TL;DR: In this article, the synchronization performance of CMOS circuits is examined theoretically and experimentally, and the phase characteristics of metastability are identified, and experimental measurements of error rate are made on a CMOS test circuit and the gain-bandwidth product for the circuit is determined.
Abstract: The synchronization performance of CMOS circuits is examined theoretically and experimentally. Criteria for maximizing CMOS gain are determined and are then compared with NMOS gain curves. The phase characteristics of metastability are identified. Experimental measurements of error rate are made on a CMOS test circuit, and the gain-bandwidth product for the circuit is determined from these data.
TL;DR: In this article, the authors present a new technique for production monitoring of VLSI thin film metallization reliability at wafer level by measuring the median energy-to-fail of properly designed structures during a ramp current stress.
Abstract: This paper presents a new technique for production monitoring of VLSI thin film metallization reliability at wafer level. By measuring the Median-Energy-to-Fail of properly designed structures during a ramp current stress, the long term electromigration reliability of metallization can be determined in a test taking a few seconds. Process variables such as metal width, thickness, step coverage, defect density, etc., have been monitored by measuring critical MEF in a production environment.
TL;DR: A comprehensive reliability program aimed at assessing and assuring the reliability of InGaAs photodiodes, and devices which are unaffected by the purge are shown by extended life tests to be robust devices which have a high probability of meeting TAT-8 requirements.
Abstract: A major objective of the TAT-8 submarine cable development program is the reliability assurance of its critical components. A relatively new device, the InGaAs photodiode, is used as the detection element in optical receivers and as the monitor of laser output in optical transmitters. In this paper, we describe a comprehensive reliability program aimed at assessing and assuring the reliability of InGaAs photodiodes. A major portion of this work has involved device operation at overstress conditions. Results to date indicate that for receiver photodiodes a device design exists which is predicted to meet a 1-FIT reliability objective. Tests of monitors of similar design are in progress. Critical to the success of the reliability assurance program is the ability to identify weak devices which are likely to fail early in the system life cycle. A conventional high-temperature burn-in is shown to be impractical, and it would not necessarily remove devices which fail by low activation energy processes. Overvoltage provides a means of accelerating such failure mechanisms. A “purge” using a combination of accelerants (high temperature and overvoltage) is shown to have considerable promise in failing weak devices, while inducing significant changes in those devices which subsequently fail early in a life test (elevated temperature at normal bias conditions). Devices which are unaffected by the purge are shown by extended life tests to be robust devices which have a high probability of meeting TAT-8 requirements.
TL;DR: In this article, a process for producing a printed circuit board characterized by forming a thermosetting resin layer on internal walls of through holes before an electroless plating treatment can produce printed circuit boards having high reliability without causing blow holes on the through hole walls.
Abstract: A process for producing a printed circuit board characterized by forming a thermosetting resin layer on internal walls of through holes before an electroless plating treatment can produce printed circuit boards having high reliability without causing blow-holes on the through hole walls.
TL;DR: The results show that the long-term degradation process is thermally accelerated, with a median activation energy of 1 eV and a standard deviation of 0.13 eV, which means that these 1.3-μm lasers have adequate life.
Abstract: To meet the stringent requirements of a submarine cable system, our 1.3-μm laser prequalification program has two objectives — first, to define the testing methodology that will accurately evaluate the potential reliability of the laser; and second, to obtain a preliminary indication of laser reliability on which the system configuration can be designed. Our testing methodology involves a combination of step-temperature, step-power, and isothermal testing over the temperature interval between 10 and 80°C and power levels of 1 to 5 mW per facet. Our results show that the long-term degradation process is thermally accelerated, with a median activation energy of 1 eV and a standard deviation of 0.13 eV. By using these activation energies, in conjunction with our measurements of degradation rates, we can project laser performance to 10°C, i.e., system operating temperature. It is estimated that the median time to failure for “light bulb” operation at 10°C is over 2 × 107 hours; and with 98-percent probability it is greater than 5 × 106 hours. Hence, when viewed strictly in terms of light bulb sources of stimulated power, these 1.3-μm lasers have adequate life. In addition, other potential operational malfunctions are being investigated, and they do not seem to change our basic conclusion about the usefulness of these 1.3-μm lasers for submarine cable application.
TL;DR: A redundant rocket engine for high reliability that incorporates a single thrust chamber and its attendent hydrogen and oxygen injectors and provides a pair of independent but indentical engine systems that provide the fuel feed and cooling systems where one system remains in a standby condition until a malfunction occurs while the other system remains operative as discussed by the authors.
