Green Car Congress  
Home Topics Archives About Contact  RSS Headlines

« Ontario Biomass Could Provide 1.6B Liters of Ethanol or 6.9 TWh of Electricity | Main | GM Launches New Advanced Science and Research Center in Shanghai; Focus on Electric Drive »

Print this post

Flybrid Flywheel Hybrid System Passes First Crash Test; Developing for Road Cars as Well

28 October 2007

Flybrid Systems has demonstrated that its high-speed spinning flywheel system can survive a crash test of the severity used for Formula One frontal impact testing.

The Flybrid kinetic energy recovery system (KERS) incorporates Continuously Variable Transmission (CVT) technology sourced from the partnership of Torotrak Plc and Xtrac. Xtrac is using Torotrak’s full-toroidal traction drive technology for use in kinetic energy recovery systems within the motorsport industry. (Earlier post.) The FIA have defined the amount of energy recovery for the 2009 season as 400kJ per lap giving the driver an extra 80hp over a period of 6.67 seconds.

The mechanical KERS system uses flywheel technology developed by Flybrid Systems to recover and store a moving vehicle’s kinetic energy which is otherwise wasted when the vehicle is decelerated. The combination of gearbox-variator and flywheel form part of the driveline assembly. Energy is received from the driveline through the Torotrak CVT as the vehicle decelerates, and is subsequently released back into the driveline, again through the CVT, as the vehicle accelerates.

Compared to the alternative of electrical-battery systems, the mechanical KERS system provides a more compact, efficient, lighter and environmentally-friendly solution.

Although flywheel systems are not new, installations tended to be heavy and the gyroscopic forces of the flywheel were significant. To overcome these issues,Flybrid uses a smaller and lighter flywheel that rotates at more than 60,000 rpm. This advance in speed has been made possible by several key inventions for which the company is seeking patent protection.

Power transmission between the flywheel and the wheels is only limited by the capability of the CVT, according to Flybrid, and it is for this reason that they sought the Torotrak solution.

We believe the Torotrak solution offers the smallest and lightest package for the power output required and that the inherent torque controlled nature of the device ideally suits our application.

—Jon Hilton, Flybrid’s Managing Partner

For F1 applications, the variator and flywheel each weigh less than 5kg in a system with a total mass not exceeding 25kg. The high level of mechanical efficiency combined with the variator’s ability to change ratio very rapidly helps to optimize flywheel performance. The transmission system selects the appropriate ratio depending on the torque demand and can change its 6-to-1 ratio within one revolution.

Flybrid has also filed various technical patents to tackle the key engineering issues of safety and noise. The flywheel is made from high-strength steel and composite material and has been designed with a high factor of safety in which the maximum stresses are significantly less than in the con-rod of a conventional internal combustion engine, according to the company.

The crash test procedure consisted of spinning the flywheel up to its full 64,500 RPM top speed before disconnecting the drive portion of the spin test rig and crashing the complete test chamber. Inside the test chamber was a F1 car representative light alloy housing within which was mounted the F1 car representative 400 kJ flywheel.

The test was performed at the Cranfield Impact Centre, a recognized F1 crash test facility, and involved a peak deceleration level of “more than 20g.” The exact level cannot be disclosed as the profile of deceleration was matched to the actual crash test data from a clients F1 car. After the test the flywheel was still spinning at high speed and was completely undamaged.

Flybrid has already secured one unnamed F1 team as a customer. The company is also well on its way to bench testing a flywheel KERS system adapted for road car applications using a Chevrolet V8 engine.

Flybrid, Torotrak and Xtrac all see the potential for wider application beyond motorsport—initially on high-performance road cars—both as an aid to performance and as a means of developing vehicles with reduced fuel consumption and CO2 levels. Applied to road cars the system supports the current motor industry trend for smaller powertrains; a lightweight kinetic energy recovery system providing a means of boosting acceleration and overall performance and economy independently of the vehicle’s internal combustion engine.

An ancillary flywheel is particularly suited to stop-start driving situations when real-world fuel economy is often at its worst. In these conditions, the variator can assist the launch of a vehicle which has slowed down or come to a standstill. In heavily congested traffic, where a car is frequently stopped and restarted, the system can help alleviate the heavy fuel consumption and emissions of greenhouse gasses normally associated with these conditions. However, unlike hybrid electric vehicles, a mechanical KERS system continues to provide the benefits of kinetic energy recovery throughout the speed range, and its benefits are maintained on the open road.

