3D Printing

New power inverter could make EVs more powerful and efficient

New power inverter could make EVs more powerful and efficient
A new 20 kW inverter packs a lot of punch in a smaller, lighter package (Photo: ORNL)
A new 20 kW inverter packs a lot of punch in a smaller, lighter package (Photo: ORNL)
View 2 Images
The inverter will be soon scaled up to half the size of a typical electric vehicle (Photo: ORNL)
1/2
The inverter will be soon scaled up to half the size of a typical electric vehicle (Photo: ORNL)
A new 20 kW inverter packs a lot of punch in a smaller, lighter package (Photo: ORNL)
2/2
A new 20 kW inverter packs a lot of punch in a smaller, lighter package (Photo: ORNL)

A new power inverter developed at the Oak Ridge National Laboratory (ORNL) marries advances in 3D printing and wide-band semiconductor technology to deliver significantly improved performance in a smaller, lighter package. With further development, it could go a long way toward helping build electric cars that are more powerful and energy-efficient.

Power inverters are an essential part of any electric vehicle, as they take the direct current stored in the battery pack and turn it into AC that feeds the motor. Making them as small and light as possible is an area well worth focusing on. Even Google's in on it, having recently instituted a US$1 million prize for the best inverter designed to take DC from solar arrays and wind turbines and convert it into AC for domestic use. Of course, reducing that footprint becomes even more important when room is at a premium, such as aboard an electric car.

The inverter designed at ORNL achieves a very significant improvement in terms of power density, weight and volume. As lead investigator Dr. Madhu Chinthavali tells us, the 20 kW device that his team designed has a total volume of only around 1,500 cc (91 cubic inches) and weighs around 1.75 kg (3.85 lb). For reference, this is over four times the already aggressive power density requirements for Google's prize.

"20 kW was the highest rating that the inverter was tested up to," Chinthavali tells Gizmag. "This is actually more than a 30 kW inverter by design; we are being conservative because of the voltage levels that the DC-link can boost up to for automotive applications."

The inverter will be soon scaled up to half the size of a typical electric vehicle (Photo: ORNL)
The inverter will be soon scaled up to half the size of a typical electric vehicle (Photo: ORNL)

This advance was achieved in great part because of the properties of silicon carbide, a so-called "wide-band semiconductor." These are high-grade materials that are very well-suited for high power applications and for working under a wider range of temperatures, which is especially relevant in EVs. As a result, the device is achieving a higher levels of efficiency, to the order of 99 percent, though this number will fluctuate depending on actual operating conditions.

The other major performance driver was the use of 3D printing to build about half of the inverter's parts, which allowed the scientists to reduce weight and be much more flexible in their design. Specifically, the researchers designed a higher-performance heat sink by placing lower-temperature components next to high-temperature ones, and allowing for better heat transfer throughout the device. Another design change was the use of an array of small, interconnected capacitors instead of a few big ones, to reduce cost and further improve cooling.

Chinthavali and colleagues are now working on scaling up the inverter so that it will be half the size of that found in commercially available vehicles (for comparison, Tesla's Model S inverter has a very beefy peak power of 320 kW) and containing an even greater number of 3D-printed parts in order to shave off even more weight.

Based on his results so far, and with a few more design tweaks, Chinthavali thinks it possible to achieve an inverter with four times the power density of their current prototype. "Mass production of the printed technology is not there yet, but things are evolving rapidly," he told us.

