Giornata di Studio sulle TURBOMACCHINE Bergamo 15 luglio 2016 STUDIO NUMERICO SPERIMENTALE DI TuRBomacchine di piccola potenza Rodolfo Bontempo Marcello Manna Raffaele Tuccillo Dipartimento di Ingegeria Industriale (D.I.I) Università di Napoli Federico II
Abstract The presentation deals with theoretical, numerical and experimental analysis of small sized turbomachines Ducted Rotors The classical actuator disk model is improved to take into account all nonlinear flow features. It is applied to the theoretical analysis and to the preliminary design of innovative devices like small ducted wind turbines. Turbochargers The steady and unsteady behaviour of automotive turbochargers are studied my experimental means. Some insights on the validity of the adiabatic flow assumption and on the statistical characterisation of the surge are provided. The analysis of turbochargers is further detailed showing the conceptual design of an innovative vaneless axial flow turbine. Finally, the design of a novel downstream volute for axial flow turbine is presented and discussed Radial flow turbine A similarity criteria approach is used to investigate the actual possibilities of adopting a radial flow turbine as an ORC expander. The results are validated by CFD. 2
References Bontempo, R., Manna, M., Solution of the flow over a non-uniform heavily loaded ducted actuator disk, Journal of Fluid Mechanics (728), pp. 163-195, 2013 Bontempo, R., Manna, M., Performance analysis of open and ducted wind turbines, Applied Energy (136), pp. 405-416, 2014 Bontempo, R., Manna, M., Effects of the duct thrust on the performance of ducted wind turbines, Energy (99), pp. 274-287, 2016 Bontempo, R., Cardone, M., Manna, M., Vorraro, G., Steady and unsteady experimental analysis of a turbocharger for automotive applications, Energy Conversion and Management (99), pp. 72-80, 2015 Cameretti M.C., Ferrara F., Gimelli A., and Tuccillo R., 2015, Combined MGT ORC solar hybrid system. PART B: Component Analysis and Prime Mover Selection, ENERGY PROCEDIA, vol. 81 (2015), pp. 379-389, doi: 10.1016/j.egypro.2015.12.107 Cameretti M.C., Ferrara F., Gimelli A., and Tuccillo R., 2015, "Employing Micro-Turbine Components in Integrated Solar MGT - ORC power plants", ASME paper GT2015-42572 Conceptual design of an axial flow turbine for advanced variable geometry turbochargers, 2016 (A. Saccomanno Dissertation, Superv. R. Tuccillo, a. Pesiridis) Fluid-Dynamic Design of an Innovative Turbine-Volute Layout for Turbochargers, 2016 (A. Ferrara Dissertation, Superv. R. Tuccillo, a. Pesiridis) 3
Summary The nonlinear actuator disk method as applied to ducted rotors UNINA Turbocharger Test Rig Conceptual design and analysis of a vaneless axialflow turbine for turbochargers Design and analysis of axial flow turbine with downstream volute for turbochargers Study of the ORC expander based on similarity criteria and CFD analysis 4
The nonlinear actuator disk method as applied to ducted rotors Ducted propellers Ducted Turbines The duct improves the propulsive efficiency The duct prevents the occurrence of the cavitation The duct improves the power coefficient ı ı (a ducted turbine can beat the Betz limit) 5
Assumptions Stationary Incompressible Inviscid Axisymmetric The rotor is modelled through an actuator disk Genesis of the method Nonlinear actuator disk Vortex Panel method Exact solution of the flow in an implicit form for a prescribed load distribution A semi-analytical and iterative procedure has been developed 6
