• Proceedings of the International Microelectronics And Packaging Society Conference
• /
• 2002.09a
• /
• pp.219-225
• /
• 2002

This paper presents a novel "bumpless flip chip package"for cost! performance driven devices. Using the conventional electroplating and etching processes, this package enables the production of fine pitch BGA up to 256 I/O with single layer routing. An array of circuitry down to $25-50{\mu}{\textrm}{m}$ line/space is fabricated to fan-in and fan-out of the bond pads without using bumps or substrate. Various types of joint methods can be applied to connect the fine trace and the bond pad directly. The resin-filled terminal provides excellent compliancy between package and the assembled board. More interestingly, the thin film routing is similar to wafer level packaging whereas the fan-out feature enables high lead count devices to be accommodated in the BGA format. Details of the design concepts and processing technology for this novel package are discussed. Trade offs to meet various cost or performance goals for selected applications are suggested. Finally, the importance of design integration early in the technology development cycle with die-level and system-level design teams is highlighted as critical to an optimal design for performance and cost.

• Journal of the Korean Society for Heat Treatment
• /
• v.36 no.3
• /
• pp.137-144
• /
• 2023

Micro electro mechanical systems (MEMS) and precision machines require excellent friction and wear characteristics to improve energy efficiency generated during sliding motion. In this study, DLC thin film with high hardness and low friction was deposited on STS304 substrate material by CVD method, and dot-shaped convex and concave micropatterns were fabricated by photolithography process. The diameter of the pattern was 20 ㎛, the pitch was 40 ㎛, and a pattern having a mesh type arrangement was fabricated and an abrasion test was performed. The results of the wear test on the micro pattern confirmed that the friction coefficient characteristics were improved compared to STS 304 and DLC thin films. In addition, in this result, the micro-pattern showed 11.4% more improved friction coefficient than the DLC thin film. The friction coefficient characteristics for convex and concave patterns of the same size showed almost similar results.

• Annual Conference of KIPS
• /
• 2008.05a
• /
• pp.73-75
• /
• 2008

현대는 시뮬레이션을 통하여 많은 모의 훈련을 대체하고 있다. 시뮬레이터는 고유의 훈련 임무를 수행하기 위하여 기획, 제작 되어진다. 하지만 현대와 같은 종합적인 입체훈련 체계는 고유의 훈련 임무 이외에 개개의 시뮬레이터의 임무가 가상의 종합 시뮬레이션 환경에 참가한 통합 훈련 체계의 필요성을 야기하게 되었다. 또한 각각의 시뮬레이터의 기능은 보다 세분화 되었으며 높은 성능을 요구하게 되었다. 현재 KA-32 헬기 시뮬레이터 영상 프로그램을 제작하고 있는 한국소프트스페이스(주)는 독립적인 시뮬레이터들에 HLA(High Level Architecture)를 적용하여 하나의 종합 훈련 시스템을 구축하였다. 고유의 임무를 지닌 시뮬레이테는 각각의 임무를 가상의 종합 훈련 환경에 참가하여 서로간의 훈련 효과를 극대화 시켜화 시켜준다. HLA 기반의 종합 훈련 시스템의 RTI(Run-Time Infrastructure)는 Pitch 사의 pRTI를 사용하였으며, 가상의 종합 시뮬레이션 환경의 핵심이 되는FOM(Federation Object Model)은 호환성을 및 모든 객체의 표현을 위하여 RPR FOM(Real-time Reference FOM)을 사용하였다. 본 종합 훈련 시스템에서는 각각의 시뮬레이션간에 최고의 성능을 내기 위하여, 기능 세부화 하여 구성하였다. Terrion 사의 FLAMES는 종합 훈련 시스템의 시뮬레이션 진행/설정 및 분석을 담당하며, Mantis는 훈련 상황을 사실감 있는 3 차원 영상으로 참가자에게 전달하여 준다.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.8
• /
• pp.427-434
• /
• 2008

In this paper, a transcoding algorithm for SMV and AMR speech coder is proposed. In the application requiring the interoperability of different networks, two speech coders must work together with the structure of cascaded connection, tandem. The tandem which is one of the simplest methods has several problems such as long delay, high complexity and the quality degradation due to twice complete encoding/decoding process. These problems can be solved by using transcoding algorithm. The proposed algorithm consists of LSP (Line Spectral Pair) conversion, pitch delay conversion, and fast fixed codebook search. The evaluation results show that the proposed algorithm achieves equivalent speech quality to that of tandem with reduced computational complexity and delay.

