Search

The module coil TOKI-MC is a twisted solenoid coil wound with an aluminum stabilized superconductor developed as an R&D program for the Large Helical Device (LHD). The TOKI-MC can simulate the conductor and winding structure cooled by pool boiling helium, the twisted winding and the large electromagnetic force of the helical coils for LHD. The TOKI-MC was designed as a cryostable coil at an operating current of 20 kA, but the coil quenched around 17 kA in excitation tests. The cause of quenches was thought to be the result of wire movement. Stability tests were also carried out and the measured recovery current was less than the designed value. The degradation of recovery current was due to the excess magnetoresistivity of the copper clad aluminum stabilizer. The stability of TOKI-MC was evaluated and compared with the data of short sample tests. >

It has become common to search necessary services and contents using the Internet, but it is difficult to find exactly what one is looking for through keywords as each service is described in just too many ways. We developed "laddering" search service system that matches the needs of the users with the search targets by communicating with the users through interviews. The system consists chiefly of the laddering dialogue engine employing "laddering method" which enables the engine to ask questions to users and extract keywords and expressions that users probably would not be able to express on their own. This paper describes the outline of laddering search service system and the experimental results of the demonstration system on the Web.

According to the 45 nm BEOL technology node, we demonstrated that a homogeneous interlayer dielectric with dielectric constant of 2.25 has a substantial advantage in terms of RC delay reduction compared to other potential architectures such as hybrid and tri-level dielectrics. Combination of the homogeneous interlayer dielectric and ultra-thinned barrier metal lowered the RC delay to 86 % compared to that listed in the ITRS 2006 update.

We tried to evaluate and predict the RC delay variability beyond the 45 nm copper interconnects technologies. The RC delay variability as a normalized delay time distribution, is caused by the line width/height variations due to the manufacturing process fluctuations. In order to evaluate the influence of the resistivity size effect precisely, we improved Fuchs-Sondheimer (F-S) and Mayadas-Shatzkes (M-S) models, in order to include the line height dependence of copper grain size, and applied it in the evaluation of the RC delay variability based on the SPICE simulation. In our results, we found that the RC delay variability in the 45nm node technology was relatively small, weakly dependent on the grid size and line height, and almost not affected by the size effect. On the contrary, in the 32 nm technology, the RC delay variability was about 2 times larger than the case ignoring the size effect and reached to the 20% of the average delay time at 3000 grid with 10% of line size fluctuation. In the 32 nm technology, the line height dependence of the RC delay variability was also strong and increased with decreasing line height. The influence of line height dependence of grain size reached about 1/5 or more of the total size effect in the RC delay variability.

In this work, a device of the atmospheric pressure plasma source was experimentally investigated, and constructed as a coaxial arrangement with a high-voltage electrode and a dielectric tube. The plasma source at the atmospheric pressure condition was formed into He or Ar gas jet exposed to dielectric barrier discharges, which operated on AC high-voltage applied. The atmospheric pressure plasma flowed out from a nozzle at the end of the dielectric barrier discharge tube. In the plasma diagnoses, an electronic temperature was quantitatively evaluated by the optical emission spectroscopy. This paper presents the electrical discharge characteristics, plasma emission photograph and the spectrum for analyzing the plasma generation mechanism. The fundamental characteristics of the electronic temperature was discussed from the atmospheric pressure plasma conditions as a function of the experimental parameter such as the gas flow rate and the distance from ground electrode.

This paper analyzes the thrust and core loss of a linear motor made of magnetic powder core, that is, soft magnetic composite material. The magnetic field is calculated by using axial symmetric finite element method and boundary element method. The difference of the thrust and core loss of the motor with magnetic powder core from those with laminated core is clarified quantitatively

This paper reports on an approach to software process modeling with its method. Our model consists of two sub models - one is for tasks which are performed in a software process and another is for resources which perform the tasks, e.g. human participants, project teams, computerized tools and so on. The formal specification language LOTOS (Language of Temporal Ordering Specification) is used to define software processes based on our modeling technique. We have developed the method to represent software processes by using LOTOS based on the methods for software specification & design. Furthermore we have applied our method to an example problem proposed for the 6th Software Process Workshop, and discussed our approach.

