« Wyoming Soliciting Coal Technologies Research Proposals | Main | Tokyo Gas, Mitsubishi Heavy Improve Hydrogen Separation Membrane Reformer »
Flywheel and Infinitely Variable Transmission for Mechanical Hybrids
3 May 2008
 |
| Schematic of FIVT. Click to enlarge. |
Researchers at the University of Warwick (UK) are developing a flywheel and infinitely variable transmission (FIVT) for application in a mechanical hybrid powertrain. The work on FIVT is the outcome of a larger mechanical hybrid project created by the university to assess the potential of mechanical hybrid systems—flywheel, pneumatic and hydraulic—to compete with electrical hybrid systems.
Andrew Bass and Alireza Veshagh developed a simulation tool to evaluate the different types of hybrid powertrains, and found that the Warwick FIVT offered a better cost-benefit to mid- and high-mileage users than any of the other hybrid systems evaluated.
The Warwick FIVT unit consists of an IVT (infinitely variable transmission) and a flywheel assembly. Any IVT can be used in the FIVT system; the Warwick researchers have been using the Torotrak toroidal IVT (earlier post) unit in their studies up to now.
In the FIVT, the input shaft of the IVT is connected via a clutch to the prime mover; the output shaft is connected to the differential. The flywheel is connected, via a clutch, to the input of the IVT unit. The FIVT unit consists of the following:
- IVT unit;
- Flywheel and flywheel clutch;
- Engine clutch;
- Vacuum pump;
- Bearings and seals;
- Speed sensor to measure flywheel speed; and
- Auxiliary gear to drive vehicle auxiliaries.
While most flywheel hybrid systems spin their flywheels at high speeds, the FIVT flywheel spins up to the maximum speed of the engine, although it can potentially reach higher speeds under braking if the IVT unit has an input speed range greater than the engine speed range.
By spinning the flywheel at lower speeds than other systems, the FIVT delivers increased efficiency with reduced stresses, according to the team. This, in turn, allows less expensive flywheel materials to be used—steel, for example, instead of carbon fibre. Denser materials also could reduce packaging dimensions. In their simulations for the US FTP cycle, the Warwick researchers used a 600 mm diameter, 9 mm thick steel flywheel which stored 400 kJ of energy.
In simulations for a 17-tonne bus and a 2.6-tonne SUV (Chevrolet Suburban), they found that the FIVT provided the greatest increase in fuel economy across different cycles, although the electric parallel, flywheel parallel and FIVT outcomes were similar. One significant downside to the FIVT is in an idling scenario; for long idles, the energy in the flywheel can be exhausted.
Barr and Veshagh note that there are two other proposed systems that integrate a flywheel with an IVT, the first developed by Leyland in the 1970s, and the second a more recent system by Kestrel Powertrains. Kestrel also uses the Torotrak IVT. Both these systems use a gearset to connect the flywheel, the researchers say, which requires extra packaging space, increases the cost, and reduces the fuel economy gains due to the additional losses in the gearset.
The researchers have developed a more detailed simulation tool to support refinement of the flywheel design (material, size, clutch, auxiliary strategy, etc.) and to develop a more detailed control strategy.
Resources
Andrew Barr and Alireza Veshagh (2008) Fuel Economy and Performance Comparison of Alternative Mechanical Hybrid Powertrain Configurations (SAE 2008-01-0083)
May 3, 2008 in Hybrids, Transmissions | Permalink | Comments (28) | TrackBack (0)
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/t/trackback/22062/28734336
Listed below are links to weblogs that reference Flywheel and Infinitely Variable Transmission for Mechanical Hybrids:
Comments
I think flywheels may have a place in stationary applications. In transportation you have the flywheel effect on handling. If you put the axis vertical it resists pitch and roll. If you put it horizontally (not recommended) you affect yawn, which is turning.
I would say that if batteries and fuel cells we not far along, then flywheels might be a way. Series hybrids, combined cycle, compounding and other methods are more practical. Maybe in large trucks and buses...we will see.
Posted by: SJC | May 3, 2008 10:04:18 AM
SJC,
Trains don't have to make sharp turns, so maybe hybrid engines could take advantage of flywheels.
Posted by: JamesEE | May 3, 2008 10:08:22 AM
Oh yeah, they have their place, but I would not want to see one mounted wrong in a Mini Cooper :)
Posted by: sjc | May 3, 2008 10:34:22 AM
SJC you can offset the Gyroscopic effect using to wheel spinning in opposite direction.
These systems can be more efficient and less costly than electric hybrid, though they don't provide the same benefit such as moving emission free at low speed in traffic jam where ICE are so polluting and inefficient.
