Propel Fuels Announces $20M Funding Round to Expand Alternative Fuel Station Network Across California
VW Unveils New Touareg and Touareg Hybrid

Porsche 911 GT3 R Hybrid Using Williams Flywheel KERS

Phantom diagram of the 911 GT3 R Hybrid. Click to enlarge.

Dr. Ing. h.c. F. Porsche AG will introduce the 911 GT3 R hybrid for production-based GT racing at the upcoming Geneva Motor Show in March. The hybrid is equipped with a flywheel energy recovery system (KERS) developed by Williams Hybrid Power initially for use in Formula One racing. (Earlier post.)

The 911 GT3 R Hybrid features an electrical front axle drive with two electric motors developing 60 kW each supplementing the 480-bhp (358 kW) four-liter flat-six at the rear of the 911 GT3 R Hybrid. Instead of the usual batteries of a hybrid road car, an electrical flywheel power generator fitted in the interior next to the driver stores recaptured energy and delivers it to the electric motors.

CAD rendering of the Magnetically Loaded Composite (MLC) flywheel. Source: WHP. Click to enlarge.

The flywheel generator itself is an electric motor with its rotor spinning at speeds of up to 40,000 rpm, storing energy mechanically as rotation energy. The flywheel generator is charged whenever the driver applies the brakes, with the two electric motors reversing their function on the front axle and acting themselves as generators. Then, whenever necessary—i.e., when accelerating out of a bend or when overtaking—the driver is able to call up extra energy from the charged flywheel generator, the flywheel being slowed down electromagnetically in the generator mode and thus supplying up to 120 kW to the two electric motors at the front from its kinetic energy. This additional power is available to the driver after each charge process for approximately 6 - 8 seconds.

The GT3 R Hybrid. Click to enlarge.

Depending on racing conditions, hybrid drive is used in this case not only for extra power, but also to save fuel. This again increases the efficiency and the performance of the 911 GT3 R Hybrid, for example by reducing the weight of the tank or making pitstops less frequent.

After its debut in Geneva the 911 GT3 R Hybrid will be tested in long-distance races on the Nürburgring. The highlight of this test program will be the 24 Hours on the Nordschleife of Nürburgring on 15-16 May. The focus is not on the 911 GT3 R Hybrid winning the race, but rather serving as a spearhead in technology and a “racing laboratory”, according to the company.

Although he energy storage system was originally developed for use in Formula One by the AT&T Williams team, Williams Hybrid Power is now focused on applications in road vehicles. In October 2009, Williams F1 and the Qatar Science & Technology Park (QSTP) formally signed an agreement to inaugurate the Williams Technology Center (WTC). The WTC is initially tasked with the progression of two Formula One inspired R&D projects with clear commercial goals.

The first is the development of an industrial-application large Magnetically Loaded Composite (MLC) flywheel—essentially a wholly composite flywheel which integrates the magnets of the electric motor into the composite. The second is the advancement of Williams F1’s simulator know-how for competition and road car application.



This car is not going to crest hills very well. Nor will it cope well with rolling back off the kerbs or banked curves at the Nurburgring.

It will turn absolutely fine around its vertical axis of rotation, but 40,000 rpm of rotational inertia is going to offer huge resistance to that car leaning.


So, are you indicating that both Williams and Porsche are complete idiots? Maybe they will come to you if (?) they find out that it does not work.


I'm sure Porsche/Williams is aware of the problem mentioned by clett, they're not idiots. I just have a hard time envisioning how they're going to deal with it.


They could have gimbal mounted the unit so reasonable angles of vehicle tilt could be accommodated with the flywheel remains flat.


They could have gimbal mounted the unit so reasonable angles of vehicle tilt could be accommodated with the flywheel remains flat.


The vertical axis battery flywheel inertia counters the the horizontal engine flywheel inertia - besides the additional 160-bhp.

