« Tesla Sues Fisker, Is Sued by Magna | Main | LDV Group Developing Electric Version of MAXUS Commercial Van »
Determining the Appropriate Design and Configuration of PHEVs
16 April 2008
A set of papers presented yesterday by engineers from Toyota, Argonne National Laboratory and GM at the SAE 2008 World Congress explored the impacts and requirements of different types of plug-in hybrid electric vehicles (PHEVs)—specifically extended range electric vehicles versus a blended power split approach.
The Toyota paper, Study on the Potential Benefits of Plug-in Hybrid Systems, started by noting that while plug-in hybrids can address the three big issues facing transportation—fuel consumption and energy diversification; greenhouse has reduction; and air quality—breakthroughs in battery energy density, reliability and cost must occur.
Toyota’s basic argument is that the costs and trade-offs of deploying an extended range electric vehicle architecture at this time outweigh the benefits, and that blended systems have greater benefit at this point in time.
Toyota considered a series hybrid with a small ICE as a range extender using an EV strategy (e.g., the Chevrolet Volt) and a parallel hybrid with a plug-in pack using a blended strategy (e.g., the Toyota PHV based on the Prius).
Using the US06 cycle to represent typical North American driving, Toyota concluded that up to 100 kW of output is required to drive that cycle in a mid-sized vehicle. The company then considered the operation of a blended system designed for charge depleting operation over the city cycle. Toyota concluded that maximum output required is approximately 40 kW, and the average was less than 20 kW.
An analysis of the impact of reducing battery power showed that while a 40 kW battery enabled all electric operation almost 100% of the time in the city cycle, using a 20 kW battery still enabled electric operation 95% of the time. An extended range vehicle, Toyota pointed out, would require a battery, motor and electrical system capable of providing maximum power (100 kW).
Toyota also concluded that while CO2 reduction increases for PHEVs with longer all-electric range, the benefit is not linear, and that as EV range increases, CO2 reduction levels off due to the high percentage of drivers with short daily driving distances.
Toyota argued that providing increased EV range increases vehicle cost due to higher battery cost; reduces luggage space; and increases fuel consumption in charge sustaining mode due to increased battery mass.
A paper from Argonne National Laboratory, Comparison of Powertrain Configuration for Plug-in HEVs from a Fuel Economy Perspective, analyzed several PHEV powertrain configurations, including series (e.g., Volt), pre-transmission parallel (e.g., the Sprinter van) and power split (e.g., Prius) with respect to component sizes and fuel economy for 10- and 40-mile all electric range (AER) applications.
Argonne sized the batteries in its evaluation (which used the Powertrain Systems Analysis Toolkit, PSAT) to follow the UDDS urban cycle while in all-electric mode, and to meet the AER targets for the vehicle. Argonne also specified that the PHEV would operate in electric-only mode at higher vehicle speed in comparison with regular hybrids.
| Component Sizes in Argonne Study | ||||||
|---|---|---|---|---|---|---|
| Parameter | 10-mi AER | 40-mi AER | ||||
| Pre-trans Par. | Split | Series | Pre-trans Par. | Split | Series | |
| Engine power (kW) | 76 | 74 | 109 | 79 | 77 | 114 |
| Propulsion motor (kW) | 48 | 62 | 90 | 50 | 71 | 95 |
| Generator power (kW) | NA | 63 | 106 | NA | 65 | 111 |
| Battery power (kW) | 58 | 52 | 55 | 61 | 64 | 58 |
| Battery capacity (A·h) | 18 | 21 | 18 | 71 | 69 | 71 |
| Total vehicle mass (kg) | 1,675 | 1,667 | 1,700 | 1,764 | 1,800 | 1,794 |
![]() |
| Simulated fuel economy results under urban and highway test cycles for the Argonne study of three different configurations in 10AER and 40AER cases. Click to enlarge. |
In terms of fuel economy, Argonne found that the split configuration provided the best fuel economy. The series configuration suffered from dual power conversion—from mechanical (engine) to electrical (generator) and back to mechanical (electrical machine). Engine efficiency was higher for the series configuration that for other configurations, and electrical efficiency was practically identical for all three configurations.
The Argonne researchers concluded that:
Although both the power split and series configurations require two electric machines and an engine, the series configuration, as expected, requires significantly higher component power as a result of the many component efficiencies between the engine and the wheel.
In terms of efficiency, all of the configurations achieve similar characteristics when operated in electric mode. Both series and power split configurations do not use a multi-gear transmission, but the parallel configuration makes up for losses by operating the electric machine at higher efficiency points. In CD mode, the power split provides the best fuel economy as a result of its dual path of power form the engine to the wheel.
