Fuel cells start to look real
More 1
2
3
4
5
6
7
8
Vehicles
Sidebars
Getting the cost out
Until recently, building a fuel cell sufficiently powerful to run a car was costly - even more than a vehicle powered by electrochemical batteries or a hybrid drive. To attain the power levels of a standard-issue IC engine in a midsize sedan, a fuel cell needs to produce from 60 to 90 kW. When NASA first started using fuel-cell technology in space, a PEM fuel cell cost about $500,000 per kW. Today that price has dropped to around $500 per kW - but that means that a fuel-cell engine still costs about $25,000, which is around seven times the price of a typical IC engine (about $3500).
Working for several years with specialists from Ford and DaimlerChrysler, Ballard researchers studied the automotive industry's needs for low-cost, high-volume fuel-cell stack manufacturing and specifically designed the Mark 900 unit to accommodate them. "The key to developing an efficient supply chain," explained Lancaster, "is to choose low-cost, readily available materials and cheap, scaleable, automated manufacturing processes. We did an actual commercial plant study for the annual production of 300,000 vehicle equivalents, considering the building, logistics, and other crucial details. Using a standard rule of thumb for value allocation in fuel-cell systems of 40% for the stack, 40% for the system components, and 20% for the electric drive and transmission, we determined that fuel-cell powerplants could be sold at around $50-60 per kW, perhaps less as volumes increase."
Facing the challenge of making economical fuel-cell units, Ballard worked with Ford and DaimlerChrysler to optimize its latest stack design for production. Said Lancaster, "Whereas the Mark 700 systems were basically hand-crafted units that needed carbon plates that were individually machined for two hours from blanks costing $100, the Mark 900 unit is made using a carbon sheet material called Grafoil which is supplied by U.S. Carbon Graftek. This soft, natural graphite material comes in rolls. The sheet is first roller-embossed, die-cut, then impregnated, and heat-treated. Now each plate costs a few dollars." The manufacturing process also reportedly employs other high-volume production processes such as injection molding. Each PEM fuel-cell stack comprises hundreds of these identical plates sandwiched between polymeric membranes.
The Mark 900 fuel-cell stack puts out about 80 kW with hydrogen fuel and 75 kW fueled with methanol reformate. "It's designed to fit within the OEMs' weight and space constraints, generally to fit under the floorboards," said Lancaster. "In addition, it is cold-start capable. We consider it the basis for our future commercial architecture, with ongoing refinement and improvement. Right now, we're testing its durability and reliability by putting it into real vehicles."
"In the Mark 900 stack, Ballard went for manufacturability and production-oriented design," noted Staley. "We expect that there'll be a couple more iterations in the design before we go into production. Of course, as OEMs, we're constantly pressing for higher power density and lower weight, higher production volumes, and lower cost."
According to Lancaster, Ballard's fuel-cell introduction strategy goes as follows: "In late 2001, we plan to start selling our first product, a portable power generator through Coleman Powermate. By 2002, we'll be producing fuel cells for transit buses. DaimlerChrysler already has gotten 33 orders. In the 2002-2003 period, stationary units for uninterrupted power supply and backup power will be introduced. By late 2003, we'll be ready for automotive applications."
Fueling and infrastructure issues
A fundamental problem with fuel-cell technology is fuel selection and storage - supplying enough hydrogen to the stacks is still a struggle. All three of the principal fuels that carmakers are considering - hydrogen, methanol, and gasoline - pose serious challenges. Though using hydrogen gas is the approach favored by environmentalists because it is the cleanest, the elemental gas takes up significant space. With "direct hydrogen" fueling, vehicles carry pressure vessels filled with this highly flammable gas. Hydrogen can also be stored as a liquid, but it must be kept at cryogenic temperatures, adding weight, complexity, and even greater safety issues than compressed hydrogen techniques.
Either way, hydrogen fueling presents problems for engineers. The two most promising alternatives at the moment appear to be miniature onboard chemical factories called reformers (or fuel cells themselves) that extract hydrogen from methanol, gasoline, or other hydrocarbon fuel. Unfortunately, a reformer adds more weight and technical complexity to a car, while in-situ reforming technology is still far off. What's more, "reformed" hydrogen is not as pure a fuel as fuel cells would like to use, and isn't likely to deliver the same performance as uncontaminated hydrogen gas.
Experts indicate that the choice will likely come down to a question of the fueling distribution infrastructure, which is going to entail a tremendous cost burden no matter how it's accomplished. "I don't think that the fuel stack is going to be the limiting factor; the real pacing item is the fuel infrastructure," said McCormick. Christian Mohrdieck, Fuel-cell Vehicles Program Manager at DaimlerChrysler's Liberty and Technical Affairs, agreed. "The main issue with fuel cells is not the fuel cell, but the fuels themselves," he said.