Abstract: A redundant rocket engine for high reliability that incorporates a single thrust chamber and its attendent hydrogen and oxygen injectors and provides a pair of independent but indentical engine systems that provide the fuel feed and cooling systems where one system remains in a standby condition until a malfunction occurs while the other system remains operative. The independent systems are comprised of all of the movable parts.
TL;DR: In this article, the reliability of laser diodes developed for undersea optical transmission systems is analyzed and the median lifetime at 10°C is conservatively estimated to be 1.6 \times 10^{6} hours.
Abstract: The reliability of laser diodes developed for undersea optical transmission systems is analyzed. Up to 1000 device samples of two types (DC-PBH and VSB) are used. An aging test with constant light power operation of 5 mW is carried out at 10, 50, and 70°C for 10 000 h. The median lifetime at 10°C is conservatively estimated to be 1.6 \times 10^{6} hours. Failure rate at 10°C is estimated at 250 FIT's at 25 years of service for the wear-out failure mode and at less than 50 FIT's for the random failure mode with a sufficient margin. Furthermore, it is determined that reliability can be further improved by selection of long-life lasers through an additional third step screening aging.
TL;DR: In this paper, the authors developed a procedure to predict the quantitative effects of nonoperating periods on electronic equipment reliability and developed a series of non-operating failure rate prediction models at the component level.
Abstract: : The objective of this study was to develop a procedure to predict the quantitative effects of nonoperating periods on electronic equipment reliability A series of nonoperating failure rate prediction models were developed at the component level The models are capable of evaluating component nonoperating failure rate for any anticipated environment with the exception of a satellite environment The proposed nonoperating failure rate prediction methodology is intended to provide the ability to predict the component nonoperating failure rate and reliability as a function of the characteristics of the devices, technology employed in producing the device, and external factors such as environmental stresses which have a significant effect on device nonoperating reliability The prediction methodology is presented in a form compatible with MIL-HDBK-217 as an Appendix to the technical report Additional keywords: Dormancy; Power on off cycling; Tables(Data); Data acquisition; Mathematical models; Microcircuits; Discrete semiconductors; Resistors; Capacitors; Inductive devices; Lasers; Tubes; Mechanical/electromechanical devices; and Circuit interconnectors
TL;DR: The transition from crystalline modules to thin film modules is comparable to the transition from discrete transistors to integrated circuits as mentioned in this paper, but the package function and design will evolve to a lesser extent.
Abstract: The reliability of crystalline silicon modules has been brought to a high level with lifetimes approaching 20 years, and excellent industry credibility and user satisfaction. The transition from crystalline modules to thin film modules is comparable to the transition from discrete transistors to integrated circuits. New cell materials and monolithic structures will require new device processing techniques, but the package function and design will evolve to a lesser extent. Although there will be new encapsulants optimized to take advantage of the mechanical flexibility and low temperature processing features of thin films, the reliability and life degradation stresses and mechanisms will remain mostly unchanged. Key reliability technologies in common between crystalline and thin film modules include hot spot heating, galvanic and electrochemical corrosion, hail impact stresses, glass breakage, mechanical fatigue, photothermal degradation of encapsulants, operating temperature, moisture sorption, circuit design strategies, product safety issues, and the process required to achieve a reliable product from a laboratory prototype.
TL;DR: In this paper, the authors consider hot-electron induced MOSFET degradation is a serious reliability concern in submicron VLSI circuits and propose a solution to improve the reliability of these circuits by understanding its physical origin and manifestation in device characteristics.
Abstract: Hot-electron induced MOSFET degradation is a serious reliability concern in submicron VLSI circuits. Better understanding of its physical origin and manifestation in device characteristics has lead to improved device structure and processing; while better understanding of its dynamic behavior and resultant impact on circuit operation will be important in setting reasonable and realistic reliability goals. Device reliability is an integral part of and tightly coupled to device design in the submicron regime. The scenario for developing a successful technology must consider all facets of the technology including speed performance, process control and its reliability goal. The final optimization should always be based on specific requirements to achieve the best overall system performance.
TL;DR: An electric appliance control system was described in this article in which a logic controlled, solid-state, electronic system was used in conjunction with a mechanical switching device in the form of a cam disc operating a plurality of cam-operated switches.
Abstract: An electric appliance control system is disclosed in which a logic controlled, solid-state, electronic system is used in conjunction with a mechanical switching device in the form of a cam disc operating a plurality of cam-operated switches. The control system combines the capabilities of performing complicated and numerous operational programs with the reliability of the cam-operated mechanical switching device.
