Present study deals with performance analysis of an inert gas generator (IGG) which is to be used as an effective mean to suppress the fire. The IGG uses a turbo jet cycle gas turbine engine to generate inert gas for fire extinguishing. It is generally known that a lesser degree of oxygen content in the product of combustion will increase the effectiveness of fire suppressing. An inert gas generator system with water injection will bring advantages of suffocating and cooling effects which are considered as vital factors for fire extinguishing. As the inert gas is injected to the burning site, it lowers the oxygen content of the air surrounding the flame as well as reduces the temperature around the fire as the vapour in the inert gas evaporates during the time of spreading. Some important aspects of influencing parameters, such as, air excess coefficient. $\alpha$, compressor pressure ratio, $ pi_c$, air temperature before combustion chamber, $T_2$, gas temperature after combustion chamber, $T_3$, mass flow rate of water injection, $M_w$, etc., on the performance of IGG system are investigated. Calculations of total amount of water needed to reduce the turbine exit temperature to pre-set nozzle exit temperature employing a heat exchanger were made to compare the economics of the system. A heat exchanger with two step cooling by water and steam is considered to be better than water cooling only. Computer programs were developed to perform the cycle analysis of the IGG system and heat exchanger considered in the present study.

The experimental results of the 17K inertance tube pulse tube cryocooler for cooling cryosensors are presented in this paper. In prototype experiments, linear compressor is driven by linear motor, and inertance tubes are inserted between one liter reservoir and pulse tube. Design of the inertance tube pulse tube cryo-cooler is conducted by ARCOPTR program of NASA Ames Research Center. To find optimal conditions of inertance tube pulse tube cryocooler, no load temperature and refrigeration capacity according to the variations of inertance tube volume, reservoir volume and charging pressure are measured. and the cool down and load characteristics at the particular conditions are presented. As the representative results, no load temperature of the cold end is 52.7K and refrigeration capacity is 5W at 72K..

In comparison with gasoline engine, LPG direct injection engine has some advantages not only in emission and fuel efficiency but also in prevention of power decrease and back fire etc. which are disadvantages of conventional LPG engine. In this study, comparision tests of the incylinder spray and combustion characteristics between of LPG and gasoline fuels were performed in the RCEM as a basic research for the development of future LPG engine with low emission and high fuel efficiency During the direct injection of LPG fuel and gasoline into the inside of RCEM, spray development characteristics according to the injection condition have been photographed by the high speed shadow graph methods. The conditions for the optimum mixture distribution of LPG and gasoline fuels are achievable at the selected ignition timing, respectively.

건물화재시의 피난자의 인명안전 확보를 위한 가압방연 설계프로그램에 필요한 건물요소의 모델링 방법, 계산방법 등을 제시하였다. 건물의 각 층에 급기용 송풍기를 설치할 수 없기 때문에 층에 따른 압력차 분포는 필연적이다. 전층가압방연 방식에서 급기덕트나 각종 유로면적이 동일하다면 송풍기에서 가까운 구획은 압력이 커지며 먼 구획은 압력이 감소할 수밖에 없다. 따라서 건물 전체가 규정범위내의 압력차로 유지되어야하기 때문에 송풍기에서 가장 멀리 떨어진 구획의 압력차를 기준으로 설계하여야 한다. 이러한 경우 과도한 송풍기의 용량으로 인한 문제에 더하여 송풍기로부터 가까운 구획에는 과압이 발생할 수도 있다. 그러므로 본 연구에서 얻어진 프로그램과 같은 설계구도를 활용하면 각 층의 공기 공급 덕트의 단면적을 달리한 경우를 계산할 수 있다. 예제에서 언급한 바와 같이 각 층 공기공급 덕트의 단면적을 달리하므로 써 송풍기의 용량을 감소시킬 수 있게된다.

Heat release rate used to characterize the fire is an important factor for determining the fire size, the fire growth and the time for suppression and evacuation. The purpose of present work is to review theoretical backgrounds and to introduce equations for estimation of heat release rate with oxygen consumption method in fires. Our work also shows the experimental results of applications for liquid fuels. The oxygen concentration is measured by the analyzer of paramagnetic type. The analyzers of Infra-Red type are used to measure the concentrations of $CO_2$ and CO gas. Time delays of analyzers are ignored.

This paper presents the status quo of the development of a piecewise bendable switch system for the urban transit MagLev. MagLev system as well as railroad requires switch systems to reach its destination. Requirements of the switch system for commercial lines are high speed operation satisfying about 2-3 minute headway and system reliability, etc. Parallel moving type switch system was installed on the test track of urban transit MagLev in KIMM. In this system, switch operation from one position to another can be done in about 90 seconds. Therefore, we concluded that this system can not satisfy the headway for the commercial lines. We decided to develop a high speed piecewise bendable switch system in which switching can be done in 20 seconds. Designed switch system is very complicated in view of operating mechanism. It consists of 11 segmented girder beams driven by hydraulic cylinders. To gain the idea of a piecewise bendable switch system, we manufactured and tested a 1/5 scale switch model. We are going to construct a full scale piecewise bendable switch system next year.

