No abstract

The paper describes the design process for a controlled pulse turbocharging system (CPT) on a 5 cylinder 4-stroke marine engine and highlights the potential for improved engine performance as well as reduced smoke emissions under steady state and transient operating conditions, as offered by the following technologies: • controlled pulse turbocharging, • high pressure air injection onto the compressor impeller as well as into the air receiver, and • an electronic engine control system, including a hydraulic powered electric actuator. Calibrated engine simulation computer models based on the results of tests performed on the engine in its baseline configuration were used to design the CPT components. Various engine tests with CPT under steady state and transient operating conditions show the engine optimization process and how the above-mentioned technologies benefit engine behavior in both generator and propeller law operation.

With the introduction of the EPA Tier 4 NOx emission limits for rail diesel engines this year, engine developers are forced to implement more advanced emission control technologies such as selective catalytic reduction (SCR) or cooled external exhaust gas recirculation (EGR). The integration and control of these systems for ensuring optimum performance throughout the operating range brings about new challenges on top of the well-known requirement for unconstrained operability in a very wide range of conditions. As a consequence, engines and their subsystems have to be designed for maximum flexibility. The turbocharging system in particular needs to be capable of dealing with extreme ambient conditions associated with high altitudes, hot summers, severe winters, tunnel operation, etc. This flexibility must be achieved without compromising reliability and while ensuring continuous in-use compliance with the emissions standards throughout the life of the installation. At the same time, engine performance should be maintained at the highest level possible. This study demonstrates that all of these targets can be met by combining two-stage turbocharging and EGR with suitable control elements. Two-stage turbocharging, which has become increasingly popular in other industry sectors due to its potential for improving the bsfc / NOx emissions trade-off when used in combination with correspondingly optimized valve actuation (Miller timing), is starting to be adopted also for rail applications. A variety of EGR concepts was proposed or put into practice over the past few years, and the most important or promising of these have been taken into consideration for this study. Extensive simulations of the resulting engine and turbocharging systems have been performed using ABB’s in-house simulation platform, based on a generic engine model that can be considered representative of the rail sector. It is shown that integration of EGR, two-stage turbocharging and appropriate control elements is highly attractive as it offers outstanding operational flexibility and very high fuel efficiency without any compromise in terms of reliability. The selection and specification of control elements and turbocharging system components depends on the EGR concept applied. As is shown below, this can be tailored to the application to ensure optimum performance and flexibility. In view of these obvious benefits, we are very confident that such integrated EGR / two-stage turbocharging systems will be adopted more widely on railway engines.

No abstract

For one-dimensional simulation of flow in ramified pipe systems loss prediction in junctions is essential. In the standard literature (e.g.[1]) total pressure loss coefficients serve this purpose. Elementary formulas follow from the principle of momentum and some simple assumptions [2]. They yield rather crude estimates, often twice as large as experimental values. Modifications described by Idelchik [3] look nice, but after all turn out to be not more accurate. We explain deviations and propose improved loss formula based on Truckenbrodt's energy and momentum coefficients [4], thus paying regard to velocity distributions over pipe cross-sections. We further discuss extraction of the coefficient's values from experiments.