For a given type and displacement of engine, it is a rule of thumb that maximum potential torque is a fixed amount – for example, all 600 cc four-cylinder, four-stroke engines with sufficient attention given to gas-exchange processes and friction will produce the same level of torque. Power depends on the torque and the speed at which it can be achieved. The same applies to electric machines, as the potential maximum torque is fundamentally a function of the electrical design choices and the construction of the motor, and the power depends at what speed this torque can be maintained. As with engines, there is a strong incentive to run an electric motor at high speed.

There are however a number of barriers to running at ever-increasing speeds, although they are very different from those in a reciprocating engine. There are no strong second-order forces and couples as we find in an engine, and fewer causes of excitation to provoke a strong resonant response.

The main enemy is centrifugal force. For a permanent magnet motor, it is not a simple task to keep the magnets in place at very high speed. The motors used on the electric machines in the exhaust heat recovery motors, driven by the exhaust turbine in Formula One, run at speeds up to 100,000 rpm. It is certainly not enough to simply rely on the magnetic forces to keep the magnets in place, and the mechanical strength of magnets leaves much to be desired – even if we were to be able to ‘stick’ them to the rotor, they are unlikely to remain intact.

It is necessary therefore to restrain the magnets radially to prevent both loss and breakage. In low- and medium-speed motors this is sometimes done with glass-fibre reinforced polymers, but for high-speed motors, higher strength composites such as carbon or PBO are used. The limitation is then the strength of the composite sleeve.

It is also important to have high-speed rotors properly balanced to a high degree of accuracy in order to prevent early bearing failures and to minimise the transmission of vibration to adjacent components. When dealing with very high-speed rotors, the machinery for balancing them can be complex.

Another problem can be maintaining the rotor magnets within their operating temperature range. At relatively modest temperatures the magnets can become demagnetised, and this is not a property that returns when temperatures fall to more desirable levels. Some companies resort to liquid or forced-air cooling of rotors in order to keep magnet materials sufficiently cool.

Written by Wayne Ward

All sorts of schemes were considered, from conventional electric hybrids to flywheels and pneumatic systems. Eventually, everyone opted to run an electric system, although more than one team seriously investigated alternatives.

High-speed flywheels often have high energy densities, but have to run in a vacuum if frictional losses aren’t to be too great. The problem comes in transferring the energy out of the vacuum chamber and to a device such as a constantly variable transmission, which allows the energy to be smoothly deployed for propulsion. There are a few alternatives to this. The two commercially successful options, both initially developed for motorsport, are now both finding a wide range of applications including racing, advanced prototypes for mainstream automotive companies, and notably on public transport.

One solution is a special type of shaft seal, developed by a British company based at the Silverstone race circuit. Although it develops hybrid systems, it might be argued that its ingenious seal is its real treasure. It is simple in principle and allows the vacuum flywheel chamber to suffer very little leakage, so the vacuum is easily maintained with a very small pump that needs to run only rarely.

The second option, which has been used very successfully in endurance racing, is to run the flywheel as an integral part of an electric motor. Energy is stored and recovered electrically, therefore a high-speed rotating shaft does not need to cross the boundary between the vacuum chamber and the outside world.

There are other, less well-developed options that have yet to be successfully deployed in racing. One of these is magnetic gearing, where there is no need for a physical coupling between the shaft in the vacuum chamber and the output shaft, which runs in ambient conditions.

There is one definite advantages to flywheel hybrid systems, whether they are fully mechanical or electro-mechanical, and that is the fact that the storage medium is not a battery. The flywheels used in motor racing and for automotive hybrid development are made from simple materials that are well understood, namely carbon fibre-reinforced polymer composites and steel.

What is more, they are not prey to the various maladies that batteries are known to suffer from, such as finite lifetime, an inability to work over the full charge-discharge range without suffering shortened life or the ‘thermal runaway’ events that sometimes make the news when a lithium-ion battery goes wrong. On the other hand, with electrical systems nobody needs a patent to exploit them, and the technology is well understood by many companies.

Written by Wayne Ward

First there is a motor/generator unit, which is used either as a propulsion motor to augment the engine’s torque or, in the case of purely electric vehicles, to provide all the motive effort. Then there is a battery, or possibly a supercapacitor, which is where the electrical energy is stored. It is possible to connect a motor to a battery without any other components, but this is generally not the case, especially not in automotive propulsion, where we require precise control. The third component, generally called the ‘power electronics’, carries out a number of very important functions.

The battery supplies a voltage and a direct current (dc); the voltage varies depending on a number of factors, as does the current. It is highly likely though that the propulsion motor is a multi-phase alternating current (ac) device, which is not compatible with a dc supply, so one of the important roles of the power electronics is to turn the ac output of the motor/generator unit, when it is acting as a generator, into dc, which can be fed to the battery. When the motor/generator is acting as a motor, the power electronics acts to convert the dc output of the battery to ac, switching it between phases very precisely so that the motor propels the vehicle forward with peak efficiency.

In the Formula One cars of 2014, the electric motors running at the same speed as the turbocharger; the speed at which the switching of current takes place is incredible, and any inaccuracy in its timing can mean a serious loss of power conversion efficiency. Where dc machinery is being run from an ac supply in a workshop, the device responsible for the conversion is referred to as a rectifier, and where an ac device is run from a battery, it is called an inverter. The power electronics combines both of these roles and has to switch very swiftly from one to the other.

Another important job for the power electronics is to supply from the battery an almost constant voltage to a number of other electronic components. This process of converting a high dc voltage to a lower one allows us to dispose of the traditional 12 V battery and run everything electronic on the car from the main propulsion battery. Components that we might need to run include starter motors, electrically powered pumps, lights and injectors, among others.

Written by Wayne Ward

Is there a real gain to be had from r&d in motorsport? Does it matter outside of the company in question? Well, yes, and people in positions of influence within governments do take notice. The British system of government, like many others, has an upper and lower chamber. The upper chamber (the House of Lords) is now populated by people who, in the main, represent political parties but who are proposed rather than elected. Two such people are Lord Rooker and Lord Astor of Heverbrook, and during a debate on the armed forces in March 2014, Lord Rooker asked a question on the UK’s use of biofuels in defence. In the question, he stated that most of the fuel and materials development in Britain stems from motorsport. In replying on behalf of the government, Lord Astor assured him that the government is working closely with the motorsport industry, and went on to give examples of our good work.

Fuels are an excellent example of how, as an industry or a business, motorsport can make itself relevant to the outside world. We can see that biofuels are going to form an important element of our liquid fuel supply in the future, so we should make rules that encourage motorsport to be early adopters of such fuels. There is no doubt that motorsport can be an extremely cost-effective way to undertake the rapid development of engineering concepts. Developing engines and fuel systems suited to the use of, for example, bio-butanol might be seen by the wider world as something very worthwhile, and technical partnerships might be forged that benefit everyone involved.

As far as the armed forces are concerned though, we aren’t going to be of much use for developing fuels for powering jet fighters, but there still remain myriad uses for piston engines – both small and large – and there is a very important role for companies with a motorsport background. What we should excel at is carrying out development for the wider automotive industry, and it seems that we are now waking up to this fact. We should take a long look at the direction roadcar fuel development is going in and get ahead of the game. If we are late to the party then we are an irrelevance – we have no role in development and derive little benefit from adopting such technologies. If we are the earliest adopters though, and can present multiple options, we assure our financial future far more easily.

Motorsport companies are always looking to do something new, especially if it can give them some advantage, but there needs to be an incentive to invest in development. We need to take a longer-term view of what future passenger car fuel trends are and move in that direction.

Written by Wayne Ward

At the moment, we are on a cusp. There is growing acceptance of hybrid powertrains, from hybrid city/commuter cars to hypercars such as the McLaren P1. Motorsport has long been aware of these powertrains’ benefits to vehicles with a need for heavy braking and rapid acceleration. From touring cars, through Prototypes to Formula One, hybrid propulsion has been embraced wherever it has been allowed. The chance to recycle some energy for propulsion that would otherwise have been lost to the atmosphere in the form of heat is an attractive proposition. At the moment, the electric hybrid, using a motor-generator unit combined with a battery is the most common choice, but there is at least one prominent alternative to this.

Beyond hybrids there is the fully electric vehicle, where all the motive power is supplied from an energy storage device. Fully electric racing is already with us and is on the rise, while fully electric roadcars are becoming a more common sight each year, at least in Europe.

What is also with us, if we are users or designers of electric hybrid technology or fully electric drivetrains, is the battery. Wherever energy is converted, it is never 100% efficient, and the process of charging and discharging batteries creates heat. Where the rate of energy conversion is high, there may be a need for cooling, and this can be as simple as ducting cooling air from the front of the vehicle where pressure is highest, through the battery to a low-pressure area to create a through-flow of air while the vehicle is in motion.

This method tends to work well only at high speed. Forcing the air to flow by using a fan allows more control of cooling, including being able to cool at low vehicle speeds. However, in order to remove serious quantities of heat, we probably need to look towards liquid cooling, simply in order to increase the mass flow of the coolant. Air is also lacking in specific heat capacity compared to liquids. The result is that liquid coolants are often more than 1000 times more effective in terms of removing heat per unit volume than air. Water is particularly effective as a coolant, but is generally not what you might want to have in contact with high-voltage batteries, so there are oils specifically designed for cooling electrical equipment such as transformers.

Although we still need to reject heat from the liquid coolant to the air via a cooler, in terms of packaging, liquid cooling of batteries can be a far more attractive proposition than air cooling. However, the practicalities in terms of battery design will be more complex than for an air-cooled installation. Leaks of any liquids are not generally welcomed, so the battery container needs to be sealed against them. There is also the issue of needing multiple electrical connections to a battery for both energy transmission and monitoring the ‘health’ of the battery. The designer will also need to consider the flow of the coolant through the battery to ensure that no cells are so poorly cooled that they are likely to overheat. In the case of lithium-ion batteries, the consequences of this can be spectacularly destructive.

Written by Wayne Ward

However, the new Le Mans regulations have, as is typical, been written in a way that is much more liberal, leaving the users free to decide what suits their application best. Although there are various choices as to the amount of energy that may be harvested during a single lap (up to 8 MJ per lap), however, there is no limit on the rate at which the energy may be harvested or deployed. Of course, there are some sensible limits as to what we might want to do with the motor: it is pointless designing a motor capable of harvesting or deploying energy at a rate that would overwhelm the tyres. Even in the case that the tyres can cope with a lot of additional torque, perhaps the gearbox would be made very much heavier to cope.

The Le Mans regulations also allow recovery of heat energy, which most easily applies to those teams using turbocharged engines. While there have been schemes for recovery of heat energy from naturally aspirated engines, those using a turbocharger can use the same turbine-driven motor/generator technology that will be used in Formula One. Porsche is thought to be using heat energy recovery on its adventurous new LM P1 power unit, which uses a V4 gasoline engine at its heart.

So, even though the Le Mans regulations are less constricting in terms of the design of the motor, they require more thought at the start of the process than in Formula One – a mistake in calculating a sensible power output can result in the motor, power electronics and energy storage being too heavy and difficult to package. Those choosing very high motor powers will force themselves into using either very high voltages or very high currents. High voltages pose problems for insulation in the motor, and high currents cause problems with heat. Using high current densities – carrying large currents per unit cross-sectional area of conductor, in order to keep the motor physically small – adversely affects efficiency. In conductors with high current density, heat losses are large and the cooling requirements are increased.

It is probable that all the Le Mans hybrid motors will have to be liquid cooled. Air-cooled motors can only carry comparatively low current densities and are therefore physically much larger for an identical power output and operating speed.

Written by Wayne Ward

There is an increasing number of electric roadcars available (in Europe at least), some of which are very agreeable, with nice styling, good performance and reasonable driving range. They are finally becoming desirable and practical at the same time. Electric motorsport is also enjoying a steady rise. The motorcycle series TT-Zero and TTX-GP have enjoyed increasing popularity and success in recent years, and electric car racing is now firmly in the spotlight, with Formula E now an official FIA series with manufacturer backing from Renault.

Away from pure electric racing, of the various hybrid technologies available, electric hybrids have proven to be the most popular solution. While most teams and car manufacturers choose to use batteries as the method of storing energy, some have chosen capacitors. Even in non-hybrid motorsport, some teams choose to replace the car battery with a capacitor.

The most high-profile user of capacitors as stores of significant amounts of energy is Toyota, which has rejected the battery in favour of ‘supercapacitors’ on its hybrid LM P1 car for Le Mans and the World Endurance Championship. The Toyota hybrid system delivers an incredible 300 bhp (224 kW) to augment the 530 hp produced by the engine – almost four times the power delivered by the Formula One systems in 2013. In the Toyota, the proportion of the power-unit performance which is electric is even greater than will be the case in Formula One from 2014 onwards. The capacitor was chosen because of its ability to charge and discharge at a higher rate than batteries can manage for the same mass. Toyota had previous experience of supercapacitor storage for racing hybrid systems, having used them in a GT car application with success in 2007.

There is evidence to suggest that Red Bull has used supercapacitors as the energy storage on its Formula One car, with the capacitor packs mounted flat on the floor of the car. There is a natural link between Red Bull and supercapacitors. Renault, the engine supplier for Red Bull, is involved in many kinds of racing and runs several one-make series. The Formula Renault 2.0 and 3.5 cars have both rejected conventional batteries and replaced these with supercapacitors; in the Formula Renault one-make cars, the capacitor supplies only the power necessary for the vehicle electrical systems, rather than providing propulsion.

Outside motorsport, capacitors for hybrid applications are becoming more popular, despite their higher initial costs because, over the life of the vehicle, capacitors are said to prove less expensive. Battery reliability and life are still in question: battery life in a pure electric passenger car can be between four and eight years. At this point, it is often uneconomical to replace the battery, so some electric cars are supplied with batteries on a lease basis.

Supercapacitors have perhaps lacked the investment that batteries have enjoyed, and as a result are at a lower level of development. They are consequently likely to improve at a faster rate than batteries and, given their current competitiveness (no pun intended), we should not be surprised to see them become used more widely for automotive and motorsport applications. 

Written by Wayne Ward

Batteries made from lithium-ion fuel cells are the most popular way to store the energy on electric cars and indeed on IC engine-electric hybrids, although there are those who prefer supercapacitors. The other alternative, especially for pure electric motorsport, is the fuel cell. These can perhaps be imagined as a cross between an engine and a battery, although they aren’t especially closely linked to either. The similarity to an engine is that while we still flow fuel into a fuel cell – in this case hydrogen – the output is electrical rather than kinetic energy, which is similar to a battery. The conversion to a more useful kinetic energy is achieved via an electric motor.

