Function to display the thin header
Arrowhead System Engineering logo
Image of WCA_Table

Worst Case Analysis

picture (WCA_Right_Side.png)
Detailed Text For Service: WCA

Analysis as an element of verification is the key to success in any implementation, whether that be hanging a picture on a wall, creating an on-orbit satellite with a 20+ year mission life, or assuring the lives of astronauts that have become national assets. Arrowhead System Engineering was founded from heritage where the founder and chief architect participated in 40 years with an organization that never experienced a flight failure that resulted in the termination of a mission. Those lessons learned early in the 1980’s built a foundation for a leader in an organization that achieved 98% on-time delivery, outstanding financial performance, innovative technology solutions, and unparalleled quality. This lead to adoption of the Arrowhead System Engineering slogan: Choose to make Mission Failure a thing of the past©.

Analysis as defined by the NASA System Engineering Handbook[1a] is "The use of mathematical modeling and analytical techniques to predict the suitability of a design to stakeholder expectations based on calculated data or data derived from lower system structure end product verifications. Analysis is generally used when a prototype; engineering model; or fabricated, assembled, and integrated product is not available. Analysis includes the use of modeling and simulation as analytical tools. A model is a mathematical representation of reality. A simulation is the manipulation of a model". The general list of established analysis are:

  • Failure Modes, Effects and Criticality Analysis (FMECA)
  • Redundancy Analysis
  • Worst Case Analysis
  • Performance Analysis with Piece-Part Parametric Variations
  • EEE Part Stress
  • Structural Stress
  • Thermal Analyses
  • Fault Tree Analyses
  • Single Event Effects / Upset Analysis
  • Parameter Trend Analyses

Analysis is a method of verification utilizing techniques and tools such as computer and hardware simulations, analog modeling, similarity assessments, and validation of records to confirm that design requirements to be verified have been satisfied. Analysis is the evaluation of the results of multiple tests and analyses at a lower level as it would apply to a higher level of assembly. Analytical methods selected for verification will be supported by appropriate rationale and be detailed in the applicable documents. Most analyses will be accomplished in the development and qualification phases of verification.

Verification by Analysis uses techniques such as statistical analysis, quantitative analysis, computer simulations, and technical evaluation of data using logic and modeling to complete calculations or make comparisons that confirm a product design satisfies the specified functional, performance, interface or design requirements. Analysis results must prove that the product design provides the functional or performance capability, margin and design features specified by product requirements

Testing is preferable, but analysis may be used upon consideration of the following criteria:

  • Operational conditions cannot be simulated adequately in a test environment on the ground
  • A rigorous, accurate and conclusive analysis is possible
  • Verifying software complies with coding standards
  • Analysis is as effective as test and costs less
  • Verification by inspection or demonstration is not adequate or appropriate

The relationship between analysis activities and the supporting data that other verification activities will provide (test data, inspection data, and other analytical data) will be defined as part of verification planning. When verification by analysis utilizes computer simulations, the model will be accredited as defined in the Model and Simulation Support Plan (MSSP)

When analysis is the selected verification method, the “test the way we fly” (TLYF) principle is applied (NASA Systems Engineering Handbook section 5.0[1b]). This principle accounts for the way the product is to be used on the mission, the environments it must face, and the conditions in which it must function. This implies that WCA is completed at worst case operational; for example a resistor ppm/°C with a worst case operational board temature of 85°C would be analized with a delta temperature of 60°, and not analized to the +125°C storage tempature or the 100°C qualification temperatue. A subset of verification by analysis is verification by similarity. Verification by similarity may pertain to characteristics such as design, material, and function or use environment. In the case of software and digital hardware, verification by similarity will be limited to those units that are non-critical. A comprehensive discussion of Analysis, as a verification method, and verification by similarity can be found in the NASA System Engineering Handbook (SP-2007-6105)[1c].

