Search NASA⌕ Search

SEARCH · Search NASA

Results for “Failure Rate”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 325 records · Page 18

A Methodology for Quantifying Certain Design Requirements During the Design Phase

A methodology for developing and balancing quantitative design requirements for safety, reliability, and maintainability has been proposed. Conceived as the basis of a more rational approach to the design of spacecraft, the methodology would also be applicable to the design of automobiles, washing machines, television receivers, or almost any other commercial product. Heretofore, it has been common practice to start by determining the requirements for reliability of elements of a spacecraft or other system to ensure a given design life for the system. Next, safety requirements are determined by assessing the total reliability of the system and adding redundant components and subsystems necessary to attain safety goals. As thus described, common practice leaves the maintainability burden to fall to chance; therefore, there is no control of recurring costs or of the responsiveness of the system. The means that have been used in assessing maintainability have been oriented toward determining the logistical sparing of components so that the components are available when needed. The process established for developing and balancing quantitative requirements for safety (S), reliability (R), and maintainability (M) derives and integrates NASA s top-level safety requirements and the controls needed to obtain program key objectives for safety and recurring cost (see figure). Being quantitative, the process conveniently uses common mathematical models. Even though the process is shown as being worked from the top down, it can also be worked from the bottom up. This process uses three math models: (1) the binomial distribution (greaterthan- or-equal-to case), (2) reliability for a series system, and (3) the Poisson distribution (less-than-or-equal-to case). The zero-fail case for the binomial distribution approximates the commonly known exponential distribution or "constant failure rate" distribution. Either model can be used. The binomial distribution was selected for modeling flexibility because it conveniently addresses both the zero-fail and failure cases. The failure case is typically used for unmanned spacecraft as with missiles.

Adams, Timothy↗

Reliability of High-Voltage Tantalum Capacitors. Parts 3 and 4)

Weibull grading test is a powerful technique that allows selection and reliability rating of solid tantalum capacitors for military and space applications. However, inaccuracies in the existing method and non-adequate acceleration factors can result in significant, up to three orders of magnitude, errors in the calculated failure rate of capacitors. This paper analyzes deficiencies of the existing technique and recommends more accurate method of calculations. A physical model presenting failures of tantalum capacitors as time-dependent-dielectric-breakdown is used to determine voltage and temperature acceleration factors and select adequate Weibull grading test conditions. This model is verified by highly accelerated life testing (HALT) at different temperature and voltage conditions for three types of solid chip tantalum capacitors. It is shown that parameters of the model and acceleration factors can be calculated using a general log-linear relationship for the characteristic life with two stress levels.

Teverovsky, Alexander↗

Analysis of Weibull Grading Test for Solid Tantalum Capacitors

Weibull grading test is a powerful technique that allows selection and reliability rating of solid tantalum capacitors for military and space applications. However, inaccuracies in the existing method and non-adequate acceleration factors can result in significant, up to three orders of magnitude, errors in the calculated failure rate of capacitors. This paper analyzes deficiencies of the existing technique and recommends more accurate method of calculations. A physical model presenting failures of tantalum capacitors as time-dependent-dielectric-breakdown is used to determine voltage and temperature acceleration factors and select adequate Weibull grading test conditions. This, model is verified by highly accelerated life testing (HALT) at different temperature and voltage conditions for three types of solid chip tantalum capacitors. It is shown that parameters of the model and acceleration factors can be calculated using a general log-linear relationship for the characteristic life with two stress levels.

Teverovsky, Alexander↗

Investigating the Interplay between Energy Efficiency and Resilience in High Performance Computing

Energy efficiency and resilience are two crucial challenges for HPC systems to reach exascale. While energy efficiency and resilience issues have been extensively studied individually, little has been done to understand the interplay between energy efficiency and resilience for HPC systems. Decreasing the supply voltage associated with a given operating frequency for processors and other CMOS-based components can significantly reduce power consumption. However, this often raises system failure rates and consequently increases application execution time. In this work, we present an energy saving undervolting approach that leverages the mainstream resilience techniques to tolerate the increased failures caused by undervolting.