Numerical techniques for the simulations of crack propagation are reviewed. This paper highlights the meshless methods as a potential method for the simulations. thus they are reviewed deeply. Especially the theoretical aspects of meshless methods are discussed. and it is shown that all meshless methods are based on the PUM and unified in GFEM even though they are originated from different sources.

In naval ships, pump machineries are the major sources of airborne and structure-borne noise. Noise is critical issue in ships not only it causes annoyance to crews, but also it can increase the underwater radiated noise, which is crucial in anti-submarine warfare. In present study, it is discussed the reduction of structure-borne noise of pump machineries by showing several examples. The most typical and effective solution is to use double resilient mount system. However, in cases where double resilient system cannot be applied due to space and weight increase, rubber pad can be used to reduce the structure-borne noise. In principle, the concept of structure-borne noise reduction is the same as that of vibration isolation.

Fatigue crack growth behaviors of Cr-Mo steels developed recently for thick-wall pressure vessel were investigated. Experiments in accordance with ASTM E647 standard were performed for 1/2 inch CT specimens of $2^(1/4)$Cr-1Mo and 3Cr-IMo steels in gaseous environments, hydrogen gas of 10 atm, 1 atm and argon gas of 1 atm. Fatigue crack growth rates were observed and effects of gaseous hydrogen and argon on the crack growth behavior were discussed.

Presented in this paper is a new method of identifying the critical speed of rotor-bearing systems without actually reaching at the critical speed itself. Using the method, it is not only possible to calculate the critical speed by measuring a series of rotor responses at much lower rotating speeds away from and without reaching at the critical speeds but also the damping ratio and eccentricity of the system can be identified at the same time. Two types of test rotors were tested on the Rotor Dynamics Test Facility at the Rotordyn-amics Lab, KIMM, and the theory has been confirmed experimentally. The method can be adopted to monitor changes of the dynamic characteristics of critical rotating machinery before and after overhauls, repairs, exchanges of various parts, or to detect trends of direction of subtle changes in the dynamic characteristic parameters over a long periods of time.

고속 Rapier loom에서 경사와 위사의 조직을 완성하여 직물을 제직하는 위입과 바듸침 기구는 캠구동 방식으로 동일축선 상에 4개의 캠이 설치되어 있다. 이들 캠들은 2개씩 한쌍을 이루면서 서로 공액운동을 수행하고 있으므로 정확한 설계에 의한 동기화운동이 필수적이다. 따라서 본 연구는 전체 거동의 파악을 위한 시스템의 정확한 해석과 컴퓨터 시뮬레이션을 통한 Rapier loom의 고속화를 목적으로, 위입과 바듸침기구의 곡선선정 및 캠의 회전에 따라 정확한 운동을 유도할 수 있도록 2가지 운동의 연관성은 물론 동시에 화면상에 구현할 수 있는 4개의 캠기구에 대한 시뮬레이션 프로그램을 개발하는 것이다.

Flow divider valve는 한 개의 공급라인에서 두 개 이상의 출력라인으로 유압유를 일정비율로 분배하는 유압제어밸브로서 하중압력이나 공급압력 등에 관계없이 항상 일정비율의 유량분배가 가능해야 한다. 현재 상용제품의 유량분할 정확도는 90~95% 수준이며, 이러한 유량분할오차(Flow dividing error)는 유압시스템에 누적오차로 작용하여 많은 문제점을 야기시키고 있어 보다 고정밀 유량제어가 가능한 Flow divider valve 개발이 요구된다. 본 연구에서는 외력을 고려한 스푸울의 거동을 수치적으로 정확하게 조사하여 Flow divider valve의 동특성을 규명함과 동시에 유량분할 오차를 감소시키는 최적설계방안을 제시하고자 한다. 동특성 해석은 일정한 하중저항을 입력신호로 작용하는 경우에 대해서 제시하였으며, 이때의 고정오리피스와 가변오리피스의 단면적 및 스푸울의 단면적 변화에 따른 동특성의 변화를 조사하였다.

The effect of ball milling conditions in the milling of aluminium foil scraps was studied. Initial foil thickness, ball size. content of oleic acid. weight ratio of mineral spirits/foil. charged amount of foil were varied in wet ball milling process. It is impossible to make flake powders by milling of foil scraps with thickness $120 \mum$. As foil thickness decreases from $60\mum$ to $6.5\mum$, Mean size of powder milled for 30 h decreases from 107 µm to 17 µm. Bigger ball is slightly beneficial for milling of foils to the flake powders due to the larger impact energy produced by them. It is impossible to mill the foil without oleic acid to fabricate the flake powder. As content of oleic acid increases from 1.5 % to 5 %, mean size of flake powder milled for 30 h is drastically decreased. For the mineral spirits content below 50 %, foil scrap was not milled because sliding motion of balls by lubricant effect between balls and wall of container. As weight ratio of mineral spirits and foil increase over 100 %, foils were milled powders with mean powder size 15 - 20 때 irrespective of mineral spirits content due to reduced lubricant effect. As charged amount of foil decreases, mean powder size decreases due to increased collision frequency between ball and foil.

