Optical fibre probes for total pressure and total temperature measurement in a turbine test rig

Author(s):  
J. S. Barton
2017 ◽  
Vol 139 (12) ◽  
Author(s):  
Sabine Bauinger ◽  
Emil Goettlich ◽  
Franz Heitmeir ◽  
Franz Malzacher

For this work, reality effects, more precisely backward-facing steps (BFSs) and forward-facing steps (FFSs), and their influence on the flow through a two-stage two-spool turbine rig under engine-relevant conditions were experimentally investigated. The test rig consists of an high pressure (HP) and an low pressure (LP) stage, with the two rotors rotating in opposite direction with two different rotational speeds. An S-shaped transition duct, which is equipped with turning struts (so-called turning mid turbine frame (TMTF)) and making therefore a LP stator redundant, connects both stages and leads the flow from a smaller to a larger diameter. This test setup allows the investigation of a TMTF deformation, which occurs in a real aero-engine due to non-uniform warming of the duct during operation—especially during run up—and causes BFSs and FFSs in the flow path. This happens for nonsegmented ducts, which are predominantly part of smaller engines. In the case of the test rig, steps were not generated by varying temperature but by shifting the TMTF in horizontal direction while the rotor and its casing were kept in the same position. In this way, both BFSs and FFSs between duct endwalls and rotor casing could be created. In order to avoid steps further downstream of the interface between HP rotor and TMTF, the complete aft rig was moved laterally too. In this case, the aft rig incorporates among others the LP rotor, the LP rotor casing, and the deswirler downstream of the LP stage. In order to catch the influence of the steps on the whole flow field, 360 deg rake traverses were performed downstream of the HP rotor, downstream of the duct, and downstream of the LP rotor with newly designed, laser-sintered combi-rakes for the measurement of total pressure and total temperature. Only the compact design of the rakes, which can be easily realized by additive manufacturing, makes the aforementioned 360 deg traverses in this test rig possible and allows a number of radial measurements positions, which is comparable to those of a five-hole probe. To get a more detailed information about the flow, also five-hole probe measurements were carried out in three measurement planes and compared to the results of the combi-rakes.


Author(s):  
J. S. Barton ◽  
J. M. Kilpatrick ◽  
W. N. MacPherson ◽  
J. D. C. Jones ◽  
K. S. Chana ◽  
...  

Optical fibre sensors offer the prospect of miniature aerodynamic probes for highly localised flow measurements in aerospace wind tunnels and turbomachines. We discuss the design and construction of optical fibre sensors for temperature and pressure. The temperature sensors consist of multilayer coatings deposited on the fibre end face from which the reflected intensity is temperature-dependent. Two sensors were incorporated in a dual heated probe to measure total temperature. The pressure sensors are miniature diaphragms in which pressure-induced deflection is measured interferometrically in reflection. We present results from initial trials made in unsteady flow in a single stage research turbine, in which total temperature data with harmonic components up to 30 kHz and total pressure signals up to 230 kHz were recorded.


Author(s):  
Mehmet Mersinligil ◽  
Jean-François Brouckaert ◽  
Nicolas Courtiade ◽  
Xavier Ottavy

This contribution addresses the possibility of exploiting the temperature dependency of piezoresistive sensors as a temperature measurement per se. This requires the characterization of the sensor, or the probe as a temperature probe, i.e., determination of the recovery factor between the sensor temperature and the flow temperature. This temperature calibration as well as the determination of the thermal response time was performed for two probe geometries: a high temperature flush-mounted and a low temperature subsurface mounted single sensor total pressure probe, both with a probe head diameter of 2.5 mm. Two applications are reported. The first application was performed with the flush-mounted sensor probe in the high-speed 312-stage axial compressor CREATE tested in the 2 MW test rig of LMFA at École Centrale de Lyon, in France. The probes were traversed at each inter-row section up to temperatures of 180°C and an absolute pressure of 3 bar. The second application was performed with the subsurface mounted sensor probe in the high-speed single stage R-4 compressor test rig of the von Karman Institute in Belgium. Both applications have shown results in extremely good agreement with simultaneous total temperature measurements with a Kiel-type thermocouple probe. They also underline the necessity of a very accurate temperature calibration. Finally, considering the fact that a simultaneous temperature measurement can be obtained at the same location as the pressure measurement from the sensor, it is possible to derive entropy generation after a blade row, based on the average pressure and temperature quantities. This unveils another extremely interesting aspect of using the fast response probe technique in turbomachinery applications.


