Question? Leave a message!




Turbines for propulsion and power

Turbines for propulsion and power 6
New Developments in Combustion Technology Part II: Step change in efficiency Geo. A. Richards, Ph.D. National Energy Technology Laboratory U. S. Department of Energy 2014 PrincetonCEFRC Summer School On Combustion Course Length: 6 hrs June 2324, 2014 Presentation Identifier (Title or Location), Month 00, 2008 This presentation Updated, expanded from 2012 CEFRC lecture: – Inherent carbon capture: chemical looping combustion (Day 1) – Stepchange in generator efficiency: pressure gain combustion (Day 2) – Frontier approach (): making oxyfuel an efficiency advantage (Day 2) Sampling Diagnostics RDC Flow Pgain rig NETL ‹› The role of capture AND generator efficiency • A simple Define: heat/energy a = (kg CO produced) / (kg fuel burned ) 2 balance defines w = (separation work, Joules ) / (kg CO ) CO2 2 CO 2 the overall efficiency h with ov a carbon separation unit. h g Generator Efficiency • Reducing the Q = m DH f W 1 W Fuel Heat o Net penalty from Gross Input Generator Output Generator Carbon carbon capture Work Separation comes from Unit BOTH: – Decreasing w CO2 – Increasing h g Approx Ranges: (30 – 60) (610) ‹› Turbines for propulsion and power Almost any fuel – even coal, via integrated gasification combined cycle (IGCC). Shale gas revolution = more turbines “…The research firm (Forecast international) anticipates that 12,054 turbines with a value of 218 billion will be sold worldwide in the coming decade…” Siemens Moves Fueled by U.S. Gas, Wall Street Journal, May 8, 2014, pp. B2 IGCC plant under construction, Kemper County, Mississippi, USA ‹› History and Turbine Efficiency Gas turbine efficiency trend • Combined Cycle Gas Turbine 70 Efficiency is today + 61 (LHV). 60 50 Linear… • Efficiency gains have occurred 40 y = 0.5x 942 with steady progress in materials, 30 heat transfer, and system design. 20 – About +0.5 per year (right). 10 0 1960 1980 2000 2020 • Impressive performance is still Year wellbelow potential: Test rig for h = 1293/(1873) = 84 Carnot 1600C advanced aerothermal State of the art turbine inlet temperature cooling development • What can be done to “jump above” the line Sources: (1) Herzog, H., Unger, D. (1998) Comparative Study on Energy RD Perfrmance: Gas Turbine Case Study, Final Report for Central Research Institute of Electric Power Industry (CRIEPI), Figure B, pp. iii. , http://web.mit.edu/energylab/www/pubs/el98003a.pdf (2) Gas Turbine World 2012 Performance Specs, 28th ed Vol 42, No1, pp 31. ‹› Combined Cycle LHV Efficiency () A stepchange in efficiency • Turbine pressureratio and firing temperature influence the combined cycle efficiency. • A combined cycle exploits the heat rejected by the “hotter” turbine cycle to the “colder” steam cycle. • If you want a “stepchange” in efficiency, it is logical to identify the biggest losses and work on those. – Where is the biggest loss of thermodynamic availability (or exergy) – An interesting example for a cogen system (e.g. no steam “bottom”, but steam heat) is presented by Bejan et al. in the table. Exergy Destruction ( of fuel Component input) Combustion 30.0 Chamber Steam generator 7.3 Turbomachinery 3.5 Gas turbine 3.1 Bejan, A., Tsatsaronis, recuperator Moran, M. (1996) Thermal Design and Compressor 2.5 Optimization, John Wiley publishing, Table Overall 46.4 3.2, page 140. ‹› Pressure Gain Combustion A different cycle Constantvolume combustion products are at a significantly greater thermodynamic availability than constantpressure. 30 bar, 30 bar, 30 bar, 100 bar, 1600 K 600 K 600 K 2000 K Conventional steady combustion Pressure gain combustion (constant pressure) (constant volume) DU = Q DH = Q C DT = Q C DT = Q v cons V p cons P ‹› Pressure Gain Combustion A different cycle Constantvolume combustion products are at a significantly greater thermodynamic availability than constantpressure…..but what happens if the pressure is bled off to the ambient Unrestrained unrestrained expansion Returns to constant pressure availability –must capture the pressure gain to have a benefit. 