Henry Gibson

The interesting facts not mentioned are that the amount of energy allowed to be stored by the rules of racing is limited. This means also that the flywheel is limited in size. There are at least five other operating flywheels in this automobile including the wheels and the traditional engine flywheel. The electric flywheel could be of such a weight and size and mounting that it exactly counter balances the traditional flywheel. Or it could be mounted to help avoid tipping over in curves. There is no reason why its rotation cannot be reversed so that it matches the curves. There was once a demonstration railroad with a single rail on the ground and flywheel stabilized waggons and locomotives.

The energy issues are more interesting. The fact that the energy storage is limited by rules is a complete and total validation of the hybrid concept. The fact that ultra-capacitors are not being used is a validation of the fact that present ultra capacitors do not store much energy per unit weight. A specially made battery pack of any chemistry, even lead, that could provide the power desired would have far too much energy capacity. EFFPOWER could have supplied a simple delightfull lead acid prototype bipolar battery that incorporated CISRO lead-acid-ultra-capacitor technology, but it would have been very difficult to make it have low enough energy to fit the rules.

It would be interesting to have a competing automobile that used Artemis technology as a hydraulic hybrid. The energy stored in such machines is stored as compressed air. There is little or no weight advantage to having a large tank compared to many small ones, so many small tanks could be fit into convenient places. I am sure that the Artemis technology can have higher power at lower cost and complexity, but then I persist in believing that the Kitson-Still diesel steam locomotive should have only slightly more modern and powerfull versions operating, not only in the UK but also throughout the world to replace diesel electric locomotives.

The ventilation of the Chunnel is quite adequate enough to allow the operation of such Kitson-Still locomotives fitted with the presently used catalytic converters so that electronic circuit breakdowns due to fine snow can be avoided. Compressed natural gas can be used to supply most of the fuel in such locomotives as it does not add much extra weight to a train to pull such CNG waggons.

Bosch please use the Artemis technology for an F1 demonstration. ..HG..


Why do you think Williams never raced with their flywheel KERS at any F1 race in 2009? It was for this reason.

Roger Pham

A simple 2-axes gimbal mount is all that would be needed to insulate the car from gyroscopic forces in the pitch and roll axes. The flywheel is electrically coupled to the drive train, not mechanically coupled, so it should further facilitate mounting and disposition of the flywheel. Flywheel energy storage or KERS, is ideal for high-performance vehicles which need to deposit and withdraw energy rapidly.


"Why do you think Williams never raced with their flywheel KERS at any F1 race in 2009? It was for this reason."

Where's the data/internet link to this statement?

Henry Gibson

If it were not for the rules, it would be better to decouple the engine and wheels, very similar to diesel electric locomotives. AC propulsion could put a motor drive on each of four wheels and supply electronic drives for each wheel motor and also an electronic drive for the engine. AC propulsion's trailer range extender TZERO was a long ago prototype of such a system. Energy would then be stored in the engine flywheel which could be beefed up with all kinds of KEVLAR and carbon fibre. Ian Wright of Wright Speed has demonstrated that such a car would be unbeatable in any race with a well trained driver with not a single cell of a lithium battery within a hundred feet except in cell phones. Fully electrically operated valves would make the engine highly economical to operate. It might be necessary to inject air and fuel directly into the catalytic converter just to keep it operating at highest efficiency. It would be best to reform the fuel into hydrogen and CO first but all of this can be computer controlled.

Such a machine would not even need a starter motor or any belt powered accessories such as alternator, water pump, air-conditioner or power steering because electrical powered versions are more efficient and now lighter weight. Starting would be done by computer calculated fuel injection, spark-activation and valve control. ..HG..


I think two of those flywheels mounted side by side, rotating in opposite directions would be more weight efficient than a gimble, and eliminate any precession forces.

Anyone have an idea how the kw/kg and kw-hr/kg compare to lithium batteries?

I like the fact that battery degradation is not an issue.


Interesting choice to use the flywheel. I wonder how much it weighs?

