EU1 EU2 EU3 EU4 5 4 120 1985 1990 1995 2000 2005 2010
Year Fig. 1: Evolution of Spray hole number and spray hole diameter
Consequently, the idea arises, to combine small spray hole diameters with a sufficient hydraulic flow, which leads to a significant increase of spray hole numbers. But not only overall hydraulic flow rate is important. If the impulse of a single separate spray is not sufficient to penetrate the complete piston bowl under high load in-cylinder conditions, poor air utilization and subsequent soot emissions can be a consequence. One approach to combine good mixture formation due to small nozzle hole diameters with adequate spray impulse is to use multi layer nozzles with nozzle holes arranged in groups. In the course of a public financed research program the potential of grouped multi layer nozzles shall be analyzed with the help of fundamental spray investigations in high pressure vessels and detailed 3D CFD spray simulations. As a first step to gain a detailed understanding of the underlying mechanisms of multilayer nozzles, the injection rate was determined separately for layer I and II of the nozzle. These investigations were accompanied by 3D CFD in nozzle flow simulations to understand the observed effects. Optical Schlieren investigations were used to validate in nozzle flow simulations and injection rate measurements.
Within the overall research project numerous multi layer nozzles with grouped nozzle holes were investigated. The chosen approach for layer assigned injection rate measurement and accompanying in nozzle flow simulation and optical validation shall be presented at the example of a multi layer nozzle with parallel nozzle holes. The main parameters of the investigated nozzle are shown in Table 1.
The realized injection rate is one of the key input parameters for spray and overall combustion simulation. For the investigated multi layer nozzles, where the nozzle holes are arranged on an upper and a lower row, the overall injection rate is not sufficient to describe the nozzle behavior as it cannot be assumed that the injection rate is comparable for holes of the upper and lower row. Especially for nozzle concepts with varying hole diameters for upper and lower row and during the needle opening phase a significant difference in injection rate can be expected. In order to analyze the impact of multi layer nozzle concepts on spray propagation, mixture preparation and finally combustion and emission formation it is necessary to measure and analyze these differences in injection rates.
In order to minimize the necessary machining effort and possible impact of any machining on the nozzle behavior, a high precision measurement tool was developed, adapted to the different nozzle geometries and calibrated. The nozzles are prepared with a 0.2 mm wide groove, which is machined on the entire circumference of the nozzle tip. The groove is located between the upper and lower nozzle layer and is a defined surface for sealing which also ensures that the injector is centered in the layer separator tool. Figure 2 shows the developed injection rate measurement tool.
The lower chamber of the injection rate measurement tool is connected to an injection rate analyzer with high resolution data logging capabilities and a precise fuel mass measurement. A comparison of injected fuel mass via upper and lower nozzle layer is shown in Fig.3.
The separate injection quantity measurements for the upper and lower layer show that especially for low rail pressure levels, e.g. 800 bar and injection durations with energizing times less than 400 micro seconds the fuel quantity that is injected via the upper row is up to 25% higher in comparison to the fuel quantity that leaves the lower layer. A measured injection rate profile for a typical pilot injection as shown in Fig. 4 with an energizing time of 300 micro seconds and a rail pressure of 800 bar confirms the results of the fuel mass measurements.
In order to better understand the experimental results a 3D CFD simulation model was created with StarCD.
To reduce calculation time and meshing effort a sector mesh for one pair of nozzle holes was set up. Fig. 5 shows the realized mesh.
Fig. 5: Sector mesh for 3D CFD Simulation
Model Steady state flow calculations were carried out for three different needle lift positions: 1) almost closed / slightly open 2) mid-open position 3) completely open As especially for low rail pressures and short energizing times, differences in injection rate and subsequent injected fuel mass could be identified for upper and lower row, high emphasis was put on simulations with a mesh corresponding to needle lift position 1). Figures 6 and 7 show the obtained velocity vectors for the simulated flow fields in the area of the upper and lower orifice. The low needle lift and subsequently reduced cross section, available for the flow from the upper part of the injector to the lower orifice in the micro sac area has a significant throttling effect. In spite of the significantly improved conditions for the flow at the immediate orifice inlet in the micro sac and the worse conditions at the valve-covered-orifice (VCO) in the upper row, the throttling results in a higher injected fuel mass via the upper orifice under these boundary conditions.
Fig. 6: Velocity Vectors for almost closed needle position in upper orifice (VCO-type),(800 bar Rail Pressure) Fig. 7: Velocity Vectors for almost closed needle position in lower orifice (Micro-Sac-type),
(800 bar Rail Pressure)