Literature Review

According to Jan Peirs et al.’s researches [1][2], a micro turbine with a rotor diameter of 10mm was used to generate electric power, and the fuel used was liquid fuel. It was well-known that the performance of both turbine and compressor were clearly limited by the maximum attainable speed. Consequently, the first goal was to increase both turbine and compressor speeds. Secondly, the turbine and compressor needed to redesign to match each other’s characteristics to increase their efficiencies.

The micro turbine developed at MIT was a radial turbine with a rotor diameter of 4mm, manufactured lithographically in Si or SiC. The micro turbine developed at Stanford University used an axial-radial turbine with a diameter of 12mm. A 10mm diameter axial micro turbine with generator was developed and successfully tested to speed up to 160,000 rpm. It generated a maximum mechanical power of 28W with an efficiency of 18%. Power and efficiency were mainly limited by the tolerable maximal speed of the ball bearings. When coupled to a small generator, it generated 16W of electrical power, and the efficiency for the total system was 10.5%. The rotor was tested up to 130,000 rpm with 330℃ compressor air. The size of the micro turbine was so small that could be widely applied, but some researchers were wondering why not

established the micro gas turbine generator systems, whose capacities were up to hundreds of kW. However, the size scale was too large to be portable. Wen [3] studied the micro turbine generator system that introduced the experimental foundation for a 375hp micro turbine generator system, which included inlet and exhaust sections, testing frame, operating control system and measurement system. Besides, the micro turbine generator system had to be coupled with the operating control system, which included starter system, fuel supply system, lubrication system, ignition system and secondary air supply system of combustor. And he completely designed and constructed the measurement devices and data acquisition system that were needed for measuring engine baseline performance. Wang  [4] studied a 150 kW micro turbine generator set with twin rotating disk regenerators by testing and analyses.

Using the PC-based data acquisition system, the engine speed, turbine inlet temperature, exhaust temperature, compressor inlet and discharge pressure, and fuel flow rate were measured. The generator set was tested by using load bank to establish the baseline performance including temperature, pressure, horsepower, fuel consumption, and speed. He used a software program (GasTurb) to predict the performance of the micro turbine generator set at different operating conditions in order to compare with the test results. The thermal efficiency of 28% was predicted at full load with regeneration, whereas in the case of no regeneration, thermal efficiency was only 14%.

Yamashita et al. [5] studied a MGT, operating with the experimental conditions using low-heating value fuels simulated in terms of the

dilution of LPG (liquefied petroleum gas) with N₂. Efficiencies at each system component were calculated from the measured temperature and pressure. The obtained results were summarized as follows: Under the condition of a fixed volume flow rate of the supplied fuel, TIT (turbine inlet temperature) is deterioration of the overall system efficiency. As the LPG-N2 mixing ratio is increased, NOx emission is decreased, and CO is increased because of the incomplete combustion occurring at the combustor. The MGT system was successful in operating at the fuel condition of 43% heating value of LPG, indicating the acceptability to such low-heating value fuels without any modification of the combustor.

Kousuke et al. [6~8] and Shuji et al. [9] studied the micromachining gas turbine, which was under development at Tohoku University. In the micro gas turbine, the heat transferred from the high temperature components, such as the combustor and the turbine to the low temperature ones, such as the compressor, was very large because the distance to separate the high temperature from low temperature components was closer than that of the general turbine. So, the temperature gradient in micro turbine became large if the highest and lowest temperatures in both turbines were the same. Therefore, the micro gas turbine could not be directly reduced the scale as the same structure of the macro-scale gas turbine because the thermal efficiency would be too low. The performance test of their gas turbine has been started, and ran up to 566,000 rpm, which is approximately 65% of the design speed.

The compressor performance has been successfully measured along a constant speed line at 55% of the design speed, and indicated that the

temperature gradient increases with a decrease of MGT size if the highest and lowest temperatures for both large and small scale gas turbines are the same. AS a result, the material of the micro gas turbine should be stronger than that of the large scale one and the MGT efficiency is expected to be reduced because the ratio of the heat loss to heat release increases. The heat loss is proportional to the surface area of the MGT’s case, temperature gradient across the wall, and heat conduction coefficient of wall material. On the other hand, the heat release is proportional to the volume of the MGT’s case and the space heating rate.