30 bar, 30 bar, 30 bar, 100 bar, 1600 K 600 K 600 K 2000 K 30 bar, 1600 K Conventional steady combustion Pressure gain combustion (constant pressure) (constant volume) Noisy, but no DU = Q DH = Q benefit C DT = Q C DT = Q v cons V p cons P ‹› Pressure Gain Combustion Cycle DP 0 • Convention gas turbines combustion results in a C T pressure loss across the combustor (Brayton cycle) DP 0 C T • Pressure gain with constant volume combustion (Humphrey cycle) – Deflagration or detonation pressure wave increases pressure and peak temperatures at turbine inlet reduced entropy production during combustion. ‹› History • The idea of capturing the available energy from confined combustion (versus constant pressure) is well recognized. – Piston engines do this already. – Early gas turbines used the concept (Holzwarth “explosion” turbine). – Compound pistonturbines have been built and flown. – Constantvolume combustion eclipsed by easier improvements THYSSENHOLZWARTH OIL AND GAS TURBINES, Journal of the American Society for Naval Engineers Volume 34, Issue 3, pages 453–457, August 1922. . From the article: “……Holzwarth turbine working with a compression of 2.2 atmospheres and an explosion pressure of 17.3 atmospheres absolute….” Photo used with permission FIG. 8. – THE 500B.H.P. THYSSENHOLZWARTH OIL from Naval Engineers Journal TURBINE, WHICH MAY BE THE POWER OF THE FUTURE FOR MERCHANT SHIPS. Napier Nomad Engine (1950) Nomad photo credit: Kimble D. McCutcheon via the Aircraft Engine Historical Society. http://www.enginehistory.org/napier.htm ‹› Why is pressuregain appealing now PressureGain Combustion for Power Generation Michael Idelchik, Vice President of Advanced Technologies at GE Research… Research…Sept 2009 interview on Pulse Detonation for Technology Review published by MIT. “An existing turbine burns at constant pressure. With detonation, pressure is rising, and the total energy available for the turbine increases. We see the potential of 30 percent fuelefficiency improvement. Of course realization, including all the hardware around this process, would reduce this. I think it (efficiency gains) will be anywhere from 5 percent to 10 percent. That's percentage pointssay from 59 to 60 percent efficient to 65 percent efficient. We have other technology that will get us close to that but no other technology that can get so much at once. It's very revolutionary technology. The first application will definitely be landbasedit will be power generation at a naturalgas power plant. “ “If we can turn 5 pressure loss in a turbine into 5 pressure gain, it has the same impact as doubling the compression ratio” – Dr. Sam Mason, RollsRoyce (2008) Quotation courtesy Fred Schauer AFRL 2012 lecture Gas Turbine World Pequot Publishing Nov – Dec 2013 issue December 2013 Pulse detonation for 65 plant efficiency Page 20 ASME Mechanical Engineering Magazine, Image used with permission Image used with permission of Gas Turbine World ‹› Current Technology Approaches Resonant Pulsed Combustion † ( deflagration) † Envisioned as a canular arrangement Detonation or ‘Fast’ Deflagration G.E. Global Research Center NASA Glenn, 2005 2005 University of Cambridge, 2008 IUPUI/Purdue/LibertyWorks, 2009 Rotating Detonation DOE National Energy Technology Laboratory, 1993 Engine (NRL) Slide provide by Dan Paxson, NASA Glenn ‹› Pulse deflagration combustion Current RD at NASA, CambridgeWhittle Past Work at NETL ‹› Aerovalved Pressure Gain Combustor ‹› NETL Atmospheric Pressure Rig (1991) • Combustor constructed with standard pipe fittings. • Allows simple changes in inlet and tailpipe geometry. ‹› OneDimensional Modeling Characteristic Timescales • Divide combustor into three distinct zones. • Solve conservation equations of mass, momentum and energy. • Provides estimation of frequency and amplitude. ‹› OneDimensional Modeling Why not CFD Characteristic Timescales 1) Hint: this was 1990. 2) No theory for initial design scaling. Nice Computer • Divide combustor into three distinct zones. • Solve conservation equations of mass, momentum and energy. • Provides estimation of frequency and amplitude. ‹› Atmospheric Pressure Rig Data NG/Air f=0.82 • Baseline geometry (L =10 cm, L =60 cm). in ex • Resonant frequency 160 Hz ‹› Optimized Geometry • Maximum of 0.45 pressure gain achieved. ‹› NETL High Pressure Rig (1994) • NG/Air up to 11 atm. • Simple nonrectified design. ‹› High Pressure Results • Pressure controlled with a control valve on chamber exhaust. • Flow rates increased linearly with pressure. • Little effect of pressure when flowrates are