World Economic Forum: Global Risks Are Outpacing Ability to Mitigate Them
2007 NAIAS Recap

The Volt May be First, But E-Flex is the Key

Although the first E-Flex concept—the Volt—is using a combustion engine genset, GM is also at work on a fuel-cell variant that will use the upcoming 5th generation stack. Click to enlarge.

While it was the introduction of the Chevrolet Volt—a plug-in series hybrid electric drive vehicle (earlier post)—that generated the most excitement at the North American International Auto Show, it is the accompanying announcement of the E-Flex system that is the key, according to Nick Zielinski, chief engineer for the Volt.

The Volt represents the first application of the E-Flex System, a developing vehicle architecture that will encompass a range of compact to intermediate vehicles with all-electric drive systems (the “E”) powered by electricity from a variety of sources (the “Flex”).

Broadly defined, the E-Flex architecture consists of an electric drive motor, on-board storage for electricity (battery or fuel cell), on-board mechanisms for producing electricity, grid charging (plug-in) capability, and the associated power electronics and control systems.

E-Flex vehicles can include the genset-powered plug-in series hybrid (such as the announced Volt), a fuel-cell hybrid, or a pure battery electric vehicle. GM envisions a range of genset options for the E-Flex vehicles, including engines optimized to run on E85 or E100 and biodiesel.

There is much overlap between E-Flex needs and work being done in other parts of GM—specifically fuel cell vehicle development and the mechanical hybrid systems. (In its evolving taxonomy of offerings, GM refers to its existing portfolio of hybrids as “mechanical hybrids”—i.e., the engine provides mechanical drive power in addition to the electric drive power.)

The drive motor in the Volt, for example, is the same system being used in the Equinox Fuel Cell Vehicle. The upcoming 5th generation fuel cell stack that will be applied in a GM fuel cell vehicle will also find its way onto an E-Flex platform.

And although it is not yet determined, it is possible that the battery pack work being done for the development of the Saturn VUE Green Line plug-in two-mode hybrid (earlier post) will also apply to E-Flex vehicles.

Of all the elements, the electric drive—the motor and the controller system—is the farthest along. The technology in the motor is already on the road in the Equinox fuel cell program. We’ve been working the details of those systems—the controls, the inverter—for the last three or four years. We’ve made major advances in motor efficiency, and also in the size of the controls and the inverter, which are substantially smaller than a few years ago. And we have plans in place [in the fuel cell program] for much more compact inverter packages.

The generator itself is technology very similar to the [drive] motor. We feel we can share a lot of the technology between the two—the electronics controls are very similar.

Least mature is the large energy store battery. The new work is more in the battery pack. The challenge is the systems integration of all those cells. One of the key elements of integrating the batter pack system is cooling, and understanding temperature deltas across the pack as your charge it.

—Nick Zielinski

In working through battery pack management and control, GM combines simulation-based analysis, hardware cycling tests and then into vehicles for road testing.

We depend very heavily on the computer simulation work, and also depend heavily on component-level testing.

—Nick Zielinski

Presumably, the work being done on integration and control—and the development of optimal operating strategies—for the battery packs in the VUE plug-in hybrid will support the more rapid deployment of E-Flex vehicles (and vice versa).

There are organizational intersections where work is done, that can merge together where it makes sense on the E-Flex. We need to come up with the most efficient and highest level of component sharing with E-Flex—and we are setting up the organization to make that happen.

—Nick Zielinski



The one architecture that I don't see in the options above is a plug in fuel cell vehicle. Unless batter costs come down tremendously I probably wouldn't be buying a car with more than about 40 miles off of batteries. My problem with ICE gensets is the problem of gas going sour in the tank for lack of use and mechanical parts seizing for lack of use. While I imagine that any left over H2 in the tank would evaporate at least it wouldn't go sour. Does anyone know if fuel cells spoil with lack of use? In a plug-in fuel cell car the stack wouldn't have to be all that large.


