The design and dimensions of the boat in this study are shown in Fig. 8. The hull size was chosen in order to be able to float six 9-V batteries. The water channel height is equal to the magnet pole width plus the thickness of the electrodes. The water channel width was determined through the optimization detailed above. The electrodes have the same width as the channel and are made of brass scraps. Copper can also be used although it suffers vigorous erosion from the chlorine ions during operation. Heavy-gauge wire connects each electrode to a pole of the battery stack. The batteries are connected in series and laid in the hull. The wire must be sufficiently thick to be able to support 1- to 3-A currents. The hull is made of styrene plastic bonded with CA (cyanoacrylate) glue. Both are available at any hobby store and can be purchased for a total of less than $10. A pair of rare-earth (Nd-Fe-B) magnets was bought online through eBay for $17.

Figure 8.

The boat along with the magnet stack was run in a small tub of salt water. The salt concentration was made as high as possible in order to reduce the resistance and maximize the current. The tub was 20 × 30 cm long and contained only sufficient water to float the boat (about 3.5 l). This minimized the amount of salt needed. Two hundred or more grams of salt were added to the water. The proper orientation of the batteries/electrodes was verified by checking that the water was ejected from the stern of the boat. A picture is shown in Fig. 9. An ammeter was connected in series and 0.8 A were measured during operation. The boat attained a speed of 0.5 to 1.0 cm/s. When a laboratory power source was connected to the boat in place of the batteries, 7 A flowed through the system and the water velocity was measured as 10 to 15 cm/s. The boat did not move, however, due to being tethered by the wires. A second boat was constructed with identical dimensions but with a water channel width of 7 mm. Under the same operating conditions, its velocity was either very small or undetectable. This would tend to support the analysis, although this boat may also have had problems with hydrogen buildup (discussed below) on the electrodes, which could result in greatly diminished current flow.

Figure 9.