OCR Text |
Show higher maintenance cost; added capital requirements for equipment and tankage; and some derating of the machines. Nevertheless, overall economics for crude and heavy oil operations were found favorable [5], and as shown in Figure 5, the loss of power output remained moderate - only about five percent - for a span of 400 hours operation sustained without cleaning the machines. Fuel cost margins, price fluctuations, and site specific factors may influence the level of actual savings, but the technical feasibility of the scheme has been strongly established in long duration commercial operations. V. FULL SCALE COMBUSTION TESTS WITH ALTERNATIVE FUELS To extend the long-range commercial experience of the BBC turbines to lower grade fuels of the future, a full scale experimental program was undertaken at BBC's Munchenstein test site in Switzerland. Fuel selection was primarily made to study the effects of low hydrogen and high fuel bound nitrogen (FBN) content. By these tests, answers were sought as to what extent partially refined lower grade synthetic fuels can be accommodated in the BBC gas turbines, with obvious implications to future synfuel refining costs. Six hundred barrels of each hydrotreated residual shale oil [6], EDS Middistillate from coal liquefaction (produced by Exxon Research Company) [7,8] and heavy pyrolysis residue (PFO) from steam cracking were combusted in the 35 MW capacity type 9 gas turbine at varying loads. Diesel fuel runs were also made simultaneously for base case data and comparison. Selected properties of the fuels are shown in Table 4. A complete description of the test site, the origin of the fuel and details of the test results are given elsewhere [10]. The project was supported by BBC, Stone & Webster and Exxon Research organizations and contribution to the program was also made by Shell Internationale Chemie, Holland. Results with regard to operation showed: • Ignition of all test fuels took place instantly and proceeded similarly to diesel oil under cold and warm starting conditions. 1.4.5 |