The use of fuel cells has been limited to date, although there has been a notably impressive Land Speed Record car, the Buckeye Bullet, which uses fuel cells and which holds the electric-powered record at more than 303 mph (487 kph) for the flying kilometre. It was mentioned in a previous RET-Monitor article, which also mentioned the Green-GT, another hydrogen fuel-cell powered car that unfortunately proved to be an no-show at Le Mans and has not been heard of since.

One arena where fuel cell-powered cars compete with other electric and IC-engine cars though is Formula Student. Delft University, for example, has competed in Formula Student with great credit, and has used its experience to create its sixth car, the Forze VI. The aim of the car is twofold – to compete in the Caterham Cup against a field of IC-engined cars and to be the fastest hydrogen-powered car around the famous Nurburgring Nordschliefe circuit.

That would be no mean feat. An Aston Martin Rapide with an engine configured to run on either gasoline or hydrogen ran a full lap of the Nurburgring on hydrogen in 2013. The estimated power output of the Aston was about 500 hp (375 kW). At 260 hp (190 kW), the Forze VI car has around half of this power but also weighs around half of the Aston’s 1600 kg and is much smaller, so may have less drag. Delft estimates the top speed to be about 137 mph (220 kph).

The beauty of a fuel cell vehicle is the energy storage density, although hydrocarbon fuels have a large advantage over batteries for this very reason. I have yet to see a car that can go 600 miles on batteries, but I see hundreds of gasolines and diesels every day that will do just that. And if there is a 600-mile battery car out there somewhere then I bet it wouldn’t be able to put in a repeat performance five minutes after draining its batteries. The Forze VI car has a fuel tank that  holds only 3kg of hydrogen. Although the tank itself and the associated plumbing may be heavy, I also bet it would be lighter than a comparable battery.

Written by Wayne Ward

Both Mercedes and Renault powertrains have been photographed numerous times (the photos can be found easily on the internet) and show clearly that the KERS motors are mounted to the front of the engine close to the crankshaft, indicating that they are probably geared directly to a gear on the crankshaft or through a single idler. It would certainly be the preference of the chassis engineers to have the motor mass close to the centreline of the car and as low down as possible, which illustrates the close relationship between the chassis and engine design departments when the systems were conceived.

There are also photos on the internet of the proposed Honda installation, although with its withdrawal at the end of 2008, a KERS-equipped Honda Formula One engine never competed. Honda had examined both flywheel and electric hybrids, eventually choosing the electric hybrid route, possibly based on its use of this technology in some of its roadcars.

Honda’s approach to the mounting of the motor was completely different from that of Mercedes and Renault. It chose to take a drive from the timing gears on the left bank of the engine and, using a number of idler gears, it had the motor mounted out to the left side of the engine. This was a highly unusual approach and was clearly based on a different logic from the Mercedes and Renault installations. It is thought that Ferrari also has its motor  motor/generator mounted in front of the engine. There are a number of photographs on the internet showing a driveshaft protruding from the front of the engine.

The other ‘headache’ that hybrid systems present is where to place the battery and power electronics. Mercedes has been the happiest to talk about its KERS system and its development, showing detailed pictures in Race Engine Technology magazine issue 67 (Dec 2012/Jan 2013). Initially its 2009 KERS system had separate enclosures for power electronics and battery, one mounted each side of the car, requiring a cross-car cable to transfer power.

By 2011, Mercedes had developed a neat ‘one box’ solution, with power electronics and battery in a single enclosure, doing away with the need for the cross-car cable. The single-box solution is neat for a number of reasons, notably the fact that only a single recess and a single set of mounting points are required in the chassis to retain these components. Honda released photos (again which can be found using a web search) of its proposed car installation, showing the power electronics mounted in the sidepod to the left of the driver, and the battery pack being mounted below the driver’s legs.

The single-box scheme also makes component cooling slightly simpler. At one race in 2013, there was TV footage of a Renault KERS module being removed, apparently full of fluid sloshing about. This shows that the battery, and possibly the power electronics, are liquid cooled. If two separate boxes require cooling, there are extra hoses and fittings to consider and an increased number of seals, each of which could potentially cause a leak. Reliability is very much a case of eliminating risks, as well as solving problems as they occur.

Written by Wayne Ward

This type of hybrid still uses an internal combustion engine, but uses it only to drive a generator; the engine itself has no direct mechanical connection to the wheels. It has been used widely and successfully in passenger applications for decades, mainly in diesel-electric trains, and this type of powertrain has much to recommend it.

Most of us will be primarily involved with race engines, although a small minority of us (myself included) have had periods in our careers during which our main occupation has been hybrid systems. The implications of using a series hybrid approach have profound implications on the design of the engine that drives it. Logically, we would use the engine only at full throttle at a single speed; we can design the engine to be optimised at this single speed, at which it would run after a short warm-up period. This should mean that the engine becomes much more efficient as a whole – everything would be optimised for this one speed. When we don’t need the engine we would simply switch it off.

The closest approach racing has had to this was the Williams FW15C CVT car. This wasn’t a hybrid but used a constantly variable transmission. If you see footage of this car in a straight-line test, it sounds very strange – there are no gear changes and no change in engine tone, just an incessant fixed sound under acceleration. A series hybrid powertrain would have the same ethos, except there would presumably be periods where the engine could be turned off completely. The single-speed engine has many advantages: it makes estimating fatigue life for each component much easier, and therefore it should be easier to arrive at an optimum design.

The electric side of the powertrain, assuming that the first series hybrid is an electric system, replaces the electrical components of the parallel hybrid and the car’s usual transmission. There is a large generator attached to the engine, linked to a motor or motors that drive the wheels. In an optimal system, we might see an independently controlled motor driving each wheel. This is certainly what we would want in a passenger car if safety is our primary concern.

If we want motor racing to remain a visual spectacle though, we might choose to have only the rear wheels driven. The battery is still required as a reservoir of energy, with the size depending on how much captured energy from braking events we would like to store. The maximum power that the battery can discharge can be added to the maximum power that the engine can develop. A large battery also requires large propulsion motors to handle the battery’s extra power. There is a trade-off between the extra mass of the battery/motors and the fuel that the system allows us to save, and the extra mass they add to the vehicle and the energy involved in braking and accelerating this additional mass.

Such a series hybrid system is an intriguing engineering challenge – one day we might drive such a passenger car. Will we ever see it in racing though? Perhaps even the technologically adventurous ACO isn’t that brave – it might ruin the visual and aural spectacle of racing irrevocably.

Written by Wayne Ward

We can recover significant quantities of braking and exhaust energy by using hybrid technology, of which there are several types used in racing. The most common is the electric hybrid, which is also the most common type found in passenger car applications. This type of hybrid system comprises three main components – one or more motor/generators, a battery and a power electronics module. In a conventional electric hybrid which recovers braking energy, the motor/generators are usually driven by the engine or transmission.

The question of where to site the battery and power electronics on the car or motorcycle is one which needs to be considered carefully; the choice is not simply one of weight distribution alone. As detailed in issue 67 of Race Engine Technology recently, when the kinetic energy recovery systems (KERS) were introduced to Formula One in 2009, Mercedes had distinct power electronics and battery modules, with one situated on either side of the car.

For 2010 there was no KERS in Formula One, but from 2011 onwards the systems have been used again. Mercedes chose to combine the power electronics and battery into a single module. If it is possible within the packaging constraints of the car to site this close to the centreline of the vehicle, the inertia relative to the roll axis of the car is reduced compared to the two-module approach. There may also be a marginal mass reduction owing to the more efficient ‘semi-detached’ module construction, eliminating one wall from the enclosures.

However, a more significant mass reduction and packaging advantage of the combined power electronics and battery module is the elimination of the cable linking the two units. Cables carrying high voltage and current are generally made of copper or aluminium, and require both electrical insulation and metallic shielding. The shielding is to eliminate electromagnetic interference (EMI); if the EMI shielding breaks down, changes in current in the cable affect the magnetic field in the cable’s vicinity and can have a significant effect on the operation of the other electronic systems; this may have a much wider effect than problems with the KERS itself.

So, amalgamating the power electronics and battery into a single module can have a positive effect on weight distribution, KERS system mass and reliability. Reducing the number of physical connections between the systems is also desirable – there is less to go wrong and there are two fewer connections that the KERS technician must make when installing the system.

There are some disadvantages though to the single-box approach. In effect, we need to have more hardware available, as a faulty battery or power electronics module in a combined unit cannot easily be replaced in the field. If a battery is suspected of being faulty, the power electronics are removed by default as they are housed in the same box, and vice versa. Batteries are also classed as hazardous freight, so it becomes difficult to get a faulty power electronics module back to the factory quickly for investigation if there is a problem. 

Written by Wayne Ward

The car-makers and various governments are taking hydrogen seriously as a fuel, with some manufacturers supplying limited production cars for road use. One of these, the BMW Hydrogen 7, has a dual-fuel capability that allows it to burn hydrogen and gasoline. Many other car-makers are intent on supplying vehicles with hydrogen fuel cells.

Such developments have not gone unnoticed by those involved in motor racing, and recently an Aston Martin Rapide raced in a 24 Hour GT race at Germany’s famous Nurburgring circuit with a dual-fuel system, allowing the V12 internal combustion engine to burn either gasoline, hydrogen or a mixture of the two. The car in question had a hydrogen storage capacity of 3.2 kg, stored in multiple specially constructed tanks comprising a 15 mm thick aluminium tank with what is described as a composite outer ‘wrapping’, but which is likely to be a filament-wound structure which is typical of many high-pressure vessels used in motorsport.

In terms of energy density, hydrogen’s is high; usually this property is known as the ‘lower heating value’. Where each kilo of gasoline contains around 44 MJ of energy, hydrogen contains almost three times this amount at 120 MJ. However, owing to its low density, hydrogen suffers relative to gasoline on account of its very low ‘volumetric energy density’ – a litre of hydrogen compressed to 700 atmospheres contains around 5.6 MJ of energy, whereas gasoline contains about 36 MJ. Also, the storage pressure being worked to as a provisional standard for automotive applications is 700 bar.

Consequently, to produce a comparable performance, hydrogen needs to be burned at a prodigious rate. The Aston Martin racer stores its hydrogen at 350 bar – about half the usual automotive storage pressure – so its 3.2 kg of hydrogen takes up more volume than it would in a passenger car application. This 3.2 kg contains the same energy as 8.7 kg of gasoline, and so appears quite attractive. However, owing to the very low density of hydrogen, the volume of charge required to enter the cylinders of the race engine required the use of twin turbochargers when using hydrogen alone. The weight penalty of the additional hardware required for the hydrogen-powered racer was about 100 kg, which would present a serious deficit in terms of lap time.

To compete with gasoline, cars will need much more specialised technology if we are to get the vehicle performance we have become accustomed to, even in a passenger car. If hydrogen is to compete with gasoline – even on a dual-fuel basis in racing – it is clear that much needs to be done to decrease the weight penalty of the hydrogen storage, and engines may well need to be boosted, as is the case with the GT Aston.

It is clear then that to replace gasoline, even on a limited scale, there is much that needs to happen, not only in terms of engine technology and on-car storage, but also in terms of fuel production and storage infrastructure. The fact that motorsport is helping to develop the vehicle side of the technology will hopefully advance our learning about how to use hydrogen at a faster rate than the mainstream automotive world can achieve alone.

Fig. 1 - The 3.2 kg of hydrogen stored on the Aston GT car is in four special tanks (Courtesy of Alset Global)

Fig. 2 - The hydrogen-powered Aston GT car in action at the famous Nurburgring circuit (Courtesy of Alset Global)

Written by Wayne Ward

Dubbed KERS (kinetic energy recovery system), it was scrapped for 2010, but has been a permanent feature since. In 2014, hybrid technologies will be a key part of the Formula One power unit, with double the power of the current systems, greatly increased energy storage allowed and a much higher duty cycle. If a driver has a KERS system failure in 2013, the penalty is serious but perhaps not disastrous. In 2014, with a great deal of the power unit’s performance coming from electric motors, the loss of the ERS (energy recovery system) in 2014 will be much more serious.

With the change in power and duty cycle for the electrical machinery in the power unit, the battery will have a very important part to play. As has been discussed previously, lithium ion cells are the usual choice for any modern electrical energy storage application. Supercapacitors have been used, but remain a niche choice.

The Mercedes KERS system was featured in RET issue 67 (December 2012/January 2013), and the pictures show that the battery and power electronics are in a single module. The equivalent battery for 2014 will be considerably larger, so they will present much more of a challenge in terms of packaging.

The amount of stored energy allowed is 4 MJ, ten times that which can be stored under the 2013 rules. If the current KERS cells are limited in terms of volume by their energy capacity, and if the same cells are used, the battery volume will be ten times the size. The larger battery and the new power electronics, combined with the complexity of the new turbocharged power unit, with its attendant charge coolers, will make packaging all of this hardware a real challenge for those designing the new generation of Formula One cars.

Lithium ion cell chemistry is the subject of rapid development. With the automotive industry pushing hard to produce long-lived cells with higher energy and power densities, racing is able to benefit from and contribute to such development. In developing cell technology, there is generally a trade-off between energy storage density and power density. With a doubling of power output and a tenfold increase in energy storage capacity, the energy density of the cell will become more important compared to the cells used in 2013.

The increasing focus on energy density may give the cells an easier life; with less incentive to charge and discharge a small cell mass at a high rate, cell lifetime may improve. Cell charge capacity is known to suffer when charged at high rates, which means that using the cells at the higher end of their power capacity will lead to a fall in their storage capacity over time.

If we can rely on the FIA to maintain a fixed set of rules for Formula One ERS in terms of the key parameters, cell development will lead to ever more compact and reliable batteries.

Written by Wayne Ward

The motor’s architecture is a variation on what is known as an axial flux, or axial gap, machine. Such motors have a notably small ratio of axial length to diameter, and are sometimes described as ‘flat’ or ‘pancake’ motors because of this aspect ratio. They develop high torque, but their speeds are limited owing to the rotor’s construction. This is not a particular problem though for ‘wheel motors’ that provide propulsion directly to each driven wheel, as they don’t need a large speed reduction. Axial flux motors also have a higher proportion of electromagnetically active material, so the percentage of winding mass which develops no torque, but which still generates losses, is lower in axial flux machines.

Most motors for hybrid systems are of the radial flux type, where (usually) a rotor fitted with permanent magnets spins within a wound stator. There is a small gap between the outer surface of the rotor and the inner diameter of the stator, so these motors are known as radial gap or radial flux machines, and can develop high power owing to their ability to run to high speeds. The output of the new Formula One engines for 2014 will be augmented by the use of sophisticated energy recovery systems; one of the electric motor/generator units will spin at the same speed as the turbocharger shaft, which might be as high as 125,000 rpm.