At this scale of a micromachined combustor, the heat loss to heat release ratio is expected to be about 5%. It is still an order of magnitude larger than that of conventional gas turbine, but is much less than that of M.I.T’s, which is about 20%. Kousuke et al. [7][8] also indicated that when the requirement a pressure ratio of 3 by an impeller of diameter 10mm is achieved, the rotor s required to rotate at 870,000 rpm, which is not tolerable by today’s ball bearing technology. Hence, air bearing should be used. Among the air bearings, hydrodynamic air bearing is preferable because the extra air supply is not required. The hydroinertia bearings are employed in both radial and axial directions. The performance of the compressor was measured at 60% (530,000 rpm) of the rated rotational speed (870,000 rpm) by driving a turbine using compressed air. The measured pressure ratio was lower than the predicted one because the impeller’s tip clearance was larger than the designed value. During an acceleration process toward the rated rotational speed, the shaft was destroyed at the bearing at 566,000 rpm. At high speed, a large bearing

gap and very high rotor balance were required to achieve low viscous loss and to prevent the rotor from hitting the bearing at critical speed. Hence, a hydro-inertia gas bearing had been selected with half-split bearing sleeves. A hydro-inertia bearing was a type of static air bearing which had large bearing clearance to generate supersonic flow in the bearing gap. At present, a rotor speed as high as 770,000 rpm has been achieved in the test.

Li [10] manufactured a model chamber, which was geometrically identical to prototype chamber, in the cold-flow tests. It was found that the pressure loss of flow through the liner hole is about 3% of the total pressure. And he also found the information about the flow rate of the holes. These are helpful to modify the chamber design. In the firing tests, it discussed the mixing effects by changing the fuel injection direction and by controlling the numbers of swirl holes. The results showed that upstream fuel injection mode dose play well in the short chamber length comparing with the side fuel injection mode.

Tsai [11] studied the gas turbine combustor design. The combustor had to design to meet some requirements, including high combustion efficiency over a wide operating envelop, stable operation, low pressure loss, low temperature pattern factor and low pollution emissions. In cold flow test, differential pressures as airflow penetrating the liner hole were measured to further understand its flow distribution. It was found that the jet dynamic pressure ratio K is found different from the one in large turbine combustors and K's value seems too small that the discharge coefficient Cd is quite sensitive to flow condition, resulting in the

variations of Cd and flow distribution.

Shiung [12] found that during the design of a micro engine combustor, it was always difficult to accurately arrange the inlet air distribution, because the suitability of the conventionally recommended Cd (discharge coefficient) based on large and small engines was questionable for micro-engine. The relationship between Cd and dynamic pressure ratio K (for micro-engines with K ranging from 1 to 4) was experimentally established. Then, with the aid of the oil flow techniques and pressure data, the inlet air distribution could be estimated.

1.3 Scopes of the present study

In this research, the original combustor of a micro gas turbine (WREN MW54) is modified first to burn the gaseous fuel, instead of the primary Jet A1, a liquid fuel. Now, the Jet A1 fuel is replaced by a simulating biomass gas fuel of low heating value, CH4 mixed with CO2, to generate the electric power. The use of mixture gaseous fuel is because the present biochemical technique cannot obtain very pure methane gas from the purified marsh gas. In other words, the bio-gas inevitably contains certain impurity gases, such as carbon dioxide (CO2), that are hard to remove completely. The quantity of impurity gas will dilute the concentration of pure methane so that the power of micro gas turbine is expected to be lower than the original designed value, burning with liquid fuel, even worst; it may not work at all.

The experimental study will vary some parameters in order to measure the power generation of such modified micro gas turbine. There

will be various mixture ratios of CH₄ and CO₂ of low-heating-value fuel used in this study. The concentrations of CH4 in fuel are changed from 90% to 50%, and the ones of CO₂ increase from 10% to 50%

simultaneously. Via this parametric study, it is intended to find the best operating parameters between pressure and flow rate to obtain the best output.