scaled linearly with pressure. • Slight gain likely due to reduced frictional and heat losses. ‹› Some challenging problems • Predicting a design that will produce oscillations. – Progress in eliminating oscillations in premixed gas turbines makes this (relatively) easy. – But, at what operating condition • Developing an oscillating design that will also have a pressure gain. – Qualitative understanding, but no fundamental criterion, theory. – Modern CFD may be the enabler • Capturing the energy of the unsteady flow ‹› Capturing the pressuregain Courtesy R.J.Miller, Whittle Lab, Cambridge University • Time resolved experimental data. • Vortexinduced separation leads to loss in Phase II. • Work in progress: some configurations avoid the loss Color corresponds to pressure gain fraction (0.1 = 10 Imposed pressure gain) unsteady jet with 23 pressure gain Phase II Phase I Pressure Large rise in Transonic test facility gain in free loss as vortex stream exits Cause of loss : Vortex interacting with vane suction surface. ‹› Work at NASA • Demonstrated pressuregain and small turbine operation. • Simulations of pulse jet using commercial CFD. Liquid fueled. Automotive turbocharger “turbine” Reedvalve pulse combustor. Experimental results: Combustor pressure ratio 1.035 at temperature ratio 2.2 Simulation of pulsejet behavior –with NOx emissions and experimental validation. Paxson, D. , Dougherty, K. (2008). Operability of an Ejector Enhanced Pulse Combustor in a Gas Turbine Environment NASA/TM—2008215169 Graphics courtesy Dan Paxson, NASA Glenn ‹› Movie of pulse jet ejector (courtesy Dan Paxson, NASA) ‹› Pulse Detonation (Tubes) The detonation essentially “traps” the combustion behind the shock. Compared to pulse deflagration, much higher pressure gains are possible. This may be the only constructive applications of detonations Pow ‹› Typical Pulse Detonation Cycle 4. Detonation wave propagates at CJ velocity with 1. Fill coupled combustion wave 5. Detonation wave exits tube. Remaining gas at elevated T and P. 2. Upstream end closes 6. Rarefaction wave propagate upstream to assist with purging burned gases 3. Detonation initiated (DDT) 7. Exhaust complete ‹› Pulse Detonation for Propulsion Lab test • Pulse detonation tube concept has been extensively studied. • “Direct” propulsion: simple – No turbomachinery. Pulse Valve Propulsion tubes Assembly – Conventional recip. engine valve assembly for inlet. – Progressed to flight demonstration. • A key scientific issue: – Optimizing deflagration/detonation transition (DDT). Flight Demonstration Photos courtesy Fred Schauer, AFRL “…The applicability of a single combustion model to cover all the regimes of turbulent flames, which are encountered in The runup to detonation sets the length. confined highspeed flame transitioning to a detonation…..is Obstacles can accelerate – but add losses. yet to be established” Tangirala et al, Proc. Combustion Institute 30 (2005) 28172842 ‹› DARPA Vulcan Project • Integration in a turbine – humphrey cycle. • Combines the PDE with turbomachinery ‹› Multitube PDCTurbine Hybrid System • Eight tubes arranged in a canannular configuration coupled to a single stage axial turbine • Accumulated 144 minutes of PDC fired operation • Turbine performance was indistinguishable between steady flow operation and pulsed flow at 20 Hz per Some work supported from: NASA Constant Volume Combustion Cycle Engine Program tube Tangirala, V., Rasheed, A. and Dean, A.J., “Performance of a Pulse Detonation CombustorBased Hybrid Engine”, GT200728056, ASME Turbo Expo, Montreal, Canada, May 1417, 2007. ‹› An instructive question A detonation is started in tubes filled with fuel/air mixture, and travel left to right. One tube is open, and one is closed at the right end. Detonation The lower sketch shows what will become constant volume combustion. Both devices will release the same heat, burn the same fuel. The top device has significant mechanical energy as the detonation gets to the end. The bottom device has significant mechanical energy as the detonation gets to the end. What happens to that mechanical energy (with high availability) in both devices ‹› Brayton and Humphrey thermodynamic cycles The ideal Brayton cycle (e.g., gas turbine) compresses the mixture (12), adds combustion heat at constant pressure (23), extracts work from isentropic expansion (34) and then rejects heat from the working fluid at constant volume (41). For the Humphrey cycle, the heat addition occurs at constant volume, (2 – 3’ ). 