Detailed Text For Service: WCA

Worst Case Analysis

Arrowhead System Engineering has specific experience with the execution and review of Worst Case analysis. As such, Arrowhead System Engineering submits that the most representative description of Worst Case Analysis (WCA) is according to JET PROPULSION LABORATORY RELIABILITY ANALYSES HANDBOOK (JPL D-5703)[2]. “WCA is an extension of classical circuit analysis, but uses a different approach and has a different objective. The most significant difference in the approach is the use of part parameter data and conditions at their extreme values rather than the nominal value. In the WCA, the classical circuit analysis is repeated for each combination of extreme values of part parameters and conditions. The objective is to verify that the circuit functions as required for all combinations of allowable part parameters and conditions. Circuits that are designed to provide the required output at nominal conditions and parameters may not meet the output requirements if the operating conditions or parameters vary from the nominal values over their allowable range. Out of specification performance is even more likely when several conditions or parameters vary from the nominal design condition resulting in excessive part variation. In such cases, fault isolation can not identify any part as failed or input as unacceptable. Thus, to assure reliable performance of spacecraft circuits, it is essential that variations in these parameters and conditions from their nominal values be addressed as the circuit design is being developed”.

Analysis by Type Table
Reference: Analysis by Type
System Box - Line Replacable Unit Card - Shop Replacable Unit

ARP4754A Verification

  • System Functional Test
  • Equipment and Item Verification
  • Non-Regression Verification
  • Observance of Unintended Functionality
  • Minimum Operational Performance Standards
  • Problem Reporting Process

ARP4754 ValidationObjectives

  1. Platform, system, item requirements are complete and correct
  2. Assumptions are justified and validated
  3. Derived requirements are justified and validated
  4. Requirements are traceable
  5. Validation compliance substantiation is provided

Platform Radiation

  • Prompt Dose
  • Total Dose
  • Latch-up/Burn-out
  • Single event upset
  • SEFI
  • Parametric shifts
  • Single event transients

Power Quality

  • E3/EMC
  • EMC/EMI Analysis
  • Lightning Analysis

Ground Operation

  • Transportation/Handling
  • Shipping Fixture
  • Storage (Maintainability)
  • Limited Life Analysis

SQRM&A

  • Human Factors
  • Reliability Analysis
  • Parts Stress Analysis
  • FMEA
  • Safety

Analog

  • Accuracy, Tolerance, Scale Factor, Threshold, Signal integrity
  • Drive analysis
  • Power-up
  • Bias, Reference, Clamping
  • Overvoltage protection
  • Fault protection
  • Delay
  • Feedback Stability (gain/phase margin)
  • ADC/mux timing (W/C)
  • Signal BW/slew rate capability (min)
  • Output stage stability (cap load)
  • Output stage loading (max)
  • BIT Thresholds (min/max)
  • CMRR (min)
  • PSRR (min)

Hybirds

  • Analog or Digital as applicable
  • Static & dynamic trim
  • ╬Žjc stack-up
  • Conformance to QML

Digital

  • Timing analysis (driven by ASIC analysis output)
  • Drive/Fan-out
  • Crosstalk
  • Decoupling
  • Metastable protection
  • DC Analysis of power and ground
  • ASIC/FPGA
  • Timing
  • Critical Path
  • Race conditions
  • JTAG test coverage

Power Supply

  • Regulation
  • Bypassing & decoupling
  • Over/User Bus Detection
  • Over Current Limit
  • Efficiency
  • Stability
  • Output level and ripple
  • Inrush Current
  • Power Bus Transients (Load/Line)
  • Hold up
  • Source Impedance
Detailed Text For Service: WCA

During the performance of the WCA, the function may be specified at a higher system block with documented performance requirements, or a complex circuits may be partitioned into smaller functional blocks. By using this approach, the analysis becomes more manageable, aiding both the analyst and the reviewer. In either case, when a circuit is reduced to these functional blocks, performance requirements for each block need to be established. Both input and output requirements should be established. These requirements will serve as the evaluation criteria for the WCA results for the functional blocks. The WCA should show compliance with all requirements, both on the functional block level and at the circuit level. Analysis is usually performed to one of the following standards:

  • Extreme Value			∑ (ΔITOL + ΔAGE+ ΔTEMP + ΔVCC + ΔRAD) for each comp.
  • Alternate Extreme Value		∑ [(1+ΔExtream) - (1+ΔITOL)(1+ΔAGE)(1+ΔTEMP)(1ΔVCC)(1+ΔRAD)] each comp.
  • Root-Sum-Square			√ [∑ (ΔITOL + ΔAGE + ΔTEMP + ΔVCC + ΔRAD for each component)2]
  • Monte Carlo			√ [∑ (ΔITOL)2 + (ΔAGE)2 + (ΔTEMP)2 + (ΔVCC)2 + (ΔRAD)2] for each comp.