On a directly comparable basis, a flywheel system offers up to twice the efficiency of a kinetic energy recovery system that stores its energy in a battery. The overall in-out efficiency of a mechanical drivetrain feeding energy into a flywheel and back out to the vehicle again via an ancillary transmission system is approximately 65-70 per cent compared with 35-45 per cent for a hybrid battery-electric system. Fundamentally, this is because a purely mechanical system doesn’t have to convert the kinetic energy into electrical and chemical energy as with a battery system.

What this means is that with a flywheel each time the brakes are applied at least 65 per cent of the energy is available to re-accelerate the vehicle, whereas the best that can be achieved with existing battery technology is 45 per cent.

—Jon Hilton

Flybrid and Xtrac will discuss the technical details of the flywheel KERS at the upcoming Global Motorsports Congress being held in Cologne on 5-6 November 2007.

October 28, 2007 in Hybrids, Motorsport, Transmissions, Vehicle Systems | Permalink | Comments (35) | TrackBack (0)

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c4fbe53ef00e54f14279b8834

Listed below are links to weblogs that reference Flybrid Flywheel Hybrid System Passes First Crash Test; Developing for Road Cars as Well:

Comments

Hybrid city buses, delivery vehicles and taxis could benefit. specially if cost is lower and efficiency is higher than with batteries.

Why not try it in the real world to see how good it really is?


Sounds like a pretty cool idea, but I wonder what the inherent parasitic losses will be? Will it be worthwhile for a consumer-market application?

Flywheel, Capacitor, Battery - all 3 are coming to road, Beginning of Electric Age.

This might work in the context of F1 racing but the flywheel bearings will be shod after each race. Since this concept sticks with a mechanical coupling to the driveshaft, friction losses will be high. The system is really only useful for recuperative braking going into and boost acceleration coming out of a very fast corner.

The only other applications for flywheel storage in vehicles that I'm aware of are Rosen Motors' gas-turbine hybrid (company folded) and the Frauenhofer Institute's AutoTram.

The flywheel weighs 5kg and can store 400 kJ.
So, a 45kg flywheel can store 3600 kJ = 1kWh.
While the principle is nice, flywheel technology is not expandable to an equivalent of 'full electric'.
Since full-electric vehicles will have in-wheel motors, it will be more difficult to include it parallel with the electric engine. ultracapacitors will probably be much easier for regenerative breaking.

This was tried & found inadequate several years ago.
It may have some value on trains & ships, but for a cars, batteries & hyper-capacitors show more promise.

ke = m v^2

So if you could keep people's top speeds down (in town) you could eliminate the need for regenerative braking.

If you reduce the top speed from 35 to 30 mph in town you reduce the energy by 36% which is about what you can expect from battery powered regenerative braking.

This could be done electronically, with the user opting in for much less cost and disruption than a regenerative braking system ( or in conjunction with one ).

The main thing is to prevent people braking hard at the traffic lights and burning off the energy they have in the vehicle.

"The overall in-out efficiency of a mechanical drivetrain feeding energy into a flywheel and back out to the vehicle again via an ancillary transmission system is approximately 65-70 per cent compared with 35-45 per cent for a hybrid battery-electric system."

That seems to be naively assuming an in/out efficiency typical of nickel-based batteries, but certainly not of ultracaps or lithium-ion batteries.

"The flywheel weighs 5kg and can store 400 kJ.
So, a 45kg flywheel can store 3600 kJ = 1kWh."

Which means ~22Whkg, which would be totally unacceptable for a battery in a hybrid road car.

However, for high power use in F1, you WOULD have to put a lot of batteries together to discharge 80hp of power. @5,000W/kg, you'd need~12kg of batteries. Add on about 12kg for the motor, and you're about twice the weight of the flywheel system before taking into account the CVT for either.

So scratch batteries. High power ultracaps could probably be competitive with the flywheel though, and require far less maintenance.

"So scratch batteries. High power ultracaps could probably be competitive with the flywheel though, and require far less maintenance."

how would ultracaps work with a CVT? An F1 car is not an electric car, nor does it have in-wheel motors.

Anytime I see technology like this appear on race cars I smile, because that's the technology test bed for what makes it into cars we drive daily. Will it work on a daily driver? Not for me to say, but I'm glad to see F1 is doing something.

