GM’s views on challenges for battery development for extended range electric vehicles
31 January 2011
|
|
| Examples of degradation effects causing Li-ion battery power or capacity fading. Source: Roland Matthé, GM. Click to enlarge. |
With the Volt extended range electric vehicle and the Leaf battery electric vehicle now on the market, joining an ever increasing array of hybrids, and with next-generation versions of all of these already in the works, automakers and battery manufacturers provided some insight at the recent Automotive Advanced Battery Conference (AABC) in Pasadena into their learnings over the requirements for and development of advanced lithium-ion battery packs targeted at the different automotive applications.
Roland Matthé, GM technical manager for the Voltec battery system, provided an overview of GM’s views on the requirements and challenges for batteries specifically for extended range electric vehicles—i.e., the Volt—but also more broadly for batteries and electrified vehicle applications in general. Different applications require different types of cells, he noted.
| Key metrics of electrified propulsion systems (GM) | ||||||
|---|---|---|---|---|---|---|
| Mild Hybrid (e.g., LaCrosse w/ e-Assist) | Full Hybrid (E.g., Tahoe two-mode) | Plug-in hybrid (announced) | EREV (e.g.,Volt) | Battery electric (in development) | Fuel cell hybrid electric (fleet test) | |
| Pure battery-electric range | na | up to 2 km at speed < 50 km/h & low acceleration | up to 20 km at speed < 100 km/h & low acceleration | 40 to 80 km, all speeds, full acceleration | > 100 km | up to 2 km at speed < 50 km/h & low acceleration |
| Total range | > 500 km | > 500 km | > 500 km | > 500 km | < 200 km | 400-500 km |
| Battery energy | < 1 kWh | 1 to 3 kWh | 5 to 10 kWh | > 10 kWh | > 20 kWh | 1 to 3 kWh |
| Battery power | < 20 kW | 20 kW to 40 kW | > 50 kW | > 100 kW | > 100 kW | 20 to 40 kW |
| Power to energy ratio | ~20 | ~20 | ~7 | ~7 | ~4 | ~20 |
| SoC window | < 20% | < 20% | < 70% | < 70% | < 90% | < 20% |
| Recharge time | na | na | 1 to 4h | 4 to 10h | 10 to 20h (Fast: 0.5h |
na |
| Refuel time | < 5m | < 5m | < 5m | < 5m | na | < 5m |
In an earlier talk describing GM’s battery life estimation process, Joe LoGrasso, an engineering manager also with the GM’s Global Battery Systems Engineering Group, like Matthé, noted that customer expectations are an important factor to consider in establishing specifications relating to battery life and battery safety. In short, he said:
- Customers expect that batteries will last the normal life of the vehicle, that expensive replacements will be minimized, and that such service will be delayed until at least 10 years of battery life have elapsed, assuming normal usage.
- Customers expect xEVs with advanced batteries and high voltage systems to provide a level of safety comparable to that present in today’s vehicles.
Achieving the first requires predictive life models and adaptive vehicle control, he noted. Achieving the second requires a comprehensive system approach to battery safety at system, pack and cell level.
The battery pack for an extended range electric vehicle such as the Volt—which runs in an all-battery powered charge-depleting mode with full speed and acceleration up to the point at which it switches to operate in charge sustaining mode faces a number of challenges based on this mixed duty cycle—i.e., part EV, part hybrid. The challenges include:
- High number of full operation charge/discharge cycles
- High discharge power during charge sustaining mode (at a low state of charge, SOC)
- High discharge power requirement for acceleration performance
- High charge power requirement for regenerative braking and charge sustaining mode (transients at both high and low SoC)
- Temperature conditions
The factors all interplay, complicating demands placed on both battery and driver. For example, the depth of pack discharge in daily use will vary, Matthé noted. With public charging or charging at work, two or more cycles per day are possible; he said that he (driving a Volt) sometimes charges 3 times day. Because of the differences in how you can use the car, he said, you have to accommodate for that in your battery life.