• Advances in nano research
• /
• v.17 no.3
• /
• pp.285-291
• /
• 2024

The paper considers one of the new applications of computer methods in music composition, using smart nanobeams-an integration of advanced computational techniques with new, specially designed materials for enhanced performance capabilities in music composition. The research applies some peculiar properties of smart nanobeams, embedded with piezoelectric materials that modulate and control sound vibrations in real-time. The study is conducted to determine the effects of changes in the length, thickness of nanobeams and the applied voltage on acoustical properties and the tone quality of musical instruments with the help of numerical simulations and optimization algorithms. By means of piezo-elasticity theory, different governing equations of nanobeam systems can be derived, which are solved by the numerical method to predict the dynamic behavior of the system under different conditions. Results show that manipulation of the parameters allows great control over pitch, timbre, and resonance of the instrument; such a system offers new ways in which composers and performers can create music. This research also validates the computational model against available theoretical data, proving the accuracy and possible applications of the former. The work thus marks a large step towards the intersection of music composition with smart material technology, and, when further developed, it would mean that smart nanobeams could revolutionize the process for composing and performing music on these instruments.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.20 no.3
• /
• pp.304-311
• /
• 2014

In this study, to apply hydrogen energy to ship engine and to generate effective hydrogen production, we investigated the effects of high temperature $H_2SO_4$ feed rate and cooling water rate to pump parts with fixed frequency needed to reciprocate motion and a simulation was conducted at each condition. In the fixed frequency and cooling water inlet flow rate of 0.5 Hz and 3.9 kg/s, we changed the high temperature $H_2SO_4$ flow rate to 47.46 kg/s (it is 105 % of 45.2 kg/s), 49.72 kg/s (110 %), and 51.98 kg/s (115 %). Also, at 0.5 Hz and 45.2 kg/s of frequency and high temperature $H_2SO_4$ flow, the thermal hydraulic analysis was performed at the condition of 95 % (3.705 kg/s), 90 % (3.51 kg/s), and 85 % (3.315 kg/s). In overall simulation cases, the physical properties of materials are more influential to the temperature increase in the pump part rather than the changes on the feed rate of high temperature $H_2SO_4$ and cooling water. A continuous operation of pump was also capable even if the excess feed of high temperature $H_2SO_4$ of about 15 % or the less feed of cooling water of about 15 % were performed, respectively. When the increasing feed of high temperature $H_2SO_4$ of up to 5 %, 10 %, and 15 % were compared with base flow (45.2 kg/s), the deviation of time period rose to a certain temperature and ranged from 0 to 4.5 s in the same position (same material). In case of cooling water, the deviation of time period rose to a certain temperature and ranged from 0 to 5.9 s according to the decreasing feed changes of cooling water at 5 %, 10 %, and 15 % compared to a base flow (3.9 kg/s). Finally, the additional researches related to the two different materials (Teflon and STS for Pitch and End-plate), which are concerned about the effects of temperature changes to the parts contacting different materials, are needed, and we have a plan to conduct a follow-up study.

• Journal of the Microelectronics and Packaging Society
• /
• v.7 no.4
• /
• pp.23-29
• /
• 2000