So, even though the casual observer might think that all electric motors used in motorsport might be very similar, there are two distinct ‘branches’ – axial flux motors that drive the wheels directly or with only a small speed reduction, as we might find in pure electric racing vehicles; and radial flux motors, which will be capable of higher speeds and which lend themselves to hybrid applications such as KERS, where they drive and take power from the engine or the transmission. Of course, dividing lines are not always absolute –  there are, for example, high-torque, relatively low-speed radial flux machines that have been used for pure electric propulsion, although these are in a minority.

It is likely that these two branches will continue to co-exist, as they use many of the same design strategies in high-performance applications. For instance, liquid-cooled stators, which allow high current densities in the motor windings, have been used in both axial and radial flux motors. Material choices are again similar – high-strength permanent magnets, high-conductivity coppers, and stator materials that saturate at high levels of magnetic flux, are required for both kinds of motor if optimum performance is to be achieved.

Written by Wayne Ward

In terms of the exact technologies applied to racing hybrid systems, however, there is a middle ground. In one corner, we have electric hybrid systems, of the type used on existing production hybrid passenger cars. In such a system, current generated by a motor under braking is used to charge a battery which, when discharged, flows current back through the motor, generating torque and adding to the performance of the powertrain. In the other corner there are mechanical systems, where energy ‘harvested’ under braking is stored as kinetic energy in a flywheel. Flywheel energy storage is not new, but flywheel hybrids have not yet reached production in passenger cars, although according to various press reports its introduction cannot be far away.

Then there is a ‘middle way’ between the pure electric hybrid system and the pure mechanical systems, and this is the electro-mechanical (EM) hybrid, as raced by Porsche in selected GT events in recent years in a specially modified GT3 car, but with great success by Audi in endurance racing, collecting a very notable win at the Le Mans 24 Hours in 2012.

So, what is this middle way, and how does it work? The basis of the system is a flywheel, similar in appearance and concept to those used in mechanical hybrid systems. The difference is that the flywheel in the EM system is ‘loaded’ or ‘doped’ with magnetic particulates that can be magnetised in certain directions. What we have in effect is a large conventional electric motor rotor, although one in which the magnets are not continuous solids but are dispersed within a composite matrix. It is of higher inertia than a conventional magnetic rotor, but this part of the system does not act as a propulsion motor; it is merely an energy storage device, so its high inertia is an advantage.

The housing of the EM flywheel also contains a fairly conventional electric motor stator. The magnetically doped composite is a filament-wound glass fibre ‘ring’ which is shrouded with filament-wound carbon fibre composite. The carbon not only provides containment for the glass fibre, it ensures it remains in compression. It also provides a very useful degree of inertia. High-strength carbon is an ideal material for this owing to its strength. Conventional flywheel materials are not strong enough to prevent bursting at the speeds required to store sufficient energy.

Rather than using a mechanical continuously variable transmission as a way to transfer power, the EM system uses some fairly conventional power electronics to transfer energy between the propulsion motor (mounted somewhere in the driveline) and the EM flywheel. This means the energy storage and propulsion motor can be mounted remotely from each other. As shown in Fig. 1 here, the EM system ‘simply’ substitutes a conventional battery with an EM flywheel motor.

In selecting one of these systems for a race vehicle, the engineer needs to weigh the packaging advantages of each system, energy density, power density, longevity and reliability.

Fig. 1 - The basic architecture of the EM hybrid system compared to conventional electrical and mechanical systems

Written by Wayne Ward

Usually in a hybrid car, the output of the engine is used to drive it, and this is augmented by energy supplied through an electric motor. More refined systems can bypass the engine at low car speed. This type of technology is called ‘parallel hybrid’. Diesel-electric locomotives are ’series hybrids’; the diesel engine is there simply as a means to power a generator, the energy from which is used for propulsion. Such series hybrids have a means of converting chemical energy in a fuel to electrical energy on board the vehicle. A number of car manufacturers are looking at series hybrids, and we are likely to see these on the roads in years to come.

Fuel cells follow this concept of onboard generation of electrical energy, but the process no longer involves an internal combustion engine. A fuel cell is similar in many ways to a battery: it has positive and negative electrodes, and an electrolyte, but it differs in not being able to store electrical energy like a battery can, relying instead on the flow of fuel into the fuel cell(s) to produce power. Fuel cell vehicles have proven themselves to be much more efficient than those with combustion engines. Fuel cells that use hydrogen as a fuel seem to be very popular among concept cars, and Mercedes, Toyota and Honda are notably pushing ahead with development and road trials.

The Le Mans race has recently granted one entry per year to a car, given the race number 56, which showcases radical new technologies. In 2012, the DeltaWing car had this entry, and in 2013 the GreenGT H2 will race as car 56. The GreenGT is a hydrogen fuel cell-powered car, and is featured in issue 65 of Race Engine Technology magazine. This car is eagerly awaited, as it is the first time that a fuel-cell powered car will compete in a top-level circuit race against well-developed conventional cars with internal combustion engines. There is potential to provide 400 kW (544 hp) to the rear wheels, using two motors. The car is at an early stage of development, and there is a lot of potential to save mass from the rather uncompetitive 1200-plus kg that it currently weighs.

That is not to say that fuel cells have not been used for competitive motorsport previously. At a lower level of cost, but involving strong engineering, Formula Student has had a hydrogen fuel cell-powered entry from Delft University, which competed at the Silverstone event this year. Formula Student engineers often go on to careers at the top levels of motorsport, and the Delft entry will be watched carefully in the future; it is unlikely to be the only fuel cell car.

Speed record competition embraces all kinds of motive power, and cars powered by fuel cells are no exception to this. The land speed record for a fuel cell-powered ’streamliner’ car stands at over 300 mph, held since 2009 by a team from Ohio State University.

There is some debate over the eco-friendly credentials of hydrogen-powered cars. While it is true that hydrogen can be formed by electrolysis of water, much industrial hydrogen is formed from a technique called methane-steam reforming, which liberates carbon dioxide. This process can only really be labelled as environmentally friendly if it is combined with CO2 capture.

Fig. 1 - The Buckeye Bullet is a hydrogen fuel cell-powered streamliner. It holds the fuel cell land speed record at over 300 mph

Written by Wayne Ward

Electric machines, whether used for direct propulsion or, as is the case with hybrids, to augment the conventional internal combustion engine, are essentially constant-torque machines. Unaffected by the vagaries of unsteady airflow and without the headache of keeping hundreds of mechanical components working in harmony, a wonderful plateau of torque remains undiminished, so long as the motors' power electronics switch power in their electromagnets at the optimum timing. Electric motors, unencumbered by heavy battery packs, would make a wonderful propulsion system for a car or motorcycle. If power is your target, and you also seek low motor mass, then rotor speed is your friend. For a given motor geometry and construction, power is directly proportional to speed.

If we ignore the actual electrical operation of the motor - which I imagine will please the mechanical engineers here - let us see how the rotor geometry affects torque. The tangential force per unit area of the rotor, acting on the rotor, is a function of the current density in the windings and the flux density in the gap between the rotor and the stator. The first quantity is a measure of the current being used, the second is a measure of the strength of the magnets and the gap between the rotor and stator.

The torque per unit area of rotor is the product of the tangential force, F, and the rotor radius, r, such that:

and from this it follows that torque is the product of the tangential force per unit area, the rotor radius and the rotor surface area.
For a rotor of radius r and axial length, l:

In seeking high torque, rotor length is important, but rotor radius is a much more potent performance factor. However, as is the case with crankshaft stroke and increased engine speeds, large rotor diameters and high rotor speeds are not easy bedfellows. Even if we assume that the rotor avoids any resonant conditions throughout its running range, centrifugal forces are a function of the rotor radius and the square of the rotor speed. In trying to remain within the mechanical strength constraints of the rotor materials, we have to play speed against rotor radius. Increasing rotor length to increase torque is reasonably efficient, but there are limits to this as well, owing to shaft dynamics. As with an engine, so there are compromises to be made with electric motors.

With comments made in the press recently about the new 2014 Formula One power units, which have an electric motor integral with the turbocharger and which will run at or around 100,000 rpm, choosing the correct rotor dimensions in order to give the correct performance while not failing under centrifugal loading will be no easy matter.

Written by Wayne Ward

The fact is that cars with better optimised engines have tended to outperform hybrid roadcars in terms of fuel consumption. The basic principle of a 'normal' hybrid system, such as that in a Honda or Toyota hybrid, is that the kinetic energy which would normally be converted to heat during braking, is stored and then re-used at an appropriate time. This is called regenerative braking, and is a laudable aim with real environmental benefits. Efficiency is increased, and less fuel should be used.

However, a driver who is genuinely concerned for the environment and seeks to maximise fuel economy will already do relatively little braking and will be very gentle on the throttle. It is quite possible then that the driver who is most likely to buy an efficient hybrid car is also the driver who is least likely to make the best use of such a system. The irony is that hybrid systems on roadcars would give the greatest improvement on those that are driven aggressively.

There are some vehicles that would benefit enormously from a hybrid system - city buses, garbage trucks, taxi cabs, some trains and racing cars. All of these have a very 'start-stop' drive cycle characterised by regular heavy braking events where a large amount of energy is converted, followed soon after by a requirement for strong acceleration. If we measure the 'success' of a hybrid system by the improvement in fuel economy rather than by the absolute fuel economy achieved, then these drive cycles are surely the natural targets of regenerative braking hybrid systems.

There is a type of hybrid system that predates the 'eco-hybrids' and one that is probably more relevant to most drivers, including those whose drive cycle is quasi-steady-state, and that is the motorway driver or long-haul trucker. Turbo-compounding is a system where excess energy is extracted from an exhaust gas turbine and transferred back to the driveline, increasing torque.

Formula One will embrace this concept in 2014, when the new engine rules take effect. The Formula One rules are published in draft form on the FIA website, and define the energy recovery system (ERS) as "a system that is designed to recover energy from the car, store that energy and make it available to propel the car", and as part of this it specifies that a "Heat Motor Generator Unit is the electrical machine linked to the exhaust turbine of a pressure charging system as part of the ERS".

The Formula One system will be able to store recovered energy rather than having to deploy or waste any energy available to it. The ERS will combine recovery of heat from the exhaust flow and from braking. Note that the heat recovery is linked to the use of a turbocharger. Unlike regenerative braking hybrids, this is not a piece of equipment that can be adapted to an existing, normally aspirated engine for an instant performance increase.

For once, motorsport will be a worthwhile development arena for relevant technologies for passenger and haulage vehicles. The FIA ought to be warmly applauded for bringing in such a brave set of rules. Unfortunately, at this early stage, involvement appears to be limited to wealthy automotive manufacturers. The costs of developing a new energy recovery system alongside a new Formula One engine to compete against the likes of Renault, Mercedes and Ferrari are immense.

Written by Wayne Ward

A great deal of press coverage has been given to alcohol fuels, and they are fast becoming an everyday reality: much of the gasoline commercially available now contains a small percentage of bio-ethanol. The 'bio' part is important. While bio-ethanol is produced from plant matter, it is no different in terms of composition or behaviour from synthetic ethanol produced industrially from petrochemical feedstocks. Bio-ethanol still burns to produce carbon dioxide and water, but the fuel is renewable, and the fuel crops remove carbon dioxide from the atmosphere. We are simply recycling the carbon.

However, ethanol as a fuel has some disadvantages, some of which were covered briefly in Race Engine Technology (issue 63, June/July 2102). It can be very corrosive, owing partly to contaminants, and this affects metal components, polymers and elastomers. Modern road engines are capable of running gasoline with around 10% ethanol, but pure ethanol can be a real problem.

Another problem is the energy density of the fuel and the air-to-fuel ratio. One litre of ethanol contains only 61% of the energy of one litre of gasoline. The air-fuel ratio compensates somewhat for this, as we need to add 1.6 times as much ethanol to the air, giving roughly the same power output from an engine. In fact, ethanol is slightly superior to gasoline in this respect, but we need a much larger tank to go the same distance.

Bio-butanol goes some way to making up for these disadvantages. In terms of energy per litre of liquid fuel, it is much closer to gasoline, containing 91% of the energy. The air-fuel ratio is 11.1:1 compared to 14.6:1 for gasoline. The result is that, with a stoichiometric mixture, 1 kg of air can liberate 3.2 MJ of energy from butanol, compared to 3.0 MJ for ethanol and 2.9 MJ for gasoline. Butanol is also less corrosive than ethanol, which is important if you are considering it as a race fuel. Methanol and ethanol engines often have to have their fuel systems purged with gasoline.

Bio-butanol is the subject of a lot of r&d by fuel companies, one of which is using a pair of American Le Mans Series LM P1 cars to promote its bio-butanol fuels. The Mazda-powered Dyson Lolas have been running very successfully on the fuel. Whether this is 100% bio-butanol or a blend is not known, but it is likely to be a blend, owing to the low octane number of butanol compared to ethanol and racing gasoline. The Mazda engines are turbocharged, and therefore high octane rating is important as it is a measure of the knock resistance of the fuel.

Butanol can be successfully blended with gasoline and ethanol. While E85 (gasoline with 85% ethanol) is gaining acceptance as a racing fuel, E85B is being proposed as a possible future 100% bio-fuel, with 85% ethanol and 15% butanol.

Fig. 1 - The Mazda LM P1 engines used by Dyson Racing use bio-butanol fuel

Written by Wayne Ward

Much is made of the advantages that electric hybrid systems will bring to production vehicles and to racing. Pure electric racing is already here: recently an electric motorcycle lapped the 37.7-mile Isle of Man TT course at over 104 mph during the TT Zero race for electric motorcycles. TTX-GP is another electric motorcycle class, with championships on three continents.

Despite having none of the highly loaded sliding contacts that characterise an internal combustion engine, or the mass of meshing gears that we find in an automotive transmission, electric motors also have losses, which have to be minimised, and which might need to be cooled. In racing applications of electric motors, we will want the most powerful motor to fit in the smallest space, and liquid cooling is quite likely to be required. In a previous article we noted that some liquid cooling strategies are more successful than others. Surely with the only mechanical relative movement being in the bearings, there should be only very small amounts of heat and losses to deal with? Unfortunately not.

There are two main kinds of losses in electric motors, which are often referred to as iron losses and copper losses. Dealing first with iron losses, these are made up of two 'components', namely eddy current losses and hysteresis losses. Hysteresis losses occur as we switch the electromagnets in our motor on and off, and depend on the magnetic properties of the iron material used in the motor.