Author(s):  
Sabine Bauinger ◽  
Franz Malzacher ◽  
Emil Goettlich ◽  
Franz Heitmeir

For this work, reality effects, more precisely backward and forward facing steps, and their influence on the flow through a two-stage two-spool turbine rig under engine-relevant conditions were experimentally investigated. The test rig consists of a HP and a LP stage, with the two rotors rotating in opposite direction with two different rotational speeds. An S-shaped transition duct, which is equipped with turning struts (so-called turning mid turbine frame or TMTF) and making therefore a LP stator redundant, connects both stages and leads the flow from a smaller to a larger diameter. This test setup allows the investigation of a TMTF deformation, which occurs in a real aero-engine due to non-uniform warming of the duct during operation — especially during run up — and causes backward and forward facing steps in the flow path. This happens for non-segmented ducts, which are predominantly part of smaller engines. In the case of the test rig, steps were not generated by varying temperature but by shifting the TMTF in horizontal direction while the rotor and its casing were kept in the same position. In this way, both backward and forward facing steps between duct endwalls and rotor casing could be created. In order to avoid steps further downstream of the interface between HP rotor and TMTF, the complete aft rig was moved laterally too. In this case, the aft rig incorporates amongst others the LP rotor, the LP rotor casing and the deswirler downstream of the LP stage. In order to catch the influence of the steps on the whole flow field, 360 deg rake traverses were performed downstream of the HP rotor, downstream of the duct and downstream of the LP rotor with newly designed, laser-sintered Combi-Rakes for the measurement of total pressure and total temperature. Only the compact design of the rakes, which can be easily realised by additive manufacturing, makes the aforementioned 360 deg traverses in this test rig possible and allows a number of radial measurements positions, which is comparable to those of a five hole probe. To get a more detailed information about the flow, also five hole probe measurements were carried out in three measurement planes and compared to the results of the Combi-Rakes.


Author(s):  
Mehmet Mersinligil ◽  
Jean-François Brouckaert ◽  
Nicolas Courtiade ◽  
Xavier Ottavy

This contribution addresses the possibility of exploiting the temperature dependency of piezoresistive sensors as a temperature measurement per se. This requires the characterization of the sensor, or the probe as a temperature probe, i.e. determination of the recovery factor between the sensor temperature and the flow temperature. This temperature calibration as well as the determination of the thermal response time was performed for two probe geometries: a high temperature flush-mounted and a low temperature sub-surface mounted single sensor total pressure probe, both with a probe head diameter of 2.5 mm. Two applications are reported. The first application was performed with the flush-mounted sensor probe in the high-speed 3 1/2-stage axial compressor CREATE tested in the 2 MW test rig of LMFA at École Centrale de Lyon, in France. The probes were traversed at each inter-row section up to temperatures of 180°C and an absolute pressure of 3 bar. The second application was performed with the sub-surface mounted sensor probe in the high-speed single stage R-4 compressor test rig of the von Karman Institute in Belgium. Both applications have shown results in extremely good agreement with simultaneous total temperature measurements with a Kiel type thermocouple probe. They also underline the necessity of a very accurate temperature calibration. Finally, considering the fact that a simultaneous temperature measurement can be obtained at the same location as the pressure measurement from the sensor, it is possible to derive entropy generation after a blade row, based on the average pressure and temperature quantities. This unveils another extremely interesting aspect of using the fast response probe technique in turbomachinery applications.


