• Nov 17, 2009
Chevy Volt going through a shake test as part of its durability evaluation - Click above for high-res image gallery

General Motors offered an update this afternoon on the current status of the testing and development of the lithium ion batteries for the Chevy Volt. Chief Engineer Andrew Farah and battery engineering group manager Bill Wallace provided a run down on the work done so far. Since work started in earnest on the production program in early 2007, GM has gone from two initial battery development partners (LG Chem/CPI and A123 Systems/Continental) to selecting LG Chem to supply cells. GM also decided to design and build the battery pack in-house rather than having CPI provide the finished pack.

Through the course of development, the team has learned that the battery management system is the key to extracting the best life and performance out of the battery. However, while the battery is the heart of the Volt, Farah emphasized that the goal was to create a great a great vehicle for customers, not just a battery on wheels. That means a lot of integration work has been necessary to incorporate the pack in the car, and develop the charge and battery management software. Read on after the jump.



[Source: General Motors]

Since the beginning of this year, GM has opened its new battery pack manufacturing plant in Brownstown, Michigan, opened a huge new battery test lab and built 80 pre-production (IVER) Volts. In addition, 300 battery packs based on the production design have been built and are being tested in those 80 pre-production cars, the battery lab and other facilities. The test vehicles are now in their third generation starting with the original engineering development cars that were based on the previous-generation Malibu body, the Cruze-based mules (like the one we drove last spring) and the current IVER units.

The next batch of vehicles are the product and process validation vehicles (PPV) which will be the first to be built at the Detroit-Hamtramck assembly plant. The PPV units will built starting in March 2010 and these will be based on the production hardware specs including the proper light clusters (unlike the IVERs). Finally, between August and November of next year, the manufacturing validation build (MVB) will occur. At that stage, everything will be done with production equipment and tooling and these are the first so-called salable vehicles. Those cars, along with the PPVs, will be used in captured test fleets for a variety of real world evaluations including durability testing. Those IVERs recently completed a 65% calibration ride and, according to Farah, none of the issues seen so far are show-stoppers. Farah and Wallace claim everything is on schedule and moving forward to production one year from now. By the end of December, GM will have completed 300,000 miles on the Volt full vehicle simulator, or about three times the design life of most components.

The first builds of packs at the Brownstown plant will begin in February to supply the PPV builds that start a few weeks later. The first of the GM-designed packs was completed in December of last year. Since the beginning of the program, over 50,000 cells have been built and tested and Wallace claims that there has not been a single cell failure to date in either cell level, module level or pack level evaluations. In the lab, the packs have accumulated over 300,000 miles of simulated customer use testing which involves cycling the batteries through routines based on real world driving and charging use. Over time, the cell chemistry has been tweaked a bit to help meet the battery life requirements without affecting performance. According to Wallace, the pack and cells are now set and will not change before production.


The battery testing involves far more than just charging and discharging. Like other parts of the car, the cells, modules and packs have to go through the full "battery" of abuse and crash testing to ensure they live up to the expectations of what might happen in customer hands and beyond. That includes crush and penetration, thermal and overcharge tests on the cells as well as the packs. Characterization tests are also done on cells in order to determine the tolerance range of performance so that the battery management software can properly estimate the state of charge of the cells and packs.

The pack is also put through crush and impact tests to insure that its structural and seal integrity is maintained. This is especially important since the pack is liquid cooled to help optimize its life and performance. The corrosion testing is also very important because the pack is on the bottom of the car. In order to speed up the testing, salt is applied directly to the pack and then it's put into a heated chamber to see what happens.

The whole car is also put through the same testing procedures as all other conventional cars including rough road, hot and cold and of course crash tests. The central location of the pack under the car has benefits in terms of both weight distribution and safety. A photo of a prototype that was run into a barrier at a 30 degree angle at 40 mph shows no intrusion into the pack area.


Speaking of weight, Wallace confirmed that the final pack weighs about 200 kg (440 pounds) but no one would reveal the weight of the car. Farah did acknowledge that the car is a bit heavier than they would like but it is not affecting the performance targets at this point. The team is evaluating weight reductions that will could be rolled out over time during the lifecycle of the car.

Perhaps the biggest concern, aside from the battery performance and life, is the noise, vibration and harshness when the range extender is running. During the tests, the engineers have worked to minimize the NVH during the charge sustaining mode. That includes developing some new bushings such as switching rubber for hydraulics for mounting the engine/generator. That work is still ongoing and includes calibration work on the engine. Other noise sources that have popped up were aerodynamics, including the full belly pan. Changes have been made to venting for pressure relief. Farah again declined to get specific on the capacity of the fuel tank and said that is one area that is yet to be finalized.