Has NASCAR gotten Cams yet?

"how would ultracaps work with a CVT? An F1 car is not an electric car, nor does it have in-wheel motors."

Just have the torque input into the CVT be an electric motor, not the flywheel.

AES stated: "Which means ~22Whkg, which would be totally unacceptable for a battery in a hybrid road car. "

This is actually very good, since the NiMh battery of the Prius can only store 33wh/kg. (40kg total pack weight for 1.3 kWh.) Worse, the battery is only charged to 70% and discharged to only 30%, thus only 40% of battery capacity is utilized, to ensure longevity. Multiply 33 Wh/kg by 40% and the Prius' battery can manage only 13 Wh/kg in real-life. And even much worse, the car also needs a motor and a power inverter to translate the battery energy into kinetic energy, which would take another 30kg (assuming a motor just big enough for the battery and no more). And even worse, if you push an electric motor or the battery to its max, efficiency drops precipitously, down to the 50%, due to Ohmic losses in the motor and battery, meaning that you will have to divide the Wh/kg by half again to adjust for the loss of kinetic energy to heat during the conversion.

So, 22 Wh/kg of pure kinetic energy from a flywheel energy storage system is VERY GOOD. 4-6 folds better than current NiMh/motor/inverter technology. Worthy to be placed in the ultimate vehicle of high performance like the F1 racer. I wish I can put one of those in the rear axle of a 200-hp Honda Civic in order to race with the big boys! I'd bet you can shave a few seconds off the already fast 0-60 mph time.

Roger- I still have concerns about putting this on a road car.

1)Putting this in a road car would require much more maintenance than an electric system, and the basic principle of the idea would be unable to evolve beyond being a mere energy recovery system, as opposed to a primary energy storage system

2) More importantly, the kinetic system is totally dependent on regenerative braking, and cannot be "recharged" while driving at highway speeds like with an electric system (i.e. a Prius acting as a "series-parallel hybrid"). So the article's statement about the flywheel providing continuous benefit is perhaps a bit misleading, as it's a transient benefit. There's also the matter of the flywheel slowing down overnight while parked.

3) TOY's Prius batteries are also fairly wimpy and low-end as far as HEV batteries go, so it's not fair at all to compare those to the top-end F1 flywheel! Cobasys NiMH cells can get up to 55Wh/kg, and the EnerDel cells at least 80. Cobasys also expects improvements in construction to further improve NiMH. Not that I'm a particular fan of this chemistry, but it still has life left in it.

If you want to do a useable energy comparison like you did earlier, and pessimistically assume that only 50% of a battery's energy is actually usable, the batteries still come out on top, and have room to grow.


For F1:
In terms of power generation, a Maxwell ultracap can put out 13, 587W/kg (18 horsepower/kg). To get an output of 80 horsepower, that's 4.44kg - easily comparable to the flywheel at "less than 5kg". The downside is less energy stored per mass.

Altair nanotech batteries manage 4 kW per kg (so 80 hp limit would weigh in at 15 kg), >98% efficient, and at 100 Wh/kg you could store (although by regulations probably not use) 1.5 kWh, or 5,400 kJ.

I think given the recent 10 year engine freeze, the FIA should allow the teams to go balls-out technical and innovative with no restrictions whatsoever on KERS output or onboard storage. That would lead to spectacular improvements for the technology that would benefit road cars no end.

As already noted, flywheels have a power and energy rating somewhere in the middle between ultracaps and batteries. Once major factor, which has been touched also is the "self-discharge" time. Both ultracaps, and even more so flywheels, have extremely high self-discharge rates. Also, the bearings of a flywheel are one very important part - large scale flywheels (like those powering the magnets in JET and upcoming ITER) also run in a hydrogen filled casing, to reduce atmospheric drag.

Furthermore, I don't think that a pure mechanical system with a flywheel as energy storage has much merit in real-world stop and go driving. In addition to the above mentioned points, the article claims that the CVT has a ratio between 1 and 6; thus you would still need a clutch to provide launch power (accelerating from a standstill), and this clutch would reduce the overall efficiency considerably.

In addition, the article claims that the flywheel could survive a frontal crash unharmed - that rules out all ultra-low friction bearings, as the acceleration forces have to be transmitted to the spinning flywheel. The article indicates, that there is only a single flywheel installed in this setup - which means that during spin-up or spin-down, there will be some unbalanced gyro forces. I guess, that they install the flywheel in such a way, that you have additional downward force on the front car axle during spinup (braking), and additional downward force during spin-down (acceleration) - meaning that the flywheel axle is aligned perpendicular to the vector of primary motion, in the horizontal plane.

