Experimental and numerical investigation of the Wankel engine and skip cycle system

Wankel engine has been an important research subject for many academics and industry from the perspective of its sealing, emission, performance, ignition, injection, construction, and material selection. On a Wankel engine, there exists a rotor and an eccentric shaft as moving parts, and the rotor is connected to eccentric journal that is on the shaft. Hence, the rotor performs an eccentric rotation inside a housing that has a shape of a trochoid with the rotation of the shaft. One cycle of Wankel engine is completed with a complete rotation of rotor within the housing. In one complete cycle of rotor, eccentric shaft turns three times. At each complete rotation of eccentric shaft, one work stroke is obtained. As opposed to conventional reciprocating engines whose work cycle is completed with the 720⁰ rotation of the crank shaft, in a Wankel Engine it is completed with the 1080⁰ rotation of the eccentric shaft. Furthermore, it takes 180⁰ CA (Crank Angle) to complete a single stroke for a conventional reciprocating engine, while this is 270⁰ ESA (Eccentric Shaft Angle) on a Wankel engine. The superior features of a Wankel engine when compared with reciprocating engines, can be listed as low vibration, lightness, fewer number of parts, and higher specific power (ratio of power/weight). Due to these advantages, Wankel engines have many application areas, such as unmanned air vehicles, boats, hybrid vehicle applications, passenger cars, usage of alternative fuels (such as hydrogen and natural gas), etc. Currently, the global automotive manufacturer Mazda works on Wankel engines, who has recently launched RX-8 model to the market which is powered by a 13B-MSP (Multi Side Port) Wankel engine. In the experiments of the current work, the same engine (13B-MSP Wankel engine) has been used due to easier access of spare parts and the compatibility of the engine size for the assembly of the measurement tools required for the experiment. Furthermore, another important parameter in choosing this engine is the exhaust ports being placed on side housings (the only Wankel engine having this feature) and the fact that rotor housing is more suitable for constructive changes. In this study, a one dimensional thermodynamics model is constructed to study the combustion chamber of 13B-MSP Wankel engine. In this model, one cycle between 0-1080⁰ ESA is investigated. In the model, governing equations are derived from the conservation energy, conservation of mass and ideal gas laws. First of all, the geometry of the combustion chamber and port area such as volume and area is obtained with preliminary calculations, which forms the basis of the model built. By using these fundamental data, compression and expansion processes are calculated (270 - 810⁰ ESA). By achieving expected results, same process is applied for fuel-air mixture. By adding the combustion process defined by Vibe to the model, heat release arising from the combustion of fuel is added to the thermodynamic cycle. Afterwards, charge exchange process is added to the model along with the calculations of leakages and crevice volumes arising from the seals. By comparing results of the model with those of previously conducted experiments in terms of torque and volumetric efficiency under full load, parameters influencing the cycle process are investigated. Furthermore, Vibe parameter and combustion duration changes are handled for each experiment point from the model outcomes obtained as being dependent on speed under full load condition in the engine. As the speed increases, the combustion duration is shortened. It is observed that the Vibe parameter, which is a function of combustion, decreases as the speed increases. In the model, by only changing the optimum values of Vibe parameter and combustion duration, their impact on engine torque are investigated. It is determined that reduction of Vibe parameter and combustion duration until 6000 rpm and increasing these values after 6000 rpm increases engine torque. Furthermore, the effect of different intake port conditions on the engine torque is investigated with the model for full load and different speed regimes. As a conclusion, it is observed that operating at low speeds (1000, 2000 and 3000 rpm) with primary intake port, provides higher torque when compared to other ports. It is also found that engine torque is higher for medium speeds (4000, 5000, and 6000 rpm) with primary and secondary intake ports and for high speeds (higher than 6000 rpm) with three open ports (Primary, secondary and auxiliary). Afterwards, friction pressure and specific fuel consumption of the engine are evaluated under full load condition at different speeds. Model results are compared with experimental data for different speeds, load, and excess air coefficient values. Conditions given for each experiment point are used in one dimension model. Various parameters such as intake temperature, absolute intake pressure, exhaust back pressure, obtained volumetric efficiency, combustion duration, and advanced spark ignition are taken as preliminary conditions for the model. It is observed that the results obtained by adapting the experimental conditions to the model, are in good match with the experimental data. In the experimental part of the study, it is initially planned to establish a test setup with 13B-MSP Wankel engine with a single rotor. Though, a prototype engine is first built by gathering the spare parts of the engine. The prototype engine is driven by an electric motor that has different pulley diameters. Hence, the friction pressures and compression testing of the prototype engine is performed at different speed. As a result of these preliminary tests, it is concluded that the engine can be converted into a single rotor engine. For the experiments, the original engine with two rotors is modified to a prototype with a single rotor, that required several alterations on various parts. Some parts are produced again in order to adapt the prototype engine with single rotor. Various mechanic parts such as eccentric shaft, intake and exhaust manifolds, and lubricating and cooling systems are adapted from the original engine with modifications in order