Editorial: Henderson's fuel cell 10X cost comments are out of context
Posted Nov 2nd 2009 11:55AM by Sam Abuelsamid
The other day we reported on an interview with General Motors CEO Fritz Henderson published in the Washington Post. While most of the discussion focused on the bailout and bankruptcy, from the perspective of this site, the main items of interest were Henderson's responses to questions relating the to the cost of the Chevy Volt and hydrogen fuel cells. Much has been made of Henderson saying that the Chevy Equinox Fuel Cell used for the Project Driveway program cost 10 times the Volt's approximately $40,000 price.
The problem here is that Henderson's numbers are taken out of context and mean nothing in and of themselves. This is a total apples to oranges comparison. The Volt is expected to carry a sticker price of somewhere around $40,000 at launch. That price will be reduced for customers by a federal tax credit of $7,500 along with whatever state incentives are available. However, that does not necessarily reflect the cost to build the extended range EV, which will likely be somewhat higher than $40k at launch. More important to this discussion is the fact that the Volt is also designed and engineered for mass production meaning that it is cheaper in all respects than the Equinox FCV. Keep reading after the jump.
The Equinox FCV is a whole different beast. The 100+ examples that have been built are essentially better finished equivalents of the pre-production prototype Volts that are currently undergoing development and durability testing. These were hand built prototypes with a lot of custom-made parts and the fuel cell system is not at all designed for mass production. The generation 4 fuel cell stack in the Equinox, in particular, is expensive for a number of reasons.
Take a look at the gen 4 and 5 stacks in the above photo. The case of the larger, older design on the right is made up of many individually machined pieces that are bolted together. Each of those pieces is fabricated from an aluminum billet and all the joints have to be sealed. Inside are tubes for the coolant passages and much larger than normal prototype hydrogen injectors. The smaller gen 5 stack has the same output while using half as much hydrogen and the case is cast as a single piece with integral cooling passages just like a conventional engine block. The overall part count of the gen 5 stack has been dramatically reduced. The catalyst coating on the stack plates is also much more even spread thanks to new manufacturing processes and uses less than 40 percent of the platinum content.
Similarly, the hydrogen storage system of the Equinox was hand built and is much more expensive than a design created for mass production. Newer designs like the Honda FCX Clarity that use a single tank rather than the three used in the Equinox are cheaper, lighter and integrate a number of components inside the tank.
Taken together, a car designed around these newer subsystems rapidly begins to approach the cost of something like the Volt if the whole vehicle was engineered for assembly line production. This is even more true if you start to remove the internal combustion range extender and expand the battery to give a purely zero emission vehicle. Getting a battery that would provide the 200+ mile range of a typical fuel cell vehicle gets enormously expensive and heavy.
The pre-production Chevy Volts currently being tested likely cost at least as much to build as the Equinox, but that's part of the development process. Building 80 pre-production cars was a process that took about four months and far more man-hours than any regular production car. That's why automakers are enhancing their up-front simulation work and compressing the number of prototype builds so they don't have to create as many.
The real reason GM is pursuing an architecture like the Volt has as much to do with politics as cost. The cost of the hardware for fuel cell and battery vehicles is converging rapidly. The problem is that there remain stubbornly few hydrogen filling stations and politicians have veered toward what is perceived as the easier infrastructure solution of the plug. That in itself is a whole other discussion.
On a recent segment of Autoline Detroit, GM's VP of research and development Dr. Alan Taub discussed batteries and fuel cells with John McElroy, Paul Eisenstien and myself. Taub reiterated what has been said by Honda, Toyota and Daimler, that battery vehicles will be best suited to small, short range, urban use vehicles with fuel cells providing the longer range full functionality that people often need. The reality is that there is likely to be little difference in the ultimate cost of either a plug or hydrogen infrastructure and we will probably want some of both in the long term. For now, though, comparing the cost of the Volt and the Equinox fuel cell is an irrelevant argument, and Henderson (or the WaPo) should have made that more clear.
