Cavitation is a phenomenon affected considerably by the underlying pressure waves that occur on similar time and length scales as the bubble dynamics. Thus appropriate representation of wave dynamics within numerical frameworks is of paramount importance for the prediction of the phase change process in the nozzle as well as the subsequent spray formation. In this paper we focus on investigating the sensitivity of the wave dynamics within a compressible Large Eddy Simulation framework with regards to downstream geometry and boundary representation. Diesel was used as working fluid and was injected at various pressures through a micro-channel. Results in terms of vapour fraction, velocity and pressure are compared with the experimental data of Winklhofer [30,31]. The downstream domain length and reflectivity properties are shown to exert a significant effect on in-nozzle processes.

This paper provides new physical insight into the coupling between flow dynamics and cavitation bubble cloud behaviour at conditions relevant to both cavitation inception and the more complex phenomenon of flow “choking” using a multiphase compressible framework. Understanding the cavitation bubble cloud process and the parameters that determine its break-off frequency is important for control of phenomena such as structure vibration and erosion. Initially, the role of the pressure waves in the flow development is investigated. We highlight the differences between “physical” and “artificial” numerical waves by comparing cases with different boundary and differencing schemes. We analyse in detail the prediction of the coupling of flow and cavitation dynamics in a micro-channel 20 m high containing Diesel at pressure differences 7 MPa and 8.5 MPa, corresponding to cavitation inception and "choking" conditions respectively. The results have a very good agreement with experimental data and demonstrate that pressure wave dynamics, rather than the “re-entrant jet dynamics” suggested by previous studies, determine the characteristics of the bubble cloud dynamics under “choking” conditions.

No abstract

The measurement of the rate of fuel injection using a constant volume, fluid filled chamber and measuring the pressure change as a function of time due to the injected fluid (the so called "Zeuch" method) is an industry standard due to its simple theoretical underpinnings. Such a measurement device is useful to determine key timing and quantity parameters for injection system improvements to meet the evolving requirements of emissions, power and economy. This study aims to further the understanding of the nature of cavitation which could occur in the near nozzle region under these specific conditions of liquid into liquid injection using high pressure diesel injectors for heavy duty engines. The motivation for this work is to better understand the temporal signature of the pressure signals that arise in a typical injection cycle.A preliminary CFD study was performed, using OpenFOAM, with a transient (Large Eddy Simulation -LES), multiphase solver using the homogenous equilibrium model for the compressibility of the liquid/ vapour. The nozzle body was modelled for simplicity without the nozzle needle using a nozzle hole of 200μm diameter and the body pressurised to values typical for common rail engines. Temperature effects were neglected and the wall condition assumed to be adiabatic. The chamber initial static pressure was varied between 10 and 50 bar to reflect typical testing conditions.Results indicate that vapour formation could occur in areas 10-30mm distant from the nozzle itself. The cavitation was initiated around 100 μs after the jet had started for low ΔP cases and followed the development period required for the formation of vortices associated with the vortex roll up of this jet. These vortices had localised sites, in their core region, below the vapour pressure and were convected downstream of their initial formation location. It was also found that vapour formation could occur at chamber static pressures up to 50 bar (the highest tested) due to cavitation in the shear layer and this vortex effect. The pressure signal received at the chamber would therefore be more difficult to interpret with additional error components. IntroductionFuel injectors are under constant scrutiny as one of the prime factors in governing the overall efficiency and emissions output from heavy duty diesel engines. Manufacturers of fuel systems are therefore pursuing increasingly challenging targets for the accurate quantification of the instantaneous mass flow rate and total fuel quantity delivered. The characteristics of the fuel delivered such as the time rate of change of mass flow for a given pressure differential (i.e. rail pressure) are as important as the total quantity delivered for a given "shot". The rate of change of fuel mass flow is an important consideration to engine developers as it allows more complex combustion strategies that can be tailored for a given engine configuration and application [1,2]. This 'rate shaping' is a useful tool in order to gain the incremental improvements of combustion that...

At the request of the L3 collaboration (one of the LEP experiments) a new facility was built close to the Electron Positron Accumulator (EPA) to provide a single electron per beam pulse at an energy which can be selected between 180 MeV and 500 MeV. The intensity of the e3ected electron beam is reduced by changing the gun parameters and by adxusting the aperture of the slits in the linac. The energy is defined by letting the beam do 1 1/4 turn in the EPA. The enery spread is limited to 0 . 5 l o -' by reducing the slits aperture in the LILlEPA transfer line. A new transfer line was built to transport the beam from the EPA ejection septum to the detector test stand. The facility was then used to test the response of the L3 electromagnetic calorimeter corrposed of Bismuth-Germanium-Oxide (BGO) crystals in the 100 MeV range.

A steady incompressible three-dimensional viscous flow analysis has been conducted for the Space Shuttle External Tankforbiter propellant feed line disconnect flapper valves with upstream elbows. The full Navier-Stokes code,

No abstract

No abstract