Unconditionally stable tracking discriminator at 35 GHz

A two-mode tuning method for injection locked oscillators at microwave frequencies is presented. This technique allows one to modify the frequency response characteristics and stability boundary of an injection locked oscillator (ILO) in a favorable direction. Using this idea, a lock-in microwave discriminator has been proposed. Both a mathematical and experimental results on the stability properties, tracking zone, and other characteristics of the discriminator are presented     

Water modelling study of the jet characteristics in a continuous casting mould

The jet characteristics and the fluid flow pattern in a continuous slab caster have been studied using a water model. The fluid jet is studied under free fall and submerged discharge conditions. In the latter case, the jet was followed by dye-injection technique and image analyser was used to find out the effect of nozzle parameters on jet-spread angle, jet-discharge angle and the volume entrainment by the jet. All free-fall jets with nozzle port angle zero and upward are found to be spinning. Some of the free-fall jets with downward nozzle-port angle are found to be spinning and rest are smooth. The spinning direction of the jets are found to change with time. The well depth, port diameter and the inner diameter of the nozzle have a clear effect on the free-fall jets with downward port angle. The jet-spread angle is found to be about 17° for smooth jets. The spread angle for spinning jet increases as the nozzle-port angle is increased from downward 25 to upward 15°. The jet-discharge angle is always downward even when the nozzle-discharge ports are angled upward. The extent of volume entrainment by the spinning jet is higher and it increases as the nozzle-port angle is increased from 25 downward to 15° upward.     

A large-bias conduction model of polycrystalline silicon films

There exists a need for a large-bias conduction model of polysilicon films used in VLSI/ULSI and in high power integrated circuits. A large-bias conduction model has been developed by extending the emission-based models of Lu et al. (1983) and Mandurah et al. (1981) valid for small-bias, small-signal conditions. The following large-bias effects have been taken into account: (1) asymmetry of potential distribution around grain boundaries and (2) avalanche multiplication of carriers in the grain boundary layers at high electric fields. Since the exact nature of the grain boundary material is not yet known, and there is no direct method for determining the model parameters relating to grain boundaries, these were extracted by the parametric fitting of resistance versus temperature data of polysilicon resistors near room temperature with the above small-signal resistivity models modified by including Fermi-Dirac distribution. The model has been validated with experimental data on the current-voltage characteristics of ion-beam sputtered polysilicon resistors of different sizes and aspect ratios. The dependence of model parameters relating to grain boundary scattering and avalanche multiplication on the dimensions of resistors have been explained physically. The increased kink effect in polysilicon TFT's may also be predicted from the present theory. Some results on the I-V characteristics of polyresistors trimmed by high current pulses have been discussed qualitatively in the light of the present model. Although the model involves numerical integrations and a few iterations, it is reasonably fast in execution     

Verifying Relative Safety,Accuracy, and Termination for Program Approximations

Approximate computing is an emerging area for trading off the accuracy of an application for improved performance, lower energy costs, and tolerance to unreliable hardware. However, developers must ensure that the leveraged approximations do not introduce significant, intolerable divergence from the reference implementation, as specified by several established robustness criteria. In this work, we show the application of automated differential verification towards verifying relative safety, accuracy, and termination criteria for a class of program approximations. We use mutual summaries to express relative specifications for approximations, and SMT-based invariant inference to automate the verification of such specifications. We perform a detailed feasibility study showing promise of applying automated verification to the domain of approximate computing in a cost-effective manner.     

Kinetics of precipitation in Al-1 wt% Mn alloy

Precipitation kinetics in high purity Al-10 wt% Mn alloy has been investigated during the early stages of isothermal annealing between 823 and 698 K by resistivity measurements. Aged specimens were also examined by transmission electron microscopy. The precipitation kinetics can be represented by an Avrami type equation with the time exponentn=2/3 during the early stages of annealing. This can be attributed to the nucleation and growth of fiat needle-like precipitates on dislocations. The precipitates were Al₆ Mn at 773 K and the metastable G₂ -phase at 773 K. After prolonged annealing times at 848 K, isometric plates of Al₆Mn were also observed and the value ofn was found to be 1.0. However, at 698 K after longer annealing times, isometric needles of Al₁₂Mn, G-phase were observed along with flat needles of G₂ precipitate. At 698 K, the value ofn tends to decrease with time from 0.67 to 0.59.     

G-protein-coupled receptor microarrays

Membrane-bound proteins represent the single most important class of drug targets. Arraying these proteins is difficult because they typically need to be embedded in membranes to maintain their correctly folded conformations. We describe here the fabrication of microarrays consisting of G-protein-coupled receptors (GPCRs)--the single largest family of membrane-bound proteins-by robotic pin-printing on slides, and demonstrate assays for screening of ligands on these arrays.     

Kinetics of copper and high Pb/high Sn bilayered Pb-Sn solder interactions

Use of bilayered Pb-Sn solders consisting of high Sn and high Pb solder compositions is an option for joining chips to organic substrates at lower temperatures in which the high Sn solder is deposited onto Cu pads on the substrates. In this work interactions between the two-layered solder and copper pads during the reflow operation have been studied for both flip chip and Ball Grid Array (BGA) applications. It has been observed that Sn from the high Sn solder migrates faster at the edges along the surface of the high Pb solder than at the interior, resulting in a non-uniform Sn concentration along the Cu-solder interface. The thickness of the intermetallic compound formed due to the interaction of Cu and Sn has also been found to be non-uniform along the solder-Cu interface. This has been attributed to the variation in the Sn concentration of the solder adjacent to the Cu pads at different positions. The intermetallic compound growth rate has been explained using a model based on Sn diffusion into copper.     

Identifying frequent items in a network using gossip

We present algorithms for identifying frequently occurring items in a large distributed data set. Our algorithms use gossip as the underlying communication mechanism, and do not rely on any central control, nor on an underlying network structure, such as a spanning tree. Instead, nodes repeatedly select a random partner and exchange data with that partner. If this process continues for a (short) period of time, the desired results are computed, with probabilistic guarantees on the accuracy. Our algorithm for identifying frequent items is built by layering a novel small space “sketch” of data over a gossip-based data dissemination mechanism. We prove that the algorithm identifies the frequent items with high probability, and provides bounds on the time till convergence. To our knowledge, this is the first work on identifying frequent items using gossip.