3’ Combustor T 3 3 2 Turbine Compressor 2 4 4 1 1 S Assume you add the same quantity of heat for both cases. 23 Conventional constant pressure heat addition 23’ Constant volume heat addition Note that how you add the heat determines state 3. How does a detonation add the heat ‹› Particle history behind a detonation For more details: Law, C.K. (2006) Combustion Physics, Cambridge University Press, pp. 656 ff provides derivation of the flow field. H = head of expansion wave, D = detonation Particle H path Time 1 D H D Time 2 H D Time 3 X H D What does the Ts diagram P look like for the particle of Pressure P U Velocity U gas (lab frame) X ‹› Time Particle history behind a detonation For more details: Law, C.K. (2006) Combustion Physics, Cambridge University Press, pp. 656 ff provides derivation of the flow field. Particle H path Reaction Shock wave D ZND Detonation Structure (Zeldovich, von Neumann, Doring) 3 2a 2 Close –up view X Time 3 22a3 Detonation 23’ Constant pressure hear addition H D 2 3” Constantvolume heat addition T 3 P 2a Pressure P 3’ 3’’’ U Velocity U (lab frame) 2 X How you add the heat determines state 3 State 2 Assume all isentropic – S What is going on in here ‹› Note: 22a is shock compression. 2a3: T can indeed have a maximum with heat addition for (gamma)(1/2) Mach 1; subsonic Rayleigh curve. Time A few comments – and more details • Pressure gain combustion Interesting analogues turbine two ways: – Constant volume combustion: Humphrey cycle Reaction Shock wave Could we – Replace the conventional constant pressure call it a combustor with a “gas dynamic engine” – a new “Gas combined cycle. Dynamic Engine” • It is useful to treat the “gas dynamic engine” as a thermodynamic cycle (why). Compress Combust Expand – Literature citations, next slide. (to thrust) – Definitions: This phrase is not used in the open literature • FickettJacobs cycle: ignores the Combust shock structure; does not account for Expand compression before heat addition. (Turbine) Compress • ZND cycle: described here. – Be very careful with cycles: • Stagnation versus static properties. • Unsteady, adiabatic flow: total enthalpy is NOT constant. Note that the “gas dynamic engine” does not have 𝐷ℎ𝜕𝑝 𝑜 constant pressure combustion as in 𝜌 = +𝜌𝑞 +𝜌 (𝒇∙𝑼 ) a turbine; the analogy is not perfect 𝐷𝑡𝜕𝑡 ‹› Anderson, J.D., Modern Compressible Flow, McGraw Hill, (1982). pp. 161 eq. 6.43 also pp. 179 References for thermodynamic cycle analysis of detonation • These references do show a modest theoretical efficiency advantage (+ 1 3 points) to the detonation heat addition versus pure constant volume heat addition. • What practical issues may limit the actual advantage The gas at the left starts as a deflagration; there is a deflagration/detonation transition (1) Heiser, W. H., Pratt, D. T. (2002) Thermodynamic Cycle Analysis of Pulse Detonation Engines, AIAA J. Prop. Power Vol. 18, No.1, pp. 6876. (2) Kentfield, J. A. C. (2002) Fundamentals of Idealized Airbreathing Pulse Detonation Engines, AIAA J. Prop. Power Vol. 18, No.1, pp. 7783. (3) Winterberger, E., Shepherd, J. E., (2006). Thermodynamic Cycle Analysis for Propagating Detonations, AIAA J. Prop. Power, Vol. 22, No. 3. pp. 694 697. th (4) Vutthivithayarak, R., Braun, E. M. Lu, F. K., (2012). On Thermodynamic Cycles for Detonation Engines, 28 Int. Symp. on Shock Waves, Kontis, K., ed Springer Berlin Heidelberg pp. 287292. (5) Wu, Y., Ma, F., Yang, V. (2003). System Performance and Thermodynamic Cycle Analysis of Airbreathing Pulse Detonation Engines, AIAA J. Prop Power, Vol. 19, No. 4, pp. 556 567 (6) Nalim, M. R. (2002). Thermodynamic Limits of Work and Pressure Gain in Combustion and Evaporation Processes, AIAA Journal of Propulsion and Power, Vol. 18, No. 6 pp. 11761182. ‹› There are more details…. • Actual detonations don’t travel as plane waves. • Detonation progress is affected by transverse waves reflected from the tube walls…and other factors. • Thermodynamic model Use a computer Detonation progress Detonation cells look like this Substantially different P, T history here at the “triple point” Transverse wave Good background images – see Austin, J.M. (2003) The Role of