Special materials have been developed to limit these losses. 'Electrical' steels with a high silicon content have low hysteresis, but are disadvantaged by having what's called a low saturation flux density. This essentially means that we need a big, heavy stator to achieve the same magnetic field as a material with a higher saturation flux density. So, for a racing electric motor, more specialised materials may be called for.

Eddy current losses depend on many variables, including the thickness of the laminations in the stator, the frequency at which the motor operates, the flux density and the material's resistivity. The only mechanical changes we can make here are to use a laminated stator and decrease the thickness of the laminations, each of which need to be electrically insulated from each other. The improvement that comes from using ever-thinner laminations follows the law of diminishing returns. The iron losses are proportional to the square of the lamination thickness.

Copper losses are generally the greater proportion of losses in an electric motor, and these are the 'I-squared R' losses we were taught about in school. Copper losses are due to the resistance of the copper conductors and the current that we ask them to carry, and they lead to heating of the conductors. Of course, because copper is heavy, we want to use as little of it as possible, minimising cross-sectional area. This increases resistance and therefore increases losses and heating. Unless we can remove sufficient heat from the conductors or limit the current to low levels to allow air cooling to suffice, they will melt, just as a piece of fuse wire melts when we exceed its current capacity.

Fig. 1 - Michael Rutter on the MotoCzysz, which holds the lap record for electric bikes around the Isle of Man TT circuit

Written by Wayne Ward

There is a third, very important, module that is part of a typical racing KERS system, and this is the power electronics. The aim here is to point out the reasons why we need power electronic modules, and look at what goes in inside one.

If you studied physics at school, you might have built a rudimentary electric motor. For a high-speed, high-torque electric motor, there is a requirement to turn the electromagnets on and off very quickly and accurately. If the point at which we choose to energise the electromagnet is calculated wrongly, we may produce lower than desired torque or possible zero or even negative torque. Just as engine torque reacts to changes in ignition timing, so motor torque reacts to changes in the 'timing' of energising the electromagnets.

Thinking back to that 'school physics motor' we probably switched the power to the coiled electromagnets via slip rings and brushes on the motor shaft. Brushed motors can cause problems at high levels of power transfer because of arcing, and also increase electrical noise, which can affect many of the other electrical systems on the car. If we wish to exercise accurate control of our motor/generator, we need to be in total control of the switching. In essence, the power electronics houses the high-current 'switches' and the electronic circuitry required for controlling these switches.

There are a number of devices capable of switching currents very precisely at high speed - two of these are the MOSFET (metal-oxide semiconductor field-effect transistor) and the IGBT (insulated gate bipolar transistor). There is a wealth of information available on both devices, but it is generally felt that MOSFETs are most suitable for low-current, low-voltage, high-speed switching, while IGBTs are more suited to high-voltage, high-current but lower speed switching.

In racing of course, we want to run the motors fast because, relative to a roadcar manufacturer, we place high value on the power density (power divided by package volume) of our system. A small electric motor that can produce or absorb a relatively modest torque at high speed can produce a very impressive amount of power for its size and mass. Typical electric motor/generators for hybrid race use have a power-to-mass ratio that is very similar to the highly developed race engine whose output they augment. When switching the power on and off to our motor's electromagnets, we want the speed of the MOSFET with the power capability of the IGBT. What we can end up with is an IGBT that struggles to cope and which is unreliable as a result. There are companies who are working on and promoting technology that can control high-power motors up to high rotational speed with a reduced rate of switching, bringing greater reliability to the IGBTs and the power electronics module.

Fig. 1 - An IGBT is a high-power switching device, and is one of the key components in the power electronics module of an electric hybrid

Written by Wayne Ward

If we take the energy storage aspect of hybrid powertrains (electrical energy only - flywheels will be dealt with in another article), we are still some way from matching gasoline, diesels or alcohols. However, electric motors are more than competitive with internal combustion engines. While not at anything like the same stage of development as a race engine, an electric motor can deliver huge power per unit mass - 5 kW per kg was not at all unreasonable even a decade ago, and we can assume that the current batch of electric motors used in Le Mans and Formula One competition are at or above the same level.

Just as an engine has losses that detract from its output, so does an electric motor. There are losses in the rotor and the stator which are apparent as heat; if the motor is not to overheat rapidly, its output must be limited or the heat must be rejected. The power is limited often by the ability of the conductors in the windings to carry the current without overheating and melting.

The limits of passive air-cooled motors - those using finned stator cases - in terms of current capacity per square millimetre is in single figures, with 5-8 A/sq mm being typical*. This may be improved by forced convection, for example using fans or air ducts to channel air from the outside of the vehicle to the motor, thus keeping the cooling fins at a lower temperature. Finned motor cases lead to a large increase in diameter, although for low-cost static machinery, this is not a concern.

Jacketed motor bodies offer an effective way of taking heat away from the stator. As with a water jacket around a single-cylinder engine, the same principles apply with an electric machine but, if the windings are to be kept dry, this method relies on the conduction of heat through the steel stator 'stack'. An accepted level of current in the windings of such machines is in the region of 10-15 A/sq mm. Complexity is added here because we now need to have a cooler in the system, and a pump.

A method of cooling the windings more directly without having them actually in contact with liquid coolant is to have enclosed coolant channels running along the stator slots and between the windings. This method most readily lends itself to conductors that are produced in rectangular section. By conducting heat through thin-walled coolant vessels rather than via the steel/iron stator, we can increase the allowable current capacity of the conductors to around 20 A/sq mm.

If we wish to go further than this limit, then we have to consider directly cooling the conductors with an electrically non-conductive coolant. There are many suitable fluids for this, ranging from pure water (deionised water is a poor electrical conductor) to special-purpose electrical oils, which are designed for cooling electric motors and transformers, and are hence often known as 'transformer oils'.

There are two common methods of cooling the winding conductors directly. The first is immersive cooling, where the winding coils are immersed in coolant which is then circulated via a cooler. The second method is to spray oil directly onto the end-turns of the coiled conductors. These are the areas which are most easily accessed, and in Fig. 1 here they are the parts of the coils at the end of this wound stator. Such cooling schemes where coolant removes heat directly from the coils allow a current capacity of about 30 A/sq mm.

* Gieras, J.F., "Advancements in Electric Machines", Springer, 2010, ISBN 9-0481-8051-1

Fig. 1 - Direct cooling of the stator windings allows high current densities to be used, resulting in a smaller motor for a given output

Written by Wayne Ward

If you want to see real diversity of technology in terms of energy recovery then you need look no further than sportscar racing, where two mechanical systems will be racing in 2012. The unique DeltaWing 'technology car' will race at Le Mans, and we hope very much that Porsche will continue to enter its hybrid 911 in selected races in 2012. Ironically, both systems have their roots in design studies for Formula One projects.

The two systems, and an unknown number that are not yet racing, are all based on flywheel energy storage. Rather than converting kinetic energy into electrical energy, these systems maintain energy in the form of kinetic energy, and this is stored/used by changing the speed of the flywheel. If you studied physics at school, you might recall the formula for the energy contained in a rotating body:

In this formula, E is the energy, I is the moment of inertia of the body and ? is the rotational speed. We can see that if we want to store a lot of energy, we need to increase inertia and speed (especially speed). In order to increase inertia, we need to either increase mass or increase the radius at which the mass is distributed. Increasing speed is something we all understand. So, it seems very easy - take something heavy, and spin it very fast.

When we do so, however, the problems soon become apparent. The centrifugal forces acting on a body (as students of physics will again recall) increase in proportion to the mass, and to the square of the rotational speed. As we get towards the centre of the flywheel, the material has to support a force due to a large wedge of material. If you work through the mathematics, you will soon see that you run into trouble at fairly low levels of energy storage. A metallic disc just won't do the job.

If you look at any of the current racing flywheel energy storage devices, you will see a metallic hub that supports a thick rim of carbon-fibre reinforced polymer (CFRP) composite. What might surprise you is that neither of these components is capable, on its own, of working at the design speed. Both would fail in a spectacular manner at a speed much lower than their design speed, due to huge centrifugal forces leading to stresses that exceed the strength of the materials used for their construction.

If ever there was a match made in heaven, it is this. The key here is to use a heavy interference, which puts the flywheel rim and the hub into radial compression. This means the flywheel can reach a high speed without the inner fibres of the rim being put into radial tension.

Fig. 1 - A typical racing energy storage flywheel, with a metallic hub and CFRP rim

Written by Wayne Ward

Permanent magnet motor technology is more efficient than the more traditional 'brushed' motors, and this can lead to the motors running cooler. They are also quieter, as there is no arcing between the stationary brushes and the rotating motor shaft (commutation arcing). This is true in the acoustic as well as the electrical sense. While we may not notice any noise from an electric motor in a racecar that is also powered by a noisy engine, 'electrical' noise - or electromagnetic interference - can play havoc with adjacent electrical components or systems.

A normal 'brushed' alternator, such as those that we might have seen on our roadcar or motorcycle, will probably have a number of windings carried on the shaft assembly. These windings are connected to different segments of the 'slip ring'. As the segmented slip ring is contacted by the brushes, the coils are energised, creating a magnetic field. These electromagnets are attracted to magnets on the stator, causing the motor to turn. Before the magnets on the stator become aligned with those on the rotor - at which point there would be no turning torque - other coils on the rotor are energised, and so the motor continues in motion.

In a typical brushless motor that we might find in a hybrid system, the wound coils are housed in the stator, and the permanent magnets are on the rotor. By energising the coils in the stator in turn, the shaft is caused to rotate. While the switching of the coils in a brushed motor happens automatically due to the mechanical connection of brush against slip ring, in a permanent magnet motor the switching has to be controlled by other means.

Brushed motors have essentially fixed 'timing', in much the same way as in a distributor and rotor arm, though without any fancy advance mechanism. However, a brushless motor, because it is controlled externally by a computer, can have its 'timing' - that is, the switching on and off of the power to the field coils - changed to suit operating conditions.

Despite the complication added by needing power electronics to allow the rapid switching of electrical power to the stator coils, brushless motors have mechanical advantages over brushed alternatives. Owing to their greater efficiency, they generate less heat, giving an easier time to mechanical components such as bearings. The absence of brushes and slip rings means there is no concern over wear of these components, or their physical breakage.

Fig. 1 - Lightening Motorcycles holds the electric motorcycle land-speed record; its bike uses a permanent magnet motor

Written by Wayne Ward

The various Formula Renault series though do provide a good graduation scheme, starting with relatively cheap cars powered by four-cylinder engines, through to the top series which, from 2012, is to be powered by a bespoke race engine for the first time. The 530 hp car further distinguishes itself from the other race series by being the first racecar to be designed without a battery for the 12 V supply on the car. I spoke to Renault-Sport's circuit technical manager Benoît Dupont about the replacement.

A decision was made to run without an onboard starter for reasons of saving mass; the current requirement on the car is much reduced once the starter motor is no longer a part of the equation. Instead a supercapacitor system will be used. This has been proven on other Formula Renault cars; the two-litre car uses a supercapacitor system to actuate the gear change. So, this is simply an extension of an existing system to accommodate the electrical requirements of the other systems on the car.

The supercapacitor system is covered in some detail in issue 57 of Race Engine Technology magazine (September/October 2011). The decision to dispense with the services of an onboard starter means there's no need for a large battery and, after examining a number of alternative options, a supercapacitor system was eventually chosen. The mass of the supercapacitor 'box' is 1.3 kg, and this incorporates the car's fuse box, which used to be additional to the 13 kg mass of the battery. Dupont says the combined mass reduction from removing the starter and using the supercapacitor system has roughly the same impact on the car performance as the switch to the significantly more powerful race engine!

So, how 'super' is the supercapacitor system? Well, a typical capacitor that you might find in an electronics kit will have its capacitance measured in microfarads or nanofarads, which are one-millionth and one-billionth of a farad respectively. The Renault system is 350 farads and operates at a nominal voltage of 15 V. In terms of capacitance it is many orders of magnitude greater than a typical capacitor.

The advantages of supercapacitors over the other obvious alternative (lithium ion batteries) are numerous. They are less affected by the number of discharge cycles, and have a much greater power density. This power density advantage allows them to accept and discharge energy at a high rate.

On the other hand, they have a comparatively poor energy density, and this means that to hold a given amount of energy, they score poorly against a lithium ion battery. These are the reasons that they are suitable for things such as gear-shift actuation, but not for energy storage for hybrid systems, where lithium ion systems are favoured.

Fig. 1 - The 2012 Formula Renault 3.5 car dispenses with a conventional battery in favour of a supercapacitor system

Written by Wayne Ward

A particular area of motorsport that offers huge scope for development here is that of land-speed record breaking. People's endless endeavour to be the fastest shows no signs of diminishing, especially where the technologies involved are within reach of modestly funded teams and small manufacturers.

In August 2011 one manufacturer of production electric motorcycles, Lightning Motorcycles, set a new land-speed record for an electric motorcycle of nearly 216 mph, apparently using less than 20 US cents' worth of electricity for a typical run.

The electric motors used for the land-speed record are the same kind used in the production motorcycle. I discussed the design of the electric motors with their supplier's director of hybrid engineering. He explained that the motor used in the record-breaking motorcycle is a three-phase alternating current motor, with a permanent magnet rotor. The stator design is patented and gives this particular type of motor its name, High Voltage Hairpin (HVH). Rather than using stranded conductors for the stator windings, the HVH stator uses solid rectangular conductors. The conductors are formed with a 180º bend, forming the 'hairpin' and inserted into the stator before being twisted together with other hairpins to form continuous windings.

While these rectangular conductors are subject to higher eddy-current losses than multi-stranded conductors at higher frequencies, the difference at low frequency is not so marked. The advantage of using the solid-section conductor is that more cross-section of copper can be used in the stator slots, allowing higher currents to be used. Without specifically stating current densities for the record-breaking motors, the company said the conductors in existing HVH motors "are capable of over 50 A per square millimetre". With such high current densities, there is a lot of heat generated within the stator, so the motors are liquid-cooled.

The heat generated in the copper conductor has two means of escape. Heat transferred to the steel stator stack via direct contact between the conductor and the stator slot walls is removed by oil flowing around the periphery of the stator stack. Oil also flows around the end turns of the conductor, transferring heat directly from hot electrical conductor to cool oil.

Not only is the stator liquid-cooled, so too is the rotor, with the aim of removing heat from the magnets.

This application is a great example of how racing can positively influence mainstream automotive development. Although my contact is not a race engineer, his words on the value of racing should hearten us all. "What racing does is help to get products developed much more quickly. The creativity of the people in racing is awesome, so some really neat designs come from this area."