Main advantages of the method The method naturally takes into account the contraction of the wake (very important for ducted and heavily loaded propellers) The method duly accounts for the nonlinear and mutual interaction between the duct and the rotor the rotor and duct flow fields are strongly coupled Non-uniform load distributions, rotor wake rotation and ducts of general shapes can be dealt with Low computational cost it is well suited to be integrated in the first stage of design systems based on the repeated analysis scheme of hierarchical type 7
Velocity contours Duct: NACA 4415; Parabolic Load Distribution; ı = ı. ı Dimensionless axial velocity ı ı /ı ı Dimensionless radial velocity ı ı /ı ı 8
Verification against CFD Duct: NACA 4415; Parabolic Load Distribution; ı = ı. ı SA CFD 9
CPU Time evaluation Ducted Wind Turbine: NACA 5415 10
Ducted Wind Turbine Versus Open Wind Turbine Duct: NACA 5415; Parabolic Load Distribution open ducted Betz Limit 11
UNINA Turbocharger Test Rig Rig Layout 12
UNINA Turbocharger Test Rig VI Layout 13
UNINA Turbocharger Test Rig VI Layout 14
UNINA Turbocharger Test Rig Applications: Steady-State Performance Maps Compressor Map Turbine Map 15
UNINA Turbocharger Test Rig Applications: Evaluation of the heat transfer effects on the performance 16
UNINA Turbocharger Test Rig Applications: Surge Characterisation Surge N/N ref = 0.6 Surge N/N ref = 1 17
UNINA Turbocharger Test Rig Applications: Surge Characterisation ı ı ı ı ı ı Surge N/N ref = 0.6 Surge N/N ref = 1 18
UNINA Turbocharger Test Rig Applications: Surge Characterisation 19
Conceptual design and analysis of a vaneless axial-flow turbine for turbochargers Cooperation with Brunel University, London (Dr. A. Pesiridis) Engine 1D modelling Preliminary design 3D design CFD investigation Off-design and map generation 20
Engine 1D modelling Simulation at 100% of load Engine performance Simulation at 75% of load 150 Load 100% 150 Load 75% Engine Power [kw] Engine Torque [Nm] 100 50 0 300 250 200 150 100 Axial turbine GT1548 0 1000 2000 3000 4000 5000 6000 7000 Engine Speed [rpm] Load 100% Axial turbine Engine Power [kw] Engine Torque [Nm] 100 50 Axial turbine GT1548 0 0 1000 2000 3000 4000 5000 6000 7000 Engine Speed [rpm] 300 Load 75% 250 200 150 Axial turbine GT1548 GT1548 100 0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 Engine Speed [rpm] Engine Speed [rpm] 5000 6000 7000 21
Preliminary design Data from 1D engine simulation of a 1.6l Ford EcoBoost Engine Speed rpm 6000 5000 4000 3000 2000 (max. load) Mass flow kg/s 0,1140 0,1148 0,0926 0,0736 0,0446 rate Inlet bar 2,268 2,377 1,967 1,669 1,324 Total Pressure Outlet bar 1,375 1,463 1,283 1,170 1,077 Static Pressure Power kw 8,66 8,11 6,01 4,14 1,51 Speed rpm 137992 146199 132084 115917 87655 Inlet Total temperature K 1050,36 1090,40 1066,15 1040,31 1039,06 22
Preliminary design Balje Diagram for axial turbines 23
Preliminary design Cordier diagram 24
Design of the volute nozzle axial flow rotor Rotor: BladeGen The fluid conditions at rotor inlet must be achieved by the volute nozzle system Nozzle: CFturbo Volute: CFturbo 25