failures↗

Scalable Energy Efficiency with Resilience for High Performance Computing Systems: A Quantitative Methodology

Energy efficiency and resilience are two crucial challenges for HPC systems to reach exascale. While energy efficiency and resilience issues have been extensively studied individually, little has been done to understand the interplay between energy efficiency and resilience for HPC systems. Decreasing the supply voltage associated with a given operating frequency for processors and other CMOS-based components can significantly reduce power consumption. However, this often raises system failure rates and consequently increases application execution time. In this work, we present an energy saving undervolting approach that leverages the mainstream resilience techniques to tolerate the increased failures caused by undervolting.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

A Nuclear Interaction Model for Understanding Results of Single Event Testing with High Energy Protons

An internuclear cascade and evaporation model has been adapted to estimate the LET spectrum generated during testing with 200 MeV protons. The model-generated heavy ion LET spectrum is compared to the heavy ion LET spectrum seen on orbit. This comparison is the basis for predicting single event failure rates from heavy ions using results from a single proton test. Of equal importance, this spectra comparison also establishes an estimate of the risk of encountering a failure mode on orbit that was not detected during proton testing. Verification of the general results of the model is presented based on experiments, individual part test results, and flight data. Acceptance of this model and its estimate of remaining risk opens the hardware verification philosophy to the consideration of radiation testing with high energy protons at the board and box level instead of the more standard method of individual part testing with low energy heavy ions.

Culpepper, William X.↗

Scalable Energy Efficiency with Resilience for High Performance Computing Systems: A Quantitative Methodology

Energy efficiency and resilience are two crucial challenges for HPC systems to reach exascale. While energy efficiency and resilience issues have been extensively studied individually, little has been done to understand the interplay between energy efficiency and resilience for HPC systems. Decreasing the supply voltage associated with a given operating frequency for processors and other CMOS-based components can significantly reduce power consumption. However, this often raises system failure rates and consequently increases application execution time. In this work, we present an energy saving undervolting approach that leverages the mainstream resilience techniques to tolerate the increased failures caused by undervolting.

energy, resilience↗

J-2X Abort System Development

The J-2X is an expendable liquid hydrogen (LH2)/liquid oxygen (LOX) gas generator cycle rocket engine that is currently being designed as the primary upper stage propulsion element for the new NASA Ares vehicle family. The J-2X engine will contain abort logic that functions as an integral component of the Ares vehicle abort system. This system is responsible for detecting and responding to conditions indicative of impending Loss of Mission (LOM), Loss of Vehicle (LOV), and/or catastrophic Loss of Crew (LOC) failure events. As an earth orbit ascent phase engine, the J-2X is a high power density propulsion element with non-negligible risk of fast propagation rate failures that can quickly lead to LOM, LOV, and/or LOC events. Aggressive reliability requirements for manned Ares missions and the risk of fast propagating J-2X failures dictate the need for on-engine abort condition monitoring and autonomous response capability as well as traditional abort agents such as the vehicle computer, flight crew, and ground control not located on the engine. This paper describes the baseline J-2X abort subsystem concept of operations, as well as the development process for this subsystem. A strategy that leverages heritage system experience and responds to an evolving engine design as well as J-2X specific test data to support abort system development is described. The utilization of performance and failure simulation models to support abort system sensor selection, failure detectability and discrimination studies, decision threshold definition, and abort system performance verification and validation is outlined. The basis for abort false positive and false negative performance constraints is described. Development challenges associated with information shortfalls in the design cycle, abort condition coverage and response assessment, engine-vehicle interface definition, and abort system performance verification and validation are also discussed.

Santi, Louis M.↗

Qualification Testing Versus Quantitative Reliability Testing of PV - Gaining Confidence in a Rapidly Changing Technology: Preprint

Continued growth of PV system deployment would be enhanced by quantitative, low-uncertainty predictions of the degradation and failure rates of PV modules and systems. The intended product lifetime (decades) far exceeds the product development cycle (months), limiting our ability to reduce the uncertainty of the predictions for this rapidly changing technology. Yet, business decisions (setting insurance rates, analyzing return on investment, etc.) require quantitative risk assessment. Moving toward more quantitative assessments requires consideration of many factors, including the intended application, consequence of a possible failure, variability in the manufacturing, installation, and operation, as well as uncertainty in the measured acceleration factors, which provide the basis for predictions based on accelerated tests. As the industry matures, it is useful to periodically assess the overall strategy for standards development and prioritization of research to provide a technical basis both for the standards and the analysis related to the application of those. To this end, this paper suggests a tiered approach to creating risk assessments. Recent and planned potential improvements in international standards are also summarized.