I had the idea of the battery dominant series hybrid June 2003. My idea was to use a turbine for the genset, but basically it was similar. People have been thinking about series hybrids for more than 100 years, nothing new, except the battery advances and high power semiconductors.


A plugin fuel cell car is probably one of the last options they would go for due to cost. Big battery pack is expensive, fuel cell & hydrogen tank are expensive. Put both together and the cost would be far beyond what the average chevy customer could afford.


Patrick: I was thinking in terms of a medium sized battery pack (20-40 mile battery) with a medium sized stack for range extension. A smaller H2 tank would be cheaper than a big one. (you might have to tell the car how far you are planning to go so it can start the fuel cell early enough).

My personal favourite architecture would be a cheaper limited range (40 mile) BEV with the option to rent a range extender (in trunk or tailer) for that once a year long trip.

Lou Grinzo

Excellent article as always, Mike.

As for the overall E-Flex architecture, I think this is a very smart approach, as it gives GM the maximum, well, flexibility in the face of multiple uncertainties over the next five to ten years.

The really big source of uncertainty is, as others have noted many times, the future cost of battery packs. My hunch is that this system component will come down a lot in price, if only because the stakes are so enormous and we've done so little (relatively speaking) on this front to date.

I'm still not convinced that hydrogen fuel cells will ever be a major player in passenger cars (unless we make one heck of a biohydrogen breakthrough), but GM's approach leaves them with other paths to pursue. Even the most basic version--flex-fuel ICE, 40 mile battery--would let most people get an insanely high number of miles per gallon of non-renewable oil, and would be light years better than what we're doing now.


If GM would allow the user to specify the battery pack size/cost as well as the ICE size/cost, I'd be all over this. For those of us who don't especially want the "performance" aspect, have a model with 60-80 mile all electric range and a 25kw genset for the same cost.


Great article, I really enjoyed reading the comments by Nick Zielinski.

As an owner of an 06 Prius, I can tell you the best thing aboutt the car is when it is running in full electric mode. Like most Prius owners, I just hate to hear that gasoline motor start up.

A 40 mile range would be enough for me to run in full electric mode 90 % of the time. My office is only 8 miles away, so it would have enough for work and back and some errands on the way home, perfect.

This car will motivate Toyota to improve the hybrid line even faster than they are already are. That is a good thing for all of us who breath air. :)

The thing that I like the most about this E-Flex is the flexiblity, to plug-in or use biodiesel or use ethanol. The biggest problem with todays transportation system is that 98% percent Oil based.

A plug-in hybrid ( 40 mile range ) and a some solar panels on the roof of the garage would be the ultimate solution to multiple problems.

Mark A

I have had a similar concept in my head for a while now. Have an affordable vehicle, with a powerplant capable of totally being replaced as a unit. My idea was a cobra inspired commuter vehicle. When a newer powerplant becomes viable, swap in the new powerplant. Turn in old powerplant for a newer one. Could start as a ICE with a genset, evolve to a rechargeable battery pack, or evolve to a fuel cell stack, all in a convenient package.

Perhaps the more mechanical of us owners could have each powerplant in his garage, and easily swap powerplants as the upcoming immediate vehicle use changes. Fully battery only for weekly commutes, and the ICE/genset for longer trips or weekend use. Easily swapable, like changing batteries in a TV remote. I know its probably pie in the sky stuff.



One thing to consider: Many people on a long trip want to stuff their trunk with luggage, presents, food or whatever they bring with them on their long trips. I would find it terrifying to have large numbers of people on the roads towing trailers. Watch how they drive without trailers, watch how they park and backup without trailers...

I would bet that the cost of the fuel cell hybrid depicted above is equal or greater than the full EV version. Replacing a $500-$1000 ICE genset with a moderately scaled down fuel cell and H2 tank would not be possible without tens of thousands of dollars (probably close to the same price difference in the BEV battery pack and the fuel cell HEV battery pack).