Usually, control of a large-scale real time system is performed by a combination of simple control rules (rulebase). With growing control rule-bases, new problems such as consistency among rules, and behavior of a physical system under the control, arise. For this reason, checking techniques for such problems are required. In this paper, we assume that a multi agent net, which is a hierarchical extended Petri net (T. Murata, Proc. IEEE, vol. 77, no. 4, pp. 541-580, 1989), model for physical system is given, and we propose a modeling technique for an integrated system model of control rule-bases and physical systems for verification.

A fault-tolerant system architecture is proposed for digital signal processing with quantum devices which make transient errors. Two techniques are employed in the proposed architecture. Firstly, there exists no circuit block to control system operation sequence. Every circuit modules in the system autonomously repeat simple operation at each clock cycle. This structure prevents system failures. Secondly, pulse density modulated single-bit data whose average represents signal level is the data form for the system. Unlike multi-bit binary data, a few bit-errors do not make data haphazard but merely let data lower in accuracy. Computer simulation of a digital filter based on the architecture demonstrated the effectiveness of these techniques.

The changes of the customers' demands become rapid and the product lifecycles have become shorter and shorter. Although batch processes have natural flexibility, addition of new products needs many modifications not only in the batch plant but also in the system for control, scheduling and so on. The flexibility of this management system must be improved to catch up the rapid changes. In this paper, multi-agent structure for batch control system is proposed to make its change management easier. Four types of agents work in communicating with each other. The recipe agents instruct the recipe to equipment agents. Equipment agents have intelligence to recognize instructions described in the recipe. The specification of the human interface to support the operator is also discussed.

Simple structure, high torque and low cost new hybrid type stepping motor (new HB-STM is used hereafter) is proposed and its torque performance is clarified theoretically. Also, this new motor is compared with conventional hybrid type stepping motor, and discussed the advantage and disadvantage in this paper.

The huge earthquake of magnitude 6.3 had occurred at Bam city of southeastern Iran on December 26, 2003. The author applied nonlinear mapping method for detection of collapsed building area due to Iran earthquake with high-resolution QuickBird satellite images. As a result, ROC evaluation based on the result of photo interpretation by operator showed that the detection ratio of the proposed method is 75-90% of changed area with 20% of false alarm.

A novel 4-operand redundant binary adder by using neuron MOS is described. Proposed adder can achieve totally parallel multi-operand addition, because four input operands can be added simultaneously without the carry propagation chain by our novel addition algorithm. The principle of this algorithm is to utilize the partial addition in every two digits block. The neuron MOSFETs are applied to the implementation of this system, accordingly the ternary operations for the redundant binary number and the multi-input operations can be simply realized. The features of the proposed adder are capability of high speed operation and less number of transistors as compared with the conventional binary one. Simulations have been made by HSPICE.

A binary-quantized cellular array model of linear diffusion systems is investigated microscopically by introducing virtual particles. A deterministic rule applied to all the cells of the array determines the motion of the particles in the array. The probabilistic distribution in terms of the displacement of the particles in the array is similar to a Gaussian distribution, and its variance increases in proportion to time. The discontinuous velocity of the particles has almost uniform frequency spectrum distribution and delta-function-like autocorrelation. Thus, the virtual particles whose behavior is deterministic possess almost the same characteristics that probabilistic Brownian particles in diffusion systems have. We conclude that the cellular array is an excellent binary-quantized and discrete pseudo-diffusion system.

We point out a problem of the congestion control mechanism of TCP on the long fat pipe network (LFN) by precise analysis using our hand-made tools, and propose a congestion control algorithm for parallel streams and packet spacing algorithm in slow start phase for LFN. This paper presents: (1) observation of TCP/IP multi-stream data transfer across 7500 miles, (2) analysis tools for congestion problem; i.e., packet logging tool and pseudo LFN emulator, (3) pseudo LFN experiments, and (4) modification of congestion control algorithm, balancing parallel streams with TCP fairness and TCP compatibility and packet spacing.

This paper proposes a norm-type discrete algebraic Riccati Equation, which is a generalized version of the well-known standard discrete algebraic Riccati Equation, and has additional norm terms. Under stabilizability and assumption that the additional terms are not too large, the existence of a positive semidefinite solution is guaranteed. Application to guaranteed cost control is given.

