Power Density Limits Research Articles

Co‐operating retrodirective system

Practical demonstration of the operational advantages gained through the use of a co-operating retrodirective array (RDA) basestation and Van Atta node arrangements is discussed. The system exploits a number of inherent RDA features to provide analogue real time multifunctional operation at low physical complexity. An active dual-conversion four element RDA is used as the power distribution source (basestation) while simultaneously achieving a receive sensitivity level of −109 dBm and 3 dB automatic beam steering angle of ±45°. When mobile units are each equipped with a semi-passive four element Van Atta array, it is shown mobile device orientation issues are mitigated and optimal energy transfer can occur because of automatic beam formation resulting from retrodirective self-pointing action. We show that operation in multipath rich environments with or without line of sight acts to reduce average power density limits in the operating volume with high energy density occurring at mobile nodes sites only. The system described can be used as a full duplex ASK communications link, or, as a means for remote node charging by wireless means, thereby enhancing deployment opportunities between un-stabilised moving platforms.

Package design and development of a low cost high temperature (250°C), high current (50+A), low inductance discrete power package for advanced Silicon Carbide (SiC) and Gallium Nitride (GaN) devices

The demands for high-performance power electronics systems are rapidly surpassing the power density, efficiency, and reliability limitations defined by the intrinsic properties of silicon-based semiconductors. The advantages of post silicon materials, including Silicon Carbide (SiC) and Gallium Nitride (GaN), are numerous, including: high temperature operation, high voltage blocking capability, extremely fast switching, and superior energy efficiency. These advantages, however, are severely limited by conventional power packages, particularly at temperatures higher than 175°C and >100 kHz switching speeds. In this discussion, APEI, Inc. presents the design of a newly developed discrete package specifically intended for high performance, high current (>50A), rapid switching, and extended temperature (>250°C) wide band gap devices which are now readily available on the commercial market at voltages exceeding 1200V. Finite element analysis (FEA) results will be presented to illustrate the modeling process, design tradeoffs, and critical decisions fundamental to a high performance package design. A low profile design focuses on reducing parasitic impedances which hinder high speed switching. A notable increase in the switching speed and frequency reduces the size and volume of associated filtering components in a power converter. Operating at elevated temperatures reduces the requirements of the heat removal system, ultimately allowing for a substantial increase in the power density. Highlights of these packages include the flexibility to house a variety of device sizes and types, co-packaged antiparallel diodes, a terminal layout designed to allow rapid system configuration (for paralleling or creating half- and full-bridge topologies), and a novel wire bondless backside cooled construction for lateral GaN HEMT devices. Specific focus was placed on minimizing the cost of the materials and fabrication processes of the package components. The design of the package is discussed in detail. High temperature testing of a SiC assembly and electrical test results of a high frequency GaN based boost converter will be presented to demonstrate system level performance advantages.

Optimization of a hybrid diesel-wind generation plant with operational options

Hybrid diesel-wind power generation systems have a great potential in the application of providing energy supply for remote communities and facilities. Compared to the traditional diesel system, hybrid power plants can offer many advantages such as additional capacity, being more environmentally friendly, potential reduction of cost, etc. One of the problems for the hybrid system is the uncertain profit due to the uncertain nature of the wind. In this paper a general model is presented based on real option theory for valuating a hybrid diesel-wind generation plant. A dynamic programming method is used to generate the optimum operational option by maximizing the net cash flow of the plant. Results show that operational options can provide additional value to the hybrid power system when this operational flexibility is correctly utilized. The optimal scale of wind turbines generators (WTGs) can be calculated as a consequence of the limitation of demand and local wind power density. This paper also provides a framework to find the optimal operating decision at each time step based on the real option model.

High Temperature Silicon Carbide Power Modules for High Performance Systems

The demands of modern high-performance power electronics systems are rapidly surpassing the power density, efficiency, and reliability limitations defined by the intrinsic properties of silicon-based semiconductors. The advantages of silicon carbide (SiC) are well known, including high temperature operation, high voltage blocking capability, high speed switching, and high energy efficiency. In this discussion, APEI, Inc. presents two newly developed high performance SiC power modules for extreme environment systems and applications. These power modules are rated to 1200V, are operational at currents greater than 100A, can perform at temperatures in excess of 250 °C, and are designed to house various SiC devices, including MOSFETs, JFETs, or BJTs.

