration is important when calculating the efficiency of the compressor. The accuracies associated with these measurements as well as the fluid properties calculations are often small, however, small errors in each variable can combine to a larger error in the computed output.

Core Engine Performance Testing

During the later parts of the development program the full core compressor is typically tested in the core engine subsystem configuration, where the high-pressure compressor, the combustor and the high-pressure turbine assembly is performance tested together. The purpose of the core engine test is to verify that the compressor and turbine are properly matched aerothermodynamically to operate at their design intent levels of performance in a subsystem environment. The core engine is likewise typically tested at corrected (referenced) inlet pressure, temperature and shaft rotational speed, since the inlet air is at ambient conditions, and not buried behind a low-pressure compressor as it would be in an engine. Aerodynamic performance data taken during core engine tests are typically the flow rate, shaft rotative speed, component overall pressure and temperature ratios and fuel flow rate.

Performance Testing Seminar

The topics covered in this blog and many more will be included in the first of a series of upcoming seminars by SoftInWay on the broad topic of turbomachinery performance testing, taking place August 25th. Learn more and register 

Meet the Author

Joseph Veras This blog was written by Joseph Veres,  a Principal Consultant Engineer and Turbomachinery Expert at SoftInWay. He has over 40 years of experience in commercial and aerospace compressor design, development testing and code development. He retired from NASA Glenn Research Center in Cleveland, Ohio, USA, where he was the Chief of the Turbomachinery and Heat Transfer branch at NASA from 2004-2009. At NASA he developed design and analysis codes for axial and centrifugal compressors and pumps. The codes were validated on performance test results from research compressor rigs of advanced gas turbine engine components. Previously, he was at Dresser-Rand where he successfully designed and tested industrial multistage centrifugal compressors for oil refineries, gas pipeline boosters and natural gas reinjection. From 1984 – 1989, he was at Teledyne CAE Turbine Engines, playing a key role in the advanced design and development group where he designed and tested numerous high-performance centrifugal compressors for small turbojet, turbofan and turboshaft engines. Joseph is a lifetime member of the American Society of Mechanical Engineers (ASME),  and has authored, or co-authored over 33 technical conference publications. He will also be leading the upcoming Performance Testing During the Turbomachinery Development Process Seminar  hosted online.  Register here

 

References:

  1. https://journals.sagepub.com/doi/abs/10.1243/0954410971532758?journalCode=piga
  2. https://www.ihi.co.jp/var/ezwebin_site/storage/original/application/2b9fb4333c5a3718efbac1c4c99ad3e4.pdf
  3. https://assets-turbomachinerymag-com.s3.amazonaws.com/uploads/2018/09/ge-turbine.jpg
  4. https://www.calspan.com/services/aerospace-test-equipment-design-fab/turbomachinery-test-rigs/
  5. https://www.semanticscholar.org/paper/Computational-Assessment-of-a-3-Stage-Axial-Which-Kulkarni-Beach/87053c6f42a939e45e807734e7ef5f539b31fee9