Another issue that the Volt and other EVs will face is cold weather performance. The Volt will have an electric heater along with heated seats during cold weather driving. According to Farah, the heated seats actually provide excellent bang for the buck in terms of energy use. Drivers will be able to use the Volt's controls to pick the desired balance of comfort and electric range. The charging system, including the standard J1772 connector, will also interact with the climate control system to pre-warm the car while it is plugged in to minimize the effect on range once the cord is decoupled.

The Detroit-Hamtramck plant is currently being re-tooled for Volt production and should be ready to go by March for the PPV build. The Volt will be one of the focuses of GM's display at the LA Auto Show next month and we expect to learn more about it at that time. Perhaps we'll even get an announcement about the fate of the Cadillac Converj. Farah called it a great concept and he looks forward to getting to work on it, but declined to say if and when it would be built.



I'm reporting this comment as:

Reported comments and users are reviewed by Autoblog staff 24 hours a day, seven days a week to determine whether they violate Community Guideline. Accounts are penalized for Community Guidelines violations and serious or repeated violations can lead to account termination.


    • 1 Second Ago
  • 19 Comments
      • 5 Years Ago
      letstakeawalk

      how can the White Zombie go from lead acid to Lithium and immediatly get better times down the track which keept getting better. It's a matter of the BMS not the batteries (I'm pretty sure) correct if I'm wrong i'm trying to learn as much as possible.

      I still feel it'd be cheaper to convert your car but, all the testing they're doing reassures me atleast it will be safe(something backyard mechanics like me can't afford to heat test shake test cold test even crash to see how they'll hold up) so even if i never plan on getting an accident (can't predict i knowi've gone 7 yrs and nothing but a little bump that i couldn't see cuz my sight was blocked by a f-150 with a cap and the driver never worned me she was coming up on my ass even though she said she saw me backing out. since then i always back in now) so for all the testing they've done IF it's all true and not just PR then they've got one sold to me... i still hope aptera but just keeps getting pushed back...
        • 5 Years Ago
        I don't really know much about "White Zombie", but I can assure you that they really aren't concerned with getting the maximum lifespan out of their battery pack.

        • 5 Years Ago
        There are a few reasons White Zombie got better times with LiFePO4 over PBA

        1) Weight. Lead acid batteries weigh a LOT. White Zombie's Lead acid pack was 884 Lbs. The Lithium Iron Phosphate batteries were only 175 LBS.The less energy it takes to get the car moving, the faster the acceleration, the better the time.

        2) Voltage. The configuration of the Lithium pack resulted in a bump up from 360 volts to 374 volts. Not a lot, but higher voltage means less amps needed to get the same work out of the motors. Or put another way, you can get more work done with the same amps. Since the controller is amps limited, more power could be pushed to the wheels with the same amps.

        3) Voltage drop. The faster you pull energy out of a battery (the more amps you try to pull) the more the voltage will sag. when the voltage sags, you lose power, which has to be made up for with more amps, which causes more sag ... Lead Acid batteries sag a lot more than lithium batteries do. The ability to pull large amounts of energy out of the batteries more efficiently means more power to the wheels and faster times.

        An ultracapacitor setup would probably be ideal for a drag strip. not a lot of energy storage, but if you only care about going a 1/2 mile (down the strip and back) it should be feasible... not very practical for a daily driver though.
      • 5 Years Ago
      Holy Shit no body has picked up on this 300000miles now that should last a hell of along time now if only the body could last that. Now i'm really considering this i take great care of my cars so to buy a new one that i could drive till the day i die (probably not but holy crap) i'd only need a few litres for the weekend drive i Barely ever drive more then 50km at a time, so 40 miles that works for me.
      • 5 Years Ago
      exciting :)
      The one thing i like about GM is that they're willing to divulge a lot of information regarding the development of their cars. As a car and electronics nerd, i like this.

      Either it's a load of crock, or they're beating/testing the crap out of this car in efforts to bring the best GM product in decades to us. Only time will tell. I hope this car saves them.
        • 5 Years Ago
        GM doesn't have a plug-in car or a super fuel-efficient car right now to sell so it doesn't hurt sales to divulge all this work on a future car.

