The STD Model I has a sac chamber with 8.5 mm diameter and 13.8 mm height. The STD Model III has the same geometry of the sac chamber as the STD Model I, but the discharge hole location in the sac chamber of the STD Model III is 7.1 mm higher than that of the STD Model I. The flow path from the needle seat to the discharge hole through the sac chamber is supposedly quite different between the STD Model I and III. The STD Model II has the same geometry of the sac chamber as the STD Model I and II but has in between discharge hole location of STD Model I and III. The height of the sac chamber of the Mini Sac Model is one half of the standard model. In the VCO Model, the needle seat directly covers the inlets of the discharge holes.
Figure 4 shows the change of the cross sectional area of the flow passage in the nozzle with a needle lift. The minimum cross sectional area changes with respect to the needle lilt. In a needle lift smaller than hn = 0.47 mm, the needle seat has a smaller cross sectional area than the discharge holes, whereas in a needle lilt larger than hn = 0.47 mm, the smaller cross sectional area changes from the needle seat to the discharge hole.
The change of the needle lift during the needle valve opening was measured from the images taken by the high-speed video camera. The result is shown in Fig.5. When the needle valve lever is released, the needle valve lifts up by the water pressure in the upstream chamber of the nozzle. In the initial stage of the injection until t = -20 ms, the needle lilts up slowly. From t = -20 ms to 62 ms, the needle lifts up more rapidly and at a constant rate. When the needle valve reaches the full lift of hn = 3.5 mm, the needle valve bounces and then maintains the full lift after t = 70 ms.
2.3 Experimental Conditions
The similarity of the flow in the injection nozzle was established by setting a Reynolds number of the flow in the discharge hole of the model nozzle same as that of an actual nozzle. The Reynolds was number defined by Eq. (1).
Re = Vidh / vi (1)
where Vi: velocity in the discharge hole
dh: hole diameter
vi: kimenatic viscosity of injection liquid
Table 1 shows a comparison of nozzle dimensions and flow conditions between the actual D.I. Diesel nozzle and the model nozzle. A Reynolds number of the flow in the discharge hole is 39800 under an injection pressure of 0.2 MPa for the model nozzle, which is the same as a Reynolds number under an injection pressure of 149 MPa for the actual nozzle. The cavitation number was defined by Eq. (2).
K = (Pi - Pv) / (Pi - Pa) (2)
where Pi: pressure in the upstream chamber of the needle seat
Pv: vapor pressure of injection liquid
Pa: ambient pressure
The larger cavitation number means the tendency for the cavitation hardly to occur. The cavitation number for the model nozzle is larger than that of the actual nozzle as shown in Table 1. This leads to a conclusion that, if the cavitation is observed in the model nozzle, the cavitation definitely occurs in the actual nozzle.
Table 1 Comparison of Nozzle Dimensions and Flow Conditions between Diesel Nozzle and Model Nozzle
3. INTERNAL FLOWS OF NOZZLE AND SPRAY BEHAVIORS DURING NEEDLE VALVE OPENING
3.1 STD Model I
The flow patterns in the sac chamber for the STD Model I during the needle valve opening are shown in Fig.6. The streamlines of the tracers are visualized for a period of 2 ms of the shutter opening of the high-speed video camera. At t = 4 ms from the beginning of the needle valve opening, the stream lines in the sac chamber are relatively smooth and bend smoothly before entering the discharge hole. The main flow to the discharge hole is, as shown by the arrows in the figure, from the needle tip to the discharge hole. Around t = 12 ms when the minimum cross sectional area of the flow passage switches from the needle valve seat to the discharge holes, the flow in the sac chamber appears to be choked. From t = 34 ms to 48 ms, this choking flow disappears gradually. Around F62 ms when the needle valve reaches the full lift, the stream lines become relatively smooth again. Around t = 80 ms, cavitation bubbles are seen in the discharge hole and grows into the thin films. They enter the sac chamber and are linked to each other. These cavitation films disappear soon. The flow pattern in the sac chamber after t = 62 ms looks like a steady flow.
The spray behaviors for the STD Model I are shown in Fig.7. The spray structure for the model nozzle is different from that for the actual D.I. Diesel nozzle for lack of the similarity law of atomization. However it is considered that a qualitative analysis can be made of the effect of the internal flow of the nozzle on the spray behavior just after injection.
In the initial stage of the injection until t = 8 ms in Fig.7, the liquid column without the breakup is pushed out of the discharge hole. After t = 14 ms, a disturbed spray issues and then passes over the initially issued liquid column. The spray with a larger spray angle issued until t = 36 ms. Around t = 50 ms, the spray angle becomes smaller gradually because of the smooth flow in the sac chamber, as shown in Fig.6 after t = 48 ms. At t = 64 ms in Fig.7, the dense jet appears in the region near the discharge hole due to the bounce of the needle valve. After t = 82 ms, the spray angle fluctuates when the cavitation films flow out of the discharge hole.
3.2 STD Model III
The flow patterns in the sac chamber and the spray behaviors for the STD Model III are shown in Figs. 8 and 9. In Fig.8 of the flow patterns, at t = 4 ms from the needle valve opening, the smooth flow patterns in the sac chamber are seen entering the discharge holes from the bottom of the sac chamber, as shown by the arrows in the figure. These flow patterns are quite different from those for the STD Model I. Around t = 12 ms when the minimum cross sectional area of the flow passage switches from the needle valve seat to the discharge holes, the flow in the sac chamber appears to be choked. After t = 34 ms to 48 ms, the choking flow near the inlet of the discharge hole gradually disappears, whereas the flow in the lower region of the sac chamber is still disturbed and does not have the main flow. At t = 62 ms, most of the seat flows enter the discharge holes directly as shown by the arrows in the figure.