"We are not really actively from a company standpoint seeking out racing endeavours to showcase our product, but the performance of the product is really allowed to shine in racing. This is the truest form of racing to street being a dual path - solid, good performance motors provide a great foundation for creative minds to take to the next level in racing - and we get input on how to make a better motor for production vehicles."

Fig. 1 - Part of the stator from a High Voltage Hairpin (HVH) motor, as used to set the electric motorcycle land-speed record (Courtesy of Remy International)

Written by Wayne Ward

There are three main subsystems of the electric KERS systems used in Formula One - the electric motor/generator, the power electronics and the battery. The battery, for both hybrid and pure electric production vehicles, holds the key to producing a light, affordable vehicle. In racing, where the mass of the car is lower and the packaging of major components is critical, the battery represents something that not only has a critical effect on the output of the combined engine/KERS unit but also on the design of the rest of the car. The Red Bull Formula One car this year would not compromise the design of the rest of the car in order to package the KERS system, and this has given the team KERS reliability headaches throughout the season.

The battery is made up of multiple cells, connected in series to give the system voltage. They are thought to be exclusively based on lithium ion technology, with each cell giving about 3.6 V output. With systems operating at hundreds of volts in order to limit maximum currents, there can be hundreds of cells in a typical KERS battery. The current 60 kW discharge limit, for a 100-cell battery, would result in an average current draw of about 160 A, with peak currents being higher than this.

I asked battery expert Nigel Vincent of ABSL Power Solutions, which makes lithium ion cells, about various aspects of the cells and their development. He explained that there is a trade-off between energy density (the amount of energy stored per kilogramme of cells) and power density (the charge/discharge power per kilogramme of cells), with differing chemistries offering different advantages.

Lithium ion cells based on LiMnO2 (lithium manganese oxide) and LiFePO4 (lithium iron phosphate) show high power density, but lack energy density compared to newer 'mixed oxide' cells which have excellent energy densities (>225 Wh/kg). These lithium ion chemistries concern the cathode but changes in anode chemistry, according to Vincent, could yield "further significant increases in specific energy". Moving away from graphite anodes to ones based on silicone or titanate could see energy densities reaching 350-400 Wh/kg in the next five to ten years.

The other important metric against which batteries are judged is the charge/discharge rate, and this is limited, otherwise damage can be caused. This limiting rate is affected by the construction of the cells, with features such as electrode surface areas being important factors, as well as the chemistry being used.

Of great importance in a race application is not only the consideration of energy/power density but also space efficiency. Traditional cylindrical cells are being displaced to a certain extent by 'pouch cells', which are essentially flat and can therefore be packaged more efficiently.

Developments in cell chemistry and manufacturing have also improved the operating life of the cells we might choose to use for race hybrid systems. Other factors affecting cell life, as discussed with Vincent, are said to be the 'depth' of charge/discharge, rate of charge/discharge, charge voltage and ambient temperature.

The advent of KERS in Formula One and hybrids in other forms of racing will, hopefully, help drive forward advances in cell chemistry and construction for the benefit of racing and also for production hybrids in future.

Fig. 1 - The 'pouch' cell offers packaging advantages over traditional cylindrical cells (Courtesy of ABSL Power Solutions)

Fig. 2 - A traditional cylindrical lithium ion cell (Courtesy of ABSL Power Solutions)

Written by Wayne Ward

Nowhere is the efficiency of energy recovery and the function of the electric motor more important though than in pure-electric racing. There are a few race series that involve only electric vehicles, and electric motorcycle racing is probably the most successful of these. This class of racing is still in its infancy, and with an all-electric race now forming part of the Isle of Man TT race programme, and TTX-GP championships running on both sides of the Atlantic, there are many variations of powertrain in use. I spoke recently to Italian TTX-GP entrant CRP about the electric motors used in its e-CRP machine.

The machine uses two 26 kW brushed direct-current (DC) motors. Brushes are the components that carry current between the stationary parts of the motor and the rotating shaft; there are brushless DC motors that have the advantage of being more efficient but require complex motor control. The speed control of a brushed DC motor can be achieved with a simple rheostat or potentiometer. Previous winners of the electric Isle of Man TT have used brushed DC motors. This, along with the good balance of cost and performance of a brushed DC motor, were stated as the main reasons for using this type of motor.

CRP uses a standard (although unspecified) DC motor. In using two motors, CRP says it is important to select pairs of motors with equal performance, so that both are working equally hard for a given controller setting. A special air-cooling system has been developed by the team in order to control the motor temperature. Francesca Cuoghi of CRP said, "Keeping low temperature of the motor means having a better output." The motors are very rugged and, as such, require no anti-vibration measures; they are mounted solidly in the motorcycle.

However, the motors are stated to be the limiting factor in the performance of this motorcycle, with the batteries able to supply more power than the motors can convert. It is true for any electric race machine, but particularly with an electric race motorcycle, that it is important not to have a large mismatch between the performance of the energy storage (battery) and the energy conversion (motor). The mass of both these components depends to some extent on performance, and any unused performance means a mass penalty for the motorcycle. A battery with more energy capacity will be larger and heavier, while a motor with greater torque will also be heavier. Any weight penalty on a race machine as small and light as this motorcycle needs to be kept to a minimum.

Fig. 1 - Detail view of the brushless DC motor used on the e-CRP 1.4 electric race motorcycle (Courtesy of CRP)

Fig. 2 - A view of the complete race machine. The electric motor in this view is installed as a mirror image of that in Fig. 1 (Courtesy of CRP)

Written by Wayne Ward

However, there is no particular requirement in the regulations of those race series that allow the use of such hybrid technologies to use electrical conversion of energy, and there are a number of alternatives which are of a purely mechanical type. A lot of publicity in the motorsport, motoring and general mechanical engineering press has been given to the concept of flywheel energy storage. If we equate such a system to the electrical type, then the battery is analogous to the flywheel (energy storage) and the electric motor/generator is analogous to a CVT (continuously variable transmission). The CVT is required to allow the flywheel rotating at any speed to accept (or discharge) energy to the vehicle transmission at whatever speed pertains at the time in question. Such flywheel-based mechanical systems generally see the flywheel rotate at high speed - both in terms of rotational speed and the surface speed of the outside diameter - in conditions of very low pressure (high vacuum).

There are a few companies marketing such systems, and a Flybrid flywheel-based mechanical system will race at Le Mans this year. Its flywheel is a filament-wound carbon flywheel with a lightweight steel hub.

The purely mechanical systems on offer are said to be more efficient, in terms of the percentage of braking energy that can be returned to the driveline, than a typical electrical hybrid system. The useful energy density of the flywheel is high compared to a battery, and this is particularly true of production vehicle hybrids where battery energy capacity is very high compared to the actual amount used, owing to battery life problems where full charge and discharge cycles are used. Race systems accept shorter battery life in exchange for improved energy density.

Flywheels are not the only method of storing energy in a mechanical system. Both hydraulic and pneumatic energy storage systems have been studied, and in these cases a pump/motor is used to compress a fluid.

With turbo-compounding likely to make an impact in both general automotive and racing at some point in the future, could a mechanical system be used for this purpose too? There are a number of schemes looking at 'turbo-generators' to electrically harvest energy from the exhaust flow downstream of the existing turbocharger turbine, and even systems that replace the turbocharger turbine with a power turbine and have the compressor driven electrically. Provided that a CVT could be made compact enough, there appears to be little reason why exhaust energy could not be extracted via a mechanical system. However, packaging such a system so that energy can be transferred to the transmission may present a challenge.

Fig. 1 - This is a flywheel from a purely mechanical KERS system. It features a lightweight steel hub and a carbon-fibre rim (Courtesy of Flybrid Systems)

Written by Wayne Ward

One very topical subject is that of the various hybrid technologies that seek to make our race vehicles more efficient by storing and re-using the kinetic energy normally converted to heat under braking. This is being very actively developed by roadcar manufacturers - for example in the Toyota Prius which, while not being the prettiest child in school, is beloved of many high-profile people who would like to portray a 'greener' image. Honda has recently introduced its latest hybrid, and this looks more like a car that people will want to drive.

Both vehicles are what are known as 'petrol-electric hybrids'. While the technology can also be applied to a diesel to improve fuel economy, the costs of a hybrid-equipped diesel have been considered to be too high, although Peugeot has recently introduced such a car.

These vehicles store energy recovered from braking in a battery, most commonly a battery made up of lithium-ion cells. This behaves much as would a normal car battery, but these hybrid batteries have a much greater energy storage capacity and run at much higher voltage. One problem associated with many types of battery is that of ageing, especially when they are cycled between full charge and full discharge. Race systems generally accept shorter life as a compromise for being able to use all of the battery's capability and therefore use the minimum mass of battery. Lithium-ion cell technology is improving all of the time, but cell ageing is one problem that continues to worry customers in the mainstream production automotive sector.

There are alternatives to the chemical storage offered by lithium-ion cells, and one of these is the 'flywheel battery'. Those who read Race Engine Technology and some other engineering publications can't have failed to notice the articles on Flybrid Systems' pure mechanical KERS system, which uses a flywheel to store energy. Ricardo also has a flywheel-based system.

An electromechanical system, linking a flywheel to a motor/generator can store kinetic energy in a rotating flywheel, and this can be fully charged and discharged without any ageing effect causing loss of performance over time. Williams Hybrid Power has a novel system where the motor/generator rotor is a filament-wound carbon fibre flywheel that is 'doped' with magnetic particles, although the more conventional application is to drive the flywheel via a conventional motor/generator.

NASA has been developing this technology for space-flight applications, as the performance of the system is, compared to battery technology, relatively insensitive to temperature fluctuations and offers an advantage in usable energy storage per unit mass. Closer to home, you may have an electromechanical flywheel storage system in the building where you work. They have been popular for many years in uninterruptible power supplies (UPS) for PC and server applications.

One disadvantage of purely mechanical flywheel-based hybrid systems on racecars is the packaging constraints, owing to the fact that the flywheel and constantly variable transmission need to be mounted to the engine or vehicle transmission, depending on where the energy is 'harvested' or redeployed. Electrical systems need only mount the motor/generator at this point, allowing more flexibility in packaging. Electromechanical systems, where the flywheel energy storage is mounted remotely, offer the combination of energy-dense storage but with flexibility in terms of car packaging.

Fig. 1 - NASA has developed 'flywheel battery' energy storage for space-flight applications

Written by Wayne Ward

Among the early pioneers in motorsport applications of so-called 'regenerative braking' were UK company Zytek. It had a regenerative braking system racing in endurance sportscars more than a decade ago in a project with Panoz, and has remained active in the development of race 'hybrid' technology since then. A large part of the Zytek group works on the design, development and manufacture of hybrid and full electric drivetrains for the mainstream automotive industry, so access to the leading technology is not a problem for the racing arm of the company.

I had the chance to see the latest Zytek race hybrid electric motor at the LMES sportscar race at Silverstone in 2010, and was startled to see the progress that has been made in motor technology since the days when I was involved in the Panoz Q9 project. The new motor is aimed at the Le Mans prototype class, but would be suited to other types of racing. A successful manufacturer-backed test in 2010 showed its potential for high-level production-based endurance racing.

The power of the new motor is stated to be 55 kW (about 74 hp), and its mass is 8.5 kg. Compared to the Q9 motor, this one produces less power (mandated by regulations) but in terms of power density it has made huge strides, being about 200% better than the Q9. The basic technology of the motor remains the same, but constant development means that real improvements have been made.

My recollection of the Q9 was the amount of effort required to turn the output shaft. It felt like winding a cam over, an effect known as 'cogging'. The large spikes in the torque required to turn the motor, requiring real physical effort, have been replaced by a motor that can be turned easily with one finger. This, I was told, is a consequence of improvements in magnet technology and the construction of the motor.

The motor operates within the context of a much lighter system overall. The Q9 battery needed an engine hoist to lift and manoeuvre it, whereas the new Zytek battery can easily be lifted by one person. Motor, battery and electronics form the major part of the system mass which comes in at 42 kg, whereas reports suggest that the weight penalty of the Q9 system was about 170 kg. Much of the difference comes from improvements in battery power density and energy density, as discussed in an earlier article.

It has recently been announced that Rangoni Motorsport is to campaign a Zytek sportscar in 2011 using this ZPH (Zytek Performance Hybrid) system.

Fig. 1 - The ZPH hybrid motor is small and compact compared to its predecessors (Courtesy of Zytek Motorsport)

Written by Wayne Ward

Flybrid Systems' approach is again based on a flywheel, but is a purely mechanical system. Again developed originally for Formula One, Flybrid's system wasn't raced in 2009. Besides the KERS regulations for Formula One, the ACO, which organises the Le Mans 24 Hour race, has supported hybrid technology for a number of years.

In 1998, Panoz entered a hybrid for the 24 Hour race, and in subsequent years there have been a number of ACO-sanctioned endurance races in which hybrid prototype cars have competed. Peugeot is known to be developing a hybrid system for its new diesel-powered car which will race under the new 2011 LM P1 rules, Porsche has the GT, and rumours abound that the new Aston Martin LM P1 will also use hybrid technology.

To this list, we must add the purely mechanical Flybrid system. In 2011, Flybrid will supply Hope PoleVision with its high-speed flywheel system, which will be installed on the bellhousing of the Lehmann-powered Oreca car's XTrac 1059 transmission. The Flybrid system is very compact, and is known to be a low-mass solution. According to Flybrid's Technical Director Doug Cross, the total mass penalty of having the system on the car is about 35 kg. Mass reduction has been an important part of the development of this latest Le Mans Prototype system.

The first Flybrid system was designed such that maximum power was available over the entire working range of the system. By comparison, an electric motor - which is essentially a constant-torque device - has a certain peak power speed, with power below this varying in proportion to motor speed. Cross says that, in attempting to supply full power over the entire range, the first system was heavier than was really required.

The application of the extra performance is controlled by computer rather than being in the hands of the driver, and the power of the system is allowed by regulation to be 100 kW in both energy release and capture modes. The energy storage is about 500 kJ. The location of the flywheel system on the bellhousing means the car be driven by the flywheel with the engine turned off, as required under the ACO rules.

The ACO rules also limit the use of the full hybrid functionality as might be available in a road application, and state that "a traction control system operating exclusively on the engine is authorised". In traction-limited situations, a hybrid can provide a degree of traction control.

The duty cycle and length of the Le Mans race is something that should not limit the performance of the Flybrid system in the same way that an electric hybrid might need to be de-rated. However, Cross says that, owing to the minimal cooling requirement for the system, the cooling would be sensitive to debris blocking the rows of the oil cooler dedicated to the cooling of the KERS system, and that "a mitigation strategy has been developed to cope with a blocked cooler".