Piecewise linear (PWL) circuits operating on sigma-delta (/spl Sigma//spl Delta/) modulated signals are proposed. The proposed circuit library includes an absolute circuit, a min/max selector and a negative resistance. As an application of the PWL circuits, a median filter is presented. It is structured with 23% of logic gates consumed by an equivalent filter with 8-bit parallel signal form. It is expected that application of /spl Sigma//spl Delta/ domain signal processing including proposed PWL operations to other modern information processing and communication systems such as neural networks and chaotic communication systems will reduce their hardware cost.

Aiming at the utilization of low temperature thermal energy, a thermoelectric generator was applied to a municipal solid waste incinerator. The generator consisted of 60 BiTe thermoelectric modules. In the system, oil was used on the hot side as the heat transfer medium and water was used on the cold side. The electric output obtained from the system was 535 W under the conditions of 34 degC at cold side (water) and 261 degC at hot side (oil). A running test was carried out with 22 times on/off heat cycles. The accumulated operating time was 115 hours. No significant degradation of the thermoelectric module was observed throughout the test period. The energy conversion efficiency of the thermoelectric generator was measured at 2.8%.

Proximity effect is a phenomenon such that an electric current density distribution becomes asymmetric when an alternating current flows in conductors arranged at a short distance. This effect possibly raises problems in the case when cylindrical conductors are arranged at a highly short distance like in compact and large capacity GIS. For example, unexpected local concentration of an electric current may cause welding or arcing especially in contact joint areas. Therefore, detailed study on the proximity effect is important. In this paper, it is shown that this effect can be quantitatively evaluated with an experiment and two kind of computations, finite element method and approximated lumped parameter circuit. Finally, ideas for improvement of GIS arrangement are suggested.

Soft fingertips can easily achieve high friction contacts but their frictional moment around the contact normals makes manipulation control difficult. We propose a control scheme to manipulate an object by a multifingered hand with soft fingertips. The proposed scheme cancels the frictional moment according to the direction of the relative rotation between the fingertip and the object. Whatever the fingertip contact type is, accurate pose measurement is very important for the object manipulation. We also propose a systematic method to estimate the object pose with a complementary use of camera images and finger joint information. The proposed control scheme and the pose estimation method were combined and implemented in an experimental system. Results of the experiment show the effectiveness of the proposed pose estimation method and the overall control system for manipulating an object by a soft-fingered hand.

The proposed method uses a measurement system to determine the parameters of the fundamental and harmonic equivalent circuits of the motor. It is shown that, by considering the frequency characteristics of rotor parameters and the magnetic saturation effect on the exciting inductance, one can predict the starting performance quantitatively. >

An electron beam direct imaging system that can image electric circuit patterns directly on printed wiring boards (PWBs) from computer-aided design (CAD) data without using film masks has been developed. The system realizes beam deflection over a large field of 100 mm*100 mm. The deflection field can be imaged in less than 1 s with an accuracy of better than +or-30 mu m. The maximum deflection speed is 254 m/s, and the beam diameter in full width at half maximum is 40 mu m. This technology can use the same kinds of resists as those for conventional photolithography. It was shown that fine patterns with line widths less than 100 mu m can be imaged for PWBs 340 mm*400 mm in size by stitching deflection fields. >

The relation between EM (electromigration) characteristics and interlayer dielectric film structure is described. It was found that a well-planarized interlayer significantly increases EM resistance for underlying Al wirings. This EM improvement is attributed to reinforcement and crack suppression at interlayer sidewalls. A constant atom flux model is proposed that explains the mechanism for this phenomenon. >

A 3 kW class Stirling engine, the NSO3T, has been developed for a gas-fired residential heat pump. The specifications and mechanical system of the engine, which is a deformed inline design with two pistons, are described. A mechanical analysis, using both experimental and computer analysis results, is presented. Also discussed are an improvement of the engine mechanical system and engine endurance test results, which clarified the relationship between an increment of a mechanical loss and endurance time. >