Design of the volute nozzle axial flow rotor Rotor: BladeGen Nozzle: CFturbo q Due to the non- Volute: uniformity CFturbo of the q fluid condition at the nozzle outlet surface, a nonperiodic solution is carried out A MRF has been used for stationary part rotor interaction (FLUENT flow solver) 26
Grid Size Sensitivity Case # Stator (*10^3) Number of cells 1 rotor passage (*10^3) Total (*10^3) η T-T [-] Mesh sensitivity analysis η T-S [-] Mach Number ex [-] CFD resluts Tot Pressure ex [Pa] Mass flow [kg/s] Computational time [h] Case 1 50 10 210 0,809 0,697 0,316 156322 0,1144 1 Case 2 200 20 520 0,818 0,704 0,322 156639 0,1158 4 Case 3 400 60 1360 0,823 0,701 0,327 157030 0,1154 8 Case 4 500 80 1780 0,824 0,700 0,328 157128 0,1152 12 Case 5 600 100 2200 0,824 0,699 0,329 157137 0,1151 15 Case 6 800 160 3360 0,824 0,698 0,330 157182 0,1151 23 Case 7 1000 200 4200 0,824 0,698 0,331 157233 0,1151 32 0,116 0,1158 Case 2 Mass flow rate [kg/s] 0,1156 0,1154 0,1152 0,115 0,1148 Case 3 Case 4 Case 5 Case 6 Case 7 0,1146 0,1144 0,1142 Case 1 0 500 1000 1500 2000 2500 3000 3500 4000 4500 Number of cells x 10 3 27
CFD analysis of the fixed part Free vortex distribution 28
CFD analysis of the fixed part Free vortex distribution 29
Rotor CFD investigation Results Power [W] Mass Flow [kg/s] η T-S η T-T Final Turbine 11340 0,115 70.0% 82,4% AxStream 12397 0,113 77,0% 86,0% HUB SECTION MEAN SECTION TIP SECTION 30
OFF DESIGN ANALYSIS 3 2,8 2,6 2,4 2,2 βt-s 2 1,8 1,6 1,4 1,2 Turbine Operating Map 55008 100015 145022 185028 210032 Tref=109 7K ; pref=2.35 bar 1 0,08 0,09 0,1 0,11 0,12 0,13 Corrected mass flow [kg/s] m (T/Tref)/((p/pref) Compressor pressure ratio [-] Turbine Speed [rpm] Total-to-static efficiency [-] 2,00 200000 0,72 0,7 150000 0,68 0,66 100000 0,64 50000 0,62 Engine load Axial 75% turbine 0,6 GT1548 0 0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 Data at engine 3000 speed 4000[rpm] 5000 6000 7000 Data at Engine Speed [rpm] Total-to-total efficiency [-] Engine Turbine Matching Load 75% Load 75% 0,84 1,80 0,82 0,8 1,60 0,78 1,40 0,76 0,74 1,20 0,72 Axial turbine Engine load 75% 0,7 GT1548 1,00 0 01000 100020002000 3000 3000 4000 4000 5000 5000 6000 6000 7000 7000 Data Data at at Engine engine Speed speed [rpm] 31
OFF DESIGN ANALYSIS Turbine 280 Operating Map 3 260 Load 75% Tref=109 7K ; pref=2.35 bar 2,8 2,6 2,4 2,2 2 1,8 Engine Torque [Nm] 240 55008 100015 220 145022 185028 200 210032 180 160 1,6 140 1,4 120 Axial turbine GT1548 1,2 100 1 0 1000 2000 3000 4000 5000 6000 7000 0,08 0,09 0,1 0,11 0,12 0,13 Corrected mass flow [kg/s] m (T/Tref)/((p/pref) Engine Speed [rpm] 32
Cooperation with Brunel University, London (Dr. A. Pesiridis) Fluid-Dynamic Design of an Innovative Turbine-Volute Layout for Turbochargers The solution proposed: Honeywell Axial Flow Turbine 33
v Kinetic energy recovery v Improved efficiency under steady and unsteady conditions due to the largely circumferential uniform flow. v Improved transient response due the shorter flow path and less exhaust manifold and turbine volute volume between the engine and the rotor. v Reduction of pulsating flow effects (phase shift) due to the shorter flow passage of the stator than that of a volute 34
Reduction of the discharge static pressure Static Pressure Outlet Turbine [bar] 1,45 1,39 Total to Static Efficiency [%] 74,07 73,01 Blade Speed [m/s] 294,1 294,1 Isentropic Velocity [m/s] 534,72 555,17 Blade Speed Ratio [-] 0,55 0,52 Rotor Mean Diameter [mm] 38,56 37,56 Rotor Blade Height [mm] 13,41 13,47 Stator Mean Diameter [mm] 38,56 37,56 Stator Blade Height [mm] 12,05 12 Maximum Rotor Diameter[mm] 51,97 51,03 Inlet Area [mm^2] 1623,6 1588,6 35
