Residual toughness of poly(acrylonitrile-butadiene-styrene) (ABS) after fatigue loading—effect of uniaxial fatigue loading

Toughness variation of non-notched poly(acrylonitrile-butadiene-styrene) (ABS) subjected to uniaxial fatigue loading was investigated. The experiments were conducted by applying fatigue loading to strip specimens first, from which dog-bone specimens were machined. The dog-bone specimens were tested to measure the strain for the on-set of fracture, named cracking strain here, thus to monitor the toughness change due to the fatigue loading.The test results showed that the fatigue loading caused the toughness drop in ABS, even before any visible crack was developed. Damage development and fracture behavior were then analyzed using scanning electron microscopy (SEM). The SEM analysis revealed that damage zones, not cracks, were initiated during the fatigue loading, and were the main cause of the toughness drop. Mechanisms for the damage initiation include matrix crazing and debonding of small rubber particles; however, large rubber particles remained intact. Based on the results, a deformation model is proposed for the damage zone initiation, which provides an explanation for the toughness change under the fatigue loading.     

Broadly wavelength-tunable external cavity lasers with extremely low power variation over tuning range

Tunable external-cavity lasers with low power variation over a broad tuning range are demonstrated using asymmetric multiple quantum-wells with a wide and flat gain. For a 2.8-/spl mu/m-wide ridge waveguide laser of cavity length 380 /spl mu/m, when used in a grating external cavity and with an antireflection coating on one facet of laser diode chip, the power variation of less than -1 dB is obtained over a range of 80 nm. This extremely low power variation is a direct result of the spectrally flat gain.     

W-band divide-by-3 frequency divider using 0.1 /spl mu/m InAlAs/InGaAs metamorphic HEMT technology

A W-band divide-by-3 frequency divider with wide bandwidth and low power dissipation is presented using harmonic injection-locking technique. A cascode FET is employed for a self-oscillating second-harmonic mixer which is injection-locked by third-harmonic input to obtain the division order of three. The fabricated frequency divider using 0.1 /spl mu/m GaAs metamorphic HEMT technology shows superior performance such as large bandwidth of 6.1 GHz around 83.1 GHz (7.3%) under small DC power consumption of 12 mW.     

2/spl times/4 optical add-drop module with no difference in propagation length

We developed a micromachined X-type 2/spl times/4 optical add-drop module (OADM) featuring no difference in propagation length. Four pairs of lensed fibers are aligned in "X" position, and four micromirrors are located between the pairs of optical fibers. The OADM was fabricated utilizing a silicon-on-insulator process. Electrostatic comb actuators can be driven up to 90 /spl mu/m to change the light path within 1 ms. The insertion loss and the on-off ratio were less than 3 and 70 dB, respectively. The loss uniformity in every channel was 1.5 dB.     

Instability of solid–liquid interface in transitional metal–carbide systems

The instability of solid–liquid interface (ISLI) during the liquid-phase sintering was studied using carbide–Ni composites. Of the various transitional metal carbides TiC of 4th period is the only carbide that exhibits a strong ISLI with negative curvatures in molten Ni. No ISLI was observed for other carbides in the 5th and 6th periods. The origin of ISLI is strain developed at the interface between the carbides and the newly formed solid solutions. The difference in the size of the atoms involved can be used to predict the formation of a carbide–Ni solid solution when the Hume-Rothery rules are applied. Aside from the size factor, other factors in the rules are not effective in predicting this phenomenon.     

Kernel Spectral Matched Filter for Hyperspectral Imagery

In this paper a kernel-based nonlinear spectral matched filter is introduced for target detection in hyperspectral imagery, which is implemented by using the ideas in kernel-based learning theory. A spectral matched filter is defined in a feature space of high dimensionality, which is implicitly generated by a nonlinear mapping associated with a kernel function. A kernel version of the matched filter is derived by expressing the spectral matched filter in terms of the vector dot products form and replacing each dot product with a kernel function using the so called kernel trick property of the Mercer kernels. The proposed kernel spectral matched filter is equivalent to a nonlinear matched filter in the original input space, which is capable of generating nonlinear decision boundaries. The kernel version of the linear spectral matched filter is implemented and simulation results on hyperspectral imagery show that the kernel spectral matched filter outperforms the conventional linear matched filter.     

Reduction of optical beat interference using gain-saturated RSOA in upstream WDM/SCM optical links

A new scheme for reducing optical beat interference (OBI) noise in optical network units is proposed for subcarrier multiplexing-based access network applications. The optical spectrum of the transmit lasers is broadened by using a radio frequency (RF) clipping tone with a modulation depth greater than one. This reduces the impact of the OBI noise. The distortions caused by an RF clipping tone are also suppressed by introducing a gain-saturated reflective optical amplifier, which shows the characteristics of high-pass filter. The proposed scheme has been verified by measuring the error vector magnitude of 16QAM signal with 20 Mbps. Error-free transmission has been achieved even when the light of OBI-noise-causing lasers is stronger than that of the signal laser by 7 dB     

Reversible energy recovery logic circuit without non-adiabaticenergy loss

The authors propose a reversible energy recovery logic (RERL) circuit for ultra-low-energy consumption, which consumes only adiabatic energy loss and leakage current loss by completely eliminating non-adiabatic energy loss. It is a dual-rail adiabatic circuit using the concept of reversible logic with a new eight-phase clocking scheme. Simulation results show that at low-speed operation, the RERL consumes much less energy than the complementary static CMOS circuit and other adiabatic logic circuits     

Numerical simulation of flying phenomena of the running tape on rotating drum

This paper analyzes the running mechanism of flexible and thin tape above rotating protrusion through a numerical simulation. The scope of analysis is confined to the phenomena of elastohydrodynamic lubrication between the rotating drum with a protrusion and the running tape. This model is based on the modified Reynolds equation and the equation of plate considering the effect of geometric nonlinearity and geometry of protrusion. Finite element method of Bubnov-Galerkin type is adopted as a numerical simulation technique to solve the above two coupled nonlinear equations. In numerical simulation, the influences of tape tension and protrusion velocity are evaluated in simple model. In complex models, the reciprocal action of two protrusions is simulated.     

Analysis of matrix damage evolution in laminated composite plates

A new model has been developed to investigate matrix cracking in laminated fibrous composite structures. The model can predict matrix cracking and its effect on stiffness reduction. It can also compute the load transfer from the cracked matrix to surrounding fibers. The model is based on the micromechanical concept of the fiber and matrix as well as the matrix material degradation concept as matrix cracking progressed. The micromechanical concept uses a rectangular cell geometry representing a fiber and its surrounding matrix while the material degradation concept uses an empirical expression of a Weibull type function. Two material constants are required for matrix cracking. The constants were obtained from an experiment on matrix cracking. With those constants, the present model predicts the matrix cracking in other cases. The predicted solutions were comparable to the experimental data.