The problem is to make these system compact enough to fit in a small car, the Torotrak look promising
Notice that hydraulic system have been proposed by Eaton that look very interesting for heavy truck or bus, but are propably less suited for car.
we'll see
Posted by: treehugger | May 3, 2008 11:35:25 AM
It would seem to me that a flywheel would be ideal for recovering the momentum that is lost breaking while compressed air or ultracap techology for longer periods of storage. Some variation of the three should find widespread use in transport of the future.
Posted by: ken | May 3, 2008 11:38:28 AM
Since we know the Lithium-ion cells will flood the market in 2010 it really doesn't make sense to do flywheels for automobiles at this moment.
I do see there is a lot of room in range extender ICE improvements, maybe even enough to offset the generator losses.
Posted by: Lulu | May 3, 2008 1:29:50 PM
So, if you have a counter rotating one maybe there place is between super caps and batteries. They do not like to sit for long, they run down. Caps for braking and take off and flywheel to buffer the currents to the batteries.
What I read, you have to have high RPM to make them pay off. That means air bearings or magnetic bearings which cost a lot. Then you have the safety aspects. Carbon fiber case to catch the pieces and keep the weight down. Does not sound like a cheat solution for every day cars.
Posted by: sjc | May 3, 2008 1:42:05 PM
The energy to accelerate to 60 mph is comparable to the energy needed to go 60 mph for 1 mile. A cheap steel flywheel can provide this. Stan
Posted by: Stan Lass | May 3, 2008 2:38:25 PM
Perhaps a carbon fiber flywheel could be used. In the event of delamination the CF would shatter into thousands of tiny fragments, absorbing much of the energy that would otherwise cause a safety hazard.
Posted by: GreenPlease | May 3, 2008 2:43:21 PM
Lulu
The statement that Li-ion will flood the market is well over stated at this point, maybe you are right but batteris have a long history of unkept promises, so let be causious before dumping alternative approaches. The problem of batteries is still they cost and their weight, as well as their life duration, so even if I believe in PHEV, I am aware that it won't be a panacea before long since history tell us that progress in battery are inherently slow
Posted by: treehugger | May 3, 2008 3:12:03 PM
Even if lithium and PHEVs happen, people do not go out and buy new cars because gas is expensive. It will take a long while to get enough of them out there to make a difference.
Biofuels, PHEVs, EVs, carpools, mass transit, telecommuting and everything else will be needed. I keep coming to the conclusion that we will have lots of options, it will hit the fan and everyone will want all of them yesterday. It does not work that way, but that seems to be the pattern.
Posted by: sjc | May 3, 2008 4:10:03 PM
400 kJ energy storage or 111 Wh is one 10th energy stored in a Prius Ni-MH battery. It's still a long way to reach a number around that realistic target. A heavy steel case flywheel with a low RPM value, it's may be too just for a clear advantage solution.
Posted by: raymond bonnaterre | May 4, 2008 12:59:30 AM
>>The energy to accelerate to 60 mph is comparable to the energy needed to go 60 mph for 1 mile. A cheap steel flywheel can provide this. Stan
-------
That is incorrect. Totally depends on the vehicle but for most it takes far more power get to 60mph then it does to stay there.
There are going to be more LIon batteries but there is a shortage of lithium so a flood may not be the right term. There may not be enough lithium to hybridize even half the car fleet currently in existence. We will need other technlogies like ultracaps and flywheels.
Posted by: hampden wireless | May 4, 2008 10:54:22 AM
The lithium is abondant enough to make 2billions PHEV cars, though the production capacity might be limited for some time. The main advantage of flywheels is that it is a much cheaper approach than a HEV with potentially similar performanace in term of efficiency
Posted by: treehugger | May 4, 2008 7:33:07 PM
I only took elementary mechanics but I don't remember that two flywheels turning in opposite directions would offset the gyroscopic effect.
Counter rotation offsets torque. That is nice but another matter.
I await the word of a mechanical expert about counter rotation.
Flywheels seem well suited to storage of power from solar. Bunker them underground.
Posted by: K | May 4, 2008 10:05:29 PM
I don't know why SJC is worried about yawn, which is a motion of the mouth. He notes that if you put the flywheel horizonal then it affects turning. Well yes; the wheels are flywheels with horizontal axes and they affect turning too.
Treehugger is worried about the energy lost breaking. I guess he means braking, ie using the brakes. Hopefully cars don't break too often. The energy lost when a car breaks in a crash isn't going to be recovered by a flywheel.