120kW for 8 seconds means it can deliver 267Wh of energy. That's enough for the 1200kg GT3 R to accelerate an extra 75kph coming out of a curve (ignoring friction losses, etc) on top of whatever power the ICE engine is providing.

It would take about 44kg worth of supercapacitors with todays production technology to provide that same energy (though there are some experimental supercaps that could cut that down to about 15kg).

For A123 lithium phosphate batteries it would only take about 2 kg to provide the energy storage, but you'd have to have about 23kg to provide the 120kW power output. And of course you'd have to deal with life cycle issues that wouldn't be a problem with the flywheel or supercaps.

Roger Pham

Without gimbal mounting, there will be crushing gyroscoping forces on the flywheel's bearings whenever the car rotates on the pitch or roll axis. This will likely put a lot of friction that will rob power from the flywheel, and if mechanical bearings are used, will lead to accelerated wear. If magnetic or air bearings are used, they must be protected from high forces, because they cannot withstand high forces.


This system is different than the one mentioned in the previous post about their flywheel because that one had a max power output of 60kW and this one has 120kW.

Maybe GdB is correct and they just put of them side by side spinning in opposite directions to help deal with some of the gyroscopic effects, but that seems like it still wouldn't help with the roll needed for the car to corner. Maybe they dealt with that as some folks are suggesting with gimbal mounts?

Regardless, the original article said that unit put out 3kW/kg and this puts out 120kW so clearly it's at least 40kg as well....about the same as the supercaps with today's production supercaps.

I would have expected a better power and energy density that would have made it better than older supercap technology, not about the same. But the supercaps have a lot of voltage drop off as they deplete their energy so maybe it's easier to handle the flywheel on the electronics side.


The flywheel apparently (from the figures) has ball bearings and no gimbals.

A change in road camber would result in very modest pitch forces on the car and negligable friction increase.

Likewise a pitch change would result in modest roll forces.

These tipping forces would obviously be no problem for the wheel bearings and likewise would be only moderately greater (because of closer coupling and higher speed) for the flywheel bearings. If there was any concern, they would use/add roller bearings.

No energy is extracted or added to the flywheel by these redirections in the angular momentum vector.

Roger Pham


Due to the small angular velocity change of the car, there is no need for a fullfledge gimbal setup. Just a set of bi-directional spring suspension mount for the flywheel unit would simulate the function of the gimbal when only small angular velocity change is occuring during normal driving. The exception is when the car violently flipped over or sideway, but then, this means GAME OVER for that race vehicle.

Even though no energy is extracted or added to the flywheel with angular change, more friction would be experienced by the flywheel bearing that would drain the energy from the flywheel. Since power equals to force times velocity, the very high rotational speeds of the flywheel means that even a small force of friction would amount to a significant power drain from the flywheel.


This thing would have an angular momentum > 3,000 kgm2s-1!


So many inertia problem comments that Porsche apparently hasn't noticed.


After the ferrari 955, Porsche ! Yeah

far away the air'car arrive ... you can order it :)

Account Deleted

if they are idiots they probably not doing such hybrid invention using williams flywheel.

Mario Gottfried

The solution is many faster, lighter, stronger flywheels for efficient KERS; shape enhanced to go faster to store more energy for the weight. A common frame for sets of 6 flywheels with motors mounted in a hard vacuum, burst safe, promising optimized flywheel performance.
I invented this new battery (US &,232,671 B1), I would love for the engineers at Porsche to want to optimize not one, or two counter rotating 3 pairs in 3D, out performing other arrangements. The invention is named KEMEA, or 3Dimensional, Counter Rotating, Kinetic Electro Mechanical Energy Accumulator "3D CR KEMEA"


Kelly - regarding inertia please see

The inertia isn't a problem. The added "gyro-effect" is probably larger from the crank shaft and engine flywheel.
A faster and smaller flywheel is even better in this sence,


I wish I could revamp my 911 to be that tech'd out. As for now, all I've been able to work on is it's Porsche exhaust.

The comments to this entry are closed.