Reader Comments (Page 1 of 3)
letstakeawalk 12:17PM (11/02/2009)
Thanks for the clarification. The automakers continue to bring the cost of fuel cells down, and the energy companies continue to develop methods of providing low cost hydrogen.
BEVs have taken well over a century to bring to market in a competitive form, so it's no surprise fuel cells still have to catch up. However, mass-production will bring the same cost decreases for fuel cells as it has for batteries.
Reply
Tim 1:03PM (11/02/2009)
H2 will ALWAYS be much more expensive energy carrier than electricity and H2s infrastructure complexity and costs make it a non-starter unless of course you’re trying to scam some idiot politician out of someone else’s (ahem, taxpayer) money.
It's simple physics.
(BTW, what is the average IQ of a congressman these days, below 80? All they have to do to win votes is to promise a “free” chicken AND a “free” pot. Public schools never taught the children that NOTHING is really free. What a shame!)
Dave 1:32PM (11/02/2009)
I have no doubt that hydrogen fuel cells are viable.
The total cost of the raw materials required to build a FCV are far cheaper than the raw materials required to build a BEV with a reliable 200 mile range. The 4th gen GM fuel cell requires less than $1,000 worth of platinum.
When mass produced, an FCV plus 200,000 miles worth of hydrogen will cost less than a BEV, unless there is a revolutionary advance in battery technology.
And FCVs actually fill the needs and wants of customers, while BEVs don't.
The truth, however, is that neither can replace liquid fueled vehicles in the USA because neither can be used to evacuate in an emergency. When the grid goes down, you can't charge your batteries, nor can you compress H2 to 10,000 psi. But you can easily hand pump gasoline, diesel, ethanol, or even methanol.
polo 3:04PM (11/02/2009)
"The total cost of the raw materials required to build a FCV are far cheaper than the raw materials required to build a BEV with a reliable 200 mile range. The 4th gen GM fuel cell requires less than $1,000 worth of platinum."
Its not the raw materials that make fuel cells rival Ferrari's in cost. Its engineering those components into a production-worthy system that can reliably sip hydrogen and last longer than 5 years. The "buuuut they're custom-built" argument doesn't account for the high cost either. The EV1 was custom-built, using state-of-the art, unique parts, by hand, and had a full cost of $87,000. and lets say they used lithium batteries instead of cheaper ones used in the EV1, making the price $100K, and lets say as the Volt is only around 40K, that mass production can chop 60% off the cost. Well....the cost for the Honda Clarity is in the 7-figure range, and a conservative estimate for the Equinox would be the mid 6-figure range. Even if you chopped the price by 60% with mass production the cars are still too cost-prohibitive to mass produce. This is why we won't be seeing any limited-production luxury hydrogen-powered equivalents to the Telsa Roadster - the cost is simply too high no matter the margins. And that is why in 2015 we'll be hearing the same arguments and promises by the hydrogen folks..with no actual production cars to show for it.
To say a car can go from the price tag of an F1 race car to a nice Hyundai with just some mass production is undeniable pure fantasy and I think Fritz Henderson's $400K FCV Equinox figure was the price AT mass production, not before.
"When mass produced, an FCV plus 200,000 miles worth of hydrogen will cost less than a BEV, unless there is a revolutionary advance in battery technology."
One huge logic bomb there......if thats the case WE WOULD ALREADY HAVE HYDROGEN CARS. Even Honda is backing out their "mid-next decade" and conceding that they'll be pursuing electrics.
Unlike FCVs, EVs are a flawless example of scale-of-economies being the primary barrier to cost-competitiveness. Produce more, the batteries get cheaper, the cost goes down, you sale more, the batteries get even cheaper, the cost goes down even more, you add more batteries, you sale more, etc. That cycle be producing some very cheaply priced batteries with good ranges by the time FCV's are promised to come out - there is simply no way they would be price-competitive with BEVs even if they ever were put into mass production (they won't be). Add in the likelihood of the non-existent hydrogen infrastructure being developed (here or globally) and you have a near guarantee these will be nothing but showroom ponies - propped up by big oil interests and automotive grant-hunger labs - into the far future.