Superfocusing of mutimode semiconductor lasers and light-emitting diodes

The problem of focusing multimode radiation of high-power semiconductor lasers and light-emitting diodes (LEDs) has been studied. In these sources, low spatial quality of the output beam determines theoretical limit of the focal spot size (one to two orders of magnitude exceeding the diffraction limit), thus restricting the possibility of increasing power density and creating optical field gradients that are necessary in many practical applications. In order to overcome this limitation, we have developed a method of superfocusing of multimode radiation with the aid of interference. It is shown that, using this method, the focal spot size of high-power semiconductor lasers and LEDs can be reduced to a level unachievable by means of traditional focusing. An approach to exceed the theoretical limit of power density for focusing of radiation with high propagation parameter M2 is proposed.

Miniaturized concentration cells for small-scale energy harvesting based on reverse electrodialysis

We describe experimental and theoretical results that demonstrate the feasibility of power generation using concentration cells based on ionic concentration gradients and reverse electrodialysis. A peak power density of 0.2 (0.7) μW cm−2 and a maximum energy density of 0.4 (0.4) mJ cm−3 delivered in 3 h to a 2 (5) kΩ resistor were recorded using a microfiltration (anion exchange) membrane, respectively. A comprehensive model is developed to predict the evolution of the output voltage with time in relation to the solute concentration in each cell and derive the power density and efficiency limits.

Customizable Domain-Specific Computing

To meet computing needs and overcome power density limitations, the computing industry has entered the era of parallelization. However, highly parallel, general-purpose computing systems face serious challenges in terms of performance, energy, heat dissipation, space, and cost. We believe that there is significant opportunity to look beyond parallelization and focus on domain-specific customization to bring significant power-performance efficiency improvement.

Near IR laser light visualizators using nonlinear GaSe and other layered crystallites

Two methods of preparation of the devices for visualization of pulsed and continuous near-IR (near infrared) are described and the results of conversion of pulsed and continuous IR (800–1360 nm) laser radiation into the visible range of spectra (400–680 nm) by using a transparent substrate covered with the particles (including nanoparticles) of effective nonlinear materials of GaSexS1 − x (0.2 ≤ x ≤ 0.8) are presented. Converted light can be detected in transmission or reflection geometry as a visible spot corresponding to the real size of the incident laser beam. Developed device structures can be used for checking if the laser is working or not, for optical adjustment, for visualization of distribution of laser radiation over the cross of the beam and for investigation of the content of the laser radiation. Low energy (power density) limit for visualization of the IR laser pulses with 2–3 ps duration for these device structures are: between 4.6–2.1 μJ (3 × 10−4−1 × 10−4 W/cm2) at 1200 nm; between 8.4–2.6 μJ (4.7 × 10−4−1.5 × 10−4 W/cm2) at 1300 nm; between 14.4–8.1 μJ (8.2 × 10−4–4.6 × 10−4 W/cm2) at 1360 nm. Threshold damage density is more than 10 MW/cm2 at λ = 1060 nm, pulse duration τ = 35 ps. The results are compared with commercially existing laser light visualizators.

Optical technologies for data communication in large parallel systems

Large, parallel systems have greatly aided scientific computation and data collection, but performance scaling now relies on chip and system-level parallelism. This has happened because power density limits have caused processor frequency growth to stagnate, driving the new multi-core architecture paradigm, which would seem to provide generations of performance increases as transistors scale. However, this paradigm will be constrained by electrical I/O bandwidth limits; first off the processor card, then off the processor module itself. We will present best-estimates of these limits, then show how optical technologies can help provide more bandwidth to allow continued system scaling. We will describe the current status of optical transceiver technology which is already being used to exceed off-board electrical bandwidth limits, then present work on silicon nanophotonic transceivers and 3D integration technologies which, taken together, promise to allow further increases in off-module and off-card bandwidth. Finally, we will show estimated limits of nanophotonic links and discuss breakthroughs that are needed for further progress, and will speculate on whether we will reach Exascale-class machine performance at affordable powers.

