Feasibility analysis of an exhaust gas waste heat driven jet-ejector cooling system for charge air cooling of turbocharged gasoline engines

Highlights

• Investigation of a compound system of a combustion engine and a cooling system.

• Determination of feasible charge air temperatures in the compound system.

• Analysis of thermal COP, hydraulic COP and cooling capacity of the system.

• Derivation of required exhaust gas conditions and resulting pressure losses.

• Evaluation of power densities, additional fuel consumption and multi-staging.

Abstract

The present paper analyzes the feasibility of an exhaust gas driven jet-ejector cooling system for charge air cooling of turbocharged gasoline engines in addition to the conventional charge air cooler to increase the engine efficiency. Thereto, steady-state experiments of a jet-ejector cooling system and an exhaust gas heat exchanger prototype working with the refrigerant R134a are used to analyze the operation and control of the compound system and determine feasible cooling capacities, charge air temperatures, thermal ${\mathit{COP}}_{\mathit{th}}$ and hydraulic ${\mathit{COP}}_h$. Moreover, the cooling system is rated regarding its power densities and engine backpressure. The exhaust gas waste heat recovery, system power densities, and engine backpressure are acceptable. However, the hydraulic ${\mathit{COP}}_h$ and the amount of reject heat need to be improved, necessitating for instance a multi-staging of the jet-ejection.

Introduction

According to Pucher and Zinner [1], only a third of the fuel energy of today’s internal combustion engines is converted into mechanical energy while one third is rejected to the ambient by means of the engine coolant circuit and the other third as exhaust gas waste heat. The coolant water is used for cabin heating at low ambient temperatures. An exhaust gas waste heat recovery on the other hand is promising due to its comparably high temperature level of about $500''–''1400\mathrm{K}$ despite a possible elevated engine backpressure due to the exhaust gas heat exchanger.

Extensive research on waste heat recovery of vehicles by means of a Rankine-cycle subsequent to the combustion engine to generate additional mechanical or electrical work began after the first oil crisis around 1970 [2], [3], [4]. Studies on the Rankine-cycle in vehicles have most recently intensified according to Wang et al. [5] due to improvements in expansion devices and an enlarged variety of new working media [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19]. Moreover, power cycles such as the open Joule or Stirling process are currently subject to research [20]. The power process efficiency ${\eta }_{\mathit{PP}}$ is defined as the ratio of the mechanical power output $P_m$ to the driving heat flow $Q_2$:

$${\eta }_{\mathit{PP}}=(P_m/{\dot{Q}}_2)$$

Endo et al. [18] presented experimentally validated process efficiencies of power processes between 0.1 and 0.18 resulting in fuel reductions of up to $6.5\%$ [21].

An alternative technology aims at the direct conversion of the exhaust gas energy into electrical energy by means of thermoelectric generators [22], [23], [24], [25], [26]. The process efficiency ${\eta }_{\mathit{TEG}}$ is defined as the ratio of the electrical power output $P_{\mathit{el}}$ to the driving heat flow $Q_2$:

$${\eta }_{\mathit{TEG}}=(P_{\mathit{el}}/{\dot{Q}}_2)$$

Reported process efficiencies range between 0.02 and 0.05.

Köhler et al. [27] rate the generation of cold for transport refrigeration with an exhaust heat driven single-stage absorption chiller to be promising in particular for constant exhaust gas heat flows. The thermal coefficient of performance of the chiller ${\mathit{COP}}_{\mathit{th}}$ is defined as the ratio of the generated cooling capacity ${\dot{Q}}_0$ to the driving heat flow $Q_2$:

$${\mathit{COP}}_{\mathit{th}}=({\dot{Q}}_0/{\dot{Q}}_2)$$

The investigated absorption chiller prototype exhibits thermal ${\mathit{COP}}_{\mathit{th}}$ of about 0.2–0.3.

The concept of generating mechanical power by means of a Rankine-cylce, electrical power by means of a thermoelectric generator or cooling for air-conditioning and refrigeration by means of a sorption cooling system recovering the exhaust gas heat of an internal combustion engines is therefore not new to literature. A novel concept, however, is to generate a cooling capacity by means of an exhaust gas driven jet-ejector cooling system in order to reduce the charge air temperature of turbocharged gasoline engines below the level attained by state of the art charge air coolers in order to increase engine efficiency. This novel concept is analyzed in the present paper.

For a given engine operation point, the effective engine efficiency ${\eta }_{\mathit{eff}}$ and the effective engine power $P_{\mathit{eff}}$ depend on the density of the charge air ${\rho }_{\mathit{ca}}$ for a given engine, fuel, and combustion method [1]. According to the ideal gas equation, i.e. ${\rho }_{\mathit{ca}}=p_{\mathit{ca}}/(R·T_{\mathit{ca}})$, an increase of the density of the charge air ${\rho }_{\mathit{ca}}$ can be achieved either by an increase of the charge air pressure $p_{\mathit{ca}}$ or a decrease of the charge air temperature $T_{\mathit{ca}}$. Therefore, so-called turbocharging, i.e. a compression of the charge air above ambient pressure by means of an exhaust gas turbine, has by now become an established technology for exhaust gas waste heat recovery in particular for diesel engines. It results in an increase of the effective engine power, allows for a down-sizing of the engine at constant power output or alternatively for an increase in effective engine efficiency.

