A Study on the Effects of Sequential Turbocharged Diesel Engines

Abstract

Sequential turbocharging is a technique where multiple turbochargers are connected to an engine. These turbochargers are switched in sequentially. The manufacturers of engines with this system claim that the part-load and transient performance is better in comparison to engines with a single turbocharger. These claims and other effects in these engines are not well documented in scientific literature. The motivation for this research assignment is the application of the Pielstick PA6B V20 STC on board of the Indonesian navy SIGMA corvettes. The sequentially turbocharged engines have been selected for this design, based on the operating envelope of the engine. At low speed, the sequentially turbocharged engines are able to deliver more torque than a normal turbocharged Diesel engine. Based on the operating envelope, the sequentially turbocharged Diesel engine is able to deliver similar benefits that are normally present in gas turbine and electric motor propulsion systems; high torque at low speed. However, DSNS prefers the use of Diesel engines for their robustness and available maintenance support. The design decision for a sequentially turbocharged Diesel engine during transient and low load operation is evaluated in this report. The TU Delft Diesel B model has been adapted to model the sequential turbocharging strategy. This has resulted in two different models, each with their own benefits and drawbacks. The first model is the Simple STC model; this model gets its name from the simple implementation in Simulink. It is able to reproduce the trends of a sequentially turbocharged engine. The Simple STC model is of the same complexity as the TU Delft Diesel B model, both in application and in computational load. The drawback of this model is that it does not correctly model the transition when the sequential switching occurs, but this error is only present for a few seconds. The Full STC model also gets its name from the implementation in Simulink; in this model the full gas exchange of the sequential turbochargers is modeled. This model is a significantly more complex model than the Simple STC model, both in application as well as in computational load. However, it does provide the possibility to model each turbocharger separately and as a result the transient switching effect is modeled more accurately. Both models have been thoroughly tested in a test-bench environment, both for steady state and transient analysis. The results have proven that the characteristics of a sequentially turbocharged engine are modeled correctly. It was proven that for the sequentially turbocharged engine, there are significant benefits for having two different turbocharger modes. Under steady state conditions, the mass supplied to the engine is of better quality in terms of mass flow and pressure for each of the different turbocharging modes in their respective working regions. This increased quality of air supply leads to lower temperatures, lower fuel consumption and higher available power for the same power rating. The Simple STC model has been applied in a model of the complete SIGMA corvette to simulate acceleration tests and to test new control strategies. This has provided insight into the limits of the engine-propeller-hull interaction during acceleration. The propeller and ships ability to absorb power are limiting the engine's available power when the results are analyzed in the operational envelope. It has been proven that the engine needs to ramp-up in speed almost instantaneously to utilize the full available power. The engine is not able to ramp-up that fast due to the dynamic interaction of the engine and turbocharger. When accelerating the engine it was found that at higher acceleration rates the turbocharger does not spin up fast enough and as a result the air excess ratio decreases during acceleration; this phenomenon is also known as turbo-lag. The maximum acceleration limit is determined to be 12.8 [rpm/s] which results in a minimum air excess ratio of 1.5. An alternative control strategy based on a minimum air excess ratio has been analyzed but this method did not result in a good control alternative. It did however give some insight into the maximum possible acceleration limits and provides a basis for future research into alternative control strategies.