High Temperature Silicon Carbide Power Modules for High Performance Systems

The demands of modern high-performance power electronics systems are rapidly surpassing the power density, efficiency, and reliability limitations defined by the intrinsic properties of silicon-based semiconductors. The advantages of silicon carbide (SiC) are well known, including high temperature operation, high voltage blocking capability, high speed switching, and high energy efficiency. In this discussion, APEI, Inc. presents two newly developed high performance SiC power modules for extreme environment systems and applications. These power modules are rated to 1200V, are operational at currents greater than 100A, can perform at temperatures in excess of 250 °C, and are designed to house various SiC devices, including MOSFETs, JFETs, or BJTs. One newly developed module is designed for high performance, ultra-high reliability systems such as aircraft and spacecraft, and features a hermetically sealed package with a ring seal technology capable of sustaining temperatures in excess of 400°C. The second module is designed for high performance commercial and industrial systems such as hybrid electric vehicles or renewable energy applications, implements a novel ultra-low parasitic packaging approach that enables high switching frequencies in excess of 100 kHz, and weighs in at just over 130 grams (offering ~5× mass reduction and ~3× size reduction in comparison with industry standard power brick packaging technology). It is configurable as either a half or full bridge converter. In this discussion, APEI, Inc. introduces these products and presents practical testing of each.

Architectures, Components, and Subsystems for Future Optical Packet Switches

Rapidly increasing network traffic is posing a challenge to the construction of future routers. While high-capacity transport has kept pace with rising traffic demands through the use of dense wavelength-division multiplexing, the scaling of core routers is slowed by power density limits and complexity and interconnectivity issues. Optical switching has the potential to overcome these scaling restrictions, and as a result, has generated great scientific and commercial interest. In this paper, we present an overview of some optical packet-switching architectures and describe components and subsystems that are required to enable this technology.

Towards 100G packet processing: Challenges and technologies

Driven by media-rich and bandwidth-intensive Internet applications, 100 Gigabit Ethernet embodies the next logical and necessary line speed following 10G/40G, although enormous challenges exist. Can today's device technologies scale, or will new, disruptive approaches be required to overcome throughput and power density limitations? The evolution of field programmable gate array (FPGA) and application-specific integrated circuit (ASIC) technology was analyzed concerning speed, density, power, and pin interfacing. While clock frequencies cannot keep up with the higher interface speeds, massive parallelism and deep pipelining are needed to scale the throughput. The state-of-the-art architectures and algorithms for every aspect of packet processing are described. In addition, we look at alternative memory concepts and cover some emerging technologies: asynchronous FPGAs as a means for boosting the system clock and serial interfaces reducing the pin count between devices. Thereby, economic considerations limit the choice between the options. We conclude that although significant effort is necessary in terms of device and board technology, economic 100G networking is viable.

Impact of EMC Filters on the Power Density of Modern Three-Phase PWM Converters

Electromagnetic compatibility (EMC) filters are typically included in offline power converters, controlling electromagnetic emissions, but adding volume to power electronic systems. During the last decades, one of the main objectives of the power electronics industry has been the increase of the power density. Thus, it is reasonable to analyze how filters affect power density values, imposing limits or ldquobarriersrdquo to it. The impact of the EMC filters on the overall volume of three-phase pulsewidth modulation (PWM) converters is studied here for converters in the range of 5-10 kW. An analytical procedure based on the volume minimization of the EMC filters is proposed to estimate the total filter volume as function of the converters' rated power and switching frequency. With this, the minimum volume for EMC filters that allow the converters to comply with EMC standards regarding conducted emissions can be estimated and volume limitations identified. A discussion about the limits of power density for the considered three-phase PWM converters for state-of-the-art power semiconductors is performed, and optimum switching frequencies are identified. An experimental verification is carried out, which validates the achieved results.

