1 i

0 200 400 600 800 1000 Current Density ASF

0 200 400 600 800 1000 Current Density ASF

Figure 1 - Baseline Hydrogen Air Performance on a finite difference model developed jointly by Analytic Power and the University of South Carolina. This flow field design has dramatically improved reactant gas distribution and cell stack performance. The bipolar plates are composite units fabricated from nitrided titanium separator plates, carbon paper flow fields and plastic ccll frames with O-ring seals. The stacks arc water coolcd.

Figure 1 shows the hydrogen/air performance of the USN and USMC stacks tested on reformed hydrocarbons. The upper line is the 10-cell USMC stack performance at 40 psig and 180°F. It is the average of all cells in the stack. Some cells reached 0.6 volts per cell at 1000 ASF. The next line in Figure 1 corresponds to a 25-cell stack built for the USN program. The performance level of this stack is about three times better than previous builds which did not have the new advanced flow fields. The 25-cell stack was operated at about 35 psig and 160°F. The lower temperature accounts for the difference in performance between USN and USMC stacks. The triangular data points were obtained from a 20 cell-USN stack operating between 130°F and 150°F at about 25 psig. The figure also shows a data point which represents the average voltage at 200 ASF when 50 PPM carbon monoxide was admitted into the cell. The average performance at 200 ASF dropped about 200 mv when the stack was exposed to carbon monoxide. As we will show, performance was completely restored by titrating the stack anode gas with air.

Figure 2 shows a 20-ceIl USN substack test on 50 PPM carbon monoxide contaminated gas. The stack was also tested at 100 PPM and 150 PPM levels. Stacks are now being endurance tested on simulated reformate. The figure shows .that the average cell voltage drops about 200 millivolts with the cell stack on a 200 ASF load. Figure 3 shows that carbon monoxide admission starts at about record number 490, near the point where performance starts to drop. The 600 records accumulated in the test correspond to about 5.25 hours of testing. Each record number represents about 30 seconds. At record number 565 we started titrating air into the anode. Figure 2 shows that performance recovers completely when air is present in the anode cavities.

Autothermal Reforming Fuel Processor

Figure 4 shows a schematic of the USN fuel processor test stand. All the fuel processor components are mounted in the test stand. Analytic Power developed a simulation program which has proven useful in correlating the experimental data of the fuel processor.

Current Density & Volts vs Record No Analytic Power Corporation


0 0

Post a comment