This would reduce power a little until such time as the debris blocking the cooler could be removed. The fondness of some teams for overnight gravel-trap sojourns, and the amount of tyre debris on the track in the latter stages of the race, means that having this strategy is important.
The Lehmann engine is a 2 litre turbo The transient response of turbocharged engines is critical for race applications, and developing an anti-lag system is an important aspect of the performance of the car. In this respect a KERS system can augment the transient response, providing an instant response to changes in throttle pedal position.

Fig. 1 - This schematic shows the position of the Flybrid KERS system in the 2011 LM P1 Oreca to be campaigned by Hope PoleVision (Courtesy Flybrid Systems)

Fig. 2 - The LM P1 KERS system incorporates Flybrid's new clutched CVT transmission (Courtesy Flybrid Systems)

Written by Wayne Ward

Williams had developed, but didn't race, its flywheel/electric system. Flybrid's pure mechanical system promises excellent efficiency, low mass and good durability, but nobody in Formula One raced its system, choosing instead to develop systems in-house, many of which never made it past initial testing and fewer still that were raced with any degree of success. Having abandoned KERS this year, Formula One is set to reintroduce the technology from 2011.

The system developed by Porsche for use in sportscar racing, and used during 2010 in a small number of races on a 911 GT3 car, is similar in principle to the flywheel/electric system developed by Williams. The kinetic energy recovered from the deceleration of the vehicle is stored in the form of rotational kinetic energy in a flywheel.

Energy is added to the flywheel or taken from it by using it as an electric motor, and hence it must contain elements that enable it to act as the rotor from an electric motor. We assume that, for reasons of efficiency, it works as a permanent-magnet machine and so the flywheel is likely to contain magnets or at least be heavily doped with magnetic materials. The rate of discharge, which is the performance boost that the vehicle feels, is of the order of 120 kW (160 hp).

The Porsche system feeds the recovered energy to the front wheels of the car using two electric motors, which means full power can be added even in situations when the rear wheels are traction limited, thus providing a useful advantage. While the extra mass of the system (estimated at about 150 kg for the Porsche system) will clearly be a penalty, there is likely to be an advantage from the extra torque available at the front wheels., and this will become more pronounced as the technology improves and matures. The power and energy densities of the system will increase as the system evolves, giving a greater advantage to a car equipped with such a system.

In respect of using two motors, the system is similar to that used by Porsche on its prototype 918 Spyder roadcar. Given that Porsche had a free hand with the design of the 918, it shouldn't surprise us that it has chosen a similar method of power delivery for the 911 hybrid.

Beyond simply adding the extra energy in equal amounts to the front wheels, it is clear that the power can be added in an uneven manner, thereby tuning the steering response of the car. Additionally, if energy is recovered at the front wheels, the braking behaviour of the car can be tuned in corners to help the car turn, as with ESP systems on roadcars.

The disadvantage of adding the extra energy at the front wheels is that the system cannot take advantage of the range of gears available in the transmission, as is the case with hybrids that add their energy to the drive before the gearbox.

At the time of writing, the Porsche 911 GT3 hybrid was competing in the prestigious Petit Le Mans 10-hour race in the US, the final round of the American Le Mans Series, in the GTH class. This class has been created especially for such cars, and a measure of the car's potential is that it has, throughout practice, been halfway up the GT2 class times. We should note, however, the class of the drivers used. Timo Bernhard, Mike Rockenfeller and Romain Dumas, who are driving the Porsche hybrid, won the 24 Hours of Le Mans outright for Audi in June this year.

Fig. 1 - The prototype Porsche 918 hybrid roadcar uses two motors to supply recovered energy to the front wheels, as per the Porsche 911 GT3 hybrid racecar

Written by Wayne Ward

There are now very few large car-makers who don't have a serious hybrid development programme, and a growing number of them feel the technology is mature enough to release series production models - notably Honda with the Insight and the Civic Hybrid, and Toyota with its popular Prius model. The seriousness with which hybrid systems are seen in motorsport is evident through the involvement not only of Formula One, whose rules are very much based around a more efficient powertrain from 2013, but also in sportscar racing where Porsche has had recent success with a hybrid.

Zytek's engineering manager Ian Lovett has been involved in the development of its hybrid systems since the time of the Panoz Q9 project, and I asked him what the main developments have been over the past decade in sportscar racing hybrid systems.

In terms of analysis, Lovett explained, "An awful lot of electromagnetic FEA work goes into the lamination, winding and magnet design, and this is where the gains and thermal performance come from." In terms of providing a low-mass system, he said, "It's down to attention to detail to create the most efficient structural and thermal paths to support the loads and cooling requirements."

Lovett provided some figures for the performance of Zytek's three sportscar KERS systems spanning the Q9 era to the present day:

Q9 is the system raced by Panoz in 1998
Q10 is the system raced by Corsa in the American Le Mans Series in 2009
ZPH is Zytek's current system, now being tested.

A measure of Zytek's progress, in terms not only of performance but also removing mass from the system over the past 12 years, is to measure the power density of the motor. In 1998, the Q9 motor had a 'power density' (output divided by mass) of 3.57 kW/kg, and in 2010 the ZPH system has a corresponding figure of 6.47 kW/kg.

Looking at the Q10, this appears to be a step backwards at 1.74kW/kg, but company owner Bill Gibson said in 2008, "The whole idea of the Q10 is to increase our rate of development of hybrid drivetrains for roadcar use", and indeed this system was closer to the level of technology found in a road-going system than something developed especially for racing.

A comparison of the battery technology is also illuminating. The Q9 battery is almost 100 kg heavier than the ZPH item, while providing only an extra 20 kW of power. The energy density of the Q9 battery is 0.09 MJ/kg, and this rose with the Q10 to 0.118, an increase of 31%. For ZPH, the figure is 0.129, representing a 43% improvement.

The power density figures are more impressive, with the Q10 having a 33% improvement compared with the Q9 (0.816 kW/kg versus 0.615 kW/kg). The ZPH battery, however, has a power density 272% greater than the Q9.

The combination of improved power and energy density means the effectiveness of the batteries is much greater. At any given time, the state of battery technology seems to be such that power density must be traded off against energy density, but as the technology gets better, both can be improved simultaneously.

Fig. 1 - The Ginetta Zytek car seen here carried the Zytek Q10 KERS system. It was the first hybrid LMP car to take a podium position. Eleven years earlier, the Q9 was the first hybrid LM GT1 car to take a podium

Witten by Wayne Ward

At the end of the 2009 race season, the media seemed to be of the opinion that the team that had achieved most from KERS, having used the most successful system, was McLaren. Its KERS system was developed in conjunction with Mercedes and Zytek.

While Mercedes was new to racing hybrid systems, the same could not be said of Zytek, which had developed a system to race at Le Mans a decade earlier. The Panoz hybrid, raced in 1998, and known as 'Q9', really was revolutionary at the time, and Race Engine Technology's editor Ian Bamsey covered this in issue 47 (June/July 2010) of the magazine.

The Panoz was quite an individual car, being the only front-engined entry racing in LM GT1, which at the time was the premier category in endurance racing. This was exclusively for roadcars and for which a number of manufacturers - including Panoz, Porsche and Mercedes - produced limited numbers of very high-specification roadcars for homologation purposes. The Panoz car formed the basis of their subsequent open-top prototype that remained competitive for a number of years.

The Q9 system was based around a liquid-cooled permanent-magnet motor-generator, designed and produced by Zytek, which also developed the power electronics for the project. The motor was mounted on the gearbox and was capable of supplying some 120-150 kW at 18,000 rpm.

Tests at Zytek on a specially developed dyno showed that this had real potential, but the battery technology wasn't mature enough to match the motor's potential performance. At the time, none of the batteries (which were assembled by Zytek from cells supplied by Varta) approached the promised 120 kW, although significant progress was made from the initial battery that was able to discharge at a rate of about 60kW. Battery technology has advanced further since then, especially given the development of hybrid roadcars by many car manufacturers and the recent foray into KERS by Formula One teams.

The weight penalty of carrying the motor, battery and power electronics was considerable, and the heat generated by the battery - which sat where the passenger seat would normally have been - meant air cooling was required. The car's cabin would have been quite an uncomfortable place to be.

An attempt to qualify the car at Le Mans during 1998 was the first time it had run in anger with the KERS, and there were serious problems with the main shaft in the motor, which broke, leaving the car to qualify with a significant weight penalty but no hybrid assistance. Further development after Le Mans involved a redesign of the main shaft in a stronger material, and the car qualified and raced strongly in the US later in the year, finishing third in class in the Petit Le Mans race.

The project was stopped at the end of that year, but Zytek has continued to develop its hybrid system and has been racing this recently in ALMS.

Fig. 1 - The 1998 Panoz Q9 was equipped with a Zytek-developed KERS hybrid system

Written by Wayne Ward

Next year, however, KERS is scheduled to make a comeback. In this article we shall examine some aspects of one of the 2009 systems.

I spoke to Tom Hyder of Magneti Marelli whose system was used in 2009. Asked about the return of KERS in 2011, he says, "Magneti Marelli has proposed its KERS system for 2011. We have continued to develop the system to increase the amount of energy it can handle in each lap, in case the regulations are changed to allow KERS to make a more significant contribution to performance or fuel economy."

The FIA regulations limit both the amount of energy storage and its rate of release, but if Formula One is to improve its green credentials it must be seen as an avenue of free development, rather than a dead end allowing only limited improvements in efficiency. In this case, the increase in available energy storage gives some scope for possible rule changes in this regard.

I asked about the advantages of a purely electrical system compared to a mechanical system, and the main advantage stated was the flexibility in packaging, and owing to the 'modular' nature of the energy storage, adding extra capacity is simple matter. There is no requirement to have all the energy storage in one place, so mass distribution can be easily tailored to suit the needs of the chassis. One major disadvantage was said to be the transport of the batteries and the cars because of the high voltages involved and the precautions that need to be taken to comply with regulations for air freight and so on.

The electric motor-generator unit (MGU) shown in fig. 1 is a liquid-cooled synchronous motor, using permanent magnets. It is geared at a fixed ratio of about 2:1 to the crankshaft speed, and we can therefore expect maximum speeds of about 40,000 rpm for the MGU.

This system will be offered to Formula One teams for 2011, but owing to the need to couple the motor directly to the engine this would require some changes to engines for any teams wishing to take KERS next year.

The company has also developed a mechanical energy storage solution for its KERS motor in conjunction with Flybrid, which has its own purely mechanical system, although this is felt not to be suited to Formula One KERS regulations.

The question was raised about the use of KERS outside of Formula One, and the answer was that "Magneti Marelli has already run a KERS demonstration at Le Mans and the company is very keen to develop systems in all arenas of motorsport without exclusions."

Clearly, the real prize is to develop systems for production vehicles, but motorsport offers both a method for very rapid development of the technology and, of secondary importance, a real possibility that it can increase the appeal of efficient technology beyond those who buy the existing, often unappealingly dull, models currently available.

Fig. 1 - This is the KCU, or KERS control unit, which deals with the many aspects of KERS control and monitoring (Courtesy of Magneti Marelli)

Fig. 2 - Formula One motor-generator unit (MGU) which runs at up to about 40,000rpm (Courtesy of Magneti Marelli)

Written by Wayne Ward

In truth, Formula One KERS still have some way to go before their advantages sufficiently outweigh the disadvantages, so it is perhaps in their wider application in motorsport that they will be fine-tuned and their advantages eventually fully exploited.

For the first time, KERS systems have been approved by the Automobile Club De L'Ouest (ACO) for use in the 2010 Le Mans 24-Hours race, though in the LM P1 class only, but there are currently no entrants who have agreed to run such hybrid cars at this year's event. Once again the reasons can perhaps be found in the benefits of KERS.

Regarding energy recovered from the brakes in the LM P1 cars for 2010, only electric hybrid systems are allowed, but the stored energy can only be returned to the rear wheels and so a four-wheel drive system is not allowed.

With only electric storage systems being allowed, this rules out one of the industry's leaders in the KERS field, Flybrid Systems. Director and co-founder Doug Cross explains Flybrid's position, "We are outlawed [this year] but we understand that flywheels are likely to be allowed from 2011, and we have seen some draft rules which say they are, so the situation is likely to change."

In contrast to the system used in Formula One, the ACO rules prohibit the use of a 'push to pass' button, thereby limiting the system to an economy function rather than an additional power boost, as stated in Article 1.13 of the regulations - "The use of such a system must not be aimed at obtaining additional power but at reducing fuel consumption."

To be considered a hybrid, a racecar must be able to move along the pit lane (a minimum of 400 m) at 60 km/h while using only the electric motor. By setting this minimum distance and speed requirement, the ACO has effectively set a minimum amount of energy you need to carry - enough to call the car a true hybrid.

On the energy front, the authorities have stipulated that between two braking events, you are not allowed to use more than 1 MJ of energy, whereas in Formula One a car is allowed only 400 kJ for the entire lap. "The Le Mans rules say you are allowed 1 MJ between two braking events but, as it turns out, that is entirely irrelevant because you just cannot capture that much energy," Cross explains.

The Flybrid system, which uses a flywheel to store energy, holds an advantage for LM P1 cars in that they are heavier than in Formula One. Where a Formula One car is limited to 60 kW (80 hp), there is no power limit for LM P1 cars and, while running at similar speed transients, the heavier LM P1 cars can capture a lot more energy. "So we would go up to 120 kW, and this makes our gearbox a little bit heavier," Cross says. "For Le Mans we would be storing and recovering about 1.8 MJ, more than four times as much energy."

ACO regulations dictate that the minimum weight of any hybrid vehicle must be identical to that of any other LM P1 cars using a conventional powertrain. With the focus clearly on fuel consumption, to achieve any benefit from a KERS system, a car would have to save one lap of fuel in every 13 laps.

With this efficiency requirement, one of the few ways to achieve this is to reduce the maximum power of the engine slightly to improve consumption. Alternatively, the strategy could be for the driver to back off the throttle towards the end of each straight, allowing the KERS to take over. Cross says, "One of the key effects you have to watch out for is that you will only save fuel if you reduce the amount of fuel you put into the engine. It seems obvious, but it isn't, and it has a massive effect on fuel [consumption] because that is the point where you are using the most power because the engine is at its fastest speed, burning fuel at the greatest rate, and so by backing off there you end up saving quite a lot. That would be one of my recommended strategies."

The flywheel-based system proposed for Formula One is built from more exotic materials, such as magnesium for the casings and titanium for the clutch baskets, than the LM P1 system, which is more cost-focused, and so uses more everyday materials such as aluminium and steels.