        I wonder how the testing at Coda, Fisker and Tesla compares with 80 integration vehicles running for months, a year of pre-production vehicle testing while the factory comes up to speed, testing 300 battery packs, etc. I guess specialty vehicle manufacturers find out in warranty claims what volume manufactures learn before they sell one car.
      • 5 Years Ago
      Car and truck OEMs will test vehicle and components a lot longer and harder than most people know or would imagine. At Arvin Meritor we've had our hybrid systems on Walmart's fleet of Class 8 trucks, and our electric traction motors on Purolator delivery trucks running field tests for many months as well.
      By the time these become broadly known to the public, typically millions of miles and many harsh tests have transpired.
      • 5 Years Ago
      Hydraulic motor mounts can't really be described accurately as "using hydraulics". They're actually more liquid-filled motor mounts than anything.
      • 5 Years Ago
      There is a big difference between testing a vehicle on some hydrolics to simulate driving on rough terrain, and actually doing it. Its nice they are testing these systems for 300K miles, but what concerns me is they say the component's design life is 100K miles. Sorry but with the battery being such a large cost of the car, knowing that your car will need a new $20K battery when you reach 100K miles doesn't make me smile. Mean while my honda with over 130K miles still runs strong, and it only cost me $10K. Unless Obama is going to manipulate the oil markets, or tax gas, The Volt will never get off the ground.
      • 5 Years Ago
      To my surprise GM are more concerned that hot weather will degrade the lifetime of the battery than cold:
      http://news.cnet.com/8301-11128_3-10399861-54.html
      If their projected 10 year lifetime of the battery in any climate proves right, this should effectively mean that in temperate climates such as the UK no replacement batteries will be needed for the lifetime of the car, if they last the same around 14 years as current vehicles.
        • 5 Years Ago
        I don't have a source but i have read in multiple articles that lithium ion batteries actually have their best performance at around 0 degrees Celsius. Yeah, at freezing. When it gets below that their performance drops dramatically though. But from what i understand, they don't actually like room temperature all that much.
        • 5 Years Ago
        Batteries use cooling systems to run at the desired temperature.
        That costs a lot of energy in very hot weather - think Phoenix, Arizona, where GM have said that range in the Volt in electric mode could be reduced, or very cold weather, where the battery is less efficient and you also need to provide power for heating.
        They are going to be totally happy and get maximum range in typical London temperatures.
      • 5 Years Ago
      i'm gonna go out on limb here and ask if u have any proof if not don't claim it... life span for thundersky(i know this is not in the volt) but 3000 to 5000 cycles to 80% then 70% is a hell of alot of charges... so maybe ya we won't know what the lg chem batteries will be like in 10 yr but a charge every day for 10 yrs is 3650 charges so you'd still be able to drive on electricity maybe only 28milesish but that will still take me around town without any fuel.. i still think tthis is the best option for now... i'd still prefer a pure BEV then i could rent a trailer generator for weekend drives i wouldn't mind.
      • 5 Years Ago
      Lithium ion batteries degrade with time so that 300,000 miles spread out over 1 year is different than if spread out over 10 years. Now that Chevron no longer controls the NiMH battery patent I'd like to see a NiMH option.
        • 5 Years Ago
        Different battery chemistries require completely different control systems to keep them thermally stable and to ensure battery lifespan. It is not possible to simply swap one type of battery for another in an EV.
        • 5 Years Ago
        Well they should have all those NiMH plans in filing cabinets somewhere from the EV-1, I'm sure they could pull them out and dust them off.
        • 5 Years Ago
        Nickel-metal hydride batteries only last 500-1000 charges. They also loose a lot of charge over time. I think they should have used Lithium-ion polymer batteries .
        • 5 Years Ago
        Actually, you don't have to believe me. Here's a quote from the latest issue of Automotive Engineering International:

        "The transition to Li-ion batteries will not depend solely on production capacity. There are a number of technical issues that must be addressed to make the transition from NiMH. The NiMH battery packs cannot be managed by the control techniques in use today.

        "We would like to establish the same basic control structure for both NiMH and Li-ion batteries," said Toshifumi Takaoka, General Manager of Toyota's Hybrid Vehicle Engineering Management Division.

        "However, the internal resistance and other characteristics of these batteries differ widely, and we think that these areas will have to be controlled using seperate modules."

        "Another challenge for engineers is to optimize systems by matching control techniques to battery compositions. Unlike NiMH, Li-ion is not just one battery type. Various compositions provide different traits, with different mixes in regard to power, cost, and cold-temperature operation, for example."
        -AEI, "Move to Li-ion Requires Many Technical Changes", Nov. 2009, pp 26-28.

        My original point:

        It isn't simple to just swap between battery types as the entire control and cooling systems have to be specifically tailored for each type.

        • 5 Years Ago
        @letstakeawalk

        Actually it really kind of is. Its just a matter of the tweaking the programming and making sure that they have the same energy output from the new pack as their old pack. If they did that, then they could put any type of cell that they wanted into the car. After all, this is just storage of electricity chemically. The car shouldn't really care so long as it comes out of the pack in the same fashion.
    • Load More Comments