Investigation of concepts for high power Stirling engines

Current research and development in the field of energy conversion as well as in the area of vehicle propulsion systems focusses on new energy conversion technologies in order to reduce exhaust emissions and energy consumption. Besides the fuel cell, there is increasing interest in the Stirling engine, which provides attractive features for these two areas of application.
In the field of combined heat and power, the Stirling engine is interesting because of its flexibility in the use of primary energy sources: apart from liquid or gaseous fuels (from fossil or regenerative sources), solar and geothermal energy may also be used. In addition, modern Stirling engines allow extremely long overhaul periods of up to 10,000 hours.
Besides the optimization of internal combustion engines as automobile propulsion systems, intensive research is being done on hybrid power trains and fuel cell-based propulsion systems, but to some extent also on external combustion engines. For this field of application, expertises have revealed the high potential of the Stirling engine as an ultra-low emission, high-efficiency primary energy converter, although considerable research and development will be necessary before series applications can be envisaged.

The most important advantages of Stirling engines compared to fuel cells or internal combustion engines are as follows:

• similarity of external to internal combustion engines allows the use of know-how and existing components from this field;
• existing fuel infrastructure as well as installation space available in current vehicles can be used without major modifications;
• considerable emission reduction already attainable with current fuels at comparatively little expenditure due to continuous external combustion;
• multi-fuel capability or even use of heat from any imaginable source possible, switch to alternative fuels at relatively little expenditure;
• high thermal efficiency (in particular at part load) due to favourable thermodynamic process;
• low noise and little vibration.

Research and development activities on Stirling engines are still necessary in the field of construction and material choice, with the aim of reaching the envisaged service life at minimum cost. Furthermore, reliable and durable sealing of the high-pressure working spaces is still a problem in some configurations. For mobile applications, however, the increase of power density and (depending on the concept) the improvement of dynamic behaviour are the main areas of research.

Operating principle and basic types of Stirling engines

1-2

isothermal compression (release of heat) in the compression space (cold space);

2-3

isochoric heat transfer from a regenerator to the working fluid during displacement from compression to expansion space (heat absorbed Q₂₃);

3-4

isothermal expansion (supply of heat) in the expansion space (hot space), release of mechanical energy;

4-1

isochoric heat transfer from the working fluid to the regenerator during displacement from expansion to compression space (heat rejected Q₄₁=Q₂₃).

Fig 1: p-V and T-s diagrams of the ideal Stirling engine process

The ideal Stirling process achieves the same efficiency as the Carnot process (optimum efficiency process for conversion of heat into work): 1 - T_min / T_max.
The realization of the idealized process in a displacer type engine is shown in fig. 2. The compression and expansion processes are generated by the movement of the power piston, while the transfer between the cold and hot spaces is caused by the displacer movement. The isochoric heat transfer processes take place inside a regenerator positioned between the cold and hot space. In most cases, the regenerator consists of a dense wire mesh capable of storing at least a major part of the thermal energy still contained in the working fluid after the expansion process. The heat stored is used during the next cycle to heat the working fluid on its way from the compression to the expansion space.

The major difference to gasoline or Diesel engines lies in the fact that the working fluid is not replaced after each process, but stays permanently inside the engine (closed cycle), being alternately heated up and cooled.

Fig 2: Realization of the ideal Stirling process in a displacer type engine

As can be seen from fig. 2, the ideal Stirling cycle requires discontinuous movements of the piston and displacer. In reality, however, a perfect realization of this process is not possible. In most cases, a number of deviations must be accepted which reduce both efficiency and power density.

The most apparent differences result from the use of continuous instead of (impractical) discontinuous volume changes (e.g. by use of a crank shaft drive) and the heat transfer using external heaters and coolers instead of the working space surfaces. The latter modification is necessary because of the lack of surface area and time for heat transfer in reciprocating piston engines. On the whole, this leads to rather adiabatic than isothermal processes inside the engine cylinders.
Further disadvantages arise from the dead space inside the engine (heat exchangers, regenerator, tubing) which reduces the compression ratio, and loss due to axial heat conduction and temperature fluctuation of the regenerator. Finally, piston seal leakage and pressure drop across the heat exchangers and the regenerator as well as losses due to mechanical friction cannot be avoided, either.