Rafael Seidl

Neil -

I don't know if fuel cells spoil with lack of use, but having one and not using it would be a very expensive hobby indeed. Moreover, the auto industry is receiving plenty of R&D money from the powers that be to develop fuel cells that use lots of hydrogen. A hybridized fuel cell would make more sense in engineering terms but advocating one in the US could cause you to miss out on the R&D funding gravy train.

In Japan, Honda's FCX was designed as an ultracap hybrid from the outset.


you might want to look up Rosen Motors on Google. They advocated a series hybrid based on a single-stage microturbine-genset plus a gimball-mounted super-flywheel plus four in-wheel electric motors and power electronics to make it all play nice together. This was back in the mid-90s. The whole thing went nowhere, partly because potential investors got wind of crash safety concerns regarding the flywheel but mostly because the company insisted on retaining exclusive production rights for certain key components. The microturbine bit of the business still exists under the brand name Capstone.

Gas turbines operate at much lower pressures but much higher average temperatures than reciprocating engines do. Small turbines cannot be cooled internally, so their max. operating temp. is limited by the available materials to 1000-1100 degC. The net result is that thermodynamic efficiency is relatively poor, especially in part load. This is essentially why earlier attempts at turbine-driven cars by Chrysler in the US and Rover in the UK failed.


Concern about fuel souring or engines failing through lack of use seem easy to overcome. An on-board computer can automatically run the engine for a few minutes every few weeks (while the car is on the road, naturally, and not garaged), if the engine was not engaged by necessity during that time. That energy can be routed to the battery so it is not wasted.

Coupled with a reasonably small gas tank, this can help run down the fuel so it doesn't sit around for too long. Alternatively -- or in addition -- fuel testing, draining and recycling can be part of regularly scheduled service. After all, responsible motorists are presently conditioned to bring their cars in every three months for oil changes. (I know I do.)


You could always run it "dry" and use all the gasoline and never fill it up until you plan to go on a long trip. I wouldn't do that during the winter (just in case you get stuck somewhere). You would also need a fuel pump not designed to use the fuel as a coolant (like most in tank fuel pumps are).

NBK-Boston: Most modern vehicles don't need their oil changed every 3 months unless you do some severe driving (towing, dusty conditions, etc.)


Great article as always...

Personaly I do not like "hibrid" solution because more parts always mean more maintenance. For know I see BEV with fast recharge option as the most promising solution.


Neil, good call. GM is very smart to satisfy its customers with such flexible options. Hopefully we'll have all 3 options--when this thing--if this thing hits the market. It's also sort of built like your computer. One component goes bad, you switch it out for a new one. Or say you need an upgrade, based on your living situation that has changed--just buy one.

Rick Leeland

The capacity and safety of newest Li-Ion battery has exceeded the spec of a typical E-Flex like Volt.

The NanoSafe(TM) Battery (which will be used in the Phoenix Motorcars SUT released next month) has the following characteristics:

o Replace anode electrode material. Use nano-Titanate to replace graphite eliminates most safety hazards.
o Power density = 4,000 W/Kg (typical: 1,000 W/Kg)
o 85% charge retention @ 20,000 cycles (typical: 1,000 cycles, full DOD)
o Operates from -50C to +75C (typical: 0C - 40C)
o Safe - will not explode and no thermal runaway (typical: may explode)


In addition, the Nano-Li titanate spinel material provides:
o Low reactivity with electrolyte - no resistive layer forms.
o Zero strain, 3D crystal lattice - stability and better kinetics.
o High surface area - fast charge kinetics
o No SEI layer - excellent thermal stability

More importantly , it is safer than previous Li-Ion. In its safety test:
o Short Circuit Test Pass, no smoke or flame
o Forced Discharge Test Pass, no smoke or flame
o Over Charge Test Pass, no smoke or flame
o Over Discharge Test Pass, no smoke or flame
o Nail Puncture Test Pass, no smoke or flame
o Crush Test Pass, no smoke or flame
o Over Temperature Test Pass, no smoke or flame
o Drop Test Pass, no smoke or flame
o Inherently Safe Battery Performance

Of course, all these claims are to be verified next month when Phoenix Motorcars releases its SUT.