An active-matrix liquid-crystal light valve (LCLV) for a high-resolution projector, using device layer transfer technology, is proposed. This LCLV will not require high-density interconnection between bus lines and Si IC drivers. To confirm the validity of this proposal, a high-performance poly-Si TFT (thin-film transistor) array with 480*768 pixels in 80 mm diagonal has been fabricated on an oxidized Si wafer. This TFT has a triple-gate structure for obtaining a high on/off current ratio and buried-isolated-pixel-electrode (BIP) structure and ITO barrier metal for reducing defects. This TFT array was transferred onto transparent glass substrate using device layer transfer technology, and the liquid-crystal projector was successfully realized. >

A plasma CVD (chemical vapor deposition) technology was developed that can deposit high-quality Ta/sub 2/O/sub 5/ thin film suitable for the stacked capacitor dielectric film of 64-Mb-DRAM (dynamic RAM) and beyond. Ta/sub 2/O/sub 5/ films as thin as 30 AA SiO/sub 2/ equivalent thickness deposited by this newly developed CVD technology have shown remarkably small leakage current, 1*10/sup -8/ A/m/sup 2/ at 3 MV/cm. The step coverage of the film is also satisfactory. Time-dependent dielectric breakdown (TDDB) measurements have shown that a polysilicon stacked capacitor using the Ta/sub 2/O/sub 5/ film (equivalent to 30 AA SiO/sub 2/) has a reliability corresponding to 3.3 V V/sub cc/ operation for 10 years and more. >

A path-induction navigation system with aural direction has been developed by using CD-ROM and aural information for versatile and safer navigation. CD-ROM ensures high-precision self-sustained navigation and provides multiple mobile information. In the several years since its first implementation, the mobile navigation system has been the core of mobile information system. The heart of system is a database containing information that does not change, stored on CD-ROM. >

In cellular radio systems, cochannel interference determines frequency reuse distance. To achieve high-capacity systems, cochannel interference reduction is considered a key item. Intrazone channel reassignment (IZCR) and adaptive transmitter power control (ATPC), are effective techniques to reduce cochannel interference. A novel IZCR algorithm is proposed that maximizes the number of channels that satisfy a required carrier-to-interference ratio (CIR). By computer simulations, the interference reduction effects by either IZCR or ATPC and by both ATPC and IZCR are clarified. With the call block rate of 0.03 and the outage probability of 0.01, the CIR margin can be reduced by equal to or more than 10 dB by adopting both ATPC and new IZCR. >

A location probability estimation method of service availability for portable units is proposed taking into consideration the given noise properties. A necessary carrier level or bit error rate at which quality is barely bearable is first defined. Location probabilities are estimated under the conditions of both Rayleigh fading and log-normal shadowing. Average carrier level required is generally larger in portables. Diversity reception is shown to be effective in reducing the required level. A forced fluctuation is suggested to be effective. >

An automatic system for GaAs-FET fine pattern defects has been developed to replace the current unreliable visual inspection by human eyes. The system uses a laser scanning microscope and information on both line width and reflected light intensity is used for the inspecting. A high-speed image processor has been developed to achieve a high inspection speed of 15 s per chip. The accuracy and the repeatability of the line width measurement are +or-0.09 mu m and +or-0.02 mu m. These values are good enough to be accepted in a production line, and the developed system has been put to practical use. >

A CMOS chip set for a 1.28-Gb/s 16*16 output buffer ATM switch has been developed. By operating at 160 MHz with 8-b parallel input/output, a line speed of 1.28 Gb/s is achieved. 20.5-Gb/s (1.28 Gb/s*16) time division multiplexing has been realized by the introduction of a time multiplexing serial to parallel converter (TMSPC) that consists of flip-flops and 2-1 selectors in a matrix formation. Two clock drivers are used in a one phase scheme with a minimum of propagation delay. ECL-compatible I/O buffers have been used to handle high-speed signals. All chips are developed using 0.8- mu m double-metal CMOS technology. >

The free ion yields from /sup 207/Bi conversion electrons were measured as a function of applied electric field using ionization chambers filled with tetramethylsilane (TMS) and tetramethylgermanium (TMG). The distributions for thermalization lengths of electrons were calculated by using Gaussian and exponential forms. The total free ion yields considering the impurity effect in liquid, and the thermalization lengths of electrons in TMS and TMG were determined to be 3.35+or-0.49, 3.42+or-0.56 and 165+or-18 AA, 185+or-24 AA, respectively. >