There are also a number of factors—high charge/discharge rate; high or low State of Charge (SoC); hot or cold temperatures—that affect degradation of power and capacity. As examples, high charge rate and cold temperatures can result in metallic lithium plating and electrolyte decomposition. Low discharge rates and low temperatures can result in a corrosion risk on the current collector. High charge rates at warm temperatures can result in electrolyte decomposition and impedance rise. Low discharge rate at warm temperatures can result in metal dissolution and the loss of active material, with an accompanying fade in capacity. Designers must strive to keep the battery functioning in the minimal cell degradation area—essentially balanced in the center between these different extremes.
You have a wide range of matrix of conditions you have to consider. Every cell is differently sensitive to that kind of behavior. So first of all, you have to understand how sensitive is your cell to that [particular] degradation mechanism. As long as you do not have fully developed physical models, ground up...have full understanding of what happens inside the cell, you have to characterize what you have in front of you.
Now we have a very deep relationship with our cell supplier. That is important to do such an endeavor. If you just take a cell you don’t know, the vendor will not tell you...might not even in a very new cell know exactly, you have to characterize it. To characterize it, you have to think about the power levels you might face in your applications, you have to think about distribution of discharge cycles you face, and you have to consider the cold and warm exposure. You develop a test matrix for your battery to get to know you battery. And in time to understand what your cell is all about.
In general what you learn is the deeper your discharge cycles the less energy you can put through over life. If you do only little cycling, total accumulated energy is 2.5 times [that possible with high levels of cycling]. The next thing is temperature. Temperature requires a sophisticated model. The effect of battery temperature on battery cycle should not be underestimated.
If you want to have consistent performance, when you discharge your battery too low, your vehicle gets slower. The problem with an extended range electric vehicle is that you still need to do passing, so you want predictable power. On the other hand, you want to maximize efficiency.
You do not do only that power profile and discharge cycling, you also have to think about that your battery might have degradation effects you haven’t dreamed about.
—Roland Matthé
He's arguing his own case for the choices GM has made, nor carrying out an impartial assessment.
The figures given for the fuel cell vehicle for instance simply assume a low power battery, when the obvious answer is to put in a more powerful one to supplement the high energy, low power fuel cell.
Similarly the remarks about battery lifetime etc are based on the choices GM has made.
The battery from SK Energy Hyundai has chosen is far more long life than GM indicates.
Posted by: Davemart | 31 January 2011 at 06:36 AM
People should be sold the merits of what we can do, not try to meet ridiculous expectations. Lots of people still expect someone to bring gasoline below $2/gallon again. It's not going to happen, and trying to meet it is literally worse than doing nothing.
Posted by: Engineer-Poet | 31 January 2011 at 08:48 AM
We're now paying $8 per US gallon here in the UK.
Do you think things would change in the US if they had to pay similar fuel prices?
Posted by: clett | 31 January 2011 at 12:05 PM
"Do you think things would change in the US if they had to pay similar fuel prices[$8/gal]?
YES, clett. EVs would replace SUVs and cargo-less pickups in a few seasons. Mass transportation would increase for remaining travel.
Posted by: kelly | 31 January 2011 at 12:29 PM
Good to hear GM discussing these issues. It's rather involved and people will have to acquire an understanding of the complications of battery life. People in the north already know that cold temps reduce cranking power at start up (ICE). But with EVs they will realize that the range of their vehicle is reduced.
People in Florida know that the heat cooks their battery and it only lasts 3 yrs. THis will be a problem with EVs.
Posted by: danm | 31 January 2011 at 12:50 PM
Ok first davemart.. fuel cells are not low power.. they all are around 90-120 kw power output and thats exactly the power output they wanted. Thats why they use a smaller low power battery its just there for exactly the same uses as in a full hybrid.
Ok ep... You just dont get it.. the customers demand 10 years of service.. if the pack only lasts 4 years they will demand free replacements at 4 and 8 years. If they dont get them they will factor in the exztra cost and time spent in the shop and.. ignore your stupid ass pile of crap car.
Posted by: wintermane2000 | 31 January 2011 at 01:09 PM
"Do you think things would change in the US if they had to pay similar fuel prices[$8/gal]?
YES, clett. EVs would replace SUVs and cargo-less pickups in a few seasons. Mass transportation would increase for remaining travel.
And we would all wouldnt have jobs, those that did would have to make 200k to be able to buy a hot dog
Posted by: ds | 31 January 2011 at 02:00 PM
@wintermane:
Fuel cells have improved in their power density, but still need the help of a battery under acceleration.