We demonstrate the fabrication method of high-density and high-quality solder bump solving a copper (Cu) cross-contamination in Si-LSI laboratory. The Cu cross-contamination is solved by separating solder-bump process by two steps. Former is via-formation process excluding Cu/Ti under ball metallurgy (UBM) layer sputtering in Si-LSI laboratory. Latter is electroplating process including Ti-adhesion and Cu-seed layers sputtering out of Si-LSI laboratory. Thick photoresist (PR) is achieved by a multiple coating method. After TiW/Al-electrode sputtering for electroplating and via formation in Si-LSI laboratory, Cu/Ti UBM layer is sputtered on sample. The Cu-seed layer on the PR is etched during Cu-electroplating with low-electroplating rate due to a difference in resistance of UBM layer between via bottom and PR. Therefore Cu-buffer layer can be electroplated selectively at the via bottom. After etching the Ti-adhesion layer on the PR, Sn/Pb solder layer with a composition of 60/40 is electroplated using a tin-lead electroplating bath with a metal stoichiometry of 60/40 (weight percent ratio). Scanning electron microscope image shows that the fabricated solder bump is high-uniformity and high-quality as well as symmetric mushroom shape. The solder bumps with even 40/60 $\mu\textrm{m}$ in diameter/pitch do not touch during electroplating and reflow procedures. The solder-bump process of high-uniformity and high-density with the Cu cross-contamination free in Si-LSI laboratory will be effective for electronic microwave application.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.1
• /
• pp.52-60
• /
• 2008

Purpose: To establish the methods for sinogram formation and correction in order to appropriately apply the filtered backprojection (FBP) reconstruction algorithm to the data acquired using PET scanner with multiple scintillation crystal layers. Materials and Methods: Formation for raw PET data storage and conversion methods from listmode data to histogram and sinogram were optimized. To solve the various problems occurred while the raw histogram was converted into sinogram, optimal sampling strategy and sampling efficiency correction method were investigated. Gap compensation methods that is unique in this system were also investigated. All the sinogram data were reconstructed using 20 filtered backprojection algorithm and compared to estimate the improvements by the correction algorithms. Results: Optimal radial sampling interval and number of angular samples in terms of the sampling theorem and sampling efficiency correction algorithm were pitch/2 and 120, respectively. By applying the sampling efficiency correction and gap compensation, artifacts and background noise on the reconstructed image could be reduced. Conclusion: Conversion method from the histogram to sinogram was investigated for the FBP reconstruction of data acquired using multiple scintillation crystal layers. This method will be useful for the fast 20 reconstruction of multiple crystal layer PET data.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.515-523
• /
• 2020

This study aims to improve the flow and noise performances of an axial-flow fan for cooling the machine room in a refrigerator by using airfoil-cascade analysis and surface ridge shape. First, the experimental evaluations using a fan performance tester and an anechoic chamber are performed to analyze the flow and noise performances of the existing fan system. Then, the corresponding flow and noise performances are numerically assessed using the Computational Fluid Dynamics (CFD) techniques and the Ffowcs-Williams and Hawkings (FW-H) equation, and the validity of numerical results are confirmed through their comparisons with the experimental results. The analysis for the flow of a cascade of airfoils constructed from the existing fan blades is performed, and the pitch angles for the maximum lift-to-drag ratio are determined. The improved flow performance of the new fan applied with the optimum pitch angles is confirmed. Then, the fan blades with surface ridges on their pressure sides are devised, and the reduction of aerodynamic noise of the ridged fan is numerically confirmed. Finally, the prototype of the final fan model is manufactured, and improvements in the flow and noise performances of the prototype are experimentally confirmed.

• Journal of Korean Society of Forest Science
• /
• v.66 no.1
• /
• pp.109-117
• /
• 1984

Allozyme study for open pollinated seeds of forty nine pitch pine families in an $F_1$-hybrid seed orchard demonstrated that allozyme variants in aspartate aminotransferase(AAT), glutamate dehydrogenase(GDH) and leucine aminopeptidase(LAP) systems are encoded by at Least eight loci; five for AAT, one for GDH and two fur LAP. Allozyme variations showed polymorphisms at seven of the eight loci, except GDH. Average number of alleles examined over six loci were 2.33 and 2.67 for maternal and progeny groups, respectively. Average heterozygosity and genetic diversity computed over six loci were, respectively, 0.235 and 5.409 for maternal tree group, 0.238 and 5.569 for progeny group. The proportion of $F_1$-hybrid seeds estimated by allozyme analysis was 0.77%. The estimated proportion of $F_1$ hybrid seeds by allozyme study is in good agreement with the value 0.73% estimated by morphological study for the proportion of pitch ${\times}$ Loblolly $F_1$ hybrid seedlings at a nursery. Indications for Wahlund effect, high levels of self-fertilization and for non-random matings in the $F_1$ hybrid seed orchard call for cautions in estimating allele frequency changes and mating probabilities for the parental and progeny groups.