Search Results

There are numerous off-road diesel engine applications. In some applications there is more focus on metrics such as initial cost, packaging and transient response and less emphasis on fuel economy. In this paper a combustion concept is presented that may be well suited to these applications. The novel combustion concept operates in two distinct operation modes: lean operation at light engine loads and stoichiometric operation at intermediate and high engine loads. One advantage to the two mode approach is the ability to simplify the aftertreatment and reduce cost. The simplified aftertreatment system utilizes a non-catalyzed diesel particulate filter (DPF) and a relatively small lean NOx trap (LNT). Under stoichiometric operation the LNT has the ability to act as a three way catalyst (TWC) for excellent control of hydrocarbons (HC), carbon monoxide (CO) and nitrogen oxides (NOx).

Increasingly stringent emissions regulations require that modern diesel aftertreatment systems must warm up and begin controlling emissions shortly after startup. While several new aftertreatment technologies have been introduced that focus on lowering the aftertreatment activation temperature, the engine system still needs to provide thermal energy to the exhaust for cold start. A study was conducted to evaluate several engine technologies that focus on improving the thermal energy that the engine system provides to the aftertreatment system while minimizing the impact on fuel economy and emissions. Studies were conducted on a modern common rail 3L diesel engine with a custom dual loop EGR system. The engine was calibrated for low engine-out NOx using various combustion strategies depending on the speed/load operating condition.

Fluid-Structure Interaction (FSI) simulation approach can be used to simulate a turbocharger. However, this predictive 3D simulation encounters the challenge of a long computational time. The impeller speed can be above 100,000 rpm, and generally a CFD solver limits the maximum movement of the impeller surface per time step. The maximum movement must be a fraction (~0.3) of the cell length, thus the time step will be very small. A Multiple Reference Frame (MRF) approach can reduce computational time by eliminating the need to regenerate the mesh at each time-step to accommodate the moving geometry. A static local reference zone encompassing the impeller is created and the impact of the impeller movement is modeled via a momentum source. However, the MRF approach is not a predictive simulation because the impeller speed must be given by the User. A new simulation approach was introduced that coupled the FSI and MRF approach.