Brazing of Fe-Cr-Al alloy was carried out at $1200^{\circ}C$ in vacuum furnace using nickel-based filler metals : BNi-5 powder(Ni-Cr-Si-Fe base alloy} and MBF-50 foil (Ni-Cr-Si-B). The effect of boron content on the stability of oxide scale on the brazed joint was investigated by means of cyclic oxidation test performed at $1050^{\circ}C$ and $1200^{\circ}C$. Apparently, the joints brazed with MBF-50 containing boron showed relatively stable oxidation rates compared to boron-free BNi-5 at both temperatures. However, it was considered that the slower weight loss of MBF-50 brazed specimen wasn’t resulted from the low oxidation rate but from the spallation of oxide layer. The oxide layer consisted of thick spinel oxide on the surface and $Al_2 O_3$ internal oxide layer along the interface between mother alloy and braze, the mother alloy was also eroded seriously by the formation of spinel oxides such as $FeCr_2 O_4$ and $NiCr_2 O_4$ on the surface, likely to be induced by the change of oxide forming mechanism due to diffusion of boron from the braze. On the contrary, the joint brazed with BNi-5 showed the good oxidation resistance during the cyclic oxidation test. It seems that the oxidation can be retarded by the formation of stable $Al_2 O_3$ layer at the surface.

In general, manufacturing processes of thermosetting composites consist of mold filling and resin cure. The important parameters used in modeling and designing mold filling are the permeability of the fibrous preform and the viscosity of the resin. To consolidate a composite, resin cure or chemical reaction plays an essential role. Cure kinetics. Therefore, is necessary to quantify the extent of chemical reaction or degree of cure. It is also important to predict resin viscosity which can change due to chemical reaction during mold filling. There exists a heat transfer between the mold and the composite during mold filling and resin cure. Cure kinetics is also used to predict a temperature profile inside composite. In this study, a new scheme which can determine cure kinetics from dynamic temperature scaning was proposed. The method was applied to epoxy resin system and was verified by comparing measurements and predictions.

A new titanium alloy system. Ti-xFe-ySi (x,y=0-4 wt%). was designed and characterized with the point at low cost and high strength for casting applications. Fe improved room and elevated temperature mechanical properties owing to solid solution hardening and beta phase stabilization. Si yielded titanium silicides and Si addition over 1 wt% resulted in poor ductility due to coarse silicide chains at prior beta boundaries. The optimum composition was found to be Ti-4Fe-(0.5-1)Si in the viewpoint of tensile strength and ductility which are comparable to the Ti-6Al-4V. The metal-mould reaction was also examined for Ti-xFe and Ti-xSi binary alloy system. The thickness of surface reaction layer w as not affected significantly with Fe content, while it was decreased with Si content. In the Ti-4Si alloy, no reaction layer was found. The depth of surface hardening layer was about $200\mum$ regardless of the mould materials.

The vacuum investment casting process for a large gas turbine component, Inner Preswirl Support (IPS), was simulated by using commercial FEM package ProCAST(TM) with view factor radiation method. The solid fraction in mushy zone was directly measured by Differential thermal analysis(DTA-DSC mode). Three types gating design. considering liquid flow and heat release through it. were proposed. Solidification had begun at the ribs or thin sections of the IPS casting and advanced further through the upper and lower gates. The computed temperature gradient G and G/R values at 70% solidified temperature were used for prediction of microshrinkage formation during casting. The effect of mold preheat on the thermal history of the casting displayed minute effect on the microshrinkage formation.

The shape and thickness distribution according to the pressure-time curve were carried out using the FEM and experiment. Also, mechanical properties were investigated. The square cup parts have been formed with pressure-time curve generated by result of analysis. The tensile strength and elongation have been investigated according to applied pressure conditions using the tensile test specimen obtained from the superplastic formed cup. We can use to predict the shape of formed part under the applied pressure using the FEM analysis. In the case of optimum pressure condition, the thickness distribution and mechanical properties were improved. From this study, we can find the important of optimum pressure-time condition. We have investigated about the forming of airplane part and fuel tank for motorcycle. If the applied load used in boundary conditions was appropriate, the simulation result coincides with the formed part. However, it is very difficult to define the pressure condition in complicated shape. Thus, it is need to develop the optimum pressure condition for superplastic forming.