On the basis of the thermal and electrical consumption analysis, series configurations appear to be an appropriate choice for vehicles designed to provide long AER because of their simplicity in terms of control and their ability to operate in electric-only mode at high vehicle speed. The power-split configurations appear to be a valid choice for vehicles based on a CD approach.
The GM paper, The Electrification of the Automobile: From Conventional Hybrid, to Plug-in Hybrids, to Extended Range Electric Vehicles, was presented in an earlier form at the 2008 SAE Hybrid Vehicle Technology Symposium in San Diego (13-14 February). (Earlier post.)
GM is currently developing both a blended strategy PHEV (the Saturn VUE Green Line PHEV is a derivative of the conventional 2-Mode Hybrid) and the extended range electric vehicle (E-REV), the Volt.
For the paper, GM simulated the performance of a mid-size sedan with a conventional powertrain; an HEV with a 40 kW electrical power constraint; a converted PHEV with a 35 mph (56.32 kph) speed constraint, a 40 kW electrical power constraint, 3.5 kWh of usable electrical energy (as opposed to total battery pack energy), and a blended operating strategy; an urban-capable PHEV with a 60 mph (96.56 kph) speed constraint, a 53 kW electrical power constraint, and 3.5 kWh of useable electrical energy; and an E-REV with 8 kWh of useable electrical energy and EV capability not limited by electric power or driving speed.
The key to the results of the simulation is the use of the operational data from 621 drivers captured in the Southern California Association of Governments (SCAG) Regional Travel Survey (RTS).
![]() |
| Net battery energy versus distance driven, compared to the requirements of the three different cycles. Click to enlarge. |
GM calculated the driving intensity—the net energy per mile (kWh/mile)—required by the urban cycle, the highway cycle, and the much more aggressive US06 cycle, then compared these to the RTS data. (See diagram at right.) They found that while only 3% of the real-world drivers fit within the urban cycle and 21% fit within the highway cycle, fully 97% fit within the requirements of the US06 cycle.
GM concluded that:
The real-world RTS data set contains widespread and significant driving at power levels and speeds beyond that represented by the urban driving schedule.
An E-REV is more than ten times as likely to finish the day as an EV than as urban-capable PHEV derived from an HEV, when operated in the actual application, as represented by the RTS data set.
An E-REV will consumer, on average, less than half of the petroleum of a PHEV in the real world, if overnight charging is assumed.
An E-REV will reduce regulated emissions that are due to initial trip starts by more than 70% when compared to a PHEV in the actual application
Electric range when operating on the urban schedule is not a direct measure of a plug-in vehicle’s ability to run with the engine off, ability to displace petroleum or ability to reduce regulated emissions in the actual application. Rather, the ability to run with full performance on electric power alone leads to improvements which would be realized in actual application.
In the event of a petroleum disruption, an E-REV could support uncompromised vehicle operation for the majority of drivers.
We conclude that electrification that enables E-REVs may be well worth the effort. Specifically designed electric powertrains, incorporating higher power motors and thermal systems, higher energy batteries and integrating them into vehicle structures specifically designed for that purpose will be rewarded with societal benefits in real world use. While PHEVs can make improvements compared to HEVs, an E-REV appears to realize a much greater portion of societal benefits.
Resources
Masayuki Komatsu (Toyota), et. al.; Study on the Potential Benefits of a Plug-in Hybrid System (SAE 2008-01-0456)
Aymeric Rousseau (Argonne); Comparison of Production Powertrain Configuration Options for Plug-in HEVs from Fuel Economy Perspective (SAE 2008-01-0461)
Edward Tate (GM) et. al.; The Electrification of the Automobile: From Conventional Hybrid, to Plug-in Hybrids, to Extended-Range Electric Vehicles (SAE 2008-01-0458)
April 16, 2008 in Plug-ins | Permalink | Comments (63) | TrackBack (0)
Comments
Posted by: JamesEE | April 16, 2008 at 11:53 AM
sry. here is the one i read this morning. it references the first link i posted. the top picture is the same, but the text is different.
http://www.evworld.com/article.cfm?storyid=1434
Posted by: PeteVE | April 16, 2008 at 11:55 AM
I know a little bit about US06, probably just enough to be dangerous....