Dave 4:45PM (11/02/2009)
"Unlike FCVs, EVs are a flawless example of scale-of-economies being the primary barrier to cost-competitiveness. "
Pure nonsense.
Batteries are already mass-produced. They are still heavy and expensive.
jake 4:48PM (11/02/2009)
@Dave
Raw materials aren't necessarily the main drivers of price for either technology. If this was true, lithium manganese batteries (used by the big automakers) should be a lot cheaper than lithium cobalt (cobalt being insanely expensive in comparison), but so far it isn't (yet).
If you want to compare raw material costs:
Lithium carbonate is about $10 per kg. ~0.6kg per kWh of battery. So a 100kWh battery that will easily give 300 miles of range will have about 100kWh*0.6*$10 = $600 worth of lithium carbonate.
Cobalt costs about $50-100 per kg.
Manganese costs about $2 per kg.
Iron & phospate (for LiFePO4 cells) is about $0.50 per kg.
Using the cost balance of Lithium cobalt cells, the cobalt needed is about 8.5x the lithium by weight. Pure lithium figure is 3x the carbonate ($30/kg vs $10/kg) so this takes the factor down to 2.83x when comparing cobalt to carbonate, let's say 3x for easier math.
Doing the math:
$9000-18000 cobalt per 100kWh
$360 manganese per 100kWh
$90 iron phosphate per 100kWh
Adding in the $600 of lithium cobalt, you have:
$9600-18600 lithium cobalt cathode per 100kWh
$960 lithium manganese cathode per 100kWh
$690 lithium iron phosphate cathode per 100kWh
For platinum, it's about $40k per kg right now, so you have 80g*$40k/1000g = $3200 now. If 30g by commercialization: $1200. By mass production, if GM can get it down to 10g it'll be $400.
It doesn't seem like it's that big of a difference in raw material cost savings even in the mass production case (only exception is cobalt, but again big automakers aren't going with cobalt), esp taking into account platinum prices being fairly volatile (b/c of it being rare). If we go with the 100 mile BEVs we have now, it'll immediately drop the raw cost estimates I have to a third. A 40 mile EREV will drop it further. You can't do the same with fuel cells since you need a minimum amount of power just to run the car.
Also you have the steady tech improvements (~8% year) that drive up energy density, which will reduce raw material per kWh over time.
(lithium carbonate cost & component from here:)
http://www.chemetalllithium.com/fileadmin/files_chemetall/Downloads/Chemetall_Li-Supply_2009_July.pdf
(other costs here:)
http://seekingalpha.com/user/290367/comments/symbol/abat.ob
On the article itself, the Volt isn't necessarily reaching mass production either, first year production is only in the 10k range (the same production volume for a "commercial" hydrogen vehicle, but not mass production). And the Voltec system is only in its first generation; the fuel cell stack is in its 5th gen already. In fact pretty much all of the EVs we see now are in its first gen (the automakers dumped their older designs from the ZEV program). Only major exception I can think of is the Prius plug-in, which is second gen (after switching to lithium). They are also using new chemistries (manganese and iron phosphate). There's still quite a bit of mass production that can be done on EVs.
I'm willing to guess subsequent versions of the Volt will be closer to the $30k range originally estimated. That's still a 50% drop, With hydrogen it's a 10x (or more) drop. Is it really possible? We'll see by 2015-2022, I guess.
Joeviocoe 11:21AM (11/05/2009)
"And FCVs actually fill the needs and wants of customers, while BEVs don't"
MYTH!!!
FCVs fill the needs and wants of the Oil companies who would love to continue to be the barons of us all.
Range and recharge times are the ONLY advantage hydrogen has over BEVs. Range is quickly being overcome. By 2012, 300+ mile BEVs will be available at reasonable cost.