14 SOLAR ENERGY↗

Biologically Effective Dose (BED) Correlation With Biochemical Control After Low-Dose Rate Prostate Brachytherapy for Clinically Low-Risk Prostate Cancer

Purpose: To assess the correlation of postimplant dosimetric quantifiers with biochemical control of prostate cancer after low-dose rate brachytherapy. Methods and Materials: The biologically effective dose (BED), dose in Gray (Gy) to 90% of prostate (D{sub 90}), and percent volume of the prostate receiving 100% of the prescription dose (V{sub 100}) were calculated from the postimplant dose-volume histogram for 140 patients undergoing low-dose rate prostate brachytherapy from 1997 to 2003 at Durham Regional Hospital and the Durham VA Medical Center (Durham, NC). Results: The median follow-up was 50 months. There was a 7% biochemical failure rate (10 of 140), and 91% of patients (127 of 140) were alive at last clinical follow-up. The median BED was 148 Gy (range, 46-218 Gy). The median D{sub 90} was 139 Gy (range, 45-203 Gy). The median V{sub 100} was 85% (range, 44-100%). The overall 5-year biochemical relapse-free survival (bRFS) rate was 90.1%. On univariate Cox proportional hazards modeling, no pretreatment characteristic (Gleason score sum, age, baseline prostate-specific antigen, or clinical stage) was predictive of bRFS. The BED, D{sub 90}, and V{sub 100} were all highly correlated (Pearson coefficients >92%), and all were strongly correlated with bRFS. Using the Youden method, we identified the following cut points for predicting freedom from biochemical failure: D{sub 90} >= 110 Gy, V{sub 100} >= 74%, and BED >= 115 Gy. None of the covariates significantly predicted overall survival. Conclusions: We observed significant correlation between BED, D{sub 90}, and V{sub 100} with bRFS. The BED is at least as predictive of bRFS as D{sub 90} or V{sub 100}. Dosimetric quantifiers that account for heterogeneity in tumor location and dose distribution, tumor repopulation, and survival probability of tumor clonogens should be investigated.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Brachytherapy or Conformal External Radiotherapy for Prostate Cancer: A Single-Institution Matched-Pair Analysis

Purpose: In the absence of randomized study data, institutional case series have shown brachytherapy (BT) to produce excellent biochemical control (bNED) in patients with localized prostate cancer compared with alternative curative treatments. This study was designed to overcome some of the limitations of case series studies by using a matched-pair design in patients treated contemporaneously with BT and external beam radiation therapy (EBRT) at a single institution. Methods and Materials: Six hundred one eligible patients treated between 1998 and 2001 were prospectively followed up in our institutional databases and matched on a 1:1 basis for the following known prognostic variables: prostate-specific antigen (PSA) level, Gleason score, T stage, the use and duration of neoadjuvant androgen deprivation therapy, and the percentage of positive tissue core samples. Two hundred seventy-eight perfect matches of patients (139 in each group) with low- and intermediate-risk cancer were further analyzed. bNED (Phoenix definition) was the primary endpoint. Other endpoints were toxicity, PSA kinetics, and the secondary use of androgen deprivation therapy. Results: The 5-year bNED rates were 95% (BT) and 85% (EBRT) (p < 0.001). After 7 years, the BT bNED result was unchanged, but the rate in EBRT patients had fallen to 75%. The median posttreatment PSA nadirs were 0.04 ng/mL (BT) and 0.62 ng/mL (EBRT, p < 0.001), which predicted a higher ongoing treatment failure rate in association with EBRT use than with BT use. Late urinary toxicity and rectal/bowel toxicity were worse in patients treated with BT and EBRT, respectively. Conclusions: BT for both low-risk and selected intermediate-risk cancers achieves exceptional cure rates. Even with dose escalation, it will be difficult for EBRT to match the proven track record of BT seen over the past decade.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Stochastic nonlinear analysis of unidirectional fiber composites using image-based microstructural uncertainty quantification