to comply with the prototype engine having single rotor. Furthermore, an electrical control unit, in which an Arduino 2560 card is used as a microcontroller, is also built to control the ignition and injection systems. For the faster and effective operation of the system, solid state relay (SSR), are used in the experimental setup. SSR is an electronic switching unit that has the same controller and power circuits, and performs the same function as classical relay and contactor. Software and hardware parts of control unit are first disassembled and they were adapted to the engine after preliminary experiments. In order to perform the experiments, maintenance and calibration of various devices present in the same room with the experimental engine such as cooling water conditioning unit, fuel conditioning and measurement device, electromagnetic engine brake and emission devices, is conducted. Since software of pressure measurement and evaluation unit Kistler is only suitable for engines with reciprocating piston, a study is conveyed in order to realize combustion chamber pressure measurement for Wankel engine and to form the cycle. In order to measure the combustion chamber pressure of the engine per ESA along the cycle, three piezoresistive and one piezoelectrical sensors are utilized at different locations on the engine. Finally, the pressure/volume cycle is completed by numerically at ESA values where there are no measurements. Since there was no experimental data about this engine in the literature; performance, emission and combustion chamber pressure parameters are evaluated under different conditions. In the engine experiments, load, speed, and excess air coefficient conditions are used as variables. Furthermore, impact of parameters such as number of spark plug, advanced ignition, ignition difference between two spark plugs, injection advance, intake port, injector, and intake channel numbers and operation of engine under idle mode are investigated. Finally, a method for skipping cycle is applied for the test engine. As a result of the literature search, no studies could be found about a method for skipping cycle that was used for engines with rotating pistons until now. In engines with reciprocating piston, the efficiency of engine at partial load can be improved by using the method for deactivating the cylinder instead of the method for skipping cycle. However, the method of skipping cycles has more significant importance at partial loads for a Wankel engine due to its structure and operation. Generally, the amount of fresh charge taken in is reduced by closing the throttle valve in spark ignition engines. As a result, the effective power performed is reduced. Though, the method of skipping cycles is based on the production of work at every two cycles of the eccentric shaft by opening and closing the intake port consecutively. Therefore, work is generated once when eccentric shaft turns two times. During the cycle in which the work is not produced, the fuel-air mixture is not allowed in the combustion chamber during the four stages of the cycle. Method of skipping cycles is adapted to 13B-MSP Renesis modified one rotor engine. First of all, a mechanism for skipping cycles is designed and built. In the theoretical part, studies related to the open port and geometry of rotating shaft are conducted. First, the sectional areas are obtained by drawing open port of rotating valve at different angles. Then, these sectional areas that remain open together based on the rotation of the valve are embedded in the mathematical model. In the model, comparison is done based on the combustion chamber mass and pressure. As in the case of original rotating valve, better results are obtained at 80⁰ opening and low speeds (<3000 rpm). Earlier opening conditions (-10⁰, -20⁰, -30⁰) for rotating valve with 80⁰ rotating angle are also investigated. Better results are obtained at 80⁰ rotating valve clearance with 10⁰ early opening at low speeds. Furthermore, investigation of rotating valve inside the high-speed intake channel in clockwise (movement of valve from bottom towards the top) and counter clockwise (movement of valve from the top towards the bottom) directions are performed. It is determined that more fresh charge is taken inside with the clockwise movement of rotating valve. Hence, the production of rotating valve is made in accordance with clockwise rotation. In the production of this mechanism, the design of rotating valve is taken as a primary concern. Due to its deviations from the original rotating valve, several holes are created on the valve circumferentially in order to enhance the lubrication of the valve with the surrounding port by blowing out some lube-oil mixture out of these holes. Another difference is that the rotating valve is produced with tight tolerances, so that it fits the intake channel. Hence, leakages are avoided during empty cycles while there is vacuum. In the mechanism, parts such as connection shaft, movement of mechanism and sealing had been resolved. The investigation of the mechanism during its operation with the engine fixed timing adjustment is performed. With the method for skipping cycles, engine is operated with full and empty cycles, respectively, while the valve is in an open and closed positions. With the method for skipping cycles, it is planned to keep the intake port of Wankel engine closed at specific ESA of the cycle at partial loads while keeping the exhaust port open. The engine was tested operated mechanically with this skipping cycle strategy. In conclusion, a commercial two rotor Wankel engine is modified into a single rotor engine and performance tests are conducted in this study. Furthermore, the engine is mathematically investigated with a one dimensional thermodynamic model. Furthermore, the skipping cycle strategy is applied on the engine under partial loads and low speeds and the mathematical solutions are also obtained for skipping cycle cases. By comparing the results obtained with the experimental data of the original engine, several conclusions are made. The unique features of the current study can be given as the one dimensional thermodynamic model built for the Wankel engine and skipping cycle experiments performed on the Wankel engine.