Chapter 2

EXPERIMENTAL APPARTUSES  

2.1 Experiment layout

All the Experiment layouts are shown in Figs. 2-1 and 2-2. The micro gas turbine (MGT) and its schematic dimension of major parts are shown in Figs. 2-3a~b to 2-8a~b. Transmission gear box and generator, which are placed on a test frame, made of 6061 aluminum alloy, are shown in Fig. 2-8a and Fig. 2-9, respectively. Each of the parts is stably fixed on the frame. The blower used to start to drive MGT is connected with the MGT air inlet by a rubber tube. As soon as MGT reaches the idle rpm, the rubber tube is removed and MGT can be inhaled by its own power.

The generator is connected to MGT by the transmission gear box, which transforms the power, produced by the MGT, to axial work. And the axial work can drive generator to generate electric power. The fuel pressure and flow rate, at inlet nozzles of combustion chamber, can be adjusted by the pressure valve and flow meter. The MGT’s rotation speed is detected by an induction tachometer installed near the compression fans. The compressor’s inlet and outlet, the turbine’s inlet and outlet, and the exhaust temperatures are measured by thermocouples set up across the MGT’s case. All measured data are transferred to the disk storage using a PC-controlled data acquisition system.

2.2Parameters of tests

The experimental parameters include fuel concentration (the volume fraction ratio of CH4 to CO2) and turbine rpm. The fuels used are 99.5%

C₃H₈, 90% CH₄ mixing 10% CO₂, 80% CH₄ mixing 20% CO₂, and 70%

CH4 mixing 30%CO2, and 60% CH4 mixing 40% CO2, and50% CH4

mixing 50% CO₂. The turbine rpm are 40000, 45000, 50000, 55000, 60000, 65000, 70000, and 75000, respectively. Each combustion test consists of five different fuels, and each fuel used is with eight different rpm. Each test condition is carried out twice for data consistence. If the low heating value fuel cannot reach up to the anticipated rpm, then, rpm range will be reduced.

2.3 The micro gas turbine system

2.3.1 Micro gas turbine

The MGT is modified from the one of MW54 MK III. The original oil fuel pipe, including the lubrication pipe, is changed to a single gas fuel pipe, and the lubrication pipe is moves to be an independent system, as shown in Fig. 2-2b.

MGT engine is optimized to generate large quantities of hot gas under specific pressure and uses these hot gases to pass through a second stationary guide vane and turbine, mounted on a separate shaft. The 2^nd turbine, only used to drive the load, is considerably larger in diameter than that of the 1^st stage, so it turns slower and with larger blades that can absorb a large amount of energy from the gases passing through it. This

enables a higher torque to be produced at an RPM which is relatively easy to provide further reduction to fit the application. This 2-shaft configuration allows considerable mismatch to be tolerated between power turbine and load.

The heart of the power unit is a single stage gas turbine using centrifugal compressor and axial turbine. Power is drawn from the exhaust gases by a second turbine, which drives the propeller via the gearbox. When the engine uses oil fuel, it is a powerful unit capable of delivering 6kW of smooth and quiet power.

2.3.2 Transmission

The power of gearbox comes from a large power turbine driven by the exhaust gas from the MGT. The power turbine is mounted on a shaft running in high temperature bearings. Lubrication and cooling for these bearings are from a fuel takeoff system supplied via a tube mounted above the engine, feeding via the rear of the gearbox. The input shaft drives a gearbox, comprising a two-stage reduction using helical gears, fully hardened for long life.

The gear box lubrication system is “wet sump,” i.e. oil, held within the gearbox, circulates by the action of the gears such that requires no external oil tank or services. The gearbox is charged with 20ml of automobile transmission oil (usually red in color). In normal use, the oil needs replacement every 20hrs running or 120 tests.

2.3.3 Fuel supply

Two 10kg steel cylinders, shown in Fig 2-10, are used. Each cylinder connects to a pressure gauge. As shown in Fig 2-11, both

pressure gauges connect to the flow meter with a three-way switch and a one-way valve. The three-way switch can connect two different fuels. The higher heating value one is easier to ignite and drive the MGT. As the ignition of MGT occurs, switch the valve to the low heating value fuel.

1/8 inch Teflon pipe is used to connect the MGT because it can resist the heat. The gas fuel pressure and flow rate can be control easily. When the fuel needs to be changed, the bottle can be easily switched.