Posted by: nospam | May 5, 2008 6:18:34 AM
SJC: 400 kJ energy storage or 111 Wh is one 10th energy stored in a Prius Ni-MH battery. It's still a long way to reach a number around that realistic target.
No.
400 kJ is a realistic target and that is the reason it was chosen as the energy stored or retrieved per lap in formula 1 - 2009 rules.
Take a 1 tonne vehicle travelling at 25 m/s.
(1/2)m v*v
= 0.5 *1000 kg * 25m/s * 25m/s
= 625/2 kJ
= 312.5 kJ
There is little point in buffering more energy than the vehicle can hold unless the engine is underpowered and the hybrid system needs to buffer up plenty of power before it is used.
Posted by: crysta1c1ear | May 5, 2008 6:36:21 AM
Yawn was meant to be yaw. People do make typographic errors from time to time and I am no different. To ridicule is a sign of a deficient personality. To discuss rationally and constructively is a sign of a well ordered mature mind.
Posted by: sjc | May 5, 2008 7:26:42 AM
Here is a prior writeup on KERS, some might find interesting: http://www.greencarcongress.com/2007/10/flybrid-flywhee.html#more
Posted by: Lad | May 5, 2008 7:51:42 AM
If you want to store a significant amount of energy in a flywheel of moderate size and mass, it has to spin very fast indeed: 50,000+ RPM. The velocity at the perimeter of the wheel approaches that of sound at normal pressure, therefore such a device would suffer massive friction heating unless it operates in a high vacuum.
The flywheel body has to be made from composites to avoid high-speed shrapnel if it ever disintegrates (e.g. during an accident). The composite construction itself has to be performed in a high vacuum to prevent outgassing during operation, which would weaken the structure. Even so, safety considerations require that an uncontrolled release of the stored energy be contained. This is not at all trivial, since the flywheel is supposed to recuperate most of the kinetic energy stored in the vehicle at urban speeds, i.e. up to ~30mph. For reference, the kinetic energy at that speed is equivalent to that required to lift the vehicle ~30ft straight up.
Operation in a high vacuum adds another complication: the very light oils required for operation at very high speed will not remain liquid. Perhaps permanently sealed roller bearings could be developed for this application, though it would be a very difficult engineering challenge because there would be no easy way to wick away the friction heat. Most superflywheel designs therefore depend on zero contact magnetic bearings. Those are fairly week and quite sensitive to vibrations, though.
Finally, there are the gyroscopic forces already mentioned by other commenters to consider. Using two flywheels spinning in opposite directions neutralizes the effect on the vehicle, but their bearings would still experience very high loads - vehicles experience high turn rates about every axis in response to steering and uneven terrain. Rotation about a vertical axis is preferred, but even then the device really ought to be gyroscopically isolated with a gimbal bount.
For all of the reason discussed above, superflywheel applications in vehicles need to use electromagnetic rather than mechanical power transfer. For details, see e.g. the CCM flywheel used in e.g. the Fraunhofer Institute's AutoTram concept.
Posted by: Rafael Seidl | May 5, 2008 10:11:58 AM
Thanks RS. I have been a fan of flywheels in the past, but a few knowing people set me straight on some of the details and I have not forgotten them.
Posted by: SJC | May 5, 2008 10:25:15 AM
Un-named major car manufacturers are already in development deals using Flybrid's very similar system.
I can't wait to see how the Flybrid system gets on next year in F1.
Posted by: clett | May 6, 2008 2:30:33 AM
This reminds me of the mechanical watch golden days. It took decades for the electronic watch to move in and replace what had taken centuries to develop.
My sun powered electronic watch has been going steady and very accurately without any maintenance for 6+ years. My older excellent gold mechanical Swiss watch required a $40+ yearly maintenance job to keep it going accurately enough.
The same will soon happen to our vehicles. The current complex, inefficient mechanical ICE will be progressively replaced with maintenace free electrified vehicles.
Improved combined modular battery-supercaps + a small compact ANG Fuel Cell range extender may be one of many interim solutions.
Eventually (by 2020+), much improved very quick charge, long lasting, lower cost 50+ KWh ESSUs will supply all the power and range required to quietly drive around without creating harmful GHG.
Meanwhile, Prius and Civic Hybrids (+ Altima & Camry hybrids) and specially the new Prius coming out in January 2009, are very good reliable starts. Many more improved models will be available by 2010.
Posted by: Harvey D | May 6, 2008 7:29:25 AM
The point of this research is to develop a simplified system of energy recovery adaptable at low cost to most vehicles. It limits total energy storage to achieve the highest benefit at lowest complexity level.