2006 ◽  
Author(s):  
Peng Sun ◽  
Guotal Feng

A time-accurate three-dimensional Navier-Stokes solver of the unsteady flow field in a transonic fan was carried out using "Fluent-parallel" in a parallel supercomputer. The numerical simulation focused on a transonic fan with inlet square wave total pressure distortion and the analysis of result consisted of three aspects. The first was about inlet parameters redistribution and outlet total temperature distortion induced by inlet total pressure distortion. The pattern and causation of flow loss caused by pressure distortion in rotor were analyzed secondly. It was found that the influence of distortion was different at different radial positions. In hub area, transportation-loss and mixing-loss were the main loss patterns. Distortion not only complicated them but enhanced them. Especially in stator, inlet total pressure distortion induced large-scale vortex, which produced backflow and increased the loss. While in casing area, distortion changed the format of shock wave and increased the shock loss. Finally, the format of shock wave and the hysteresis of rotor to distortion were analyzed in detail.


1983 ◽  
Vol 105 (2) ◽  
pp. 223-230 ◽  
Author(s):  
I. Ariga ◽  
N. Kasai ◽  
S. Masuda ◽  
Y. Watanabe ◽  
I. Watanabe

The present paper concerns itself with the effects of total pressure (and thus velocity) distortion on performance characteristics and surge margin of centrifugal compressors. Both radial and circumferential distortions were investigated. The performance tests as well as the velocity measurements within the impeller passages were carried out with a low-speed compressor test rig with the inlet honeycomb as the distortion generators and compared with the case of “no distortion” as a datum. The results indicated that the inlet distortion exerted unfavorable influences on the efficiency and the surge margin of the given compressor, though the influence of the radial distortion was much stronger than that of the circumferential one. Various distortion indices were further examined in order to correlate the performance to the inlet distortion.


Author(s):  
Andrea Arnone ◽  
Ennio Carnevale ◽  
Michele Marconcini

The NASA Rotor 37 has been computed by several authors in the last few years with relative success. The aim of this work is to present a systematic grid dependency study in order to quantify the amount of uncertainty that comes from the grid density. The computational domain is divided onto several regions (i.e. leading edge, trailing edge, shear layer …) and for each of them, the impact of the grid density is investigated. By means of this analysis, substantial improvement has been obtained in the prediction of efficiency and exit angle. On the contrary, the improvement achieved in total pressure and total temperature ratio is less remarkable. It is believed that only after a systematic grid dependency study can the contribution of turbulence modeling, laminar-turbulent transition, and boundary conditions be analyzed with success.


2012 ◽  
Vol 571 ◽  
pp. 701-705
Author(s):  
Hui Yang ◽  
Feng Li ◽  
Bai Gang Sun ◽  
Dan Dan Tian ◽  
Yao Ying Song ◽  
...  

Structures of liquid-fuel injection in supersonic crossflow is studied experimentally. Schemes of flush-wall injector and aviation kerosene are selected. The conditions of the supersonic freestream are kept constant (total pressure is 0.5MPa, total temperature is 500K and Mach number is 2) and the diameter of the injector is fixed as 0.5mm, while seven scenarios of injection angle and three scenarios of injection driven pressure are discussed. Both methods of schlieren and planar laser induced fluorescence (PLIF) techniques are implemented to obtain the visual images of the liquid-fuel injection. The penetration height of fuel is analyzed quantitatively with the aid of Photoshop and Origin. The results serve not only the future computational simulation but also combined scheme of flush-wall injector and other combustor configurations.


2016 ◽  
Vol 16 (17) ◽  
pp. 6596-6603 ◽  
Author(s):  
Leonhard Polz ◽  
Andreas Zeisberger ◽  
Hartmut Bartelt ◽  
Johannes Roths

Sign in / Sign up

Export Citation Format

Share Document