It's a rigorous, 10-minute test (600 seconds). The specs of the test dictate the required speed of the vehicle for each second of the test. There are starts and stops and large accelerations. Top speed is near 80 MPH, which seems odd for a US test (speed limit is 70 MPH). Anyway, so it is possible a car could "fail" the test if it could not keep up with the accelerations dictated by it. Large power need would be for one of the high acceleration events. Note that large braking events also dictate large power "sinks", which many hybrids cannot fully absorb, and thus will lose that energy recovering opportunity.
Posted by: Jim | April 16, 2008 at 11:59 AM
Oh,
Appropriate battery size is obviously a huge issue with PHEVs, as it can so highly affect price. It is reasonable that no one really knows what the best value of this should be, be it 5 miles, 10 miles, 20 miles, etc. of all-electric range. I think the Toyota 7 miles is on the low side.
Posted by: | April 16, 2008 at 12:01 PM
PeteVE is most likely referencing this EV World Article:
http://www.evworld.com/article.cfm?storyid=1434
Posted by: Wayne | April 16, 2008 at 12:07 PM
it seems that the comment that they found those deposits a few years ago should be tracked. and i apologize for it.
Posted by: PeteVE | April 16, 2008 at 12:17 PM
"It is reasonable that no one really knows what the best value of this should be, be it 5 miles, 10 miles, 20 miles, etc. of all-electric range."
Agreed. The design specifics get REALLY interesting when it comes decision time for what the nominal charge level will be for the battery. Charge the battery all the way up and there is no where to put braking energy. Charge it only part way and you can recover braking energy at the cost of reduced all-electric range.
Eventually, we will probably see adaptive battery charge-profile controls that "learn" from individual driving habits over time. Or that can dynamically anticipate the road terrain ahead from GPS/3D-mapping cues. Even real-time traffic info for maximizing start-stop performance. And all of this can be done with software. As it's just information, no resources will be used for these benefits.
We're only at the beginning edge of a Cambrian explosion in automotive design. Car repair-technicians are going to have a real adventure trying to stay current on all the differentiated new systems.
Posted by: Hybrid fan | April 16, 2008 at 12:44 PM
You can add that discharge threshold is also important to save battery life. 100% of depth of discharge will limit the number of cycles and then of course, increase the total price per mile of the vehicle.
I’m just dreaming of the next Jurasic period without transmission but motor wheels, without ICE because of good batteries or fuel cells, without all these complex software to manage power… I call it a nice value crunch in car market.
Posted by: Fabio Ferrari | April 16, 2008 at 01:12 PM
The new test is a result of GASP them actualy testing road conditions and actual real world real driver driving needs. Becuase many us freeways are criminaly designed massive avvelleration is needed by far too many NOT for fun NOT for thrill but just to avoid becoming a stat as 25 tons of big rig doing 90 chases your ass out the merge lanein a blind double merge that happens on a CURVE going UPJILL over a big bump that then swerves under a bridge and ANOUTHER FREAKING DOUBLE LANE BLIND MOTHER FREAKING MERGE!!!!!
Posted by: wintermane | April 16, 2008 at 01:26 PM
JamesEE:
The new improved Prius coming our in January 2009 will almost fit your requirements. It may only do 7 or 8 miles on the all electric mode, instead of the 10 miles you would like to have, but it is rather close.
An improved battery pack, another 2 years down the road, will give you close to 20 miles.
I put my name on the unofficial waiting list for the new generation Prius. It may be the best compromise for the next 2 or 3 years.
Posted by: Harvey D | April 16, 2008 at 01:27 PM
US06 maximum delta V is 8mph per second.
@Jim
You can't "fail" the US06 test. It is a test originally devised for emissions testing to determine mpg (part of the US EPA five-cycle test). If your car cannot perform to the profile of the test then the EPA accomodates such issues.
From 77908 Federal Register / Vol. 71, No. 248 / Wednesday, December 27, 2006 / Rules and Regulations:
[regarding specifically US06 testing]:
"Additionally, the manufacturers may use other methods based on good engineering judgment, with EPA review and approval, as long as these methods achieve equivalent or better, technically valid results based on manufacturer submitted data."
Speculation: Such engineering judgment could include submission of US06 results even if the vehicle cannot match the US06 profile. Alternately an OEM may be able to estimate the results from information gained through other testing (such as results obtained with same vehicle body, but tweaked powertrain).
Posted by: John | April 16, 2008 at 01:27 PM
John, do you know how are they going to deal with PHEV? Starting in electric mode, enough battery to do the complete cycle, result will be a ZEV???
Posted by: Fabio Ferrari | April 16, 2008 at 01:44 PM
@Harvey D:
"GM may have to curtail or postpone mass production of the Volt PHEV until batteries are improved and their price come down (2012/14?)."