However, that would result in interesting forces during cornering.

There is one more interesting story I want to share: according to what I have read once, McLaren and BMW were already doing tests of kinetic energy recovery sytems in their mid-1990 F1 cars. However, due to complains by the other teams (most notable Ferrari), the FIA outlawed all such attempts at the time, stating that this would give an unfair advantage in the F1 racing. I find it very gratifying, that finally, Max Mosley and these guys found their senses again.

In order to push competition further into the "green" aspect of motorsports, it might also be nice to have a long-term rule in place, reducing the allowed fuel per race from year to year. Already there are regulations requiring that a single engine lasts at least two races (qualification and the race) - just compare that to how long a decent internal combustion engine has to operate in the real world...

If someone could find that article about late 1990 FIA :)

Williams were also looking at installing ultracaps and motor in the late '90s too. However, they found that the power advantage (back then at least) was not as good as the weight saving from not having the system. I think it would be a different story today - for example, Razertech have a motor the size of Prius MG1 that puts out over 500 hp.

Incidentally, some teams are also looking at hydraulic accumulators for their KER systems.

You guys saying that "if 5kg = this, then 45kg = that" are funny. You can store 10x as much with the SAME 5kg just by putting it further away from the axis. Same RPM, same weight, but longer moment arm and you increase the energy storage. The only limitation is the strength of the materials holding it all together. Flywheels are very much NOT a matter of weight = energy the way chemical batteries are. The stronger the materials are, the greater the energy storage becomes at the same weight and rpm by simply changing the shape of the flywheel.

The only limitation is the strength of the materials holding it all together.

That is the key limitation. Energy storage flywheels are engineered to max out the structural strength of the material (with a safety margin, of course). Move the mass farther from the axis or increase the RPM and your wheel will fail.

22 Wh/kg of pure kinetic energy from a flywheel energy storage system is VERY GOOD.

22 Wh/kg was just the flywheel. System energy density is 4.5 Wh/kg (111 Wh/25 kg).

Sid:
You are wrong - the amount of energy storable in flywheel is pretty independent of diameter, just depends on tensile strenght of material. Larger diameter - less rpms.
Small diameter flywheels are favourable because they have lower angular momentum for the same energy, thus lower gyroscopic effect.

Energy stored in a flywheel is 1/2 * J * omega^2, with J the polar moment of inertia and omega the radial frequency.

J is the integral of distance squared over infinitesimal mass, so mass at twice the distance from the axis of rotation will generate 4x the rotational inertia. Centrifugal forces increase linearly with the distance, but so does circumference. Assuming radial stresses are low, which is the case in a spirally wound composite, hoop forces therefore increase with the square of the distance.

The theoretically optimal shape is the one in which all parts of the material are stressed equally. In an isotropic material, the result is a Laval disk. In a composite structure based on constant-diameter fibers, it is not possible to achieve an optimum in this sense. For manufacturing and packaging reasons, a cylindrical drum is usually chosen instead.

The outer diameter of the flywheel represents a compromise between total mass of flywheel plus containment structure, required capacity and power, rotational inertia, aerodynamic friction losses and gyroscopic forces on the bearings. The latter are proportional to both the radial velocity of the flywheel about its own axis and, to the component of gross motion radial velocity of the containment structure perpendicular to that axis.

These trade-offs will yield different results for e.g., automotive applications and stationary ride-through solutions for electricity grids.

@realalarms

Presumably the flywheel will slow down quite quickly if the stored energy is not used? The flywheel will slow down due to friction in the bearings and seals but because it runs in a vacuum the losses are low. We estimate that for the flywheel to get from full speed to half speed will take around an hour.

from the Flybrid FAQ:
http://www.flybridsystems.com/FAQ.html

@KS: 3600 sec energy half-live for a zero-load flywheel (even running in a vacuum) energy storage system is indeed a very extreme example of a short-duration energy storage system. I hear that modern ultracaps have an energy half-live in the order of tens of hours (tenfold less energy loss), and a NiMH battery, as used in current production vehicles, the energy half-live is measured in months (about 4000-5000 fold better). Remember that not even the hyped hydrogen economy`s fuel tanks - especially the cryogenic kind - have an energy half-live in the order of months. Hydrogen is quite inventive in finding ways to leak (tunnel) even through the most dense materials...