Instability in Gaseous Detonation, Ph.D. thesis, California Institute of Technology, Pasadena, CA. ‹› A different approach • Wave Rotor Pressure Gain Combustor. • Developed by RollsRoyce Liberty Works, IUPUI, and Purdue Zucrow Lab • Tubes on a rotor spin past inlet and exit ports – containing combustion. • Does not require detonation – just rapid flame propagation. Indiana University Purdue University at Indianapolis Benefits: Almost steady air flow Steady torch ignition Balanced thrust load Challenges: Sealing Weight (for flight applications) The channels in the sketch at the right represent the tubes in the rotor at the left but “unwrapped” at a moment in time. The rotor revolution is driven by a motor at a speed selected to allow the flame to complete the channel combustion within a rotation. Successful test of wave rotor pressure gain combustor (2009) All photos and graphics: courtesy Dr. Phil Snyder (RR)and Professor Razi Nalim (Purdue) ‹› Understanding the flame propagation • Simulation development from basic studies (right) leads diagnosis of experiments for pressure rise and flame propagation. Rotor motion Snapshot of rotor tubes “unwrapped” (simulation). Study of flame propagation in a channel experiment Experiments and simulations used to establish design rotor with a moving entrance. speed and flow rate for fuel conversion and pressure rise. All photos and graphics: courtesy Dr. Phil Snyder (RR) and Professor Razi Nalim (Purdue) ‹› Rotating Detonation Wave Engine • Objective: detonation pressure rise with steady output. • Rotating detonation idea has been in the literature since 1950s. • Recent studies have demonstrated new potential for the concept. Higher pressure, steady flow to turbine Experiment at AFRL Courtesy Fred Schauer Rotating Detonation Inlet from compressor Simulation results courtesy K. Kailasanath, U. S.. Naval Research Laboratory see Kailasanath, K. (2011). The RotatingDetonation –Wave Engine Concept: A Brief Status Report ,AIAA 2011580. ‹› RDE photo Photo courtesy Scott Claflin Aerojet Rocketdyne ‹› End view From tests at AFRL Side View Movie Side Experiment at AFRL Courtesy Fred Schauer Rotation rate Movie 5000 Hz Simulation results courtesy K. Kailasanath, U. S.. Naval Research Laboratory End ‹› What is the thermodynamic history of gas particles How does energy transfer go into pressuregain “Unwrapped” view of annular simulation Image from Karnesky, J. and Schauer, F. (2013). Flowfield Characterization of a Rotating Detonation Engine, AIAA 20130278, courtesy Fred Schauer. See also Nordeen, C. A., Schwer, D., Schauer, F., Hoke, J., Cetegen, B., Barber, T. (2011). Thermodynamic Modeling of a Rotating Detonation Engine, AIAA 2011803 ‹› Some work in progress at NETL Simulations to evaluate turbine integration, effect of pressurized Pressurized test rig for stationary turbine operation (S. Escobar, I. Celik, West Virginia University). applications – emissions and integration with coal syngas (H2) , natural gas fuels. Mach number contours for Pout=0.11atm (top) and Pout=2atm (bottom) Schematic of simulation of interaction of RDE and integrated turbine exit. Graphic courtesy Todd Sidwell NETL ‹› Discussion/Thinking Questions: Pressuregain Combustion • What are the combustion research issues associated with different types of pressuregain • In your opinion, what is the greatest challenge to development of the pressure gain technology for power production combustor inside a gas turbine (i.e., not a simple cycle thrust device) ‹› Left blank to write notes from question slide Power generation classes: Turbines and Reciprocating Engines Reciprocating engines 75 Combined cycle turbines 50 Simple cycle turbines 25 Relative sizes scales Small turbines are approximate 1 10 100 MW Power • Where would a +5 efficiency boost get the most interest • What would the cost of development be for each class ‹› Efficiency Summary of Pressuregain combustion • Potential for an efficiency breakthrough. • Similar past concepts recognized; eclipsed by “conventional” improvements. • Successful demonstrations for direct propulsion tubes. • Promising work on turbine applications: – Pulse deflagration – Detonation tubes integrated with engine – Constant volume combustion waverotor – Rotating detonation wave combustor • Combustion and thermal science research needs discussed. ‹›