GM will most likely use the lower cost battery from A123Systems through BAK in China (See ), and it does have some characteristics of the NanoSafe(TM), but it will be a mistake not to keep an eye on the NanoSafe(TM) from Altair Nano.

From this point on, it's mostly about production management and cost control, and find a balance between cost (advantage to A123Systems) and features (advantage to Altair Nano).


While this is a series hybrid, what is to stop them using the ICE to power the e-motor in parallel with the battery.
Then you just need a larger motor which should be no big deal.
The lack of a plug in option is a shame, but presumably one that would be easy to rectify.
I would be in favour of even smaller (15 - 30 mile) range plug in's if they had an ICE generator.
But in general, it sounds like a good solution or set of solutions - as long as they build it.


mahonj, to my understanding what you've described is exactly what it is. It has a plug in option, and the small 1L supercharged engine is just for a generator. I read a lot about this vehicle on the popular mechanics website. here are links to a couple of articles they have on the volt.


Oh sorry, i misunderstood the first part, however i agree what they did with just having the motor there to charge the battery, and nothing else. However i still believe that it is a plug-in.

john galt

Have any economist/engineers read GCC tinkered with a marginal utility function of high capacity battery packs for use in vehicles such as the GM Volt? For example, what is an expected OEM cost to produce at various outputs, e.g., 1 million batteries (battery pack), 5 million batteries, 10 million batteries, etc? Annual US auto and light truck sales were 17 million in 2005 (
Thus, thinking of world wide demand increases chances of real scale economies. Perhaps current makers of small electronics battery sets (e.g., laptop, mobile phone) can be induced to contribute expertise. To reiterate the well known, most GCC contributors believe battery pack cost is the most significant barrier to wide spread adoption of PHEV, EV, and other battery assisted vehicles.


re: sour gas and seized up genset:

Thanks for your replies all! If they do make the Volt then I would hope that either the genset is optional (I'll rent for a long trip) or the genset is accessible enough that I can decommission the engine and drain the tank between uses. (I did like the idea of halving the engine kick in every so often to keep it from clogging up)


I was not thinking of a gas turbine genset but rather a Tesla vapor turbine. Higher efficiecy than a gas turbine.

Roger Pham

It is in economical to develop a common basic electrical drive platform that can serve different models or fuel types. However, the E-Flex platform does not have to exclude the use of the more energy efficient and lower-cost Serial-Parallel PHEV (SPP)architecture. This SPP architecture will have a 8-10kwh battery and a 80kw motor instead of GM's planned 120kw motor and 16 kwh battery. The 53kw engine in the removable genset will supply on average about 30kw of direct torque power during acceleration to the 80kw motor to make up for the smaller size of the motor. At high speed, the SPP layout will have 138 kw total power at the wheel for autobahn passing power or police evasion maneuvering.

I envision an 80kw motor on the right side of the engine bay connected to the differential in typical fashion. Then, on the left side and adjacent to the 80kw motor is the "range-extending device" that can be either an ICE-genset, FC-genset, or more battery-power-inverter drop-in package.

For use with an ICE-genset, just drop the genset into a rail or groove on the left side of the car's engine bay, with the 80kw motor sits opposite to the genset, and the axis of the motor to line up with the rotational axis of the engine of the genset. A hydraulic or mechanical mechanism is used to move a clutch plate (splined to the engine shaft) a short distance from the engine to the clutch plate of the motor for reversible torque coupling of the engine to the motor. At rest, the engine is not coupled to the motor,so the entire genset can be removed. No gear whatsoever is needed to connect the engine to the motor, just direct drive via clutch plates.

