Miniature HCCI free-piston engine compressor
Project / Leader
2B.2 - Prof. David Kittelson (U of Minnesota)
Title
Miniature HCCI Free-Piston Engine Compressor
Statement of Project Goals
The goal of this project is to continue development of a compact high efficiency air compressor providing 10 W of pneumatic power that can be used for the Test Bed 6, ankle foot orthosis (AFO), and for other applications needing a tiny power supply. The power source is a free piston engine operating in a homogeneous charge compression ignition (HCCI) two stroke cycle, integrated with a free-piston air compressor.
Description and Explanation of Research Approach
Hydrocarbon fuels have a distinct mass (46 MJ/kg) and volume (39 MJ/liter) density advantage over batteries (0.7 MJ/kg and 1.5 MJ/liter for state-of-the-art lithium-ion). Extremely poor efficiency engines (the Cox 0.010 two-stroke model aircraft engine has an efficiency of about 4%) still win on power density over a battery-motor supply. The ability of compressed air to be converted directly into mechanical work through ordinary, light weight cylinders further distances an engine-compressor-fluid-power approach from a battery-motor-mechanical transmission approach. Because fluid power can be transmitted with very little loss through small, flexible hoses, there is more flexibility to locate components on an AFO to meet customer demands for aesthetics.
The Engine Compressor Concept
Miniature size engines are not just scaled down versions of full scale engines. They have many unique characteristics. As engine scale goes down, the surface to volume ratio increases. This leads to increased heat loss, makes ignition more difficult and flame quenching more likely. Smaller scales also make friction more important. Sealing also becomes problematic since clearances cannot scale down linearly with piston and cylinder scale. Consequently a larger fraction of the charge is lost to leakage in small scale engines. In addition many other components, like spark plugs, fuel injectors, sensors, actuators and valves, are difficult to scale down.
A free-piston, HCCI two stroke cycle design has been developed for this project to address these difficulties. The use of HCCI combustion minimizes problems associated with wall quenching. In addition to solving the HCCI ignition timing problem, the free piston approach reduces friction because there are no side thrusts on the pistons. The two-stroke cycle offers many advantages for a small scale system. Compared with their four-stroke counterparts, simple crankcase scavenged two-stroke engines are suitable for those applications where simplicity, compactness, reliability and power density are required, while efficiency, emission and noise regulations are not so stringent. The smallest scale conventional engines (model engines) are crankcase scavenged two stroke engines.
HCCI is a combustion process distinctly different from spark ignition (SI) or compression ignition (CI) in conventional engines. The working fluid before ignition is a fuel and air mixture, as in an SI engine, but the compression ratio is considerably higher. The compression process, which is nearly adiabatic, raises the temperature and pressure to levels sufficient for auto-ignition and the fuel is consumed almost instantly, instead of relatively slowly by flame-propagation (SI engine) or diffusion burning (CI engine). HCCI has many advantages over conventional combustion. First of all, its efficiency is higher than SI due to its higher compression ratio and minimum throttling loss. Secondly, HCCI circumvents flame quenching problems. Flame quenching has been observed with conventional ignition in combustion chambers as large as 1.5 cm wide, but Aichlmayr's experimental work demonstrated the ability to run HCCI in a 3 mm bore cylinder. Thirdly, HCCI does not need ignition devices like a spark plug or a fuel injector as in conventional engines. In small scale, implementation of these devices is problematic.
HCCI combustion in conventional crankshaft engines needs sophisticated schemes to control the start of combustion, such as thermal conditioning intake air, EGR control and altering fuel content. Precise combustion phasing is needed because it must fit with the piston's motion, which is restrained by crankshaft. In contrast in a free-piston engine, the piston's reciprocating motion directly drives the load, thus the piston is free of the kinetic restraint of the crankshaft. Its motion is determined by the forces acting on the piston. The piston keeps compressing the in-cylinder charge until it undergoes auto-ignition, after which the cylinder pressure rises drastically and pushes the free-piston back. There is no fixed compression ratio in a free-piston engine so that compression will continue until auto-ignition occurs as long as sufficient compression energy is available.
Figure 1: Miniature HCCI Free-Piston Engine design concept
Some preliminary simulations have been done based on this design concept. Details of these simulations can be found in our publication presented at the recent Small Engine Technology Conference (SETC). Some of the conclusions of these simulations are: 1) to achieve the goal of 10 W fluid power output 1, the engine bore size should be about 8 mm with a displacement of about 0.4 cm3; 2) leakage is significant in engines of this size so that the engine has to run at high speed to minimize leakage; 3) HCCI combustion chemical kinetics are related to the fuel auto-ignition quality, engine compression ratio and engine speed and high engine speeds require easily ignitable fuel. Figure 2 is a plot of power output and efficiency based on these calculations. It can be seen that the operating range is limited by blow-by leakage at the low end and ignition at the high end to between about 18,000 and 45,000 rpm (300 to 750 Hz). We have selected a design speed on the lower end of the range, 300 Hz, to minimize wear and noise.
Figure 2: Engine performance with speed
These initial engine calculations were used to make preliminary performance estimates for a coupled free piston engine compressor using a one dimensional model of engine performance and MATLAB SIMULINK solver. Perhaps the most important result of the initial one dimensional modeling is that it indicates that with suitable choice of the compression spring constant and compressor load, the engine compressor system quickly establishes stable operation at 300 Hz with an efficiency from fuel to cool compressed air of about 6%. This model approach is useful for determining initial engine parameters such as spring constant and piston set mass, and for exploring the engine characteristics. But for more accurate predictions, experimental data on very small engines is required and little is available in the open literature. Consequently a small engine test system has been built to support the work.
Engine test facility
Figure 3 shows the small engine test setup. It consists of a small dynamometer with motoring capability, fuel and air flow measurement systems, and in-cylinder pressure and crank-angle measurement systems. The setup will allow determinations of engine power and efficiency, scavenging efficiency, friction loss and, with the aid of simple models, rates of combustion and leakage.
First generation engine
Based on the information gathered from preliminary simulations and testing of model engine components, a first generation prototype was designed and fabricated. The prototype is based on a small, two stroke model airplane engine with glow plug ignition and a bore size similar to the initial design value of 8 mm. The engine cylinder liner, engine cylinder head, silencer, carburetor, engine piston and piston pin, compressor piston and piston pin and compressor cylinder liner are components of model engines. This eliminates the cost of manufacturing components with very complex geometry and high tolerance requirement. Glow plug ignition utilizes a glowing catalytic filament to initiate combustion. It is like HCCI in that it requires no outside trigger but relies on flame propagation after ignition. Thus, it shares some features of HCCI but is easier to start and regulate and operates at lower compression ratio. The completed prototype is shown in Figure 4.
Figure 3: Model engine test bench
Figure 4: First generation prototype