That is why Toyota have a 21kw battery in addition to their 90kw fuel cell stack, and operate it as a hybrid using the same system as their other hybrids:
http://www.tunemytoyota.com/forum/showthread.php?t=1996
It is also an error to assume that all batteries will have a short life. Both Toshiba's SCiB battery and SK Energy's battery for Hyundai have batteries which should comfortably outlast the car.
Here is the Hyundai:
http://www.popularmechanics.com/cars/reviews/hybrid-electric/2011-hyundai-sonata-hybrid-test-drive?click=pm_latest
300,000 miles is not too bad.
Posted by: Davemart | 31 January 2011 at 02:11 PM
"Do you think things would change in the US if they had to pay similar fuel prices[$8/gal]?
YES, clett. EVs would replace SUVs and cargo-less pickups in a few seasons. Mass transportation would increase for remaining travel.
I am not so sure about the EVs.
Most of Europe has gasoline in the $7-8 region, and there are practically no EVs around.
It's mostly small gasoline cars and diesels.
Plenty of small SUVs - nearly all diesel, many 2 wheel drive.
I looked up Ford Galaxy MPVs in a secondhand (Irish) website [2010] - 31 diesels, 0 petrol.
(Japan has gone Hybrid).
Posted by: mahonj | 31 January 2011 at 02:28 PM
Davemart the 20 kw battery is a hybrid battery for the excact same reasons again as any hybrid battery. Its very useful to get the milage up and it does reduce spike loads on the fuel cell just as it does on ice engines alowing you to downsize the fuel cell.
Posted by: wintermane2000 | 31 January 2011 at 03:25 PM
@wintermane
Fuel cells are good at constant load and have high energy density compared to batteries.
But 'reducing spike loads' as you put it is simply another way of saying that the power density is not high, and you would have to greatly overspecify the stack to cope.
It is not a big problem, as you can simply boost it with a battery, but it is the way fuel cells are.
Posted by: Davemart | 31 January 2011 at 03:31 PM
Fuel cells: http://www.evworld.com/article.cfm?storyid=730
Posted by: ai_vin | 31 January 2011 at 04:12 PM
ai vin:
You are using a 2004 article to evaluate present fuel cells? I know this is an article about the theory, but the practise also counts.
And in practice a big heavy vehicle like the Toyota SUV can go 431 miles on 6kg of hydrogen, roughly equivalent to 6 gallons of gasoline in energy terms and about 72 miles/kg.
Allowing for a 70% conversion efficiency from natural gas that comes to about 50 miles per gallon equivalent.
Sounds pretty good to me.
Posted by: Davemart | 31 January 2011 at 04:31 PM
Allowing for a 70% conversion efficiency from natural gas that comes to about 50 miles per gallon equivalent.
That sounds about right. From the second page of my link we have this:
"Practically we see this fact being born out in the case of some vehicles that use Diesel or high-compression alcohols: The Toyota ES3 diesel-electric hybrid concept vehicle achieves a fuel economy of 103 MPG; the Honda FCX hybrid squeezes out roughly 61 MPG. Clearly, the ES3 is a different car -- lighter in weight, more aerodynamic and uses a fuel of higher energy content per gallon-but even when we make allowances for these facts, it becomes very clear that fuel cells may not be all that much better overall-perhaps worse in terms of efficiency. Certainly, the claim that “fuel cells are at least 2-3 times more efficient than a heat engine” needs to be looked upon with serious skepticism."
the claim that “fuel cells are at least 2-3 times more efficient than a heat engine” needs to be looked upon with serious skepticism.
Posted by: ai_vin | 31 January 2011 at 11:50 PM
Ah davemart I see your confusion.. thats not a power density issue they use a battery pack for two main reasons. It improves fuel economy. It reduces wear and tear on the stack.
They dont realy need it for power issues as thier stack is already able to burp out more power then the motors consume.
And ai vin.. the main concerns early on were getting enough fuel onboard for a solid range.. they have that now.. and getting the stack to fit in a small enough place.. they have that.. and finaly getting it to last.. the newer designs have enough of that and then some for the rollout.
The final and key thing they needed was to get the price of the car and fuel cell down low enough they could afford a rollout.. They are on track for that and in fact it sounds like some are ahead of schedual and might roll out sooner then expected.
