Since air pressure and temperature (Pc=4.6MPa, Tc=420℃) in the RCM combustion chamber are lower than those of a highly turbocharged diesel engine at high load, the test condition represents low load. This has two effects. First, because of the low air pressure air density is also low and therefore spray velocity is higher. Second, with lower air temperature ignition delay is longer. Both effects increase the possibility of liquid parts to impinge onto the piston.
Obviously, in a real engine most of the liquid part evaporates and burns after ignition because of the high flame temperature. But the liquid fuel on the cooler piston surface is considered to be the cause of the high PM emission. This corresponds with the result of higher PM emission at low load as shown in Figure 2.
3.3 Combustion Characteristics of Bunker Fuel Spray
Figure 4 shows photographs of the flames taken through a transparent piston of the visual engine. The seven cases A to G shown in Figure 4 are discussed in detail in the following five sections. In cases A and B, GO and BFO spray combustion are compared under the same conditions. In the other cases BFO spray combustion is examined under a higher air pressure than in case A and B. The influence of air temperature is examined in cases C and D, while in cases E to G various measures to improve spray combustion are tested.
3.3.1 Comparison between GO and BFO (cases A and B)
The combustion of two fuels is compared according to cases A and B of Figure 4. As expected BFO has a longer ignition delay (6°after injection start) than GO (5°). The photographs on the left show the flames just after the ignition. While in case A all single sprays burn well, in case B some sprays show a poor combustion. One cause is surely the poor ignition characteristics of BFO. The other photographs (center and right) show the flames at 5°to 10°ATDC. Here the difference in soot formation at the impingement point can be seen clearly.
When watching the moving picture, the continuous formation and disappearance of soot at the point of impingement can be observed. This can be explained as follows. When the fuel-rich zone of the spray impinges onto the cool piston surface soot is formed. This soot is at once pushed away by the following spray part and when reaching the air-rich zone it re-burns.
A comparison between case A and B shows clearly that in case B (BFO) more soot is formed at the point of impingement. That means the distribution of fuel and air in the BFO spray is less homogeneous than in the GO spray. This is caused by the poor evaporation characteristics of the residual portion in the BFO as already mentioned in section 3.2.
3.3.2 Influence of air temperature on BFO combustion (cases C and D)
The poor evaporation characteristics can also be observed in Figure 4, cases C and D. These cases show the influence of combustion chamber air temperature at the time of ignition. No difference in soot formation can be observed at 8°ATDC. However, in case D at 26°ATDC, after the end of injection white flames can be seen at each point of impingement. The reason for this can be explained as follows. Since the air temperature at the time of ignition is low (400℃ to 420℃) the evaporation characteristics is especially poor and ignition delay is also longer. Therefore, quite much liquid impinges on and clings to the piston and continues to burn after the end of injection. This situation might occur also at low load in a real engine.
3.3.3 Influence of air swirl on BFO combustion (cases C and E)
Figure 4E shows the case when air swirl is applied (swirl ratio 2.5) by means of inclined scavenging ports in the cylinder liner. (In all other cases of Figure 4 swirl is not applied.) The other conditions are the same as in case C.
At 8｡?TDC the combustion looks more active in case E than in case C.
However, at 18｡?TDC, at the end of injection more soot remains at the point of impingement in case E. Further, at 26｡?TDC much more stripe shaped clouds of soot are visible in case E. Such soot remaining after the main combustion is less likely to re-burn and therefore more likely to be emitted as exhaust smoke.
According to an earlier research [2], over-swirl state (a sudden increase of exhaust smoke when too strong swirl is applied) occurs with smaller swirl intensity when using BFO than when using GO.
Based on these two observations, the following situation is considered. As explained above, some liquid parts of the BFO spray cling to the piston surface. When now swirl is applied the flame from the neighboring spray can easily overlap this liquid. This is not the case when using GO, because then no liquid clings onto the piston. And this might be the reason why BFO is more easy to fall into over-swirl state.
However, if swirl intensity is not so strong that over-swirl occurs, then surely BFO spray combustion is improved [2]. So, selecting the correct swirl intensity is very important when mainly BFO is to be used.
