The temperature of the clean air to cool the sampled exhaust gas was controlled at 25℃.
As the aim of this experiment is to discuss about the measuring procedure, the test engine was operated at a constant load (50% load).
3.2 Experimental Result
Fig.7 shows the measured PM data (g/kWh) with a wide range of the filter loading keeping the dilution ratio at 8.5. According to the EPA recommendation, the filter loading should be more than 5.3 mg, that means more than 5.3 mg of DM should be caught on the primary filter (70 mm dia.). However according to the ISO recommendation, 1.3 mg is enough. In Fig.7, the deviation of PM data is much larger in the range of filter loading less than 4 mg, compared with the range over 4 mg.
Fig.8 shows the change in the measured PM data with the dilution ratio. Seeing this figure, it is possible that the PM data with the smaller dilution ratio than 15 show higher value than the true.
The PM emission was measured again selecting the dilution ratio of 15.5. Fig.9 shows the result. The deviation of PM data in this figure is much smaller than the result with the dilution ratio of 8.5 in Fig.7. It is considered that the filter loading of 2 mg is enough for the accurate measurement if the dilution ratio is more than 15. It leads to save the measuring time on board.
4. HORSEPOWER MEASUREMENT ON BOARD
Additional experiments were carried out to examine the accuracy of engine load. Measuring the accurate engine horsepower on board is as important as measuring the accurate PM emission itself, because the PM emission should be evaluated with the unit of g/kWh. ISO regulation prescribes that the permissible deviation of engine power should be within 2%.
For that reason, the accuracy of the measured horsepower by the shaft torque meter in the ship was confirmed. Fig.10 shows the measured SFC (specific fuel consumption, g/kWh) of the experimental ship. The fuel consumption was measured by a flow meter whose accuracy is ±0.5%. So, if the measured horsepower is correct, SFC g/kWh is also correct and vice versa. The test record, which is the data taken on the test bench before installing the engine in the ship, are also plotted in Fig.10. As the gap between the SFC data measured on board and the test record is little, it is considered that the horsepower of this engine is measured correctly.