Here, we present a data-driven nonlinear uncertainty quantification and propagation framework to study the microstructure-induced stochastic performance of unidirectional (UD) carbon fiber reinforced polymer (CFRP) composites. The proposed approach integrates (1) microscopic image characterization, (2) stochastic microstructure reconstruction, and (3) efficient multiscale finite element simulations enabled by self-consistent clustering (SCA) analysis. To model the complex microstructural variability, the proposed UQ methods take the non-Gaussian uncertainty sources into account through a distribution-free sampling approach leveraging nonparametric and asymptotic statistical tools. A hierarchical conditional sampling strategy enables the simultaneous sampling of multiple sources of uncertainties. Our approach provides insights into the impact of microstructural variabilities, which are shown to have an increasing impact on the nonlinear responses of UD CFRP parts under progressive compression loading and ultimately on the failure rate over time. We discover that before CFRP parts start to fail, a characteristic time period emerges with distinctive uncertainty distributions specific to the microstructure variability and the probability of failure. Identifying the failure time period is crucial to the reliability prediction, which is an essential component of CFRP design.

36 MATERIALS SCIENCE↗

Modeling and Simulating Multiple Failure Masking enabled by Local Recovery for Stencil-based Applications at Extreme Scales

By obtaining multi-process hard failure resilience at the application level is a key challenge that must be overcome before the promise of exascale can be fully realized. Some previous work has shown that online global recovery can dramatically reduce the overhead of failures when compared to the more traditional approach of terminating the job and restarting it from the last stored checkpoint. If online recovery is performed in a local manner further scalability is enabled, not only due to the intrinsic lower costs of recovering locally, but also due to derived effects when using some application types. In this paper we model one such effect, namely multiple failure masking, that manifests when running Stencil parallel computations on an environment when failures are recovered locally. First, the delay propagation shape of one or multiple failures recovered locally is modeled to enable several analyses of the probability of different levels of failure masking under certain Stencil application behaviors. These results indicate that failure masking is an extremely desirable effect at scale which manifestation is more evident and beneficial as the machine size or the failure rate increase.

97 MATHEMATICS AND COMPUTING↗

OPERATING EXPERIENCE AND ITS RELATIONSHIP TO EMBEDDED DIGITAL DEVICES IN SAFETY SYSTEMS

Interest in embedded digital devices (EDDs) for monitoring and providing control of components is increasing and the added functionality digital brings such as diagnostics, displays, control functions is welcomed in both the nuclear and other industries. As a result, pure analog-based components continue to disappear from the industrial I&C marketplace. Equipment with EDDs, also known by related terms such as smart devices, intelligent devices, and digital devices of limited functionality, can include sensors, breakers, priority logic modules, time delay relays, pumps, valve actuators, motor control centers, and uninterruptible power supplies. Not only is equipment consisting of analog and older digital technology being replaced with commercial grade products containing EDDs, but EDDs can also be retrofitted to work with existing components. EDDs may be installed physically within a component, such as in a relay. There are also external devices that some would not consider an EDD, but that operate in almost all technical ways identically to one internal to a component such as in a valve controller. In fact, the digital portions may be located hundreds of feet from the actual component.Due to the potential of being used in an identical manner and without alteration from their application in other industries, operating experience (OpE) for EDDs may be much more applicable than in the case of other more system level devices. Clear separation and independence of functionality provided with EDDs may also allow greater applicability of existing OpE from industry for changes that can be shown not to affect its safety function. However, OpE also demonstrates issues with different conditions between other industries and nuclear power plants, such as issues with undeclared changes or undeclared digital devices and the new failure modes they may introduce, issues with the reporting of OpE in industry, and a lack of transparency into OpE for EDDs.Understanding and addressing these issues will be key in the future use of EDDs in nuclear power plants, but failure rate data is difficult to obtain for EDDs, and lessons learned from the use of EDDs are difficult to collect and identify.Key words: embedded digital device, smart device, operating experience, failure modes