This just underlines my point that pure mechanical systems won't have any significant real-world applications (when seen in context with the other limitations). Sure, certain niche markets, like Formula-1 or grid power conditioning, do exist, but these are low-volume.

Having that said, I`m still looking forward to the `09 F1 tour, when the races might become more interesting again. With such an short-lived energy storage system, the tactics of the individual driver might finally start to make a real difference again. With the current state of F1 affairs, those races are a rather dull and unthrilling expirience...

Reducing the primary energy allowed for each racecar on a long-term basis would, I believe, start investigation into real energy efficiency measures.

@KS: 3600 sec energy half-live for a zero-load flywheel (even running in a vacuum) energy storage system is indeed a very extreme example of a short-duration energy storage system. I hear that modern ultracaps have an energy half-live in the order of tens of hours (tenfold less energy loss), and a NiMH battery, as used in current production vehicles, the energy half-live is measured in months (about 4000-5000 fold better). Remember that not even the hyped hydrogen economy`s fuel tanks - especially the cryogenic kind - have an energy half-live in the order of months. Hydrogen is quite inventive in finding ways to leak (tunnel) even through the most dense materials...

This just underlines my point that pure mechanical systems won't have any significant real-world applications (when seen in context with the other limitations). Sure, certain niche markets, like Formula-1 or grid power conditioning, do exist, but these are low-volume.

Having that said, I`m still looking forward to the `09 F1 tour, when the races might become more interesting again. With such an short-lived energy storage system, the tactics of the individual driver might finally start to make a real difference again. With the current state of F1 affairs, those races are a rather dull and unthrilling expirience...

Reducing the primary energy allowed for each racecar on a long-term basis would, I believe, start investigation into real energy efficiency measures.

This is valuable research and could make possible dynamic braking energy recovery for every future vehicle.

HEVs and PHEVs have been seen to need large banks of electrical storage. There's an alternate approach that mimics energy recovery in animals.

Humans recover 70% of the energy required for walking. Kangaroos have a 90% recovery rate for hopping. Northern pike recover the energy from the vortex generated by their tails. What these and many more examples from nature demonstrate can be thought of as short-duty energy recovery. Energy is recovered, stored and released in one propulsion cycle. There is no long-term storage, as has been the norm for HEVs and PHEVs.

The great benefit of hybrid vehicles in stop-and-go city driving has been their dynamic braking capability. But they have been saddled with heavy, complex and/or expensive energy storage banks.

It's the efficiency of the whole system that counts. Electric drives and fuel cells may be efficient, but their energy delivery systems are not. ICEs, for all their failings, are power dense and allow the development of much lighter weight vehicles requiring less overall power.

This type of flywheel energy storage, which is short-duty cycle unlike past systems which attempted to be long-duty cycle, is light, compact, comparatively low-cost and low-tech, and could be adapted to any vehicle design. It's a paradigm shift in energy storage, similar in concept to the way nature usually does it.

While I agree that using mechanical energy storage is a step back on the road to BEVs, I struggle to see why people on this board are negative on flywheel KER.

Both CVTs and flywheels are already mass produced. Both are relatively cheap (compared to a full HEV) and existing manufacturing infrastructure should adapt to the different specs of a KER relatively easily.

If a flywheel based KER can improve city cycle fuel economy by 25% while only adding $500 to vehicle cost, why complain? Shouldn't we embrace it?

Such a system, imo, could allow for substantial engine downsizing. When a car is initially started the engine could be allowed to temporarily rev to build up energy in the flywheel. That energy could then be released during initial acceleration. The idea would be to provide additional performance while maintaining a small engine (hey, American consumers always want more performance, eh).

In my mind, a 1L turbo 4 putting out ~100hp would be combined with a downgraded F1 system (40,000rpm, same weight, hopefully reduced rpm would increase bearing life, but I'm not a ME) to provide ~140hp for the first 6 seconds of acceleration. The downsized engine would provide substantial fuel economy savings at cruise while the flywheel KER would provide a performance boost and increased fuel economy in city driving.

What's not to love? Where am I off?

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Green Car Congress © 2017 BioAge Group, LLC. All Rights Reserved. | Home | BioAge Group