I don't think GM will go that way considering they have declared the Volt a "Manhattan Project" and written a blank check for its introduction. Both the doped lithium phosphate from A123 and the manganese spinel from Compact Pwr. LG Chem, are running ten year lifecycle tests. According to GM the tests are on track meeting all expectations. If they meet a first year 2010-2011 production run of up to 10,000 units that should satisfy early adopters and be a grand slam home run (Toyota sold 6,500 Prius in North America first year.) If the batteries run for ten years that keeps more ICE off the road and further strengthens demand.
Not only does GM have two competent battery suppliers - there is also the lithium titanate spinel from AltairNano hovering in the background. Should GM choose to take an investment position in Altair - the cost of their technology could be brought down more rapidly.
http://gm-volt.com/2008/04/15/even-chevy-volt-battery-suppliers-surprised-at-gms-volt-aggressiveness/
As to lithium supplies here's some work on extraction from seawater:
http://www.batteriesdigest.com/lithium_materials.htm
Posted by: gr | April 16, 2008 at 02:21 PM
@Fabio
In short, I don't know. I have read the laws and regulations and ruling several times (as part of my job), but it is beyond me to know whether the industry and the EPA will see eye-to-eye (or come to an understanding without lawsuits) over cases that are arguably not yet addressed.
The 5-cycle test was developed with existing HEVs of the time(MY2006, see footnote 29 of the EPA Final Ruling listed in my earlier post): Honda Insight, Honda Civic, Honda Accord, Toyota Prius, Toyota Highlander/Lexus RX400h, Ford Escape/Mercury Mariner, and Chevrolet Silverado/GMC Sierra pickup truck.
For the uninitiated, Fabio is bringing up the point that the EPA tests are relatively short in nature and they may not cause a representative usage of the ICE as found in real driving over the time between plug-ins of a PHEV.
It seems to me that the EPA has not considered the accounting for a vehicle that may perform the 5-cycle test in a drivetrain mode that is not representative of real world driving (e.g. 5-cycle conducted in all electric mode when a real person would drive the vehicle enough to deplete charge and require the ICE or other fuel burning device to engage). There was a lot of deliberation in the final ruling on the idea that an HEV ICE may not be properly warmed up during the warm phases of the FTP test, but I have not concerned myself with that portion of the ruling to any great detail.
Although the EPA has Carbon based measurement rules for mpg of alternate fuel vehicles, I do not understand them to account for the Carbon released in a place other than the car (a situation that would occur in a PHEV or full electric vehicle with the fuel being burned at the electric generation plant or even not at all for a nuclear plant).
Posted by: | April 16, 2008 at 03:06 PM
Oops. Last comment to Fabio was by me.
Also, my 8mph per second comment in my first post was acceleration.
Posted by: John | April 16, 2008 at 03:11 PM
Thanks for your answer John. For your information, the max acceleration of the NEDC cycle is 2.33 mph per second (1.04m/s-2). I'll try to find what they will do in Europe for PHEV...
Posted by: Fabio Ferrari | April 16, 2008 at 03:34 PM
Lithium from Seawater - they got 30g of LiCl from 140,000 litres of seawater. That is 5g of Lithium. From the same amount of Atacama brine at 3000ppm at 50% recovery one would expect to get 210kg of Lithium metal. 42,000 times as much.
Mother Nature does not produce Lithium Carbonate from these Chilean brines. She does more or less produce LiCl which has to be chemically treated to produce Lithium Carbonate.
Posted by: Emphyrio | April 16, 2008 at 03:53 PM
There is no such thing as an e-rev other than a marketing term that GM is using to brand its potential Volt PHEV-40.
As Andy Frank, who has built more PHEVs than anyone else, pointed out the issue is that none of these models have been optimized for components particularly CVTs.
Posted by: ED Centre for Sustainable Transportation | April 16, 2008 at 06:30 PM
It's so obvious from this article that Toyota are suffering from an inferior battery technology.
Their aim to limit battery output to just 20 kW for "cost" reasons is really because they want to baby their pack as much as possible because it just isn't good at cycling.
If Toyota (well, Panasonic EV) had chosen a more appropriate electrode material for the project, like GM has, and not old-skool cobalt oxide they would be saying exactly the same thing as GM: Big power, big advantage.
Posted by: clett | April 17, 2008 at 03:44 AM
One of our problems in even discussing electric transportation issues is that everyone involved keeps inventing new short terms and fails to even begin to explain them.