TL;DR: In this article, an improved monolithic, temperature compensated voltage-reference diode is realized by creating a tub of epitaxial semiconductor material in a substrate of opposite conductivity type and creating a voltage reference junction at a surface of the tub.
Abstract: An improved monolithic, temperature compensated voltage-reference diode is realized by creating a tub of epitaxial semiconductor material in a substrate of opposite conductivity type and creating a voltage reference junction at a surface of the tub. The junction between the tub and the substrate forms the forward-biased, temperature compensating junction of the device. The dopant concentration is varied during growth of the epitaxial material to provide a relatively low resistivity at the voltage-reference junction and a higher resistivity at the temperature compensating junction. The method described offers significant improvement over prior methods of manufacturing such devices in the area of cost and reliability.
TL;DR: In this paper, a short-channel JFET under the sidewall oxide at the drain end in series with the intrinsic MOSFET was proposed to overcome some of the reliability problems of LDD devices.
Abstract: In LDD-like MOSFET structures previously studied [1-3], narrow, self-aligned n- regions are introduced between the channel and the n+ source-drain regions. In this paper we describe a new n-MOS device structure which attempts to overcome some of the reliability problems [4] of LDD devices. Instead of a pure n- region, we introduce a short-channel JFET under the sidewall oxide at the drain end in series with the intrinsic MOSFET as shown in Fig.1. Both 2-D device simulations and experimental results are shown to demonstrate the operation of this device and its potential for VLSI applications.
TL;DR: In this paper, the reliability of laser diodes developed for undersea optical transmission systems is analyzed and the median lifetime at 10°C is conservatively estimated to be 1.6 × 106hours.
Abstract: The reliability of laser diodes developed for undersea optical transmission systems is analyzed. Up to 1000 device samples of two types (DC-PBH and VSB) are used. An aging test with constant light power operation of 5 mW is carried out at 10, 50, and 70°C for 10 000 h. The median lifetime at 10°C is conservatively estimated to be 1.6 × 106hours. Failure rate at 10°C is estimated at 250 FIT's at 25 years of service for the wear-out failure mode and at less than 50 FIT's for the random failure mode with a sufficient margin. Furthermore, it is determined that reliability can be further improved by selection of long-life lasers through an additional third step screening aging.
TL;DR: In this paper, the authors proposed a scheme to improve the reliability of an automatic train control system and reduce the size and weight of the system by providing switching means for switching a switching command from a drive cab switch to a set of receivers and composing of other receivers of mere amplifiers.
Abstract: PURPOSE:To improve the reliability of an automatic train control system and to reduce the size and weight of the system by providing switching means for switching a switching command from a drive cab switch to a set of receiver and composing of a set of other receivers of mere amplifiers. CONSTITUTION:One of two sets of systems is composed of a mere amplifier 6, and a switching command of a drive cab switch 1N and an output signal of a train element 2N are inputted to the amplifier 6. On the other hand, as the other set of the receiver switching means for operating by a switching command from a driver cab switch 1P is added to provide a receiver 7, the output signal of a train element 2P and the output of the amplifier 6 are inputted to the receiver 7, and they are composed to switch and output by the switching means. When thus constructed, the reliability of the system can be improved and the size and weight of the system can be reduced.
TL;DR: In this paper, a simple measurement method for the extraction of the increase of digital switching time due to channel hot-electron stress is described, and the authors obtain good agreement between this method and a direct measurement of the increased switching time of a CMOS inverter.
Abstract: Channel hot-electron (CHE) injection poses a reliability problem for n-channel field-effect transistors with small design rules. One often assesses the reliability of a particular fabrication process and design by subjecting individual transistors, instead of an entire circuit, to continuous or pulsed voltage stress. Simple inspection of the linear region transconductance degradation and threshold voltage shift of the individual devices, however, does not yield direct information on the circuit performance impact of channel hot-electron stress. In this letter we describe a simple measurement method for the extraction of the increase of digital switching time due to channel hot-electron stress. One performs this measurement on discrete transistors so that reliability tests still employ these individual devices. We obtain good agreement between this method and a direct measurement of the increased switching time of a CMOS inverter. We also find that the switching time changes negligibly even when the linear transistor characteristics are severely degraded if this swiching delay is measured with the same source and drain terminals as were employed for the hot-electron stress.
TL;DR: Microprocessor control now provides the capability to coordinate all the operations of the irradiation facility including the accelerator, the material handling system, various auxiliary services as well as the personnel safety system.