Muhlheim, Michael↗

A unified method for evaluating real-time computer controllers: A case study

A real time control system consists of a synergistic pair, that is, a controlled process and a controller computer. Performance measures for real time controller computers are defined on the basis of the nature of this synergistic pair. A case study of a typical critical controlled process is presented in the context of new performance measures that express the performance of both controlled processes and real time controllers (taken as a unit) on the basis of a single variable: controller response time. Controller response time is a function of current system state, system failure rate, electrical and/or magnetic interference, etc., and is therefore a random variable. Control overhead is expressed as a monotonically nondecreasing function of the response time and the system suffers catastrophic failure, or dynamic failure, if the response time for a control task exceeds the corresponding system hard deadline, if any. A rigorous probabilistic approach is used to estimate the performance measures. The controlled process chosen for study is an aircraft in the final stages of descent, just prior to landing. First, the performance measures for the controller are presented. Secondly, control algorithms for solving the landing problem are discussed and finally the impact of the performance measures on the problem is analyzed.

Shin, K. G.↗

A unified method for evaluating real-time computer controllers and its application

A real time control system consists of a synergistic pair, that is, a controlled process and a controller computer. Performance measures for real time controller computers are defined on the basis of the nature of this synergistic pair. A case study of a typical critical controlled process is presented in the context of new performance measures that express the performance of both controlled processes and real time controllers (taken as a unit) on the basis of a single variable: controller response time. Controller response time is a function of current system state, system failure rate, electrical and/or magnetic interference, etc., and is therefore a random variable. Control overhead is expressed as a monotonically nondecreasing function of the response time and the system suffers catastrophic failure, or dynamic failure, if the response time for a control task exceeds the corresponding system hard deadline, if any. A rigorous probabilistic approach is used to estimate the performance measures. The controlled process chosen for study is an aircraft in the final stages of descent, just prior to landing. First, the performance measures for the controller are presented. Secondly, control algorithms for solving the landing problem are discussed and finally the impact of the performance measures on the problem is analyzed.

Shin, K. G.↗

Synthesizing a New Launch Vehicle Failure Probability Based on Historical Flight Data

New launch vehicles have historically had significantly higher failure rates in early flights than what has been predicted using Probabilistic Risk Assessment - PRA. This is because PRAs typically model a mature vehicle where a significant portion of the early failure probability contributors have been eliminated due to testing and improvements after actual field operation. To capture a more accurate early flight failure probability estimate, this paper develops a method that estimates ascent failure probability starting with the first flight based on historical launch vehicle records. With new launch vehicles being developed, such as the Space Launch System - SLS, a PRA model must be extended to cover early flight failure probability contributions that are either not covered in the mature-vehicle PRA or are underestimated. These failure probability contributions include design errors, quality control deficiencies, installation errors, and environmental impacts. There are also failure dependencies due to systemic errors that still exist due to limited entire-system testing.

Cross, Robert B.↗

Retail Hydrogen Station Reliability Status and Advances

One of the most vital emerging commercial markets for hydrogen is fuel cell-powered mobility including cars, trucks, and buses. These vehicles are refueled via a network of hydrogen fueling stations, with the highest number of U.S. stations being in California. The numbers of both fuel cell electric vehicles (FCEVs) and hydrogen stations have increased in the last two years, with anecdotal information from FCEV drivers indicating that station reliability is hurting the consumer acceptability of FCEV technologies. Therefore, this study benchmarks the current state of hydrogen station reliability in practice and presents on-going research that is investigating the failures that are contribute to hydrogen station reliability issues. This is accomplished with an analysis of operation, safety, and maintenance data from hydrogen stations and fuel cell electric vehicles to benchmark the maintenance and failure of hydrogen stations and their components. This analysis, of over 5,000 station maintenance events, presents the leading maintenance categories and failure rates, and is a prerequisite to the development of data-driven reliability improvement plans. We present a reliability growth analysis and on-going research into the root causes of failure for dispensers, a particularly failure-prone subsystem.

08 HYDROGEN↗