Project Seeks to Optimize Composition of Magnets for Traction Motors to Improve Economic Competitiveness
08 July 2009
A research project currently underway at St. Pölten University of Applied Sciences (Austria), in cooperation with the University of Sheffield (UK), is exploring the ideal composition and structure for high-performance permanent magnets intended for use in hybrid and electric car motors—specifically, how the proportion of dysprosium can be reduced without compromising the thermal stability of the magnets. By optimizing magnets, the researchers suggest, hybrid and electric cars can be made economically competitive.
Overall, an electric or hybrid drive contains around 2 kg (4.4 lbs) of magnetic material in their motors. At present, neodymium iron boron magnets form the basis of this. These have considerably less mass than conventional magnets, but deliver the same level of performance. In order to ensure the magnetic properties are retained even at high temperatures—such as those that occur within a car—the rare earth element neodymium is partially replaced by dysprosium, another rare earth element. This increases the coercive force of the magnet—its stability against demagnetization. However, notes Prof. Thomas Schrefl, project leader:
Compared to neodymium, the proportion of dysprosium in the ore is less than 10 percent. However, the high-performance permanent magnets currently used for hybrid and electric cars contain up to 30 percent dysprosium. In the long term, this will prove problematic when it comes to raw materials, particularly if you consider that, in just a few years, all new cars will be fitted with a hybrid or electric drive.
—Prof. Schrefl
The research project is applying computer simulation to examine how the chemical composition and structure of a magnet influences its performance. This information will then be used to identify ways to optimize the magnetic material so that it requires fewer expensive raw materials, yet continues to deliver the best possible performance.
Applied in conjunction with the finite element method, simulation is helping to reveal the internal workings of a magnet. The computer is used to break down complex structures into individual elements so that they can be evaluated.
We reconstruct the magnet on the computer and break the granular structure of the magnet down into finite elements. By breaking down the microstructure into millions of tetrahedrons and prisms, we can recreate the spatial distribution of the metallic phases within the magnet in a computer model. We can then use the computer to simulate the effect that changes in the proportion of dysprosium have on the coercive force of the magnet.
—Prof. Schrefl
The finite element method has already been applied in the automotive industry for carrying out computer-based crash tests and wind tunnel tests.
I am thinking that the motors will get more powerful over time and AC 3 phase will be the method chosen. AC Propulsion, Tesla and Rasor have shown that they work, are lower cost, even lighter, smaller and more powerful than PM motors.
Posted by: SJC | 08 July 2009 at 04:33 PM
Yes SJC. Some of the Rasor e-motors have reached 98%+ efficiency. Unless the price and weight can be reduced, it will be difficult to do much better.
Posted by: HarveyD | 08 July 2009 at 06:12 PM
Ac Induction has its place, but claims of 98% are wrong for two reasons if even true.
First what matters in an actual vehicle are the losses over the entire driving cycle. This means some low speed high torque points, and some high speed low torque points along with an occasional peak efficiency point.
Secondly, the 98% efficiency claims are for the motor only, but what matters is the electric drive efficiency. Total conversion from DC in a battery or DC link to tractive effort on the road include inverter, AC cabling, Motor and gear reduction losses.
Bottom line, a 98% peak number is almost meaningless, and a comparison of AC induction to PM type motor electric drives is done as a total electric drive system over a relevant driving schedule. A great PM motor does about 2% better than a great AC induction motor when considered this way.
Hey, 2% is not insurmountable, and AC induction is the sensible back-up motor technology. So if the world runs out of Dysprosium or Chinese decide to hoard it all for themselves (it all comes from China) electrification will proceed with AC induction.
By the way, Raser is all hype when it comes to motors. They can design an induction motor, but nothing special about it. Better designs can be had from all the regular auto parts makers (Conti, Bosch, Remy, the usual Japanese guys,etc). Raser is about 90% hype.
Posted by: frankbank | 09 July 2009 at 07:56 AM
A couple of things jump out at me on this article.
(1) It is certainly valuable work to find less expensive magnets.
(2) Good ac induction design is very competitive with good pm motor design, but in general, pm motors allow for reduced inverter current ratings, when pm motor fault modes are handled well. This translates to a cost savings that can offset the magnet cost.
(3) FEA methods have been used for far more than crash and wind tunnel simulations in the auto industry, for decades.
(4) Raser is certainly high on hype, if not all hype - its motors are no better than other good IMs. Last time I checked even their website videos were 2nd rate.
(5) Frankbank's comments on cycle-averaged efficiency are dead on. A peak number may be good for advertising, but not much else.
Regards,
Posted by: Roy Davis | 09 July 2009 at 10:19 AM
... and ...
(6) PM motors have a weight (power density) benefit over induction motors, especially when including the inverter, and are generally a few percentage points better in cycle-averaged efficiency. That said, they suffer from worse fault modes and are somewhat more expensive, mainly due to magnet costs and rotor construction.
Regards,
Posted by: Roy Davis | 09 July 2009 at 10:29 AM
Switched reluctance motors are very efficient and very light weight for the power. The advantages of hybrid or electric cars do not depend on the efficiency of the electric motor, but on the ability to to reduce or eliminate the oversized internal combustion engine. Diesel-Electric locomotives have now shown this to be true for nearly a century.
A very efficient traction motor drive system without a single semiconductor device was used in the flywheel electric locomotives used in the UK sixty years ago when motors and generators with brushes were very adequate. Brushes are still adequate for most uses and much energy could have been saved if all subway cars had been fitted with the flywheel drive system. This would allow the use of single cars in the District of Columbia and most other systems, and give the cars the ability to creep slowly into the nearest station in the case of power failure.
Start stop vehicles, like garbage collection trucks, could use a similar flywheel system with a directly connected diesel engine to a heavy flywheel with motor generators and electric wheel drive. No fuel would need to be injected whilst the flywheel was at speed and the vehicle stopped. The engine could be much lower rated because the flywheel would provide starting torque. Compression relief valves etcetera would allow longer term flywheel idling and starting the engine. Small shunting locomotives could use the same principle without a single power semiconductor or magnet to fail. ..HG..
Posted by: Henry Gibson | 11 July 2009 at 09:21 AM