If a removable fuelcell genset is used, it would be of 30kw max power rating and has a 30kw motor built in to the genset to supplement the power of the 80kw motor, for a total of 110kw maximum power. When this genset is dropped into the grove in the engine bay and secured, then the torque of the 30kw motor will be locked via a manual mechanism to the 80kw motor of the engine to give a total of 110 kw power output.
A turbine engine genset is not recommended due to the poor efficiency of the microturbine.

If a pure BEV layout is used, then the drop-in range-extender on the left consists merely of more battery, plus an optional 30kw motor for sporty performance, and more power inverter to control additional motor power. Same coupling mechanism can be used to couple the 30kw motor to the 80kw motor as when the FC genset is used. Perhaps another 8kwh battery will fit within the drop-in space on the left of the engine bay, for total of 16-18 kwh capacity and a range of 40-50 miles, for those desiring to use this vehicle strictly for fixed-distance predictable daily commute. If charged twice daily, a commute range of 80 miles is possible with enough reserve power to deal with cold weather when battery capacity will drop. Then, before a long-distance trip, the vehicle can stop by a dealer and have the battery range-extender package swapped for a ICE genset. This ICE genset installation also comes with a 5-gallon gasoline or diesel tank located on the right side of the car's engine bay, above the 80kw electric drive motor. 5 gallons at 60-70 mpg is good for 300-350 miles before the next fill-up. This tank may be made of flexible but rupture-proof material similar to those used in race cars, or strong carbon fiber re-enforcement that will not rupture in a survivable frontal collison.

Beautiful vision of the future, ain't it? Can't wait until this thing will come out!

Norm Conwill

I believe you can lower the cost of this vehicle and extend battery life, by using NiCd batteries.
I would guess your choice of LiIon is, at least partially due to the NiCd's notorious memory effect.
If you could switch from ordinary DC charging current to a pulse charging process, the memory effect would be non-existent, and the battery bank would last 3 to 5 times longer.
A pulse charging process such as that developed by Electronic Power Technology in the mid-90's proved both benefits are valid, and available today.


Hi All,

Eflex is obvious and elegant. I am glad it finally made it through the corporate engineering committees! Its not ideal for traditional vehicle engine technology, which is oversized to deal with the poor low rpm performance of these engines. A series-parallel system like the Prius is better for that. But that assumes we will always be using engines that owe their heritage to being shaft coupled to the car wheels. With an Eflex, other engine options are available.

I think one reason it has not come out sooner is that NiMH Batteries patents are reportedly now controlled by a company unfriendly to fuel savings. Lithium batteries have made it out of the lab, but not through the vehicle manufacturer testing. Also, the Plug-in imperative makes the inefficiency of the energy conversions in a series Hybrid less of an impact, since it runs 80 % of the time on Plug-in energy.

Think about an EV1, with a stretched body, with a prime-mover bay. At sale time you take your pick of prime-movers to plug into the bay . Tubogenerator (4 inch ID) that runs continously (and on ANYTHING THAT IS FLUID AND BURNS), small turbo diesel, fuel cell, Miller cycle gas engine for the really cold climates or urban polution reduction zones (those will come, especially in China!), large charging power supply, or even a lithium air energy module (non rechargable but massive energy density).

The adaptability of this concept is its great strength. With a turbo generator, one might even pull the thing, and save the momentum during the summer driving season, and plug it back in when the temps get down below freezing, or for that long range trip. Or another idea, since these are expensive, would be just rent one when you need it, and run on battery the rest of the time.


Mr. Galt,

You use "believe" in a rather odd sense.  Summing up the costs of various parts of a good EV, the battery is by far the one with the biggest price tag and shortest life.  EV advocates have been complaining about the inadequacy of batteries since the end of the nineteenth century.

We don't need belief, we know.

The comments to this entry are closed.