In total, these deviations actually lead to a process which is significantly different from the ideal process, cf. fig. 3. The green curve shows the ideal process, considering the maximum and minimum temperatures occuring in the cycle. The red curve has been calculated using a simulation program which takes into account the most important losses (adiabatic processes, continuous volume changes, pressure loss and non-ideal regenerator / heat exchanger behaviour).

Fig 3: Comparison of the ideal and the real Stirling process in p-V diagram (simulation results)

Apart from the displacer engine shown in fig. 2, where power piston and displacer share the same cylinder (such engines are also referred to as the 'beta' type), there are two other basic configurations of Stirling engines which can be distinguished by the arrangement of their working spaces (cf. fig. 4):

In 'gamma' type engines, the piston and displacer operate in separate cylinders, which facilitates sealing and allows the use of less complicated drive systems for the displacer rod (which no longer passes through the piston). However, dead space increases and therefore efficiency and power output are lower than those of comparable beta configurations.

Stirling engines using a second piston instead of the displacer are commonly referred to as 'alpha' type machines. They may be built in single-acting as well as in double-acting configuration; in the latter case, another Stirling process is being carried out in the space below the pistons. Single-acting engines share many standard components with current mass-produced reciprocating engines or pumps and can therefore be manufactured at relatively low cost. Furthermore, such engines are very flexible in the arrangement of the heat exchangers. On the other hand, compression ratio is low compared to beta machines, and sealing is difficult (two pistons per process). Double-acting machines are more complicated, but also more compact and light-weight, and they show a favourable relation of friction loss to shaft power.

Beta engines, on the contrary, allow higher compression ratios (less dead space) and, as a result, show an improved power density. Such engines therefore allow a comparatively compact layout, but require complicated drive systems for the coupled and strictly linear movements of the piston and displacer (e.g. rhombic drive).

Fig 4: The three basic mechanical arrangements of kinematic Stirling engines: alpha, beta, gamma type

The rotary piston Stirling engine as a new high power concept

Apart from reciprocating engines, the Stirling process may also be realized in rotary piston engines. Although there is a great number of concepts for rotary Stirling engines, not much is known about existing machines. Fig. 5 shows in simplified form the concept of a rotary piston Stirling engine based on two z=3 machines (similar to the well-known Wankel internal combustion engine). The concept displayed is similar to a multiple-acting alpha type reciprocating engine, with one unit acting as a (cold) compression machine and the other as the (hot) expansion space and regenerators placed inbetween (of which only one is shown in fig. 5 for reasons of lucidity). One working space of the cold and another of the hot side respectively have to be coupled with a suitable phase angle, thus providing three independent Stirling engine units undergoing two processes each per piston revolution. Following the common classification scheme, one could call the rotary piston Stirling engine a triple-acting alpha type, which can be expected to have a correspondingly high power density.

Fig 5: Basic arrangement of a rotary piston Stirling engine

The use of a rotary piston arrangement for the realization of a Stirling process brings about numerous interesting features: Distributing the cycle to two separate units, with one of them operated at the upper and the other at the lower process temperature, leads to constant wall temperatures along the whole circumference of both rotor housings, which greatly facilitates sealing in comparison to internal combustion rotary engines. Nonetheless, as the rotor seals cannot be oil lubricated due to regenerator sensitiveness with respect to impurities of the working fluid, and the temperature in the expansion space may reach comparatively high levels, considerable research is still necessary in the field of sealing.

The strict separation of hot and cold zones, on the other hand, allows a minimization of heat loss and thus an increase in overall efficiency. A further advantage for thermal efficiency is the enhanced approach of the processes inside the working spaces towards isothermal compression and expansion. This is possible because at least a major part of the heat transferred to and from the cycle can be transmitted across the rotor housing due to the enlarged surface-to-volume ratio of the rotary engine in comparison to reciprocating engines. By reducing the size or completely eliminating external heat exchangers, dead space is also reduced (increased compression ratio), and weight and bulk space decrease. In combination with the highly compact design of a rotary piston engine, this concept promises significant improvements of both efficiency and power density.