Figures 8(a) and (b) show, as an example, time histories of gas temperatures "Tb", "Tc", "Tu" and the NO-formation rate "dNO/dθ" calculated by the two-zone model. It is seen in these figures that the NO-formation rate decreases by reducing orifice size due to a slight decrease in the maximum heat release rate. Figures 9(a) and (b) also show time histories of the soot-formation rate (dS/dθ)f, the soot-oxidation rate (dS/dθ)b and the cumulative soot-formation "∫dSn". With respect to the heat release curve shown in the top of Figs. 9(a) and (b), the approximation was conducted by means of the least square method using the Wiebe's function for the diffusion combustion period between the crank angle of the maximum premixed heat release rate and that of the latter zero-cross point of the apparent heat release curve. Furthermore, the final value at the end of combustion on the cumulative soot formation history (mg/cycle/cylinder) was identical to the measured smoke value.
Using the values of the calculated soot size and the measured soot, the number of soot particles is calculated by the following equation (4).
S = ρs(p/6)ds3Ns (4)
Figure 10 shows the calculated soot size, and Fig.11 shows the calculated number of soot particles under various engine operation conditions. The soot size indicates, generally speaking, a tendency to decrease with decrease in orifice size except for the cases of 0.18 mm orifice at the low load.
According to the spray model proposed by Tanazawa & Toyoda [12], the mean droplet size is proportional to orifice size and is reciprocal to the injected velocity. If the orifice size is decreased without changing the number of orifice, injection pressure increases due to a decrease in the total orifice area. In the present experiment, the mean droplet size is expected to reduce by 45% due to the above mentioned two effects, however, the calculated soot size is decreased as small as 15% as shown in Fig.9.
It is seen in Fig.9 that the soot size is affected largely by the engine load and the injection timing compared with the effect of orifice size. And, in addition, the soot size is large in a few cases with the orifice size smaller than 0.20 mm especially at the low load, which might be caused by that the heat release rate curve varied significantly in these cases compared with the other cases. If combustion time history were varied monotonously in these cases, soot size were monotonously decreased with the decrease in orifice size. These phenomena suggest that a change in combustion, for instance, a change in ignition delay due to decrease in orifice size, which is the secondary factor, gives a larger effect on soot size than the effect of orifice size on soot size which is the primary factor.
On the other hand, as shown in Fig.10, the number of soot particles varies clearly from a decrease to an increase at the orifice size of 0.22 mm as the orifice size decreases, which tendency is similar to that of the measured soot quantity, not shown here, and it seems to be expected that there is a good correlation between the two kinds of variation of fuel consumption and the number of soot particles.
