Search Results

As the engines of today decrease in displacement with unchanged power output, focus of today's research is on transient response. The trend of today is to use a turbocharger with high boost level. For SI-engines a regular WG-turbocharger has been used, but in the future, when the boost level increases together with higher demand on the transient response, a Variable Nozzle Turbine (VNT) will be used together with Variable Valve Timing (VVT). As the degree of freedom increases, the control strategies during a transient load step will be more difficult to develop. A 1D simulation experiment has been conducted in GT-Power where the transient simulation was “frozen” at certain time steps. The data from these time steps was put in a stationary simulation and the excessive energy was then bled off to obtain the same conditions for the engine in the stationary simulation as if the engine where in the middle of the transient.

The paper describes a measurement procedure to measure complete 2-stage turbo systems (TST - Two Stage Turbo) in a turbocharger test rig. The possibility to measure the complete range of operation of the series-sequential mode of the 2-stage system is proven. Also the performance is broken down for the two turbochargers individually; both when operating as a part of the TST system and measured as completely separated individuals. The maps measured individually were then used to calculate a composed TST-map as if there would be no losses between the two stages, and this composed map was compared to the measured TST-map. The difference between these two maps is interpreted as interstage losses when packaging the two turbochargers as closely as is necessary to fit in the car. The breakdown showed that the entire difference is not solely due to losses of total pressure interstage.

In this paper, the performance of a radial turbine working under pulsatile flow conditions is computed with two different modeling approaches, time resolved 1-dimensional (1-D) and 3-dimensional (3-D) CFD. The 1-D modeling approach is based on measured turbine maps which are used to compute the mass flow rate and work output from the turbine for a given expansion ratio and temperature at the inlet. The map is measured under non-pulsatile flow conditions, and in the 1-D method the turbine is treated as being a quasi-stationary flow device. In the 3-D CFD approach, a Large Eddy Simulation (LES) turbulence approach is used. The objective of LES is to explicitly compute the large scales of the turbulence while modeling the effects of the unresolved scales. Three different cases are considered, where the simplest case only consist of the turbine and the most complex case consist of an exhaust manifold and the turbine.

In this work a comparative study between 1-D and 3-D calculations has been performed on different junctions. The geometries are a 90° T-junction with an area ratio of unity and a 45° junction with an area ratio of 1.78 between the main pipe and the side branch. The latter case had an offset between the centerlines of the main and the branched pipe. The 3-D modeling framework uses the Reynolds Averaged Navier-Stokes (RANS) equations with the k-ε model both for the steady and the unsteady flow cases. The comparison is made both under steady and pulsating flow conditions. The aim has been to assess the 1-D/3-D differences in terms of measures for flow losses. For the steady flow cases it is shown that there is a large difference between the 1-D and 3-D computed losses for both junction geometries. The differences are largest in the junction and right downstream of it.

A standardized measurement setup and definition of compressor surge is yet to be established in the automotive community. As a consequence, compressor comparisons with regards to compressor operating map width is practically impossible today. This paper presents a possible solution to the described problem by presenting a suitable instrumentation and a correlation between measured values and actual NVH limits normally encountered in turbocharged vehicle applications. The work has been performed in laboratory conditions in a compressor test rig with both steady and pulsating flow as well as in a gas stand with steady flow. The test rig design, measurement setup and error sources are described. Some effects of different boundary conditions to the compressor and how they affect the measurement method are also presented.

This paper is a continuation of SAE-Paper 2005-01-0222 presented at the 2005 SAE World Congress, denoted Part 1 in this text. In Part 1 three turbines were selected from calculations and three manifolds with different geometries were designed. This paper, Part 2, covers the results from engine-simulations and measurements on these nine different combinations of turbines and manifolds. It was shown that the possibility of maintaining isentropic power was the most important property, overshadowing any differences in turbine efficiency. The isentropic power was inversely dependent on both manifold volume and turbine throat area. The GT-Power models of all nine setups were calibrated against the measured data. The need for efficiency and massflow multipliers is described. The efficiency multiplier depended on mass flow through the turbine, with a distinct minimum value (0.7-0.8) around 0.03 kg/s and higher around that.

In this study, the transient response of a turbocharged engine is minimized by using 1D engine simulation. A method is developed where sequential fractional factorial designs are utilized to first determine the relative importance between a wide range of design parameters and then successively come closer to an optimal setup for minimized transient response. This methodology is first developed and applied to a 2-liter standard production car engine and then it is used for optimization of the 2004 KTH Racing Formula Student engine. The race engine was first designed entirely by using steady-state 1D simulations. These simulations are described and the design explained. The philosophy behind turbocharger selection, manifold tuning etc. is described in detail. After the initial design the engine was optimized for transient response. Measurement data from steady state measurements on the dynamometer is showed for verification as well as pressure data logged in the car on the racetrack.

The paper treats pulsating flow through the turbine of SI-engine turbochargers. In engine design, 1D engine-simulations are very convenient tools for optimization and concept studies. However, they have drawbacks in certain areas. The accuracy, when predicting turbocharger turbine power, is lower than desired. The reason for that is a lack of knowledge about the phenomenon of pulsating flow through the turbine. The background to the problem is described in the paper. This investigation aims at learning more about this unsteady, pulsating flow, on the engine. The method used is to do large parameter changes to several parameters in turbine and manifold designs such as A/R and trim in the turbine and also volume and length of the exhaust manifold. For selection of A/R and trim, as well as an aid in the analysis of measured data, the meanline turbine design software Rital from Concepts NREC [1] was used. Three different turbines were investigated, all with the same mass flow capacity.

Traditionally, heat losses from the turbine are neglected in turbomatching calculations as well as in engine simulations [1]. On the SI-engine, with it's high exhaust temperatures, this assumption will lead to errors in the calculations. Significant amounts of heat are dissipated from the turbine through several mechanisms. This paper contains measurements of the different heat loss mechanisms from the turbine during full load operation on a 4-cylinder SI-engine. The largest loss components are convective and radiative. The heat losses to cooling water and lubrication oil were approximately 3-5% of the total heat loss from the turbine. In addition to heat losses to the surroundings, heat flux is also present internally in the turbocharger. Heat flux from the turbine to the compressor can deteriorate the efficiency of the compressor.

In this paper results from two different 1D engine simulation software (GT-Power and Virtual Engines) are compared to measured data. A list of simulation output properties is suggested, the order of which should be followed when calibrating the model, to make the simulation work as accurate and simple as possible. The model buildup is described in detail and the focus is on the turbocharger. For submodels aside from the turbocharger, the simplest possible options were selected, such as Wiebe model for the combustion, imposed measured wall temperatures for intercooler and manifolds. For the turbocharger model the paper describes in detail the adjustments and hands-on work that were necessary to achieve results close to measurements. Both transient simulations of the engine as well as simulations of thermocouple output are covered.

In this investigation the influence on knock from the residual gas in the cylinder is investigated. Gas was sampled from inside the cylinder prior to ignition, the Residual Gas Fraction, RGF, was determined and the Knock Intensity, KI, was measured. By altering the exhaust backpressure the RGF was changed. By measuring the knock intensity for different RGF the influence on knock from residual gas was investigated. It is shown that with increased residual gas fraction the knock propensity of the engine is increased, and subsequently, decreased RGF gives lower knock intensity. This is showed by the fact that, with maintained knock intensity at 30 kPa, the ignition timing can be advanced as much as 5 Crank Angle Degrees, CAD, if the RGF is reduced with 15%.