In the 5-10 year after rollout timeframe all that realy matters is a tank of fuel both cost low enough and get you far enough.. and that the car both last and cost the right amounts. They have all of that well in hand. Its very unlikely they will fall short on any of it.
Posted by: wintermane2000 | 01 February 2011 at 03:12 AM
@wintermane:
You seem unable to understand the implications of what you write.
The motor and the fuel cell stack both have a power output of 90kw.
However it is rarely advisable to run a stack full out, hence it would cause extra wear and tear as you yourself write, and so in practise you are power limited by the stack.
In addition it is possible to run electric motors for short periods at over rated capacity without adverse effects.
All this is why they have put in a 21kw battery, to provide extra power.
They could of course have put in a bigger stack, but the economic choice is to put in a substantial battery.
All of this shows that the fuel stack at any acceptable price is power limited, which for some odd reason you are reluctant to accept.
Posted by: Davemart | 01 February 2011 at 03:34 AM
Wintermane had his eyes fixed, but his vision problems are on the other end of his optic nerves.
Posted by: Engineer-Poet | 01 February 2011 at 05:54 AM
Davemart.. the 100 kw is its peak power not its normal power... and its not even that.. the gm car used a 93 kw peak motor then changed over to a 95 kw one the old motor was less then 50 kw running power the new one 71 kw the honda uses a 97 kw peak motor. Also oddly enough the honda uses a 30 kw battery.
And why in hell do I remember all that? I cant remember if I fed the dog but I remember this junk?
Posted by: wintermane2000 | 01 February 2011 at 06:15 AM
@wintermane:
:-)
Posted by: Davemart | 01 February 2011 at 06:27 AM
Interesting discussion. Field temperature's effect on AER is coming into focus as the Tesla and Volt pave the EV path. One begins to suspect that the Nissan engineering team did not anticipate the impact of cold/hot weather effects on their Leaf battery. Without thermal pack management the Leaf battery may be a good part of the reason for rollout delay:
"Nissan's director of product planning for the U.S. Mark Perry responded by saying:
We don't need thermal management for the U.S., but we are looking at the technology for Dubai and other locations like that.... We've gone on the record saying that the pack has a 70 to 80 percent capacity after 10 years."
http://green.autoblog.com/2010/01/25/is-the-nissan-leaf-battery-pack-under-engineered/
Posted by: Reel$$ | 01 February 2011 at 11:06 AM
Actualy real$$ the trick nissan is using is the same one toyota used with the prius. They know a good number of packs will fail but they are willing to eat it to get out there first. They will just quietly replace packs that show unacceptable wear rates and work as hard as they can to improve the pack.
Posted by: wintermane2000 | 01 February 2011 at 01:40 PM
"We've gone on the record saying that the pack has a 70 to 80 percent capacity after 10 years."
This statement highlights something most people don't get. When battery/BEV manufacturers say their batteries will last X number of years they don't mean the batteries will quit working when the batteries are X+1 years old. They measure battery life by how long it takes to have its capacity reduced to 80%. A 100km BEV at end of battery life can still go 80km!
If that 100km BEV has batteries with a 10 year lifespan it will still be able to go 64km when its 20 years old and over 50km when it's 30.
'How old was your last ICE car when you replaced it, and how many kilometres did you drive it in a day?' These are the questions people should ask themselves when they think about buying a BEV.
Posted by: ai_vin | 01 February 2011 at 02:09 PM
Wintermane, I think you nailed it with Nissan. They made a simple business decision that it was more important (and possibly cheaper) to push a simple configuration out the door and replace the ones that wore too fast in a few years as prices drop for their manufacturing.
It allows them to have lower up front cost, a simpler system with less to break, and gets them to market faster. The downside is they may be replacing a lot of batteries in 5-7 years.
Posted by: DaveD | 01 February 2011 at 03:28 PM
"The downside is they may be replacing a lot of batteries in 5-7 years."
And quite possibly souring the public's acceptance of EVs as a viable transportation alternative. Five years and I gotta replace the battery??
Posted by: Reel$$ | 01 February 2011 at 07:28 PM
Not exactly real$$ a hidden silent replacement of packs they find are aging too fast would go completely unnoticed.
Posted by: wintermane2000 | 01 February 2011 at 07:31 PM