Stator Annulus Stator Profile, Hub Stator Profile, Mean Stator Profile, Tip Rotor Annulus Hub Profile Mean Profile Tip Profile Solidworks 3D representation 36
3D turbine representation from BladeGen Solidworks 3D representation 37
Different Configurations 11 Stator Blades/ 12 Rotor Blades/ Increased Thickness 19 Stator Blades/20 Rotor Blades Configuration Original New Δ% Total To Static Efficiency [%] 70,9 74 4,18 Total To Total Efficiency [%] 85,2 82,8 2,8 Power [kw] 14,9 13 14,61 Mass Flow Rate [kg/s] 0,134 0,1136 18,58 38
Different Configurations MEAN 19 Stator Blades/20 Rotor Blades 39
Different Configurations HUB 19 Stator Blades/20 Rotor Blades 40
v Isentropic Flow v Free Vortex Law v Mass conservation The design of diffuser and volute Cross Sections Calculation: Free Vortex Law 41
The concept of a symmetric volute Use of two «semivolutes» to reduce the clutter of the last sections and accomodate the radial flow exiting the diffuser 42
Streamlines in the innovative volute 43
Different configurations of the innovative volute System Efficiency 82 80 80 79,29 78,7 78,73 78,73 78 76,5 77 76,59 77,19 77,38 77,675 76 % 74 72 70 70,73 68 66 1 2 3 4 5 6 VOLUTES Total to Static Efficiency 44
Mesh Sensitivity of the ultimate systems Innovative Volute System Elements Stator 105946 Rotor 148990 Diffuser 155678 Volute 524670 Classic Volute System Elements Stator 107140 Rotor 156942 Diffuser 123270 Volute 620345 45
Performance Maps Classic Volute Innovative Volute ı ı ı ı ı ı ı,ı ı ı ı ı ı ı ı ı ı,ı ı ı ı ı ı ı ı ı ı ı ı ı,ı ı ı ı ı,ı ı ı ı ı ı ı ı ı ı ı ı ı,ı ı ı ı ı,ı ı ı 46
Turbine Speed [rpm] 180000 160000 140000 120000 100000 80000 60000 40000 20000 0 1000 3000 5000 7000 Data at Engine Speed [rpm] Turbine-Engine Matching Turbine with innovative volute Turbine with classic volute Compressor Pressure Ratio 2 1,9 1,8 1,7 1,6 1,5 1,4 Turbine with 1,3 innovative 1,2 volute 1,1 Turbine with classic 47 volute 1 0 2000 [rpm] 4000 6000 8000 Studio Numerico Data at Engine Sperimentale Speed di Turbomacchine di Piccola Potenza Turbine Power [kw] 14 12 10 8 6 4 2 0 Turbine with innovative volute Turbine with classic volute 0 2000 4000 6000 8000 [rpm] Data at Engine Speed
Effect of Turbine-Engine Matching 140 Engine Power (Maximum Load) 120 100 80 60 40 20 Power [kw] Engine speed [rpm] Turbine with innovative volute Original Turbine 0 1000 2000 3000 4000 5000 6000 7000 260 250 240 230 220 210 200 190 Turbine with innovative 180 volute 170 Original Turbine 160 1000 2000 3000 4000 5000 6000 7000 Torque [Nm] Engine Torque (Maximum Load) Engine speed [rpm] 48
THE DOWN-SIZED RADIAL TURBINE AS AN ORC EXPANDER The radial flow turbine from the well consolidated technology of down-sized turbochargers appears to be worthy of particular attention as an effective ORC expander An analysis to investigate the actual possibilities of adapting the radial flow turbine to a different application: Similarity criteria CFD validation Working Fluids considered: R245fa (higher molecular mass compared with air and steam) R134a Ammonia (lighter working fluid ) 49