Recent progress in high power diode arrays include higher power output levels (>100 W/cm), higher efficiency (>50%), longer lifetimes (10/sup 11/ shots QCW, 10000 hrs CW), high temperature operation (>70/spl deg/C), and expanded wavelength coverage (670 nm to 2.0 /spl mu/m). In addition to advances in device performance, new package designs have been developed which provide for higher average power capability and optical systems have been developed to reformat the output in ways that are more convenient for many practical applications. >

Prototype continuous pellet injector has been designed and is under construction for the Large Helical Device at National Institute for Fusion Science in Japan. This injector is a centrifuge type, aiming at long-pulse pellet injection. A centrifuge pellet injector is a mechanical device that utilizes centrifugal force to accelerate solid hydrogen-isotope pellets using a high-speed rotating rotor. The injector consists of a hydrogen-isotope filament extruder, a pellet forming device, and a pellet accelerator. It is expected that pellets of 1 mm and 1.5 mm in diameter and length will be accelerated up to 800 m/sec in this system.

Object movement detection by high order correlation analysis of optical sensor array signals is introduced. The optical sensors observe the moving object surface which is assumed to be a non-uniform speckle-like texture. The measurement system is applicable to general robotic movement detection because: it employs a noncontact measurement method, the system can be made very compact, and it enables approximation of the movement trace with a sequence of arcs instead of the conventional connection of simple line segments. The authors looked into estimation of the running trace of an autonomous vehicle by observing the ground pattern. >

This paper presents an improved design scheme of MRAC for the electropneumatic servo system with additive external forces. In this design scheme, a difference process of the identification signals is adopted in the identification mechanism to eliminate the influence of the constant disturbances caused by the additive external forces. Furthermore, the controller of this design scheme is constructed with a compensator to overcome the influence of the disturbances. As a result, it can be possible to accomplish the satisfactory control performance.

This paper presents an approximate design scheme using delta-operator of multirate model reference adaptive control (MRAC) for an electropneumatic servo system. Since the discrete-time plant of the electropneumatic servo system often becomes nonminimum phase, it is impossible to construct the conventional design scheme of MRAC for such a plant. Adopting a delta-operator and neglecting the small coefficients of the plant, the approximate design scheme of MRAC is constructed to overcome the above problem. In this design scheme, the update-time of the controller parameters is chosen to be integral times the sampling-time to supply the lack of the update-time. Effectiveness of this design scheme is confirmed by experiments using the existent electropneumatic servo system.

Although a lot of people are interested in astronomical objects, it is difficult to teach astronomy in school. One of the reasons for the difficulty is that students can not observe astronomical objects in school. We consider that the Internet and its multimedia applications can resolve this problem. On November 19th, 1995, we had a very unusual astronomical phenomenon. We broadcasted the real time video of this astronomical phenomenon to the Internet. This experiment shows that the Internet has the advantages for classroom lessons in astronomy. We consider classroom lessons of astronomy using the Internet and its multimedia applications.

The recent trend of size reduction in various kinds of electronic equipment demands the high density mounting of high lead count, fine lead pitch components. Conventional hot-bar soldering causes problems such as solder bridges and weak bonds, and is not adequate for the mounting of fine pitch components like tape carrier packages (TCPs). We have developed an outer lead bonding (OLB) apparatus with a focused YAG laser beam for high quality TCP bonding. This apparatus solders leads one by one by scanning the focused YAG laser beam, which does not damage the printed circuit board (PCB). The apparatus uses two galvano mirrors to rapidly scan the laser beam and a flat-field lens (f/spl theta/ lens) to focus the laser beam to as small as the lead width regardless of the deflection angle. A through the lens (TTL) monitoring system, consisting of the above galvano mirrors and fB lens, is used to monitor the laser beam irradiation area. This construction ensures the monitored area coincides with the irradiation area at all times, and facilitates the precise directing of the laser beam on a single lead and its pad on PCBs. The system measures the displacement between the lead and matching pad by image processing and when the displacement is larger than one third of the lead width, soldering is not conducted. The test results show that the apparatus has realized high quality soldering with no-clean flux for TCPs having 576 leads and a 250 /spl mu/m pitch. >

We report 1 cm wide monolithic arrays of semiconductor laser diodes emitting output powers up to 90 W CW. Bars with lower fill factors were found to fail by catastrophic optical damage (COD) whereas the highest fill factor bars were limited in output power by heating effects.