Recharge times ONLY comes into play during road trips when you don't want to spend 4 hours charging for 3 hours of driving. Understandable, but instead of spending billions on this hydrogen infrastructure that will ultimately cost drivers more money, 3rd party rentable towable range-extenders will do just fine.
Peter 12:18PM (11/02/2009)
Sure, the Equinox is low volume and the Volt is high volume, but that's not enough to overcome a 10x price gap. Pagani made 15 Zonda S supercars priced at $350k, and I doubt the price would be $35k even if they produced 15 million. Some designs and materials (i.e. platinum) just don't come down in price significantly at high volume.
Reply
polo 3:08PM (11/02/2009)
Its less the platinum than the engineered components that basically lock-in FCVs as forever prohibitively expensive...unless they radically re-eingineer the tech, and good luck ever pricing them to get back the cost of investment in either case.
Reality Hurts 12:18PM (11/02/2009)
Very informative article. I still stand by Hydrogen for larger trucks, ships, semi's, etc. I think that the future will have both plug-in cars and hydrogen vehicles and I am excited that GM is heading toward both.
Though the cost of an H2 vehicle may not make sense for the family sedan, when you start getting into buses and semis, where the price tags are already pushing in the hundreds-of-thousands, H2 begins to make sense.
Reply
BlackbirdHighway 12:20PM (11/02/2009)
"...there is likely to be little difference in the ultimate cost of either a plug or hydrogen infrastructure... "
Bullfeathers! I was starting to go along with this until that phrase, then it lost all credibility for me.
Reply
Chris M 11:06PM (11/04/2009)
So, I'm not the only one that noticed that bit of nonsense. Even if we assumed far fewer H2 fillup stations than charging outlets, the fact remains that electrical wiring and the other components for recharge outlets are very inexpensive, compared to the requirements for H2 production, compression, storage and extremely high pressure H2 rated hoses and nozzles. The result is that for the cost of one H2 station we could install several hundred charging outlets, and install those outlets in places not suitable for H2 station, like parking garages and shopping mall parking lots.
Lower construction and installation costs, lower real estate costs, more convenient locations, plug-ins have all the advantages.
Joeviocoe 12:01PM (11/05/2009)
The biggest misdirections of the FCV vs. BEV debate is the Battery costs vs. Fuel Cell cost.
Look at the facts and the bigger picture. Since technology advancements both now and in the future influence cost, there is very little reliability to the predictions of cost.
Both have been heavily invested in and both will advance. But that is the misdirection!
The problem is the infrastructure. H2 cannot be as efficient as pure electricity could be. Yes, you can set up the energy conversion paths to favor hydrogen like many people have done. But bottom line is that the cheapest/best way to make electricity will ALWAYS be more efficient and cleaner than the cheapest/best way to make an equivilant amount of H2.
So what IS THE REAL DEBATE?
Each mile driven with hydrogen MUST PAY for the infrastructure!
You cannot get around the fact that stations would have to be retrofitted, rebuilt or built brand new. Pipelines will need to be run because petroleum pipes cannot hold hydrogen. High pressure systems, distrobution facilities, new employees, new natural gas reformation plants to handle the demand, the yearly maintenance associated with all that, etc.
Getting high voltage charging outlets are soooo much easier. Little to no maintenance. You can even build them on the curbs of regular streets. There isn't alot of money involved in building up a BEV infrastructure. The current one works already for those EVs currently on the road. The charging outlets will grow in parallel to the sale of BEVs.
But hydrogen stations would have to preceed FCV by a good margin by the time the FCVs hit the road. Or else you will be "Out Of GAS"!
Doug 12:31PM (11/02/2009)
The bottom line problem with hydrogen is the expensive infrastructure required for it compared to alternatives. They should continue to develop the technology until either direct alcohol fuel cells or on-board reformation of liquid fuels is practical.
Reply
Patrick 11:41AM (11/04/2009)
Actually, the costs for expanding the hydrogen infrastructure are more affordable than most people think--and about the same cost per vehicle as battery vehicles.
http://www.h2gen.com/Uploads/file/Battery_vs_FuelCell_EVs.pdf
You're going to have to invest in something if you want to move away from gasoline.
