To get around the problem of a nonexistent hydrogen infrastructure, a research firm examined reformer systems for fuel cells that produce hydrogen from gasoline.
Directed Technologies Inc. (DTI), Arlington, Va. (directedtechnologies.com), recently completed a multiyear study funded by the Dept. of Energy.
To compare several onboardhydrogen-system designs, DTI used Design for Manufacture and Assembly (DFMA) software from Boothroyd Dewhurst Inc., Wakefield, R.I. (DFMA.com). The software provides metric tools for costing, analyzing, and evaluating product designs at the early stages of development.
Fuel systems the company examined had several major subassemblies including fuel cell stacks and an autothermal reformer, which extracts hydrogen from gas by burning a portion of the hydrogen-bearing fuel inside a vessel filled with a catalyst.
DTI compared systems including one in which each fuel-cell stack produced maximum power at 0.6 V/cell and another generated 0.7 V/cell. The trade-off was the 0.6-V cell offered higher power density, which meant that a smaller cell produced the same power as the higher-voltage cell. The 0.7-V cell was more efficient converting hydrogen to electricity. "An important part of the study involved pinning down cost trade-offs for the different operating voltages," says project engineer Gregory Ariff.
Team members first established the steps in the chemical process, from hydrogen extraction to the creation of electricity and modeled the process in simulation software. Using the chemical-process model as a guide, the engineers created a single bill of materials for each reformer and fuelcell system, which included the reformer assembly, the fuel-cell stack, an air compressor, and filters, valves, fittings, and stainless-steel tubing.
Team members entered part costs and production volume into DFMA software, which produced a detailed estimate, including assembly times. Engineers evaluated the BOM at annual production volumes of 500 to 500,000 units, getting baseline costs for the 0.6 and 0.77-V systems. The engineers also used DFMA methodology to examine relationships between components, eliminate unnecessary parts, and rate each component on ease of orientation and assembly.
The team, which included chemical, mechanical, and aerospace engineers, then used the software to choose the most cost-effective shape-forming process for parts and to modify part features to reduce manufacturing costs.
The software also let them explore hypothetical designs. One focused on the bipolar plates that supply air and hydrogen to each cell. The plates conduct electricity out of the cell and contain tiny conduits to channel gas flow in and out of the cell. Typical plate materials were metals and graphite,
which are costly to machine. The team explored other materials and selected an injection-molded, carbon-blackfilled composite that required no machining.
Although the study showed onboard reformer systems did not meet DoE cost goals, even after DFMA redesigns, according to Ariff the findings did provide one more piece of a larger
puzzle. "We learned where to betterdirect our efforts," he says. For example, an infrastructure of smallscale stationary reformers may provide a more economical hydrogen source for cars in the near term.