Source: ORNL

7 comments
7 comments
zevulon
HOLY SMOKE. this is a big deal for electrics. slowly but surely, the future of electrification is arriving.
the real question is which vehicle segments are the first to fully and totally displace traditional combustion only segments.
my guess is the first series hybrid motorcycle is going to completely eliminate gasoline ONLY motorcycles sooner than people think. probably 40 to 50 years from now, people won't remember when their motor was connected directly the motorcycle powertrain, as it will only be used to generate backup electricity for the battery bank.
think this is crazy? consider that batteries are going to charge 10 times faster than they do now at some point withint the next 20 to 30 years or less.
10 times. the tesla is crap. it takes a whole hour to 'supercharge' 85kwh battery back weighing 1500 pounds'.
ten times means that's 6 minutes to charge it. ten time menas you don't need 85kwh for a battery . you can use 21 kwh battery , which increases your overall vehicle range per kwh (by reducing weight and increasing efficiency) while reducing your charge time to 1.5 minutes + overhead of plugging vehicle in and paying.
you will get 1/3 of the range you have now, using less electricity at a cost of 1/50 the amount of time it currently takes to charge your vehicle.
shrinking the inverters and the power electronics, while increasing the cooling abiilities for the motor and electronics and battery-------all of that. will yield stunning performance from electric vehicles. it simply takes time. the pieces are indeed coming together , more slowly than many people like, but i see a juggernaut of technological patchwork relentlessly moving forward to a steep cliff like tipping point for combustion powered drivetrains ( i also see an excellent future for onboard multifuel range extending combustion generators to shrink and become ultra robust and efficient)
Mindbreaker
@ zevulon, fast charging is not the only thing that maters. Those 10x charging speed batteries are already here: http://www.gizmag.com/dual-carbon-fast-charging-battery/32121/ No, what is needed is range and most importantly cost. I predict that the range will grow to eclipse any car you can buy today, likely beyond 1,000 miles in less than 10 years and possibly as much as 5,000 miles. Recharging once a month is going to be very attractive...though it may only be a luxury feature. As for cost, maybe 1/4 current prices in 10 years (inflation adjusted). There is a limit, if lithium continues to be the go to material. If it is zinc or some other common cheaper material, then the price can cease to be a major component of the vehicles, and may undercut the cost of internal combustion vehicles even without considering operating cost.
SamB
"... designed to take DC from solar arrays and wind turbines and convert it into AC for domestic use."
Surely the wind turbine is not putting out DC.
Graeme Harrison
Yes, good work. And the scaling-up is needed for some applications. However, with more EVs going to individual electric motors in the wheel (or directly coupled to the wheel), it would seem that 30kW is actually sufficient for automotive use, as one could have a 30kW electric motor connected to each of four wheels.
But the biggest improvement we can make for EV acceptance (while battery technology/costs improve) is to allow EV exemptions on vehicle design rules. Let's face it, if Mr Benz, Daimler, Rolls, Royce et al had faced the myriad of design constraints now placed on vehicles when they were working out their commercialization plans 120 years ago, they'd have each given up before producing a horseless carriage.
We need to allow EVs with speed limiting function of (say) 50mph/70kph to be registered as 'commute vehicles' provided they have seat belts, lights and indicators/blinkers. They should not need to pass any crash tests, or have airbags etc. Currently only $10b+ corporations can face the $500m+ costs of getting a design into production. That means we are getting only the incumbents to work on the job of replacing themselves. But as Microsoft eclipsed IBM, and Google eclipsed Microsoft etc, it is always outside disruptors who add the most value. Encumbents move slowly, and usually get it wrong.
We need to allow small startups, using just mountain bike technology (think of two aluminum mountain bikes joined together with struts to form a minimalist commute vehicle, with slung-seats and electric wheel-hub motors.. and bicycle LED headlights.... for a total vehicle weight of around 100-200 pounds. That would be entirely suitable for sitting on slow commutes for many major metro areas. Such vehicles should not be allowed on 60mph/100kph+ highways... but they could be great city-only commute vehicles. With the current approach, we will see conventional auto-makers trim a few pounds of weight, and try to push large/heavy vehicles with small battery packs. But if we allow the innovative bicycling fraternity onto the road with light alloy commute vehicles, existing battery packs could give a range of hundreds of miles. In short, we need to reconsider whether the constraints on internal combustion vehicles is restricting innovation in EVs, and if we ought allow a newer lighter approach proceed, and then add additional safety tests, only if these vehicles seek permission to increase their top speed, or travel on unrestricted highways.
Martin Winlow
@zevulon - Please! 10 times Tesla's SuperCharger power is 1.2 MEGA WATTS. Where the Dickens are you going to find that sort of power supply? And what is it likely to cost? And how would you connect it to the car? And why would you need it anyway? What sane and sensible person would want to drive for hours on end with no break? Personally 3 hours is more than enough and Tesla's tech can do that (and more) right now, off the shelf - why the never ending hoop-la about yet more range requirement? Especially when the average daily US range is barely 40 miles! Let's get real, folks! MW
Don Duncan
G.H. has just highlighted the reason why Aptera is not available, i.e., govt. regulations. Customer freedom to choose is preempted by bureaucratic centralized control. This arrogant, one-size fits all, "we know best" attitude is forced on us with the lame excuse of "public safety". When regulation is shown to be counter productive to safety, or not cost effective, or technologically suffocating, the fall back justification is that govt. is not perfect, but will somehow "protect" us by protecting the established industries from competition, aka, the need to progress.
This view of govt. as necessary to stamp out greed, i.e., ambition/innovation is killing the American Dream and progress everywhere.
Jane Leonard
Inverters work with an incredible technology and they do make EVs more efficient for its functions. This technique is more welcoming to drive innovation run vehicles and motors to serve many purposes.