MCF > Detailed Design and Analysis > Electrical >

Reliability Analysis

Scope and Description

This topic covers the reliability analysis of electrical systems for spacecraft. The goal of a reliability analysis is to obtain a probability of failure and a mean time between failures (MTBF) for a given system. This analysis can be used to identify components (e.g. microcircuits, discrete semiconductors, passives, connectors, soldered assemblies, electro-optical/mechanical devices) driving reliability. This knowledge can be used to inform design updates.

Resources under this topic provide reliability analysis methods, explanations of different prediction models, and guidance for adapting traditional prediction models for use with modern electronics.

Best Practices and Lessons Learned

  • When conducting a reliability analysis using the "Part Stress Analysis" method of MIL-HDBK-217, it is helpful to calculate probability of failure subtotals for each component type (i.e. capacitors, resistors, transistors, etc.) as well as for each unique identifier (i.e. part number). An analyst can then create a pie chart breakdown of the overall failure rate. This makes it easy to visualize results and identify higher risk components.
  • It's helpful to perform the "Part Stress Analysis" method of MIL-HDBK-217 so that a probability of failure subtotal can be computed for each circuit card in the subsystem. This allows the analyst to specify bounding conditions (i.e. interface temperature, etc.) more appropriately and improve the reliability prediction.
  • As with any analysis, be sure to clearly document assumptions. It is also helpful to indicate how each assumption influences the analysis. This way, if the assumption is found to be incorrect during review, it can be easily adjusted.
  • For microcircuit analyses requiring a "Years in Production" metric, the initial release date of the datasheet or source control drawing (SCD) per the revision history table is generally a conservative number to use.
  • For newer, high-speed digital devices (FPGAs, CPLDs, ASICs, memories, etc.), traditional reliability prediction methods result in unrealistically high failure rates. In order to obtain a realistic failure rate for such devices, take the MTBF rating furnished by the manufacturer and normalize or adjust it as appropriate for the operational environment.
  • Early in the design cycle, it is advantageous to conduct the MIL-HDBK-217F "Part Count" method of analysis to determine the overall part pedigree needed to achieve the desired level of hardware reliability. When completing this analysis, incorporate component costs in order to allow stakeholders to directly trade budget and mission assurance.

Resources

White Paper
NASA

This page in the NASA Public Lessons Learned System describes the importance of establishing a "mandatory ... closed-loop system for detailed, independent, and timely technical reviews of all analyses performed in support of the reliability/design process." These reviews are important for detecting design defects.

Software Tool
Vanderbilt University

SEAM is a web application for modeling assurance cases integrated with system models. SEAM supports the ... Goal Structuring Notation (GSN) standard and a subset of the SysML system modeling standard. Documentation and video tutorials are provided.

This document provides an electronics failure rate prediction standard and establishes a Community of ... Practice. It addresses the limitations of existing prediction practices with a series of subsidiary specifications that contain the "best practices" within industry for performing electronics failure rate predictions. The development of ANSI/VITA 51.0 and the subsidiary specifications is an effort to give the mean time between failure (MTBF) calculations consistency and repeatability.

This standard provides defaults and methods to adjust the models in MIL-HDBK-217F Notice 2. This is not ... a revision of MIL-HDBK-217F Notice 2 but a standardization of the inputs to the MIL-HDBK-217F Notice 2 calculations to give more consistent results. This standard also provides a means by which to normalize or adjust manufacturer furnished MTBF ratings as appropriate for the operational environment.

White Paper
US Department of Defense

MIL-HDBK-217F Notice 2 is a widely used DoD resource which provides two methods for estimating the inherent ... reliability of electronic systems. The first prediction method, "Part Stress Analysis" in Section 5 through Section 23 is the more detailed and accurate of the two and is better suited for use in the later stages of a design. The "Part Count" method in Appendix A requires fewer inputs and is more appropriate for use during early design phases or while generating a technical proposal.

This document presents a physics-of-failure approach to microelectronics reliability modeling. Section ... 1 describes in detail the various approaches to conducting a reliability prediction and is a good introduction to reliability engineering. The approaches discussed include: MIL-HDBK-217, Telcordia, PRISM, FIDES, Physics-of-Failure, RAMP, FaRBS, and MaCRO. Several of these approaches are also presented alongside the physics-of-failure approach in subsequent sections of the document.

White Paper
US Air Force

This study was conducted to update MIL-HDBK-217 failure rate prediction models and can be used as a supplemental ... resource while completing reliability predictions via MIL-HDBK-217. An example of this resource's utility is seen in Table 4.3-5. This table provides an additional base failure rate for the switching transformer type which is not included as an option in corresponding MIL-HDBK-217 model.

This document provides established instructions and requirements for the selection, screening, qualification, ... and derating of EEE parts for use in space missions. It aims to allow development teams to meet mission reliability and performance objectives while staying within their allotted budget and schedule - making this resource particularly helpful to smallsat teams. EEE-INST-002 processed parts are screened to three different levels to support missions ranging from high reliability to those where the use of high-risk components is acceptable but where a purely COTS approach is insufficient.

Get Involved