2.3.4 Start system

Blower shown in Fig 2-12 is used to push the MGT’s fan to rotate, and supply the MGT enough air to mixing with fuel. When MGT is turned off, the blower is also used to cool the temperature down. The spark plug is modified to easily ignite the fuel, and is connected with a 1.5V~3V battery or a voltage regulator that can supply enough electric power. If the voltage is too large, the spark plug might be burnt out.

2.3.5 Generator

As shown in Fig 2-9, a 170W alternating current generator (5TY 00 made by T-MORIC), whose resistance of coils are 0.56~0.84Ω at 20°C, is used to generate electric power by the axial work from the gearbox. It can output 14V 170W @ 5000RPM, and the regulator (SH640E-11 anti voltage 200V) can commutate AC to DC.

2.3.6 Lubrication

MGT can be worked at a very high rotational speed, so the lubrication of bearings is crucial. In the original system, the lubricating oil is pre-mixed with the oil fuel. When the fuel is emitted into combustion chamber, the lubricating oil and gas will splash to the

bearings.

When the MGT is modified, the fuel used becomes the gas, so the lubricating oil can no longer pre-mix with the oil fuel. The lubricating oil pipe now should be an independent component as shown in Fig. 2-13.

The lubricating oil is emitted to the bearings directly by a pump that is controlled by a program. The program is design to control the pump to provide the lubricating oil every few seconds in order to protect the bearings.

2.4 Measurement instrumentation

2.4.1 Temperature measurement

The thermocouples shown in Fig 2-2a and Fig. 2-14 are installed at the MGT compressor’s inlet and outlet, turbine’s inlet and outlet, and exhaust in order to monitor the combustion situation. The highest temperature of the MGT engine is about 600-800°C at the inlet of turbine, so K-type thermocouples, ranged from -200 to 1260°C, are used.

2.4.2 Turbine RPM measurement

An electromagnetic induction kind of tachometer shown in Fig. 2-15 using hall-effect sensor is used to measure the rpm. When the generator turbine blades pass the sensor, which is a magnet in front of the blades, will produce a pulse, and the pulse can be delivered to the data acquisition system. Using a conversion program, the pulses can be transformed to rpm of engine.

2.4.3 Fuel pressure measurement

A mass flow meter shown in Fig. 2-16 has been installed on the fuel line ahead of the injector of the combustor chamber. The flow meter is calibrated to an accuracy of 1% of the full scales, and its acquiring data are volume flow rate which are already calculated by mass divided by density (1 ATM, 0℃).

2.4.4 Fuel flow rate measurement

The fuel supply system is the main component to start the MGT, and the flow rate of the fuel not only needs to be measured but also to be controlled. A mass flow controller (MFC) is a closed-loop device that sets, measures, and controls the flow of a particular gas or liquid. In this research, a HM5000 MFC produced by Tokyo Keiso Company is adopted and is shown in Fig. 2-17. The MFC is calibrated to measure and control the methane fuel.

2.4.5 Generator output Measurement

The output voltage of generator is too large for a data acquisition, so this work uses the cement resisters to cascade with the generator output, and the decrease of the voltage across the resistor can indicate the current of the generator’s output. Also the potential difference is small enough to be accepted by data acquisition. Then, the power is obtained by the multiplication of current and voltage.

2.4.6 Data Acquisition

Data acquisition can automatically gather signals from analog and digital measurement sources, such as sensors and devices, under tests.

Data acquisition uses a combination of PC-based measurement hardware

and software to provide a flexible and user-defined measurement system.

Usually, the researcher must calibrate sensors and signals before a data acquisition device acquires them. The specifications of these modules of National Instruments are shown in Table 2-2. National Instruments, a leader in PC-based data acquisition, offers a complete family of proven data acquisition hardware devices and the powerful and easy-to-use software that can extend to many languages and operating systems. NI CompactDAQ delivers fast and accurate measurements in a small, simple, and affordable system. A CompactDAQ Chassis shown in Figure 2-18a, a product of NI, is adopted because of the following advantages:

plug-and-play installation and configuration, AC power supply and USB

plug-and-play installation and configuration, AC power supply and USB