Doug 1:46PM (11/04/2009)
I wouldn't exactly call that an unbiased report. That section 2.7 "Fueling Infrastructure Cost" ignores some basic issues. Namely, the infrastructure for PHEVs is already in place and many people already have an outlet in their garages from which they could charge a car. But really, if you're looking at the cost to set up home charging, then that should really just be rolled up in the purchase cost of the vehicle. I don't think we'll be able to buy a HFCV for under $100K anytime soon. The GM Volt is available next year.
Also that per vehicle cost comparison requires a critical mass of HFCV owners to use a single station (the article suggests 2300 vehicles). It may take a good while before there are that many HFCV owners. The PHEV owner doesn't have to wait for a bunch of other owners or anyone to build a new infrastructure.
Fuel cells as a technology do have their place and I would love to see them progress to the point where it makes sense to use them in places where we typically use combustion to extract energy from hydrocarbons. But really, pure hydrogen is an extremely poor energy carrier. I just think it's silly to build a whole transportation infrastructure for it given the alternatives.
Fuels cell vehicles should adapt to be able to use the existing liquid fuel infrastructure (like PHEVs are able to) before they're seriously considered for private transportation.
Stan Peterson 12:41PM (11/02/2009)
FCEVs are a desirous want of the Quacks at CARB who enter some religious ecstasy, from the point that the only emission from a FCEV are lots of the greatest global warming gas in our atmosphere.
So all the automakers must genuflect in the direction of FCEVs, with CARB still holding the ability to foul up their production and create multi-billion dollar losses.
No matter how "productionizing" the FCEV will reduce costs, the FCEV will still be 10x or more the cost of EREVS. I can agree with Dr Taub that EREVS and BEVs are more applicable to the smaller end of the Transport sector. Where they are not appropriate is EXACTLY where dual-mode hybrid PHEVs will dominate and suffice.
But everyone seems to think that ALL transport needs to migrate from fossil fuels. Nonsense! We can easily the reduced demand, and satisfy all our truck Transport sector needs from only a portion of domestic production for hundred or thousands of years. Indeed it becomes entirely feasible to displace all Fossil production with manufactured fuels from Bio sources, if all that is needed is to meet commercial truck, locomotive, maritime and aviation needs.
Reply
ShaunneyCakes 1:00PM (11/02/2009)
Wow, you are REALLY underestimating the amount of fuel comsumed by commercial means. The drop in fuel prices this year was caused by the shock based on a 10% reduction in deisel demand in the commericial sector, gas demand did not decrease during the recession as much as many believe, the reason we are at $2.65 a gallon is because of the hiccup in the commercial sector.
Though I agree with you that we do not HAVE to move away from fossil fuels, I feel it should still be a top priority since oil has so many uses that do not involving transport or being burned.
paulwesterberg 1:26PM (11/02/2009)
@Stan: You are assuming that other countries with emerging economies will also reduce their use of fossil fuels. Good luck with that.
One gallon of gasoline has the energy equivalent of 14 days of heavy manual labor - I don't think that many people will seriously reduce the use of fossil fuels until the price becomes much more expensive.
http://valuesystem.livejournal.com/9846.html
paulwesterberg 1:15PM (11/02/2009)
Rampimg up volume wont happen if:
-fuel cell units are too expensive
-fuel cells only last 80-120k miles
-infrastructure is non-existent/lacking/expensive
-fuel is more expensive than gas(how do you invest $$$ in stations and keep fuel prices low with low initial sales volumes)
Per unit prices may not be the problem holding back hydrogen, but you have to solve all the other problems first(simultaneously!) and then you might have a product that people want to buy and can talk about production volumes and economy of scale.
Hydrogen infrastructure is the biggest gamble, who wants to front billions of dollars for something that will lose money for many many years and probably wont ever be successful based on the poor efficiency of hydrogen energy conversion.
Reply