The changes in both rotational speed and mass flow rate are mainly governed by the molecular mass of the new working fluid, while the turbine pressure ratio varies in accordance with the isentropic exponent and the isentropic efficiency Reference conditions: Air expansion Total-to-static pressure ratio β ORC γ = γ ORC ref 1 γ ORC 1 1 γ 1 η ref γ ORC ORC β 1 + 1 ref γ ref ηref Rotational speed N ORC = N ref SF ( m) ( m) ref ORC γ γ ORC ef T T ORC ref Expected mass flow rate m ORC = m ref SF 2 T T ref ORC p p ref ORC ( m) ( m) ref γ γ ORC ref ORC 50
Reference Conditions vpressure ratio = 4.1 vangular speed = 20000 rad/s. Rotor Geometry üexternal diameter = 47.8 mm üaxial displacement = 19.9 mm ü 9 swirled blades Three dimensional views of the radial flow turbine rotor The ORC condensing pressure was estimated at 40 C m / m ref The use of ammonia, even if more favourable in terms of cycle efficiency thanks to its higher superheating temperature, would imply an increase in rotor speed Operating Conditions and Properties of the several working fluids Air R245fa R134a Ammonia P 01 (bar) 4.1 6.315 28.64 47.66 T 01 (K) 473 439 455 699 P 2 (bar) 1.013 2.52 10.17 15.55 β tt 3.17 2.19 2.39 3.07 (expected) 1.0 3.01 12.18 6.84 ω (rad /s) 20000 7850 9500 29600 γ 1.392 1.077 1.153 1.234 Inlet Sound Speed (m/s) 435.3 161.3 180.6 635.0 µ (Pa s) 2.62 x 10-5 1.54 x 10-5 1.91 x 10-5 2.56 x 10-5 ν (m 2 /s) 8.92 x 10-5 6.28 x 10-5 2.16 x 10-5 18.0 x 10-5 51
Since the values of the mass flow delivered by the radial turbine should meet those resulting from the energy balance of the heat recovery boiler a further transformation has been also considered for this component by applying an appropriate scaling factor (SF) according to the well known relationships: SF m = ORC DORC Dref D 2 ORC 1 SF 2 2 ; ωorc ; H ωorcdorc For the T100-R245fa cases, a scaling factor of 1.525 was established the C30-R134a case implied a reduction in mass flow rate that was obtained by applying a scaling factor of 0.592 52
53 A simultaneous reduction of these parameters can be obtained by means of an iterative procedure that leads to a decrease in the angular velocity (and, therefore, in the rotor flow capacity) and in the pressure ratio, in accordance with the additional relationship: = γ γ β η γ γ 1 1 1 1 T R H
CFD analysis of the radial flow turbine The CFD analysis of the radial flow rotor was carried out in a threedimensional periodic sector of 40 with a tetrahedral mesh of 330000 elements (with an average volume of 1.89 x 10-2 mm 3 and an average edge size of 0.27 mm). Time marching solution of the transonic flow Reynolds stress (7 eqs.) model for the viscous flow 54
Averaged meridional contours of relative Mach number 55
A second step of this investigation was the off-design study that is discussed, as an example, for the C30-R245fa case that did not require any adjustment in terms of scaling factor. Characteristic curves of the radial flow rotor (Working fluid R245fa) It is worth-noting that the rotor appears to cover a wide range of mass flow rates, even if acceptable values of the efficiency can be found in a more restricted interval 56
Relative Mach number contours at mid-span (Working fluid R245fa) 57
Micro Gas Turbine Laboratory, Ce.S.M.A, Napoli, ITaly The MICRO GAS TURBINE CAPSTONE C30 Fossil and Biogas fuelling Matching with bottoming ORC 58
Thanks for your kind attention Rodolfo.bontempo@unina.it Marcello.manna@unina.it Raffaele.tuccillo@unina.it 59