Fuel cells challenge diesels

A breakthrough in fuel cell control technology is set to make them odds on favourites to be the power source of choice for mobile phone base stations in Africa and other locations without a mains power supply. Efficiencies are 75% under no load and 55% on full load, with 90% on the horizon. Maintenance costs are much lower than those of conventional diesel generator sets and the new units should run for 100,000h without maintenance – more than offsetting the higher capital costs. Ron Hodkinson, managing director of Watford-based Fuel Cell Control, set out the pros and cons of fuel cells to Eureka, while sitting in front of a working 2.5kW prototype. Continuously running diesel generating sets tend to need an oil change every 250h, with a complete overhaul every 10,000h. Diesel generators are also noisy and often polluting, and are less than welcome neighbours in town or village centres. Another factor that has to be taken into account is that diesel oil tends to get stolen! On the other hand, fuel cells, despite their great promise as one of the major energy sources of the future, are costly to make with most designs require substantial amounts of very expensive platinum for their catalyst. Despite much hype, the typical working lives of many of those under development is only about 500h. However, they are both quiet and highly efficient. Fuel Cell Control has, therefore, devoted itself to establishing why they fail and to the development of a control system which avoids damaging conditions. One cause of failure of Proton Exchange Membrane (PEM) fuel cells is for the humidity to go outside the optimum 80%±5%. Another is by working at plate current densities of 1 to 2A/cm2, as in most of the fuel cells being researched for automotive use. The Fuel Cell Control cells work at 100mA/cm2. Fuel cells can also fail through the use of impure fuel. Most of the successful cells developed so far combine hydrogen with oxygen from the air to make water and electricity. Commercially available hydrogen tends to be obtained as a by-product of sodium hydroxide manufacture from sodium chloride in Castner-Kellner cells. The cathode electrode used is mercury, traces of which poison catalysts, not to mention people. In the long term, fuel cell hydrogen will probably be obtained by electrolysing water. At present, the best source of hydrogen is that obtained as a by-product of oil refining, the bulk of which is normally used to produce more petrol, with a small amount sold to hydrogenate vegetable oils for margarine manufacture. Gas fed to the cell stacks also has to go through sub-micron filters to remove particulates. For the African project, the hydrogen will be supplied as balls of sodium hydride enclosed in polyethylene. To extract the hydrogen, the balls are inserted into special machines where they are cut open and reacted with water. The cells the company has been working with are alkaline cells, based on potassium hydroxide solution electrolyte, as used in the original NASA developments for the space programme. One of the attractions of the electrolyte is that it does not freeze until –40°C. Another attraction is that alkaline fuel cells can be made to work with silver or rare earth catalysts. Hodkinson adds: "Fuel cells are extremely complicated things; something which needs to be taken into account if they are ever going to be reliable. A lot of people out there think they know all about fuel cells but don't. This is a big part of the problem." Output voltage varies from 33V at no load to 22V at full load – insufficient to charge a storage battery working at 27.6V. The team has therefore turned to another development from the Apollo Space Program, the Cuq converter. Originally developed by Dr Slobodan Cuk at Caltech, owner of the TeslaCo power conversion company, one of the virtues of the Cuk converter is that it is able to buck and boost output voltage with only a single transistor and without any high frequency ripple on the fuel cell input. The fuel cell stack has four plates in each module, and the control system monitors every plate. If the output voltage falls to 0.45V, indicating a blockage, a purge cycle initiates. In this cycle plates that are not blocked discharge into the plate that is blocked. A pressure rise from 36 to 70mbar for more than 5s triggers the release of a pulse of hydrogen, the aim being that it carries away any particulates with it. The system is managed by a Mitsubishi PLC. One of the FX1N series, designed especially for this task. The Fuel Cell Control units generate power for telephone exchanges, and create pure water and waste heat, as by-products, which are used to power the exchange air conditioning. The fuel cell has to combust the hydrogen with oxygen from the air, and sodium hydroxide is used to scrub out carbon dioxide so it cannot harm the potassium hydroxide electrolyte. The team is currently researching the magnetic enrichment of oxygen from air. The first ten 2.5kW units are being supplied to the USA at a price of £27,000 each. Efficiencies demonstrated to Eureka were 75% at low load and 55% at full load. The intention is to produce 3,000 10kW units per year. Prices are expected to be $45,000 each, falling to $30,000 each. The main customer is Mobile Systems International, which, having been established in Africa only three years ago, now has 10 million lines with more being added. Mobile lines are particularly attractive in Africa since fixed lines are highly unreliable and insecure. Ultimately, it is hoped to get the cost of the fuel cells down to $200/kW with $30/kW catalyst cost and output efficiencies as high as 90%. However, as Hodkinson adds: "Fuel cells today are about where aeroplanes were in 1907. Fuel cells for cars will probably be the last market they will conquer." Fuel Cell Control Ron Hodkinson Fuel cell control system allows them to run for 100,000h without maintenance Efficiencies are 75% on no load and 55% on full load Capital costs are presently high, but the cells are being manufactured commercially with considerable opportunities for cost reductions are volumes increase