OEMs are pumping ever bigger sums into delivering EVs; Volkswagen has increased its budget to $100bn, General Motors is investing $35bn, and Toyota is putting in the same amount. Daimler is spending $47bn and Hyundai is pouring $80bn into electrification programmes.
One of the current issues is the rate at which battery technology is improving to deliver the range consumers feel is adequate for their main mode of transport. And, importantly for OEMs, how much that technology costs.
But EV efficiency isn’t just about improvements to battery technology. Yes, it’s incredibly important, but electric powertrains can still be tweaked to eke out even more range regardless of what happens in the battery space.
Sprint Power is an automotive engineering firm that works to improve efficiency across propulsion systems. Its engineering director, James McGeachie, says: “If we’re looking at efficiency, really you’re tackling two things: mass and thermals, because ultimately the heavier it is the more energy it takes to cart around and the hotter it gets.”
The obvious place to make improvements is to look at how better integration could streamline EV powertrain design, and offer a solution to the issues of mass and thermal characteristics. And that’s something that McGeachie and his team have looked at.
“Sprint Power’s 48V power module unit has both 12V, 48V and the battery management system integrated into it, and, as a combined device, it’s smaller and weighs less. And with some of the technologies we’ve used, we also have thermal efficiency improvements too,” says McGeachie.
Semiconductor advances
Semiconductors are a big part of the improvements, and should help more EVs become ever more efficient.
Both silicon carbide and gallium nitride are emerging technologies, and silicon carbide is now becoming more widely commercially available.
“Because they operate at higher switching frequencies, higher voltages, they’re more efficient. With that you’re getting less thermal rejection and they require less cooling,” says McGeachie.
As new semiconductor technologies come online it could be tempting to simply suggest that increasing voltages is the key to unlocking greater EV efficiency. Yes, the more you increase the voltage the greater efficiency you can reach, which can then translate into smaller, lighter components, but there is a definite ceiling to what is feasible, and development teams can’t simply just keep cranking up the voltage.
“What you start to bump into are things like semiconductor and filter devices are typically rated at 600V or 1,200V and are readily commercially available at affordable prices. It’s a hard stop there almost. That’s not to say that new devices couldn’t be developed, but the commercial ones at the moment, those are the ones that make most financial sense.
“You’re starting to set yourself an upper limit. Typically, you’ll see an 800V application with 1,200V semiconductor devices; 800V is a nominal situation and when max voltages occur it makes sure things are robust,” says McGeachie.
Volkswagen is investing €2.3bn in technology to produce EVs, including the ID.Buzz (Credit: Volkswagen AG)
Increasing voltages also impacts on the motors, where you could need thicker insulation and more windings. But the motors are an area where huge, relative improvements could be made.
“There are different configurations in motors today, depending on what it is you want to achieve. They all have pretty good heat efficiency one way or another; somewhere on their efficiency curve will be 96%, 97%, but the king is keeping it in that most efficient point,” says McGeachie.
Software tools
While huge investments are being made in hardware improvements – specifically battery systems, but across the entire electric powertrain – it isn’t the only avenue to improving EV efficiency and extending range. Software is an incredibly important tool too, especially in delivering even greater control of the motors that turn the battery’s stored energy into motive power.
In a strange turn of fate, while we start to reduce our reliance on combustion engine technology, the systems designed to improve gasoline and diesel efficiency are helping to make EVs more capable.
Tula Technology made a name for itself working with Tier One supplier Delphi to bring its Dynamic Skip Fire (DSF) software to the market. It’s a technology that’s now used in over two million GM vehicles, helping them make 15% efficiency gains by dynamically shutting down cylinders to keep the combustion engine in its most efficient profile for as long as possible. And the control philosophy that underpins DSF is now being used to better control electric motors.
Tula has developed Dynamic Motor Drive (DMD) to keep electric motors operating in the sweet spot for as long as possible – exactly what McGeachie suggests needs to happen – actively pulsing the motor to produce the optimum amount of torque, reducing the number of peaks and troughs in the delivery.
The expected gains of introducing DMD to an all-electric powertrain are between 2% and 3.2%, according to Tula. WLTP drive-cycle projections are for the lower end of that scale. That may not sound a lot but when you consider that by 2030 the global EV fleet is estimated to consume 750 billion kilowatt hours of electricity, even a 3% efficiency improvement would save 22.5 billion kilowatt hours of electricity.
Research and development, including aerodynamic testing, is delivering improvements (Credit: Shutterstock)
Torque modulation
And for the end consumer an additional 3km of range for every 100km could be the difference between completing a journey or not.
“Although all motors are designed to be as efficient as possible, in substantial regions of operations the highest efficiency isn’t achieved. DMD achieves highest efficiency throughout the entire operating regime by dynamically shifting, via torque modulation, the operating points to the areas of most efficient operation,” says John Fuerst, senior vice-president of DMD and engineering at Tula.
The operating strategy eliminates the system’s core and inverter losses, and can reduce motor losses by as much as 25%.
“The system works at 20Hz to 35Hz, delivering 5Nm of torque or less, with 70Nm pulses,” says Fuerst.
“DMD can be used wherever motors are driven by a software-based inverter or drive. Efficiency gains, cost advantages and environmental benefits are greatest when DMD is applied to rare earth-free motors.”
And that is the system’s current limitation but also why software control may become even more important.
The automotive sector currently uses internal permanent magnet (IPM) motors. Tula’s DMD could be used in conjunction with the IPM motors but the gains would be much smaller – less than 1% – simply because IPM motors are already incredibly efficient at what they do.
But there could well be a shift away from IPM motors towards externally excited synchronous motors (EESM) as OEMs realise that using rare-earth materials isn’t sustainable. At present few firms use EESMs but a lot are developing the technology; Renault, Nissan, BMW and Mahle are just some of the names working on them, but, if and when that number increases, software control will be needed to increase efficiency to IPM levels, and beyond.
And that’s a given, according to Fuerst: “We see them replacing the permanent magnet machines, maybe to the point of a 50/50 split in 2030. It’s getting a lot of attention, significant cost reduction by getting rid of the rare-earth materials,” he says.
But, as McGeachie explains, there is only so much that software can do to increase efficiency if the motor technology itself isn’t fundamentally solid: “You can’t overcome the fundamentals of the motor technology link. It really is about selecting the right system,” he says.
So how much more efficient could EVs become if more emphasis were placed on finding gains, no matter how small?
Search for efficiency
It’s been tried with combustion, and the general consensus is that 50% is the limit. In a fuel cell that figure jumps up to a possible 80%, but, considering that even the worst EV is roughly 80%-85% efficient, the limit could be far higher.
McGeachie says: “It’d be difficult to see how an EV drivetrain as a whole would get beyond 95%. Some of that is because you’re still relying on some mechanical losses, because they’re always ultimately mechanical leaks. I suppose you could argue, if you had in-wheel motors that might remove some of those mechanical links, but it’s going to be pretty unlikely that’s going to be enough to reach 100%. My gut feel would be you wouldn’t go beyond 95%.”
As the EV market expands, improving vehicle efficiency will be vital. OEMs are spending huge sums to make sure their EV strategies meet demand, but improvement isn’t only about better batteries, it’s about making sure the whole powertrain is as efficient as possible, no matter how small the gains are.
Navigate a turbulent future by attending Aerospace & Defence (28 November – 2 December). Register for FREE today.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.