Any circuit which does not meet its attributes at 3σ extremes cannot be considered high reliability in the functional sense. To achieve the project benefits from performing a Worst Case Analysis, the commitment must be mission wide to prevent any "weak links" in the performance chain. For critical circuitry, preliminary analyses may be required to validate a conceptual design approach at PDR. The typical period of maximum benefit is to apply WCA concurrently with the detailed design phase and have it completed in advance of, and in support of, the CDR. Similarity is a form of analysis that is acceptable for qualification (where it is shown that the article is similar in design, manufacture, manufacturing process, and quality control to another article that has been previously qualified to equivalent or more stringent criteria). However, additional analyses may be performed to support acceptance of larger assemblies and integration activities. The analyses, grouped by function type, are listed in the table below as a reference for those selected analysis (or updates) that will be detailed in the task tables per SRR, PDR, and CDR phases.

Detailed Text For Service: WCA

IEEE Parts Stress Analysis

Contemporary system analysis uses Component Stress Analysis to create a product critical design that has a higher probability of outperforming (e.g. exceeding) the product's life expectancy when used at component specified worst case operational limits. Arrowhead System Engineering agrees with the definition of Reliability as a metric which is a quantifiable requirement and includes the probability of survival, duration, environment, and function. Many space programs have used published component failure rates to create a Mean Time Between Failure (MTBF) failure rate prediction for the system using a uniform method for estimating the inherent reliability (i.e., the reliability of a mature design) of military electronic equipment and systems) such as MIL-HDBK-217[3].

Arrowhead has experience in both design and analysis of Worst Case Component Stress Analysis using MIL-STD-975[4] and MIL-STD-1547[5]. This includes justification of minor deviations and an understanding of how to use actual performance data from existing systems. Arrowhead System Engineering recommends that Component Stress Analysis be completed using worst case operational conditions. That is to say, part temperature is calculated using the bounds of the performance temperature, not the specified qualification temperature. Worst case performance analysis is still done using the qualification environments.

Detailed Text For Service: WCA

Tools

Arrowhead supports the following electrical design tools (other tools are supported, but are not part of the Arrowhead System Engineering tool package):

  • Cadence: PSPICE
  • MathCad, MatLab, Simulink
  • MicroCap: Analog Simulation
  • Excel: Various

References


  1. a b c NASA/SP-2007-6105 Rev1, NASA Systems Engineering Handbook www.nasa.gov/sites/default/files/atoms/files/ US Government - NASA. Retrieved 12 July 2018.
  2. JPL-D-5703, RELIABILITY ANALYSES HANDBOOK (JUL 1990). Everyspec.com. US Government - NASA. Retrieved 12 July 2018.
  3. MIL-HDBK-217F (NOTICE 2), MILITARY HANDBOOK: RELIABILITY PREDICTION OF ELECTRONIC EQUIPMENT (28 FEB 1995). Everyspec.com. US Government - NASA. Retrieved 12 July 2018.
  4. MIL-STD-975M (NASA), MILITARY STANDARD: NASA STANDARD ELECTRICAL, ELECTRONIC, AND ELECTROMECHNICAL (EEE) PARTS LIST (5 AUG 1994). Everyspec.com. US Government - NASA. Retrieved 29 July 2018.
  5. MIL-STD-1547B (NOTICE 2), MILITARY STANDARD: ELECTRONIC PARTS, MATERIALS, AND PROCESSES FOR SPACE VEHICLES (20 OCT 2008). Everyspec.com. US Government - NASA. Retrieved 29 July 2018.