Kinetics of light-cure bracket bonding: Power density vs exposure duration

Recent technologic advances make it possible to increase the light power density to reduce the necessary exposure duration. The kinetics of polymerization are complex. The special case of indirectly curing the thin layler of composite below the metallic bracket base further increases this complexity. It was hypothesized that the concept of "total energy,"--the reciprocity between power density and exposure duration--does not hold for orthodontic light-cure bracket bonding. Stainless steel brackets were bonded on deciduous bovine incisors with a standard light-cured composite. A calibrated, powerful halogen lamp allowed for modification of power density from 300 to 3000 mW/cm2. Metallic brackets were bonded in 8 groups of 20 incisors each with various combinations of power densities and exposure durations to obtain 3 levels of energy density (6000, 12000, and 24000 mJ/cm2). Another group of 20 incisors served as the positive control with a conventional powerful halogen lamp (1000 mW/cm2) for 40 seconds. After storage for 24 hours at 37 degrees C in water, the bracket shear bond strength (SBS) and the adhesive remnant index (ARI) were measured. It was confirmed that bracket SBS mainly depends on the energy density of the light cure. All groups with an energy density of 6000 mJ/cm2 had significantly lower SBS than the groups with higher energy densities (P <0.01). The dependence of SBS on exposure duration for the same energy density followed an exponential model of nonlinear regression (r2 = 0.98). The concept of "total energy" does not hold for orthodontic light-cure bracket bonding. An exposure time of less than 4 seconds, irrespective of the power density, cannot guarantee sufficient bracket bond strength. There seems to be an advantage of power density over exposure duration in the context of metallic bracket bonding. These results show that, for an efficient light-cure bracket bonding, there is an absolute lower limit of exposure duration (4 seconds) and an upper limit of useful power density (3000 mW/cm2).

Power Density and Efficiency of Thermophotovoltaic Energy Conversion Using a Photonic-Crystal Emitter and a &lt;formula formulatype="inline"&gt;&lt;tex&gt;$\hbox{2}$&lt;/tex&gt; &lt;/formula&gt;-D Metal-Grid Filter

Three-dimensional metallic photonic crystals have been suggested to be ideal candidates for thermal emitters in thermophotovoltaic energy conversion systems due to their high emission power in the allowed passband and also their emission suppression in the long-wavelength tail. In this paper, calculations were performed to compare a photonic-crystal emitter and a blackbody emitter in the thermodynamic limit of the maximum power density and efficiency that are achievable with an ideal photovoltaic cell. Additionally, realistic-source and GaSb photovoltaic-cell optical characteristics and a novel metal-grid filter are incorporated into the calculations. With an ideal antireflection coating, the measured photonic-crystal radiation was found to give a source electrical power density of 7.77 W/cm2 and a system efficiency of 47.58% with an effective source temperature of 1535 K.

Specific absorption rate assessment near a base-station antenna (2,140 MHz): some key points

In this paper, a numerical dosimetry study for a typical base-station antenna used in third generation system is presented. This study is based on the limits defined by the International Commission on Non-Ionizing Radiation Protection. The primary goal is to analyze the relationship between the antenna input powers required to reach the specific absorption rate (SAR) limit and those ones which are required to reach the power density limit. The secondary goal is to underline the key points which impact the local SAR in the case of such a directive antenna. Moreover, for these studies, two numerical approaches are used: the first one is only based on the finite difference time domain method and, in the second one, this algorithm is coupled with a method of moments simulator via a Huygens box. Investigations are made to ensure that a more suitable numerical approach is used.

PWM Converter Power Density Barriers

Power density of power electronic converters in different applications has roughly doubled every 10 years since 1970. Behind this trajectory was the continuous advancement of power semiconductor device technology allowing an increase of converter switching frequencies by a factor of 10 every decade. However, today's cooling concepts, and passive components and wire bond interconnection technologies could be major barriers for a continuation of this trend. For identifying and quantifying such technological barriers this paper investigates the volume of the cooling system and of the main passive components for the basic forms of power electronics energy conversion in dependency of the switching frequency and determines switching frequencies minimizing the total volume. The analysis is for 5 kW rated output power, high performance air cooling, advanced power semiconductors, and single systems in all cases. A power density limit of 28 kW/dm3@300 kHz is calculated for an isolated DC-DC converter considering only transformer, output inductor and heat sink volume. For single-phase AC-DC conversion a general limit of 35 kW/dm3 results from the DC link capacitor required for buffering the power fluctuating with twice the mains frequency. For a three-phase unity power factor PWM rectifier the limit is 45 kW/dm3@810 kHz just taking into account EMI filter and cooling system. For the sparse matrix converter the limiting components are the input EMI filter and the common mode output inductor; the power density limit is 71 kW/dm3@50 kHz when not considering the cooling system. The calculated power density limits highlight the major importance of broadening the scope of research in power electronics from traditional areas like converter topologies, and modulation and control concepts to cooling systems, high frequency electromagnetics, interconnection technology, multi-functional integration, packaging and multi-domain modeling and simulation to ensure further advancement of the field along the power density trajectory.