The charge air temperature $T_{\mathit{ca}}$ increases during the non-isothermal compression, which counteracts the increase of the charge air density ${\rho }_{\mathit{ca}}$. In addition, the thermal stress in particular of the turbocharger turbine increases with elevated charge air temperature as the temperature level of the power cycle depends on the temperature of the gas in the cylinder at its intake and therewith on the temperature of the charge air [1]. In gasoline engines, an elevated charge air temperature might furthermore lead to an uncontrolled combustion, the so-called engine knocking. Therefore, charge air coolers subsequent to the compressor are nowadays state of the art [28].

The hypothetical lower limit of the charge air cooling is the ambient temperature. In reality, the charge air temperatures are well above the ambient temperature due to necessary driving temperature differences or limited heat exchanger area. Kadunic and Zegenhagen [29] have experimentally demonstrated an efficiency increase of a turbocharged gasoline engine of up to $18\%$ for a further reduction of the charge air temperature from $333.15\mathrm{K}$ to $273.15\mathrm{K}$ due to the ability of reduced fuel enrichment and advanced ignition timing for selected steady-state operation points at high engine loads.

This potential improvement can only be exploited by cooling down the charge air with a heat pump. Guhr [30] for instance analyzed the potential of a further charge air cooling by means of a mechanically-driven compression chiller to down-size the engine. However, the additional fuel consumption necessary to drive the compression chiller must be overcompensated by the benefit of the charge air cooling.

An alternative concept for an additional charge air cooling is a cooling system driven by the engine exhaust gases as depicted in Fig. 1.

The driving heat flow ${\dot{Q}}_2$ of the thermally-driven cooling system (TDCS) is transferred from the exhaust gases on the highest process temperature level $T_2$ via an exhaust gas heat exchanger (EGHX). The cooling capacity ${\dot{Q}}_0$ is generated downstream of the conventional charge air cooler (CAC I) in a second charge air cooler (CAC II) on the lowest temperature level $T_0$ further reducing the charge air inlet temperature. The sum of the driving heat flow and cooling capacity, ${\dot{Q}}_1$, must be rejected to the ambient in a heat rejection unit (HR) on the intermediate process temperature level $T_1$.

Heat pumps driven by the engine coolant water are often investigated for vehicle air-conditioning [31], [32], [33], [34], [35], however, some exhaust gas driven systems for air-conditioning purposes or transport refrigeration are also subject to research [36], [37], [38]. The mentioned thermally-driven cooling systems are exclusively sorption systems which are widely disseminated in stationary applications. Currently the highest thermal ${\mathit{COP}}_{\mathit{th}}$ of single-effect sorption units is about 0.8. However, fundamental improvement in process engineering is still needed to establish the sorption processes as cooling technology in vehicles due to their low power densities as a result of their system complexity [35], [39]. The experimental gravimetric and volumetric system power densities range between only $0.015\dots 0.033{\mathrm{kW}}_{{\dot{\mathrm{Q}}}_0}{\mathrm{kg}}^{-1}$ and $0.006\dots 0.013{\mathrm{kW}}_{{\dot{\mathrm{Q}}}_0}{\mathrm{dm}}^{-3}$ [33]. Options for flexible system integration and the choice of the working media, especially for evaporation temperatures below the freezing point of water, are limited. Experimentally demonstrated thermal ${\mathit{COP}}_{\mathit{th}}$ of sorption systems sized-down to the cooling capacity range of up to $10\mathrm{kW}$ adequate for the application in vehicles amount to 0.1 and 0.45 [27], [32], [33], [35], [38].

An alternative to the sorption processes is the jet-ejector cooling process for which the working medium can be chosen almost freely. High heat fluxes or small specific heat transfer areas, respectively, and thus high power densities can be achieved with the jet-ejector process due to high heat transfer coefficients during evaporation and condensation. As in sorption systems, besides a pump, no moving parts are required. In contrast to the latter, only one working medium is used and there are less heat exchanging devices. This results in a comparably low system complexity. The heat exchangers and components can be arranged freely since piping pressure losses are less disadvantageous at the prevailing high pressure as compared to the vacuum process of most sorptive systems. Therefore, the jet-ejector cooling system has been identified as promising for the charge air cooling of turbocharged gasoline engines despite its lower thermal ${\mathit{COP}}_{\mathit{th}}$ in comparison to the stationary sorption processes [39].

In the present work, a jet-ejector cooling system sized down to the necessary cooling capacities and working with the refrigerant R134a is analyzed for charge air cooling of turbocharged gasoline engines. Alternative fluids may also be employed.