Academic excellence for business and the professions Energy storage flywheels for vehicle application Keith R Pullen, Professor of Energy Systems Department of Mechanical Engineering and Aeronautics School of Mathematics, Computer Sciences and Engineering
Presentation outline Why do we need flywheel energy storage? Overview of important design considerations Rotor constructions Low Cost Laminated Electric Flywheel - Low cost with inherent safety
Why do we need flywheel energy storage? Mitigate weaknesses in batteries Mitigate weaknesses in prime movers Poor part load efficiency of ic engines Ease transient response issues of ic engines Ease transient response issues of fuel cells For ic engines, can be downsized and extended onoff implemented
Weaknesses in batteries Battery life limited and affected by cycles and DOD Battery life and performance affected by temperature High power draw and charge decreases η and increases cooling demand particularly if battery capacity is small Cost is high for high power particularly if replacements are needed over life Lithium Battery for Nissan Leaf DOD is depth of charge Source: http://fclsw.com/archives/108 However, flywheels are not always a direct alternative to batteries but may work in tandem
How much energy is typically needed? Recovery of otherwise wasted braking energy 1500kg car to 30 mph = 270 kj 18000 kg loaded bus to 30mph = 3200 kj 270kJ can power a passenger car at 30 mph needing 2kW to overcome drag and rolling resistance for 135 sec or 1.78 km Energy requirements likely to be lower from powertrain system analysis
Basic principles of flywheels Similarly to E = ½ m veh v veh2, E = ½ I f ω 2 f where I f is moment of inertia If the rotor is a thin cylinder, I=m f r 2 f E = ½ m f r f2 (v f / r f ) 2 = ½ m f v 2 f Since v f >> v veh m f << m veh Velocity
Design considerations for flywheels Specify: Energy storage kj or kwh E = ½ I f (ω f,max2 -ω f,max2 ) Maximum power kw (independent of energy capacity!) Losses typically in % loss per minute Transmission (Mech) or electrical machine type Rotor construction Bearings and lubrication system More details given in: Pullen K and Dhand A. Mechanical and electrical flywheel hybrid technology to store energy in vehicles in Alternative fuels and advanced vehicle technologies for improved environmental performance, ed R. Folkson, Woodhead Publishing Limited, 2014
How is the energy transmitted in and out of the flywheel to and from the vehicle? Mechanically using a mechanical continuously variable transmission (CVT) [mkers] Electrically via electric motor-generator to traction electric motor-generator [ekers] Electromechanically via a magnetic gearbox and then CVT [Ricardo Kinergy System] Combination of electric motor-generator and epicyclic gearbox [emkers]
Mechanically using a CVT Advantage of keeping energy in mechanical form Can keep bearings in atmospheric conditions Disadvantage of needing a rotating seal or seals Disadvantage of mechanical link when idling http://www.racecar-engineering.com/articles/f1/flywheel-hybrid-systems-kers/
Electrically via M/G to traction M/G [ekers] Advantage of no rotating seal Can position anywhere Disadvantage of energy transformation http://www.busworld.org/articles/detail/2269
Electromechanically via a magnetic gearbox and then CVT Seal eliminated Speed reduced Additional cost http://www.ricardo.com/en-gb/news--media/pressreleases/news-releases1/2014 http://www.busworld.org/articles/detail/1273
Rotor construction E max can easily be calculated - simply moment of inertia mainly x ω max 2 Speed limited by allowable stress σ max, proportional to ρω 2 (ρ = density) Materials with best specific strength σ max /ρ theoretically better Material and type Density (kg/m 3 ) Design stress (MPa) Peripheral velocity (m/s) Rotor mass (kg) Aluminium cylinder T7075 2800 350 354 7.8 Titanium cylinder 4430 520 374 7.0 Steel cylinder 7800 720 304 10.6 Aluminium solid T7075 2800 350 550 6.5 Titanium solid 4430 520 582 5.8 Steel solid 7800 720 473 8.7 Glass Epoxy cylinder 2150 1100 715 1.9 Carbon-epoxy cylinder 1670 1500 938 1.1 * * For 489 kj storage
Can we use steel the first choice in low cost for automotive? As well as the apparent weight disadvantage a solid rotor would be a major hazard if it fails Solution is to laminate the rotor which additionally allows construction from sheet steel off the rolling mill A novel bolting technique avoids any stresses higher than would occur in a solid disc without any features
Rotor comparison for 500 kj usable (50% to full speed) Carbon fibre/steel 130 mm Steel laminate 650 MPa max design stress 76 mm 3 mm thick 12 mm 13 mm thick ø200 mm Rotor mass = 5.4 kg Rotor volume = 4.1 lit Speed = 60,000 rpm Periph. speed = 630 m/s ø170 mm Rotor mass = 13.7 kg Rotor volume = 1.7 lit Speed = 50,000 rpm Periph. speed = 445 m/s
Comparison Weight is 2.5 times CF rotor but less volume by factor of 2.3 A more compact CF rotor can be made by adding more steel at the expense of weight or higher speeds Based on the designs shown windage for CF rotor at a given speed will be 3 times higher or vacuum must be increased by factor for 3 for same loss Bearing losses and life similar for rolling element bearing with higher rotor weight and larger bearings offset by lower speed given viscous loss dominates. For 50 kw, 500 kj total weight with motor generator ~ 35 kg Not heavy!
Benefits of using the laminated steel approach Raw material lower cost and fully recyclable Fully scalable technology discs can be added or scaled in diameter for same tip speed Can be analysed relatively easily and with full confidence FEA, statically and dynamically with repeatable results Not affected by temperature Inherently safe by design No outgassing problems which occurs with non metallic components Viable for mass market automotive at all vehicle sizes
IDP8 - Disruptive technologies in low carbon vehicles II Project Title: Low Cost Laminated Electric Flywheel TP101571 Project co-funded by:
Low Cost Laminated Electric Flywheel TP101571 Project overview Target to develop a low-cost Flywheel Energy Storage System (FESS) for mass production targeting significant reduction in CO 2. Prototype designed by Dynamic Boosting Systems to Nissan developed specification Tata steel to develop flywheel steels and material characterisation City University to develop analysis tools and burst test samples capturing the results using an ultra high speed camera
Burst analysis FEA emulation of the result of Hagg and Sankey tests Ref: Hagg, A.C and Sankey, G.O., The containment of Disk Burst Fragments by Cylindrical Shells, ASME Paper No.73-WA-Pwr-2, 1973.
Burst analysis Simulation of a laminated rotor
Burst testing Several samples will be tested Failure event to be captured by Photron High speed camera Containment to be demonstrated Event to be replicated in dynamic FEA City University s high speed machine laboratory (Double concrete room with blast doors)
Thank you for listening Questions Keith R Pullen, Professor of Energy Systems k.pullen@city.ac.uk