Have a look at the new and unfamiliar terms in just this one page ~>
PHEVs, EV, ICE, HEVs, PSAT, UDDS, AER, CD, VUE, E-REV, SCAG, RTS, US06, RTS,
Newbies to the field are unlikely to even sort through this mess of new and unfamiliar unexplained terms.
Someone somewhere needs to start a "dictionary of new and improved electric/alternative fuels car shortcut terms" (perhaps GCC could start a handy update reference page).
Meanwhile,
The level of disinformation foisted off by various parties involved is outrageous, and always slanted towards a particular favored sector involving their own product.
Is any one surprised that Toyota discovered the Prius was better, while GM discovered the new Volt will be an improvement?
Posted by: John Taylor | April 17, 2008 at 04:16 AM
So, why don't they focus on building "retrofitable" series hybrids, that can easily be upgraded to plugin-series-hybrids.
Once we get the batteries or ultracaps, you simply add those in to the already existing car, and you're good to go.
Posted by: GreyFlcn | April 17, 2008 at 05:28 AM
A standardized "battery slot" form factor for that would be rather nice also, to help scale up battery production.
As well for legally! allowing for factory-certified after market retrofits to not void the warranty.
Posted by: GreyFlcn | April 17, 2008 at 05:33 AM
@John Taylor
Thanks for your comments, John. You are correct that people within certain groups develop and take for granted their group-specific vocabulary.
I can shed light on the terms I used:
FTP - Federal Test Procedure. A very broad term regarding emissions testing drive cycles the US EPA uses.
US06 - The US EPA recently expanded the emissions testing that they use for determining the mpg rating that goes on those labels posted on new cars. This is one of the five tests. (see www.epa.gov for exhaustive information)
HEV - Hybrid Electric Vehicle.
PHEV - Plug-in Hybrid Electric Vehicle. An HEV that allows battery pack charging from the electric grid as a method of reducing vehicle fuel consumption.
ICE - Internal Combustion Engine.
delta V - Velocity change. Another way of saying acceleration.
OEM - Original Equipment Manufacturer. (e.g. Ford, Chrysler, GM, Toyota, Nissan, Honda, etc.....)
I'm hoping that I don't need to explain EPA, US, or mpg.
Posted by: John | April 17, 2008 at 05:53 AM
"The Toyota paper, Study on the Potential Benefits of Plug-in Hybrid Systems, started by noting that while plug-in hybrids can address the three big issues facing transportation—fuel consumption and energy diversification; greenhouse has reduction; and air quality—breakthroughs in battery energy density, reliability and cost must occur."
This entire premise is patently FALSE and thus obviates any conclusions that the paper wishes to elucidate.
No breakthroughs, no reliability issues and no cost hurdles need overcome. NONE
NO ONE knows this better than Toyota since they have already produced the great Toyota RAV4 EV, which vehicle proves every single day that auto manufacturers can give us what we need, TODAY.
Posted by: jabroni | April 17, 2008 at 06:30 AM
jabroni: While I agree with you 95%, I must admit that in todays reality, the cost of batteries is still an issue. At 4->500$/kwh, with 2000+ life-cycles, I would consider the issue solved. At 300$/kwh the world would change in a hurry.
Posted by: Neil | April 17, 2008 at 06:47 AM
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c4fbe53ef00e551d83cb78833
Listed below are links to weblogs that reference Determining the Appropriate Design and Configuration of PHEVs:

Twitter headlines


No Name posted: "Besides that, one aspect of this article that really bugs me is the use of "kW" when they should be using "kWh"."
Actually, I think they use kW when they refer to the power capability of the battery, and kWHr when they refer to the energy (correctly). In the table they use A*Hr for battery capacity, without stating the voltage, which is probably about 275-300 V. So they are using the correct units.
I believe Toyota's point is that "blended systems" require much less peak power capability from the motor and battery than extended range EVs. They believe that to meet customer expectations at least 100 kW of power is needed for acceptable acceleration. They conclude the extra weight and cost of the motor and batteries is not worth it. They prefer to achieve the acceleration by using some ICE power when needed. They've already shipped a million vehicles with that strategy and have numerous patents, so it's not surprising.
I love my Prius, and I think their approach has a lot of merit. If the next Prius continues the blended system, but with more powerful (kW) batteries with more energy (kWHr) I believe it could get up to 75-100 MPG. A 10-mile AER wouldn't be cost-prohibitive, and would get people to the supermarket and back with zero gasoline. I'd buy one.
On the other hand the GM Volt is an incredibly interesting product. It's much more challenging. I hope they can pull it off -- it really is a game changer for GM.