Fig. 1 - Flybrid flywheel
Fig. 2 - Clutches in Flybrid system
Fig, 3 - Clutched flywheel transmission (CFT) assembly

Written by Glen Smale

In sports car competition, Zytek produced a hybrid system that was used by the Corsa Motorsports team in the latter stages of the 2009 American Le Mans Series, albeit to a mere fraction of its potential.

While Corsa were not seen on the grid for the first race of the year in Sebring, Zytek is "currently in negotiation with Corsa Motorsports with regards to their 2010 programme," Christopher Foster told me.

Zytek continues to develop its hybrid system and although it is not ready for usage at this time, the UK-based firm is succinct in its desire to see its Zytek Performance Hybrid (ZPH) system utilised throughout all forms of modern motor sport.

The ZPH uses the lessons learned with the firm's early experience with the Q10 sports car project, "But will use bespoke motor racing parts in contrast to the heavily road-based Q10 system," Foster stated. The rationale is to enable a more efficient hybrid system, with double the power density of Q10's componentry.

Still in its infancy, the Zytek Performance Hybrid system is being cultivated strictly as a "bolt on" hybrid, in order that it can be fitted to a wide variety of motorsport platforms - including oval racing," RET-Monitor was told. "The system will be ideally suited for one make formulae, GT and Touring Car championships, Rally cars, single seat racing or any other series or manufacturer looking to utilise hybrid power in a cost effective manner."

In fact, Ben Bowlby, designer of the DeltaWing chassis that's in the running for use as the 2012 tub in the IZOD IndyCar Series is attuned to the system. "The DeltaWing concept is a platform for showcasing technology that enhances performance through improvements in efficiency, cost, safety, recyclability and longevity," he told RET-Monitor. "The goal is to allow race team, supplier, sponsor and fan directed development within a sustainable and relevant manner."

When fully developed, Zytek expects the ZPH hybrid system to weigh approximately 50 kg and produce about 40 kW of power.

Fig. 1 - Zytek Hybrid Technology KERS system

Written by Anne Proffit

Epicyclic gearing is a gear system that consists of one or more planet gears, rotating about a central sun gear. Typically, the planet gears are mounted on a movable arm or carrier which itself may rotate relative to the sun gear. Epicyclic gearing systems may also incorporate the use of an outer ring gear or annulus, which meshes with the planet gears.

The three basic components of the epicyclic gear system are:
Sun gear: This is the central gear.
Planet carrier: Holds one or more peripheral planet gears, of the same size, meshed with the sun gear.
Annulus: An outer ring with inward-facing teeth that meshes with the planet gear or gears.

In many epicyclic gearing systems, one of these three basic components is held stationary. One of the two remaining components is described as an input, providing power to the system. The other moveable component is described as an output, receiving power from the system. The ratio of input rotation to output rotation is dependent upon the number of teeth in each gear, and upon which component is held stationary.

The gear ratio in an epicyclic gearing system is not intuitive, because there are several ways in which an input rotation can be converted into an output rotation. One situation is when the planetary gear carrier is held stationary, and the sun gear is used as input. In this case, the planetary gears simply rotate about their own axes at a rate determined by the number of teeth on each gear.

If the sun gear has S-teeth, and each planet gear has P-teeth, then the ratio is equal to S/P. For example, if the sun gear has 24 teeth, and each planet gear has 16 teeth, then the ratio is 24/16 (or 3/2). This means that one clockwise rotation of the sun gear produces one and a half (1.5) anti-clockwise rotations of the planet gear.

The rotation of the planet gears, drives the annulus in a corresponding ratio. If the annulus has A-teeth, then the annulus will rotate by P/A turns for each turn of the planet gears. For example, if the annulus has 64 teeth, and the planet gear has 16 teeth, one anti-clockwise turn of a planet gear results in a ratio of 16/64, or one quarter (1/4) of an anti-clockwise rotation of the annulus.

Therefore, with the planet carrier locked, one rotation of the sun gear results in S/A rotations of the annulus. That is if the annulus has 64 teeth, and the sun gear has 24 teeth, one clockwise turn of a sun gear results in 24/64, or 3/8 of a rotation of the annulus.

The annulus may also be held fixed, with input provided to the planet gear carrier. Output rotation is then produced by the sun gear. This configuration will produce an increase in gear ratio, equal to 1+A/S.

Alternatively, if the annulus is held stationary and the sun gear is used as the input, the planet carrier will be the output. The gear ratio in this case will be 1/(1+A/S). This would be the lowest gear ratio attainable with an epicyclic gear train.

Fig. 1 - The diagram shows an epicyclic gear system.

Written by Eric Smart

The only mechanism for controlling energy into or out of the flywheel is by controlling the ratio of the CVT. The CVT is responsible for the smooth variation of ratios. The CVT may sometimes be referred to as a Toroidal Continuously Variable Transmission

(TCVT), due to the shape of the rotating discs. The main components that make up the CVT are: the rotating discs, rollers, carriages, and the pistons (levers).

Each roller is mounted in a carriage and attached to a hydraulic piston. The pressure in the pistons can be increased or decreased to create a range of reaction torque within the CVT. The movement of the hydraulic pistons alters the angle of the rollers, where the angle of the rollers in relation to the centreline of the CVT controls the transmission ratio. This ratio affects the torque transferred through the CVT.

The operation of the CVT may be explained briefly as follows:

Between each pair of rotating discs there are three rollers, each roller is mounted in its respective carrier. When the vehicle brakes the electronic control unit (ECU) engages the clutch and this allows the vehicle to drive the CVT. This leads to rotation of the input discs, which transfers this motion to the rollers, and the rollers then transfer this movement to the central disc. If the rotation of the input disc is clockwise, the rotation of the central disc is anti-clockwise.

In terms of the rollers and the discs there is never any metal to metal contact as a film of lubricating oil separates the discs and rollers. Transfer of power through the CVT takes place via this film of oil which is fed onto the surface of the discs to form a contact patch. The clearance between the rollers and the discs is sufficiently small, such that, the pressure between the roller and the discs at this contact point, greatly increases the viscosity of the lubricating oil. This oil film is highly resistant to the shearing action of the rotating disc and allows power to be transmitted between the disc and roller without metal to metal contact. This feature may lead the CVT being referred to as a traction drive, and the lubrication oil as a traction fluid.

The CVT transmits power across the lubricating oil film. This oil is a special fluid which becomes highly viscous under pressure. This means that as pressure is exerted at the contact points between the rollers and the discs, the oil resists the tendency to slide and transmits the power effectively.

The flywheel is connected to the vehicle via the CVT. Control of the energy storage in the flywheel and recovery of energy from the flywheel is managed by controlling the torque transferred through the CVT. This torque is controlled by the angle of the rollers.

Fig. 1 - Mechanical KERS unit showing the CVT

Written by Eric Smart

E = ½ I?2
Where:
E = flywheel energy
I = moment of inertia of the flywheel (ability to resist changes in its rotational velocity)
? = the rotational velocity (omega)

The flywheel as used in a mechanical KERS system works by accelerating the flywheel to a very high speed and maintaining the energy in the system as rotational energy. When energy is extracted from the system, by pressing the KERS button (on the steering wheel) the flywheel's rotational speed is reduced as a consequence of the principle of conservation of energy. Energy added to the system, under braking, results in an increase in the speed of the flywheel.

The flywheel is a composite construction, made of high strength carbon-fibre materials. This provides the flywheel with a high strength-to-density ratio. The use of lightweight modern materials allows the flywheel to rotate at speeds in excess of 70,000 rpm and these high rotational speeds occur whilst running in a vacuum enclosure. The benefits of running in a vacuum, are that aerodynamic drag and windage losses are eliminated, allowing the rotor to spin for longer, whilst it can also reach its operational speed quicker. The system as operated by Flybrid can come up to its operational speed of 64,000 rpm in less than one second, however at these high operational speeds, the flywheel undergoes an increase in diameter.

Carbon fibre composite flywheels have a higher tensile strength than steel and are substantially lighter. One of the primary limits to flywheel design is the tensile strength of the material used for the flywheel. Generally speaking, the stronger the disc, the faster it may be spun and the more energy the flywheel system can store. When the tensile strength of a flywheel is exceeded the flywheel will shatter, releasing its stored energy all at once. Fortunately, composite materials tend to disintegrate quickly into red-hot powder once broken, instead of large chunks of high-velocity shrapnel. Nevertheless the flywheel used in KERS has a strong containment vessel as a safety precaution.

Tensile strength is indicated by the maximum value of its stress-strain curve and is generally indicated when necking of the test specimen occurs. The value of the tensile strength obtained this way, does not depend on the size of the test specimen. It is, however, dependent on the preparation of the specimen and the temperature of the test environment and material.

Tensile strength is an important parameter for engineering materials used in structures and mechanical devices. Tensile strength is specified for materials such as alloys and composite materials to name but a few.

Flywheels are not affected by temperature changes as are chemical rechargeable batteries. An advantage of a flywheel energy storage system; is that by a simple measurement of the rotational speed, it is possible to know the exact amount of energy stored.

Written by Eric Smart.

KERS is a collection of parts which takes some of the kinetic energy of a vehicle under deceleration, stores this energy and then releases this stored energy back into the drive train of the vehicle, providing a power boost to that vehicle. For the driver, it is like having two power sources at his disposal, one of the power sources is the engine while the other is the stored kinetic energy.

What is kinetic energy (KE)

Kinetic energy is the energy of motion. If a body is in motion it has kinetic energy but if a body is not moving, its kinetic energy is zero. Mathematically kinetic energy is defined as:

Where:
M = mass of the body in motion
V =linear velocity of the body

Normal brake operation

In general, to slow down a moving vehicle or bring it to a stop the friction brakes are pressed. The vehicle is decelerated by converting the kinetic energy of the vehicle into heat at the surface of the brakes which is then given up to the air flowing over the brakes. That is, there is a conversion of kinetic energy into heat energy, but in this process the heat energy is lost and is not recoverable.

Kinetic Energy recovery

The principle of how kinetic energy recovery works may be displayed by the following example: a vehicle is travelling along the road in 4th gear and to slow the vehicle down, the driver mistakenly selects 1st gear instead of 3rd gear. The result is that the front of the vehicle lunges downward under severe engine braking. The speed of the engine experiences a severe and quick run up, for example from 3000rpm to 6000rpm. This is the effect of kinetic energy being recovered from the forward motion of the vehicle, directed through the transmission and into the engine. In this example, the vehicle is slowed down quite aggressively.

This in principle is how KERS works under braking; except that on braking, the energy is transferred into the storage device and not into the engine.

There are three main types of KERS systems, these are: Mechanical; Electronic and Hydraulic. The mechanical system uses a flywheel to store the energy, the electronic system uses a battery to store the energy and the hydraulic system uses some form of fluid containment, to store the energy. Due to the size of fluid storage required, it is not expected that hydraulic KERS will be used in Formula One and it is therefore expected that only mechanical and electrical KERS will be developed for Formula One. Both of these systems store energy that would otherwise be lost when braking, allowing that energy instead to be re-used when accelerating.

Williams Hybrid Power has a flywheel system, however in this article we are looking at the mechanical system as designed by Flybrid.

There are three main parts to the flybrid system: a continuously variable transmission (CVT); a flywheel; and an electronic control unit (ECU). For Formula One, flybrid developed two versions of the system, one which had the system mounted behind the engine on top of the gearbox, and the other which had it mounted at the front of the engine on the crankshaft. When mounted on the gearbox, connectivity is via a fixed ratio gear connection as follows: a gear is attached on the output shaft of the flybrid unit, which meshes with a gear on the output shaft of the vehicles gearbox. It is expected that this fixed ratio gearing is external to the gearbox.

When gearbox mounted, the KERS unit is activated when the driver applies the brakes. At this time the ECU allows the gearbox output shaft to drive and energise the flybrid unit. When the driver needs the stored energy to add to the engine power output for additional acceleration, the driver presses a button on the steering wheel and the ECU allows the flybrid unit to drive the output shaft of the gearbox.

With mechanical KERS, it is possible to modulate the energy delivery from the flywheel in short bursts of one or two second duration, as well as, all at once. However, during wheel spin, no KERS power will be delivered.

An added benefit of KERS in Formula One, is that it can be used to brake later. In the same way that shifting gear from 4th to 1st in a road vehicle causes severe engine braking, so too does KERS have the same effect on a Formula One car during braking. Hence, KERS can provide an additional 80 horse power of braking.

Written by Eric Smart.

The ACO (Automobile Club d’Ouest, organisers of the Le Mans 24 Hours) currently incorporate the following section into the technical regulations for LMP1 cars in the ALMS and at Le Mans itself (see end of feature).

KERS is currently only utilised in an ACO sanctioned series by the Corsa Motorsports Ginetta Zytek 09 HS LMP1 car in ALMS, which uses the obligatory electrical KERS system.
The ACO continue to be directed down the KERS route by pressure on two fronts.

Firstly, Peugeot, one of two major car manufacturers who are currently heavily committed to Le Mans, have long since announced their 908 HDi-FAP prototype and are apparently keen to see it enter Le Mans in 2011 at which point hybrid cars will become eligible to score championship points.

Professional motorsport always has been a marketing tool for the major road car manufacturers and Peugeot will doubtless have enviously eyed the PR successes that Audi have had in recent years with their FSI direct injection petrol technology and the TDi diesels.
The French marque is known to be bringing out hybrid production cars in the near future and we can be sure that motorsport will be a key part of that marketing strategy.

Secondly KERS is being used in Formula One and the ACO will not want to be seen as behind the times.

In Formula One McLaren Mercedes have shown that electric KERS can be deployed to good effect although there is not much doubt that the regulations (which limit the amount of energy recovery to a relatively small amount) currently dictate that the effort and cost involved in implementing it are not commensurate with its benefits.

Whilst Formula One as a whole may not be impressed by the system, what the FIA have done is to bring the systems to the attention of the racing public. So will KERS ever be more widely utilised at Le Mans than it has been in Formula One?

The ACO have made it clear teams will not be permitted to use the system to significantly reduce lap times; the implication being that if that happens then either restrictor sizes will be reduced (which would require a re-calibration of the engine) or (more likely) weight penalties would be introduced.

The spirit of the regulations seems clear; teams may use KERS only to reduce fuel consumption, with the added benefit of reducing engine load.

The regulations further limit teams to electrical KERS systems. Energy recovery systems are in general much better suited to slower, heavier vehicles than racing cars; a classic example would be a heavy bus or lorry operating in a mountainous region. Even in that case though the energy storage potential of batteries is very limited; such a case would more probably utilise a pneumatic or hydraulic system.