Neutronic and thermo-mechanic calculations for the design of the TRADE spallation target

The spallation target of the TRADE Project—at present in stand-by waiting for further decisions on budget availability—is based on the concept of a solid target cooled by forced convection. The solid target approach presents a maximum power density limitation due to the high thermal stresses which are generated by the huge differential temperatures in the material. These at their turn stem from the proton energy loss in the volume of the target material and the water cooling on the external surface. The maximum operating beam power and the proton beam profile were imposed by experimental and geometrical considerations; the peak power density of the target material (Tantalum) turned out to be about 8E09 W/m 3. The power generation of the reference configuration of the spallation target was accurately calculated by the code MCNPX, which allows to simulate individual particle tracks in 3-D geometry, starting from the accelerated 140 MeV protons down to all the relevant secondary particles. The temperature distribution and the stress and strain maps were calculated by the thermo-mechanical ANSYS code. The main calculation results are reported. The elastic-plastic analysis of the target material is performed in terms of maximum allowable deformation, fatigue and progressive plastic deformation.

ARE WE AT THE END OF CMOS SCALING?

CMOS scaling enabled by advances in lithography has been behind the information revolution. Over the last 15 years, there has been a new CMOS technology node approximately every two years. The key feature of every node has been 2X density shrink and ~35% performance gain per technology node. At 90 nm node a number usual knobs that have enabled the scaling have approached their limits. Furthermore chip power (both active and stand-by) has been increasing rapidly, approaching air cool limit. Chip stand-by power, which was negligible a few years ago, is now about the same order of magnitude as the active power in high end microprocessors. In this talk it will be argued that because of power density limitation of 90 nm, 65 nm, and beyond nodes, performance and ability to shrink are more than ever linked, and in fact if the performance gain would significantly slow down (for the designs that operate at the existing cooling limit). It is more than ever critical to come up with technology features that will enhance the performance, at a given device leakage.

On whitespace and stability in physical synthesis

In the context of physical synthesis, large-scale standard-cell placement algorithms must facilitate incremental changes to layout, both local and global. In particular, flexible gate sizing, net buffering and detail placement require a certain amount of unused space in every region of the die. The need for “local” whitespace is further emphasized by temperature and power-density limits as well as the increasing use of buffered interconnect. Another requirement, the stability of placement results from run to run, is important to the convergence of physical synthesis loops. Indeed, logic re-synthesis targeting local congestion in a given placement or particular critical paths may be irrelevant for another placement produced by the same or a different layout tool. In this work we offer solutions to the above problems. We show how to tie the results of a placer to a previously existing placement, and yet leave room for optimization. In our experiments this technique produces placements with similar congestion maps. We also show how to trade off wirelength for routability by manipulating whitespace. Empirically, our techniques improve circuit delay of sparse layouts in conjunction with physical synthesis. Our proposed techniques can be implemented using existing commercial placement tools without source code modifications and with modest overhead. They can also be integrated directly into min-cut placers with negligible overhead. We consider in particular detail the problem of scaling existing IP blocks to increase their porosity. Indeed, the need for additional repeater insertion when migrating a block to a newer process node often implies re-optimizing the layout. Our techniques for achieving placement stability allow one to rescale an existing layout with different minimum local whitespace requirements. In contrast to current ECO techniques, our rescaling method is not restricted to small changes of the netlist and layout, but will attempt to keep the relative placements similar if that is possible.