A novel architecture for an SRAM-based user-reprogrammable gate array which consists of programmable logic elements (PLEs), switching stations (SSs), wirings, and input/output blocks (IOBs) is discussed. The SS is designed to connect its neighboring PLEs and/or IOBs through only one NMOS pass transistor to maintain a sufficient signal level and reduce the delay in signal propagation. Hidden interconnection networks which directly merge neighboring PLEs were adopted to expand the number of inputs and product terms of combinational logic, and allow the construction of a parallel-to-serial/serial-to-parallel converter without using SSs. A fabricated CMOS prototype chip includes an 8 Kbit 50 ns temporary storage memory which is programmable to the random access mode or the FIFO mode with a word length of 4 bit or 8 bit. >

A 40000 RPM vector-controlled induction motor drive is introduced. In a previous paper by T. Kume et al. (see Proc. IPEC 1990 conf., Tokyo, p.1132-1136, 1990), conventional vector control with a shaft encoder was employed in a low-speed range, while constant V/f (volt-per-hertz) PWM control was employed in a high-speed range. The two control modes were switched at approximately 20000 RPM. By employing phase compensation and instantaneous current limit, the dead-time between two modes was minimized. This technique is further improved by the authors. The constant V/f PWM control in the high-speed range is replaced by vector control without an encoder. The current control loop is not interrupted for mode changing, and thus continuous operation over the entire range is established. >

Summary form only given. Laser opto-microactuators are discussed in this study. The actuators are rotated by the effects of molecular gas dynamics such as the forces by a thermal creep force and a radiometric force. The effects are induced by heating of the rotor surface of the actuators with a laser beam. It is elucidated that the maximum torque of the actuators occurs at the Knudsen number of about 0.2 and the maximum rotational rate of the actuators is at the Knudsen number of 0.5. The Knudsen number is defined as the ratio of the mean free path length of the gas in the chamber to the rotor blade length of the actuators.

We propose a design scheme which combines model reference adaptive control with a neural network for the electropneumatic servo system. In this design scheme, the role of the neural network is to compensate for constructing a linearized model of the nonlinear plant of the electropneumatic servo system. The role of the robust adaptive controller is to perform the model-matching for the uncertain linearized model of the plant to a given linear reference model. The effectiveness of the proposed design scheme is confirmed by experiments using the existent electropneumatic servo system.

For the development of energy saving processes for producing thermoelectric materials, the PIES (Pulverized and Intermixed Elements Sintering) method is known as a suitable one. In this study, a high figure of merit p-type Bi-Sb-Te based thermoelectric material was obtained by the oxygen free PIES method. The electrical resistivity of p-type (Bi/sub 2/Te/sub 3/)/sub 0.175/(Sb/sub 2/Te/sub 3/)/sub 0.825/ material was reduced to 0.8/spl times/10/sup -5/ ohm m and Seebeck coefficient was obtained 2.4/spl times/10/sup -4/ V/K by fabrication in the oxygen free atmosphere process. The maximum ZT value was almost 1.0. Moreover, it was demonstrated that the improvement in figure of merit value was obtained even if the intermixed powder of elements was prepared by a conventional ball milling process in stead of the high energy ball milling process. The PIES methods using a low energy ball milling process is suitable for mass-production because of its cost efficiency.

This paper presents the applicability of a commercially available thermoelectric generator for waste heat recovery in a processing plant of combustible solid waste. Oil heat transfer medium is utilized for heating the commercially available thermoelectric generator employed in this study so that the generator can be operated at a much lower pressure than that using water. Low pressure operation is not only reliable, but it should also be safe and have low maintenance cost. Each generator provides more than 500 W and achieves a relatively high efficiency with a temperature of 220/spl deg/C hot oil and around 30/spl deg/C cold water. The result of the survey of a fitted position of a thermoelectric generator to a processing plant of combustible solid waste is summarized in this paper. Generator economics and goals of a thermoelectric generator to produce power at similar rates to those of a steam turbine are also summarized in order to discuss the applicability of a commercially available thermoelectric generator.