The ACO is at the very least to be credited for allowing KERS into Le Mans; the race has a proud record of being a proving ground for production car technology and regenerative braking and exhaust system heat recovery will continue that tradition.

Whilst trying to encourage teams to come forward with workable solutions is a positive step outlawing non-electrical energy storage systems such as inertial, pneumatic and hydraulic does seem somewhat of a shame when the technology is in its infancy in motorsport applications.

It also seems somewhat disappointing to have regulations which limit the level of energy storage whilst at the same time promising to peg back anyone who gains a lap time advantage; under current regulations the only reason to develop a system would be if a team could go a lap (or two!) extra between pit stops.

Of course some teams may currently be able to run 13.1 laps between stops and whilst some teams can do 13.9 laps; in which case the second team has a much better incentive for incorporating KERS.

So it would seem that whilst the regulations continue to prohibit the development of an inertial / hydraulic / pneumatic system which could store sufficient energy to repay its development costs in bang per buck terms, the answer is that KERS will probably be an option only for high budget teams.

The Zytek electrical KERS of course remains; but whilst it acts as a fine showcase of the company’s engineering talents their successes will still be determined by the regulations.

1.13 - Energy Recovery System (LMP1 only):
The ACO wants to give to the manufacturers the greatest possible freedom to develop and use such systems while taking a certain number of measures to control them. Energy recovery systems are free, provided they respect the following rules:

- Recovery of energy from the brakes on the four wheels or from the heat of the exhaust fumes.
- Only the rear wheels can be used to propel the car.
- Regarding energy recovery from the brakes, only electric systems are allowed.
- Only the storage of electric energy is permitted.
- The car’s minimum weight is identical to that of the other LM P1s using conventional powertrains (petrol or diesel): 900 kgs.
- The combustion engine and the electric motor must be controlled by the driver using the accelerator pedal (push to pass buttons forbidden).
- The quantity of usable energy stocked on board the vehicle must not exceed 1 MJ.
- Cars must be fitted with homologated sensors which provide all necessary signals to verify the power input and output of motors/generators and the energy released from the motors/generators in one lap

Safety rules that will be imposed by the ACO:
The use of such a system must not be aimed at obtaining additional power but at reducing fuel consumption.
The ACO may adjust the performance of any car using such a system, should it enable the vehicle to improve its lap times in a significant manner.
Competitors who want to develop and use such a system must inform the ACO beforehand and provide all relevant information as to how it works, its use, the performance expected, the safety systems installed etc.
The ACO must be kept informed throughout the development of the system and the car.
It may demand additional information and carry out any checks it deems necessary.

Written by Tom Sharp.

On 26th July 2009, Lewis Hamilton won the Hungarian Grand Prix which made the MP24-4 the first ever KERS-equipped Grand Prix winner, which is quite an accolade for MB HPE and Zytek.

Norbert Haug, Mercedes’ Motorsport Manager, recently revealed details of the team’s KERS device:

“The systems is a pure electric one in which the motor / generator is mounted in front of the engine and driven directly from the front of the crankshaft. The power control unit is mounted low down in the left hand sidepod and the battery pack is mounted low down in the right hand sidepod.

The entire system weighs 25.3 kg and can be fully charged in under half a second during braking. Haug stated;

“From what we hear, this is the lightest system available, and our pace depends on it. The McLaren cars have it integrated and we would certainly be slower without KERS".

When they were asked about the safety of the system, Norbert Haug confirmed that Mercedes has not yet experienced any KERS related problem like Ferrari by stating that their batteries have not exploded before.

"The team has done its very best in terms of safety, but obviously we cannot build a harness around the batteries as that would be too heavy. Even though our carbon fibre is extremely conductive, we have done the maximum to isolate the battery pack."

Haug concluded by saying that the Red Bull RB05 is fast because of its great aerodynamics and because they have their ballast to play with: "All non-KERS teams have 30-40 kg more ballast to play with, but the way I know Adrian [Newey], Red Bull have far more than 30 kg to spare."

Written by Top Sharp.

RET

Of the total 87 teams from 16 countries, seven were in Class 1A, of which four were from the UK. Class 1A is the competition’s low carbon category. Teams are encouraged to use green technology and alternative fuels to reduce their CO2 emissions. In an additional challenge, the teams calculate the CO2 and energy that are used during manufacture of the car. Teams compete in all dynamic and static events but are judged on their car’s sustainability rather than cost.

Under Formula Student rules the first 50% of brake pedal travel may be used for energy recovery. This is potentially important in Class 1A as the cars performance in the endurance race is evaluated by measuring the mass of CO2 produced.

The three teams which carried batteries for on-board energy storage all had the potential to use regenerative braking. University Of Hertfordshire ran a fully electric car in 2009 but decided not to use regenerative braking during the event despite having the potential to do so.

James Major, Herts’ Team Leader in the Class 1A car, told RET Monitor;

“We decided in advance not to use regenerative braking as we calculated that due to the low mass of the car (Herts’ Class 1A car came in at an impressive 260 kg), and the low speeds attained during the event, the potential amount of recoverable energy would not justify the time required to develop and test the system. This is despite the fact that our controller is very capable of managing a regenerative braking system.

“Post event analysis of the on board data acquisition system indicated that we could have recovered around 3% of the total energy which was consumed during the event. Whilst that doesn’t sound like much, 3% is a very worthwhile gain in such a closely competitive environment – imagine if one of the Class 1 cars was allowed a 3% increase in restrictor size!

“Really though, the main benefit for us would be if regenerative braking allowed us to reduce the on board battery capacity, as that would obviously reduce vehicle mass.”

Imperial Racing Green, from Imperial College, unfortunately ran out of time to pass scrutineering and therefore were not allowed to race their extremely innovative and ambitious Hydrogen fuel cell car.

Siten Mandalia, Chief Engineer, told RET Monitor;

‘We run an electric motor on each wheel and a major reason for doing this is energy recovery from braking. Our go kart project uses regenerative braking on the rear wheels only; experience with this tells us we should be looking to recover 20 to 30% of total input energy – although it’s very difficult to calculate accurately and would need track testing. Safety rules limit the rate at which we can charge batteries, so for next year we aim to have a bank of super capacitors in place.”

Oxford Brookes University’s petrol-electric hybrid Class 1A car also ran without regenerative braking. James Larminie, an Engineering Lecturer in OBU’s Mechanical Engineering Department explains;

“Since only the back axle can be used, the benefits are very limited, and there is great danger of locking the back axle. I think this is liable to be so for all very light vehicles. This was shown at the all electric motorcycle race which was part of the Isle of Man TT. I don't think any used it, except for one, who crashed. Regenerative braking only really comes into its own for heavier cars, buses and delivery vehicles.”

In Class 1, the premier category, in which cars are powered by 600 cc four stroke engines fuelled by either petrol or E85, current rules do not allow energy recovery systems.

Formula Student teams have clearly come to the conclusion that energy recovery is, as the regulations stand, of limited benefit and therefore does not justify the investment of time and money.

Written by Tom Sharp.

The 2009 Formula One regulations restrict the energy available from KERS to 400 kJ per lap – which equates to roughly 80 bhp being available for a total of just under seven seconds. The total weight of the current systems appears be range around 30 kg and as it is well known that Formula One cars carry at least this much weight as ballast, so in theory it is possible to ‘build in’ an 80 bhp ‘boost button’ for no weight penalty. Sounds like a good deal…

80 bhp equates to a little over 10% peak power, which is significant, and we can get a very good feel for the actual benefit by watching the TV footage. With the 400 kJ limit, in practice it is only of real benefit on a reasonably long straight, and it is important to consider the effect under braking. Invariably the increased acceleration is used in a passing manoeuvre, but if this also results in a higher terminal velocity, and if the two cars carry similar downforce and drag and weigh the same, then the KERS car will have to brake earlier.
Although in theory the KERS car should still be ahead at the corner apex, in practice this is not always the case. Whereas activating the ‘press to pass’ button doesn’t make a great call on driver skill, modulating braking certainly does, and we have seen several instances where the non KERS car slips back through again either under braking, or on corner exit as the KERS car scrabbles around the corner off the ideal line.

There may be an additional factor in this. Again, in theory there is an advantage to be gained with KERS under braking, where charging the motor results in a degree of regenerative braking effect. However, it might be better to characterise this as a boost in engine braking, rather than mechanical wheel braking, as it will not be applied in the same proportion between front and rear wheels as when pressing the brake pedal. Although a sophisticated control loop system would be capable of producing such an effect, it is not allowed by regulation, and the actual effect will vary depending on the ‘harvesting’ control algorithms. Above all, a driver needs to feel absolute confidence in the consistency of the brake system to make maximum utilisation of it, and if there is an unpredictable outside influence on this he is likely to be cautious.

Turning now to the 30 kg system weight, it is obvious that it will raise the overall centre of gravity. This may be by perhaps 10 mm or 4-5%, but it will also decrease the amount of front axle weight by perhaps 1.5%-2%. These figures sound small, particularly the front axle weight reduction, but they run against current tyre trends and also put a lower limit on the amount of front downforce that can be used and retain a balance in the car. The systems also require additional cooling, in much the same way that an engine giving an additional 80 bhp would require larger radiators. So although there may not be an overall weight penalty, there might well be issues in transient conditions and aero losses.

We can begin to see that although there are obvious advantages, their practical implementation relies on maintaining a balance and driveability in the car sufficient for the driver to fully exploit its theoretical limits.

Many years ago in Formula One, there were strong theoretical arguments for using V12 engines , or even H16 engines rather than V8s. And yet time and again the V8 cars would be in front of the V12s – not because they were always lighter, but because they were better balanced. Keep it light, keep it low, keep it simple…

This is not to decry the concept of KERS for racing cars, rather it is a case of there needing to be a higher proportionate energy discharge relative to engine power, and the implementation of more sophisticated control logic, and in the case of the braking system, control valves before there is unquestionable reason for running with it.

Written by Peter Elleray.

Magneti Marelli is Scuderia Ferrari’s technical partner for electronics and to get an insight into how such a system works we talked to the company’s motorsport director Roberto Dalla. As electronics is its business, it made sense for the company to investigate developing an electrical KERS solution for use in Formula One. But Dalla sees it as part of a bigger plan to develop a marketable KERS solution for road cars. He admits that Formula One is a long way removed from road cars but he says lessons learned in motorsport’s premier division would find their way into road car technology. Dalla also believes that, of all the solutions, an electrical system has the greatest potential for development.

Magneti Marelli has spent two years working on its system and is proud to be collaborating with a number of Formula One teams. From the start it was the intention to offer a turnkey solution that could be incorporated as a whole. But in time this strategy had to be revised. Because the battery is the biggest and heaviest part of the system, most teams have preferred to manage the development of that part of the system themselves. Consequently, Magneti Marelli has changed its approach to supplying discrete parts that customers can mix and match to suit their own solutions.

The Italian company’s hardware can work equally well with any form of electrical storage whether that be batteries, super-capacitors or an electrically driven flywheel. However, its favoured solution is Lithium Ion batteries. The other two major parts of the system are the motor/generator and the management electronics. The former is a synchronous AC motor/generator with permanent magnets yet this weighs only 4 kg and is about the size of a typical bottle of mineral water. A bigger challenge was the electronics. As Dallo points out, you are mixing high current, high voltage systems with microprocessors running on very low current and low voltages. Plus you are doing it in an environment with high electromagnetic interference and would be handing responsibility for installation to a third party who has many other considerations to accommodate.

Magneti Marelli’s solution was to tackle most of these challenges itself by incorporating everything into a single box. As Dalla points out, you can deliver a predefined solution and what works on the bench will work on the car. Also, shorter cable runs reduce weight and power loss as well as the risk of electromagnetic interference. The resulting unit is about the size of a shoebox but Dalla believes it offers the most efficient and reliable solution. The take up of the technology in Formula One may so far have been slow but Dalla insists that Magneti Marelli is able to supply a mature, reliable solution that can be adapted to any customer’s requirements. Understandably, he is tight lipped about who is using what Magneti Marelli parts in their Formula One cars but insists that the lessons being learned now will find their way into products marketed by the parent company for road cars.

Written by Charles Armstrong-Wilson.

For now, the flywheel is the route Williams Formula One has chosen. It created its own KERS department by adopting a company set up by Ian Foley to pursue the technology. The result is now called Williams Hybrid Power and its brief extends beyond just supplying this technology to its parent company, but includes selling the technology both inside and outside motorsport. Likewise, Flybrid chose the flywheel solution for the package it set out to market within Formula One, but has now found enthusiasm from non-motorsport customers.

Both of these solutions rely on their flywheels spinning at extraordinarily high speeds to store the required energy in a sufficiently small and light package. Foley admits to somewhere between 50,000 and 100,000 rpm while Flybrid’s John Hilton is more open and reveals an operating maximum of 64,000 rpm. Unfortunately these high speeds generate an enormous amount of heat due to friction with the air and they both opt to run their flywheels in a near vacuum. This presents the problem of how do you get the energy in and out of the flywheel without venting air into the vacuum?Foley’s solution was to use a motor generator on the drivetrain as a means of creating electrical energy and then to convert that energy into the kinetic energy of flywheel rotation so as to store it. However, rather than have a second motor generator on the flywheel shaft, he turned the flywheel itself into a motor generator. The spinning wheel forms the rotor while the stator is part of the housing around it, which creates the partial vacuum. The wheel has been made from magnetically loaded composite that is wound from fibre embedded with magnetic particles. This allows it to interact with the stator as an electric motor rotor that can be spun up to speed, so as to convert the electrical energy fed to the stator into kinetic energy of rotation. To reverse the process, the spinning wheel can be made to induce a current in the stator to thereby retrieve the recovered energy and feed it back to the motor generator at the drivetrain as electrical current. Essentially the system is an electro-mechanical battery.

By contrast Flybrid did not wish to suffer the losses associated with creating and converting electrical energy and instead opted for a purely mechanical flywheel system whereby the flywheel shaft is connected to the drivetrain and consequently must pass out of the housing creating the partial vacuum. Having thus grasped the nettle of a mechanical link Flybrid dealt with the leakage problem by employing a particularly clever seal design. It has twin lips running on the shaft and the gap between the two is filled with oil at atmospheric pressure. Because there is no pressure difference between the oil and the atmosphere, no air is drawn in. There is, though, a huge pressure difference across the inner lip between the oil and the near total vacuum in the flywheel chamber. However, the oil molecules are much bigger than air molecules making it much harder to draw them past the lip than would have been possible with air. The result is very low leakage; just 2.0 cc in eight months of testing. Once inside the chamber, the oil is a special grade for use in vacuums that does not evaporate and can be cleaned out when the unit is serviced.