TL;DR: In this article, computer analyses were used to predict thermal-hydraulics, flow-induced vibration, and fretting-wear damage in shell-and-tube heat exchangers.
Abstract: This paper outlines computer analyses to predict thermal-hydraulics, flow-induced vibration, and fretting-wear damage in shell-and-tube heat exchangers. The analytical techniques are briefly described and the results arc illustrated by example. It is concluded that computer techniques can do much to improve the reliability and performance of heat exchangers.
TL;DR: The results presented in the chapter illustrate the types of aging mechanisms that should be considered for lasers used in transmitters for optical telecommunication systems and the various components of the functional test set.
Abstract: Publisher Summary This chapter evaluates the functional reliability of optical transmitters operated at 45 Mbits/sec using proton-bombarded lasers. The results presented in the chapter illustrate the types of aging mechanisms that should be considered for lasers used in transmitters for optical telecommunication systems. The chapter describes the various components of the functional test set and discusses the parameters it can record and the type of analyses applied to the raw results. The chapter discusses three second-generation experiments: (1) an investigation of the effects of different biasing conditions in a transmitter circuit, (2) a study of the temperature dependence of transmitter reliability, and (3) a life-test study of transmitters with a redesigned optical feedback scheme. It reviews data by other workers on a single-mode AlGaAs laser transmitter for space communication in the chapter. Test-set design for functional life testing can be quite simple or very elaborate, depending on the number of units and parameters to be measured. However, to ensure flexibility in the design of experiments and to minimize ambiguity during data analyses, temperature control and provisions for measuring both static and dynamic properties are basic requirements. The chapter illustrates a test set designed for evaluating the end-of-life criteria for a transmitter using a 1.3- μ m single-mode laser as the source.
TL;DR: In this paper, the authors proposed a method to maintain the reliability of the manufacturing process by confirming the result of quality judgement at the time of characteristic inspection on the basis of signs by a method wherein signs indicating the apearance of semiconductor elements are marked on the surface of the semiconductor element.
Abstract: PURPOSE:To maintain the reliability of the manufacturing process by confirmation of the result of quality judgement at the time of characteristic inspection on the basis of signs by a method wherein signs indicating the apearance of semiconductor elements are marked on the surface of the semiconductor element. CONSTITUTION:An InPn single crystal ingot 1 is sliced into a wafer 2 to be processed, the wafer 2 being severed and contained in a liquid epitaxial device, and rectangular wafers 4 being then produced. Next, each wafer 4 is subjected to growing treatment and treatments such as partial diffusion and electrode formation, resulting in the formation of a plurality of semiconductor element regions 5 in n colums and m lines. At this time, an address 6 consisting of the colum number 7 and the line number 8 of the wafer 4 is marked in each region 5. Then, a semiconductor laser element 9 is formed by separation of the wafer 4 at the boundary of the regions 5 so that the end surface of a resonator may become a mirror surface. The result of quality judgement at the time of characteristic inspection is confirmed on the basis of the address 6, and accordingly the reliability of the manufacturing process for the semiconductor device is maintained.
TL;DR: In this paper, failure rates and unscheduled maintenance frequencies for liquid metal valves were compared to those calculated for valves from light water reactors, and two possible reasons for the trend of lower failure rate and maintenance frequencies were discussed.
Abstract: In this paper, data from the Centralized Reliability Data Organization (CREDO) are used to calculate failure rates and unscheduled maintenance frequencies for liquid metal valves. These failure rates and unscheduled maintenance frequencies are then compared to those calculated for valves from light water reactors. The liquid metal valves showed, in general, failure rates and unscheduled maintenance frequencies that are significantly lower than those of water valves (sometimes as much as an order of magnitude). Two possible reasons for the trend of lower failure rates and maintenance frequencies for liquid metal valves are then discussed.
TL;DR: In this paper, a critical factor in the research and development of a new insulation system is the accurate assessment of its reliability and expected lifetime during service, and the methods to determine reliability generally attempt to accelerate the degradation processes that normally occur in electrical insulation over an extended operating time period.
Abstract: A critical factor in the research and development of a new insulation system is the accurate assessment of its reliability and expected lifetime during service. The methods to determine reliability generally attempt to accelerate the degradation processes that normally occur in electrical insulation over an extended operating time period. To date, most accelerated aging studies have featured test voltage frequencies and/or temperatures that far exceed typical operating levels. However, these efforts have not been completely successful due to the lack of knowledge concerning the relationships between the temperature, frequency of applied voltage, and the physical mechanisms responsible for dielectric deterioration and breakdown.