A Prius uses high-quantity energy storage to do three tasks: dynamic braking, electric operation at very low loads, and power supplement to the low power Atkinson cycle engine. This system adds a lot of weight, cost and complexity to the basic vehicle.
Limiting energy storage allows the use of low-cost, recyclable, high mass/low speed flywheels that don't run into the problems mentioned by Lad above. Using parsimony as a design tool returns the highest benefit/cost ratio for a system that could be easily adapted across a wide range of vehicles without resorting to major redesign of the entire vehicle.
Posted by: fred schumacher | May 6, 2008 8:35:40 AM
If you want to store a significant amount of energy in a flywheel of moderate size and mass, it has to spin very fast indeed: 50,000+ RPM.
The whole point of this project is to avoid all the exotic technology and problems associated with superflywheels. This 20 kg steel flywheel rotates at only 9000 RPM and stores 400 kJ (111 Wh), enough to accelerate a Prius-sized vehicle from 0-55 mph. The rim is subsonic (280 m/s) so moderate vacuum and mechnical bearings should suffice.
The Prius battery weighs twice as much yet delivers much less surge power. The Prius pack is rated at 1300 Wh but most of this is kept off-limits to preserve battery life. Usable energy is 400-500 Wh and actual usable energy during a 0-60 accel run is less than 100 Wh. Prius regen efficiency is weak, flywheel should be very good. On paper the flywheel looks great. There are issues to resolve, but it's certainly worth pursuing.
Posted by: doggydogworld | May 6, 2008 9:11:14 AM
In the UK there will be the opportunity to ride in one of two flywheel Vehicles in revenue service that use a heavy steel wheel. See Parry People Movers. In terms of cost and performance and energy storage per unit volume, mass produced composite flywheel systems will always be cheaper than Ultra capacitors, but groups of Ultra capacitors can be constructed in small quantities, but there is virtually no energy in a whole array of Ultra capacitors. A single lithium cell can have more total energy, but far less instant power. Soon lithium batteries will make ultra capacitors obsolete for vehicles. EFFPOWER, ATRAVERDA and FIREFLY have lead-acid technology, also, that is superior for vehicles in price and performance compared to ultra capacitors.
There were some UK electric locomotives that used two one ton flywheels, so they could pass through the gaps in third rail electrification without pause. They could even start a full train from a full stop in the gap.
..hg...
Posted by: Henry Gibson | May 10, 2008 10:21:33 PM
One point conspicuously missing from this whole discussion is that it is inherently inefficient to convert mechanical energy to electrical energy in order to store it, then convert it back to use it. The people producing the Flybrid KERS system claim that the energy efficiency (i.e. percentage of braking energy available for re-use)is about 74%, compared to only 34% for an electric hybrid like the Prius. It does this for less than half the weight, half the space volume and one quarter the cost.
The Prius is presently an expensive joke, less fuel efficient than a well designed Diesel.
http://www.engine-expo.com/forum_2008/pdf/day_2/jonhilton.pdf
Posted by: BillyP | Jun 16, 2008 7:43:43 AM
Every gasoline and diesel car already has a flywheel. Even with the best materials, flywheels will not cost less for energy storage than batteries, but they have the advantage of high power for brief periods and that is what they have been used for now for hundreds of years. The strength of a material divided by its density will give you the maximum energy that any material can store in any flywheel configuration and any speed. Each material has a definite rim velocity limit no matter the diameter of the wheel. Wood happens to be better than cast iron. Big wheels must run at slower revolutions. The temptation to store the most energy in a flywheel does not extend to regular automobile engine flywheels so people do not worry about them bursting and very few, if any, have. Flywheels have a definite use in hybrid cars because smaller more efficient engines can be used to get the same acceleration, and if the system is so designed, braking energy can be recovered, but this is only a small part of the efficiency gain. The efficiency of energy storage and transfer of a flywheel is not as important as is the cost of the system. Even if used for 10 years, Priuses are not cost effective in India compared to the TATA $2500 car. Because of the cost of the motors and drive electronics and batteries, high performance electric cars will remain uneconomical. You can buy a lot of ethanol for the additional price of a Prius over that of a used car.
We would not be using IBM compatible computers if using the best technology was most important; IBMs power and money people decided to gain the home computer market and they did and not with the best technology and not even with their own technology.
If un-encased super flywheels were used in every car, cellphones would be a far bigger cause of accidents, and not fastening seatbelts a much bigger cause of death.. ..HG..
Posted by: Henry Gibson | Jul 10, 2008 2:23:05 PM
